Annotation of imach/src/imach.c, revision 1.303
1.303 ! brouard 1: /* $Id: imach.c,v 1.302 2020/02/22 21:00:05 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.303 ! brouard 4: Revision 1.302 2020/02/22 21:00:05 brouard
! 5: * (Module): imach.c Update mle=-3 (for computing Life expectancy
! 6: and life table from the data without any state)
! 7:
1.302 brouard 8: Revision 1.301 2019/06/04 13:51:20 brouard
9: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
10:
1.301 brouard 11: Revision 1.300 2019/05/22 19:09:45 brouard
12: Summary: version 0.99r19 of May 2019
13:
1.300 brouard 14: Revision 1.299 2019/05/22 18:37:08 brouard
15: Summary: Cleaned 0.99r19
16:
1.299 brouard 17: Revision 1.298 2019/05/22 18:19:56 brouard
18: *** empty log message ***
19:
1.298 brouard 20: Revision 1.297 2019/05/22 17:56:10 brouard
21: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
22:
1.297 brouard 23: Revision 1.296 2019/05/20 13:03:18 brouard
24: Summary: Projection syntax simplified
25:
26:
27: We can now start projections, forward or backward, from the mean date
28: of inteviews up to or down to a number of years of projection:
29: prevforecast=1 yearsfproj=15.3 mobil_average=0
30: or
31: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
32: or
33: prevbackcast=1 yearsbproj=12.3 mobil_average=1
34: or
35: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
36:
1.296 brouard 37: Revision 1.295 2019/05/18 09:52:50 brouard
38: Summary: doxygen tex bug
39:
1.295 brouard 40: Revision 1.294 2019/05/16 14:54:33 brouard
41: Summary: There was some wrong lines added
42:
1.294 brouard 43: Revision 1.293 2019/05/09 15:17:34 brouard
44: *** empty log message ***
45:
1.293 brouard 46: Revision 1.292 2019/05/09 14:17:20 brouard
47: Summary: Some updates
48:
1.292 brouard 49: Revision 1.291 2019/05/09 13:44:18 brouard
50: Summary: Before ncovmax
51:
1.291 brouard 52: Revision 1.290 2019/05/09 13:39:37 brouard
53: Summary: 0.99r18 unlimited number of individuals
54:
55: 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.
56:
1.290 brouard 57: Revision 1.289 2018/12/13 09:16:26 brouard
58: Summary: Bug for young ages (<-30) will be in r17
59:
1.289 brouard 60: Revision 1.288 2018/05/02 20:58:27 brouard
61: Summary: Some bugs fixed
62:
1.288 brouard 63: Revision 1.287 2018/05/01 17:57:25 brouard
64: Summary: Bug fixed by providing frequencies only for non missing covariates
65:
1.287 brouard 66: Revision 1.286 2018/04/27 14:27:04 brouard
67: Summary: some minor bugs
68:
1.286 brouard 69: Revision 1.285 2018/04/21 21:02:16 brouard
70: Summary: Some bugs fixed, valgrind tested
71:
1.285 brouard 72: Revision 1.284 2018/04/20 05:22:13 brouard
73: Summary: Computing mean and stdeviation of fixed quantitative variables
74:
1.284 brouard 75: Revision 1.283 2018/04/19 14:49:16 brouard
76: Summary: Some minor bugs fixed
77:
1.283 brouard 78: Revision 1.282 2018/02/27 22:50:02 brouard
79: *** empty log message ***
80:
1.282 brouard 81: Revision 1.281 2018/02/27 19:25:23 brouard
82: Summary: Adding second argument for quitting
83:
1.281 brouard 84: Revision 1.280 2018/02/21 07:58:13 brouard
85: Summary: 0.99r15
86:
87: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
88:
1.280 brouard 89: Revision 1.279 2017/07/20 13:35:01 brouard
90: Summary: temporary working
91:
1.279 brouard 92: Revision 1.278 2017/07/19 14:09:02 brouard
93: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
94:
1.278 brouard 95: Revision 1.277 2017/07/17 08:53:49 brouard
96: Summary: BOM files can be read now
97:
1.277 brouard 98: Revision 1.276 2017/06/30 15:48:31 brouard
99: Summary: Graphs improvements
100:
1.276 brouard 101: Revision 1.275 2017/06/30 13:39:33 brouard
102: Summary: Saito's color
103:
1.275 brouard 104: Revision 1.274 2017/06/29 09:47:08 brouard
105: Summary: Version 0.99r14
106:
1.274 brouard 107: Revision 1.273 2017/06/27 11:06:02 brouard
108: Summary: More documentation on projections
109:
1.273 brouard 110: Revision 1.272 2017/06/27 10:22:40 brouard
111: Summary: Color of backprojection changed from 6 to 5(yellow)
112:
1.272 brouard 113: Revision 1.271 2017/06/27 10:17:50 brouard
114: Summary: Some bug with rint
115:
1.271 brouard 116: Revision 1.270 2017/05/24 05:45:29 brouard
117: *** empty log message ***
118:
1.270 brouard 119: Revision 1.269 2017/05/23 08:39:25 brouard
120: Summary: Code into subroutine, cleanings
121:
1.269 brouard 122: Revision 1.268 2017/05/18 20:09:32 brouard
123: Summary: backprojection and confidence intervals of backprevalence
124:
1.268 brouard 125: Revision 1.267 2017/05/13 10:25:05 brouard
126: Summary: temporary save for backprojection
127:
1.267 brouard 128: Revision 1.266 2017/05/13 07:26:12 brouard
129: Summary: Version 0.99r13 (improvements and bugs fixed)
130:
1.266 brouard 131: Revision 1.265 2017/04/26 16:22:11 brouard
132: Summary: imach 0.99r13 Some bugs fixed
133:
1.265 brouard 134: Revision 1.264 2017/04/26 06:01:29 brouard
135: Summary: Labels in graphs
136:
1.264 brouard 137: Revision 1.263 2017/04/24 15:23:15 brouard
138: Summary: to save
139:
1.263 brouard 140: Revision 1.262 2017/04/18 16:48:12 brouard
141: *** empty log message ***
142:
1.262 brouard 143: Revision 1.261 2017/04/05 10:14:09 brouard
144: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
145:
1.261 brouard 146: Revision 1.260 2017/04/04 17:46:59 brouard
147: Summary: Gnuplot indexations fixed (humm)
148:
1.260 brouard 149: Revision 1.259 2017/04/04 13:01:16 brouard
150: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
151:
1.259 brouard 152: Revision 1.258 2017/04/03 10:17:47 brouard
153: Summary: Version 0.99r12
154:
155: Some cleanings, conformed with updated documentation.
156:
1.258 brouard 157: Revision 1.257 2017/03/29 16:53:30 brouard
158: Summary: Temp
159:
1.257 brouard 160: Revision 1.256 2017/03/27 05:50:23 brouard
161: Summary: Temporary
162:
1.256 brouard 163: Revision 1.255 2017/03/08 16:02:28 brouard
164: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
165:
1.255 brouard 166: Revision 1.254 2017/03/08 07:13:00 brouard
167: Summary: Fixing data parameter line
168:
1.254 brouard 169: Revision 1.253 2016/12/15 11:59:41 brouard
170: Summary: 0.99 in progress
171:
1.253 brouard 172: Revision 1.252 2016/09/15 21:15:37 brouard
173: *** empty log message ***
174:
1.252 brouard 175: Revision 1.251 2016/09/15 15:01:13 brouard
176: Summary: not working
177:
1.251 brouard 178: Revision 1.250 2016/09/08 16:07:27 brouard
179: Summary: continue
180:
1.250 brouard 181: Revision 1.249 2016/09/07 17:14:18 brouard
182: Summary: Starting values from frequencies
183:
1.249 brouard 184: Revision 1.248 2016/09/07 14:10:18 brouard
185: *** empty log message ***
186:
1.248 brouard 187: Revision 1.247 2016/09/02 11:11:21 brouard
188: *** empty log message ***
189:
1.247 brouard 190: Revision 1.246 2016/09/02 08:49:22 brouard
191: *** empty log message ***
192:
1.246 brouard 193: Revision 1.245 2016/09/02 07:25:01 brouard
194: *** empty log message ***
195:
1.245 brouard 196: Revision 1.244 2016/09/02 07:17:34 brouard
197: *** empty log message ***
198:
1.244 brouard 199: Revision 1.243 2016/09/02 06:45:35 brouard
200: *** empty log message ***
201:
1.243 brouard 202: Revision 1.242 2016/08/30 15:01:20 brouard
203: Summary: Fixing a lots
204:
1.242 brouard 205: Revision 1.241 2016/08/29 17:17:25 brouard
206: Summary: gnuplot problem in Back projection to fix
207:
1.241 brouard 208: Revision 1.240 2016/08/29 07:53:18 brouard
209: Summary: Better
210:
1.240 brouard 211: Revision 1.239 2016/08/26 15:51:03 brouard
212: Summary: Improvement in Powell output in order to copy and paste
213:
214: Author:
215:
1.239 brouard 216: Revision 1.238 2016/08/26 14:23:35 brouard
217: Summary: Starting tests of 0.99
218:
1.238 brouard 219: Revision 1.237 2016/08/26 09:20:19 brouard
220: Summary: to valgrind
221:
1.237 brouard 222: Revision 1.236 2016/08/25 10:50:18 brouard
223: *** empty log message ***
224:
1.236 brouard 225: Revision 1.235 2016/08/25 06:59:23 brouard
226: *** empty log message ***
227:
1.235 brouard 228: Revision 1.234 2016/08/23 16:51:20 brouard
229: *** empty log message ***
230:
1.234 brouard 231: Revision 1.233 2016/08/23 07:40:50 brouard
232: Summary: not working
233:
1.233 brouard 234: Revision 1.232 2016/08/22 14:20:21 brouard
235: Summary: not working
236:
1.232 brouard 237: Revision 1.231 2016/08/22 07:17:15 brouard
238: Summary: not working
239:
1.231 brouard 240: Revision 1.230 2016/08/22 06:55:53 brouard
241: Summary: Not working
242:
1.230 brouard 243: Revision 1.229 2016/07/23 09:45:53 brouard
244: Summary: Completing for func too
245:
1.229 brouard 246: Revision 1.228 2016/07/22 17:45:30 brouard
247: Summary: Fixing some arrays, still debugging
248:
1.227 brouard 249: Revision 1.226 2016/07/12 18:42:34 brouard
250: Summary: temp
251:
1.226 brouard 252: Revision 1.225 2016/07/12 08:40:03 brouard
253: Summary: saving but not running
254:
1.225 brouard 255: Revision 1.224 2016/07/01 13:16:01 brouard
256: Summary: Fixes
257:
1.224 brouard 258: Revision 1.223 2016/02/19 09:23:35 brouard
259: Summary: temporary
260:
1.223 brouard 261: Revision 1.222 2016/02/17 08:14:50 brouard
262: Summary: Probably last 0.98 stable version 0.98r6
263:
1.222 brouard 264: Revision 1.221 2016/02/15 23:35:36 brouard
265: Summary: minor bug
266:
1.220 brouard 267: Revision 1.219 2016/02/15 00:48:12 brouard
268: *** empty log message ***
269:
1.219 brouard 270: Revision 1.218 2016/02/12 11:29:23 brouard
271: Summary: 0.99 Back projections
272:
1.218 brouard 273: Revision 1.217 2015/12/23 17:18:31 brouard
274: Summary: Experimental backcast
275:
1.217 brouard 276: Revision 1.216 2015/12/18 17:32:11 brouard
277: Summary: 0.98r4 Warning and status=-2
278:
279: Version 0.98r4 is now:
280: - displaying an error when status is -1, date of interview unknown and date of death known;
281: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
282: Older changes concerning s=-2, dating from 2005 have been supersed.
283:
1.216 brouard 284: Revision 1.215 2015/12/16 08:52:24 brouard
285: Summary: 0.98r4 working
286:
1.215 brouard 287: Revision 1.214 2015/12/16 06:57:54 brouard
288: Summary: temporary not working
289:
1.214 brouard 290: Revision 1.213 2015/12/11 18:22:17 brouard
291: Summary: 0.98r4
292:
1.213 brouard 293: Revision 1.212 2015/11/21 12:47:24 brouard
294: Summary: minor typo
295:
1.212 brouard 296: Revision 1.211 2015/11/21 12:41:11 brouard
297: Summary: 0.98r3 with some graph of projected cross-sectional
298:
299: Author: Nicolas Brouard
300:
1.211 brouard 301: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 302: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 303: Summary: Adding ftolpl parameter
304: Author: N Brouard
305:
306: We had difficulties to get smoothed confidence intervals. It was due
307: to the period prevalence which wasn't computed accurately. The inner
308: parameter ftolpl is now an outer parameter of the .imach parameter
309: file after estepm. If ftolpl is small 1.e-4 and estepm too,
310: computation are long.
311:
1.209 brouard 312: Revision 1.208 2015/11/17 14:31:57 brouard
313: Summary: temporary
314:
1.208 brouard 315: Revision 1.207 2015/10/27 17:36:57 brouard
316: *** empty log message ***
317:
1.207 brouard 318: Revision 1.206 2015/10/24 07:14:11 brouard
319: *** empty log message ***
320:
1.206 brouard 321: Revision 1.205 2015/10/23 15:50:53 brouard
322: Summary: 0.98r3 some clarification for graphs on likelihood contributions
323:
1.205 brouard 324: Revision 1.204 2015/10/01 16:20:26 brouard
325: Summary: Some new graphs of contribution to likelihood
326:
1.204 brouard 327: Revision 1.203 2015/09/30 17:45:14 brouard
328: Summary: looking at better estimation of the hessian
329:
330: Also a better criteria for convergence to the period prevalence And
331: therefore adding the number of years needed to converge. (The
332: prevalence in any alive state shold sum to one
333:
1.203 brouard 334: Revision 1.202 2015/09/22 19:45:16 brouard
335: Summary: Adding some overall graph on contribution to likelihood. Might change
336:
1.202 brouard 337: Revision 1.201 2015/09/15 17:34:58 brouard
338: Summary: 0.98r0
339:
340: - Some new graphs like suvival functions
341: - Some bugs fixed like model=1+age+V2.
342:
1.201 brouard 343: Revision 1.200 2015/09/09 16:53:55 brouard
344: Summary: Big bug thanks to Flavia
345:
346: Even model=1+age+V2. did not work anymore
347:
1.200 brouard 348: Revision 1.199 2015/09/07 14:09:23 brouard
349: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
350:
1.199 brouard 351: Revision 1.198 2015/09/03 07:14:39 brouard
352: Summary: 0.98q5 Flavia
353:
1.198 brouard 354: Revision 1.197 2015/09/01 18:24:39 brouard
355: *** empty log message ***
356:
1.197 brouard 357: Revision 1.196 2015/08/18 23:17:52 brouard
358: Summary: 0.98q5
359:
1.196 brouard 360: Revision 1.195 2015/08/18 16:28:39 brouard
361: Summary: Adding a hack for testing purpose
362:
363: After reading the title, ftol and model lines, if the comment line has
364: a q, starting with #q, the answer at the end of the run is quit. It
365: permits to run test files in batch with ctest. The former workaround was
366: $ echo q | imach foo.imach
367:
1.195 brouard 368: Revision 1.194 2015/08/18 13:32:00 brouard
369: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
370:
1.194 brouard 371: Revision 1.193 2015/08/04 07:17:42 brouard
372: Summary: 0.98q4
373:
1.193 brouard 374: Revision 1.192 2015/07/16 16:49:02 brouard
375: Summary: Fixing some outputs
376:
1.192 brouard 377: Revision 1.191 2015/07/14 10:00:33 brouard
378: Summary: Some fixes
379:
1.191 brouard 380: Revision 1.190 2015/05/05 08:51:13 brouard
381: Summary: Adding digits in output parameters (7 digits instead of 6)
382:
383: Fix 1+age+.
384:
1.190 brouard 385: Revision 1.189 2015/04/30 14:45:16 brouard
386: Summary: 0.98q2
387:
1.189 brouard 388: Revision 1.188 2015/04/30 08:27:53 brouard
389: *** empty log message ***
390:
1.188 brouard 391: Revision 1.187 2015/04/29 09:11:15 brouard
392: *** empty log message ***
393:
1.187 brouard 394: Revision 1.186 2015/04/23 12:01:52 brouard
395: Summary: V1*age is working now, version 0.98q1
396:
397: Some codes had been disabled in order to simplify and Vn*age was
398: working in the optimization phase, ie, giving correct MLE parameters,
399: but, as usual, outputs were not correct and program core dumped.
400:
1.186 brouard 401: Revision 1.185 2015/03/11 13:26:42 brouard
402: Summary: Inclusion of compile and links command line for Intel Compiler
403:
1.185 brouard 404: Revision 1.184 2015/03/11 11:52:39 brouard
405: Summary: Back from Windows 8. Intel Compiler
406:
1.184 brouard 407: Revision 1.183 2015/03/10 20:34:32 brouard
408: Summary: 0.98q0, trying with directest, mnbrak fixed
409:
410: We use directest instead of original Powell test; probably no
411: incidence on the results, but better justifications;
412: We fixed Numerical Recipes mnbrak routine which was wrong and gave
413: wrong results.
414:
1.183 brouard 415: Revision 1.182 2015/02/12 08:19:57 brouard
416: Summary: Trying to keep directest which seems simpler and more general
417: Author: Nicolas Brouard
418:
1.182 brouard 419: Revision 1.181 2015/02/11 23:22:24 brouard
420: Summary: Comments on Powell added
421:
422: Author:
423:
1.181 brouard 424: Revision 1.180 2015/02/11 17:33:45 brouard
425: Summary: Finishing move from main to function (hpijx and prevalence_limit)
426:
1.180 brouard 427: Revision 1.179 2015/01/04 09:57:06 brouard
428: Summary: back to OS/X
429:
1.179 brouard 430: Revision 1.178 2015/01/04 09:35:48 brouard
431: *** empty log message ***
432:
1.178 brouard 433: Revision 1.177 2015/01/03 18:40:56 brouard
434: Summary: Still testing ilc32 on OSX
435:
1.177 brouard 436: Revision 1.176 2015/01/03 16:45:04 brouard
437: *** empty log message ***
438:
1.176 brouard 439: Revision 1.175 2015/01/03 16:33:42 brouard
440: *** empty log message ***
441:
1.175 brouard 442: Revision 1.174 2015/01/03 16:15:49 brouard
443: Summary: Still in cross-compilation
444:
1.174 brouard 445: Revision 1.173 2015/01/03 12:06:26 brouard
446: Summary: trying to detect cross-compilation
447:
1.173 brouard 448: Revision 1.172 2014/12/27 12:07:47 brouard
449: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
450:
1.172 brouard 451: Revision 1.171 2014/12/23 13:26:59 brouard
452: Summary: Back from Visual C
453:
454: Still problem with utsname.h on Windows
455:
1.171 brouard 456: Revision 1.170 2014/12/23 11:17:12 brouard
457: Summary: Cleaning some \%% back to %%
458:
459: The escape was mandatory for a specific compiler (which one?), but too many warnings.
460:
1.170 brouard 461: Revision 1.169 2014/12/22 23:08:31 brouard
462: Summary: 0.98p
463:
464: Outputs some informations on compiler used, OS etc. Testing on different platforms.
465:
1.169 brouard 466: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 467: Summary: update
1.169 brouard 468:
1.168 brouard 469: Revision 1.167 2014/12/22 13:50:56 brouard
470: Summary: Testing uname and compiler version and if compiled 32 or 64
471:
472: Testing on Linux 64
473:
1.167 brouard 474: Revision 1.166 2014/12/22 11:40:47 brouard
475: *** empty log message ***
476:
1.166 brouard 477: Revision 1.165 2014/12/16 11:20:36 brouard
478: Summary: After compiling on Visual C
479:
480: * imach.c (Module): Merging 1.61 to 1.162
481:
1.165 brouard 482: Revision 1.164 2014/12/16 10:52:11 brouard
483: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
484:
485: * imach.c (Module): Merging 1.61 to 1.162
486:
1.164 brouard 487: Revision 1.163 2014/12/16 10:30:11 brouard
488: * imach.c (Module): Merging 1.61 to 1.162
489:
1.163 brouard 490: Revision 1.162 2014/09/25 11:43:39 brouard
491: Summary: temporary backup 0.99!
492:
1.162 brouard 493: Revision 1.1 2014/09/16 11:06:58 brouard
494: Summary: With some code (wrong) for nlopt
495:
496: Author:
497:
498: Revision 1.161 2014/09/15 20:41:41 brouard
499: Summary: Problem with macro SQR on Intel compiler
500:
1.161 brouard 501: Revision 1.160 2014/09/02 09:24:05 brouard
502: *** empty log message ***
503:
1.160 brouard 504: Revision 1.159 2014/09/01 10:34:10 brouard
505: Summary: WIN32
506: Author: Brouard
507:
1.159 brouard 508: Revision 1.158 2014/08/27 17:11:51 brouard
509: *** empty log message ***
510:
1.158 brouard 511: Revision 1.157 2014/08/27 16:26:55 brouard
512: Summary: Preparing windows Visual studio version
513: Author: Brouard
514:
515: In order to compile on Visual studio, time.h is now correct and time_t
516: and tm struct should be used. difftime should be used but sometimes I
517: just make the differences in raw time format (time(&now).
518: Trying to suppress #ifdef LINUX
519: Add xdg-open for __linux in order to open default browser.
520:
1.157 brouard 521: Revision 1.156 2014/08/25 20:10:10 brouard
522: *** empty log message ***
523:
1.156 brouard 524: Revision 1.155 2014/08/25 18:32:34 brouard
525: Summary: New compile, minor changes
526: Author: Brouard
527:
1.155 brouard 528: Revision 1.154 2014/06/20 17:32:08 brouard
529: Summary: Outputs now all graphs of convergence to period prevalence
530:
1.154 brouard 531: Revision 1.153 2014/06/20 16:45:46 brouard
532: Summary: If 3 live state, convergence to period prevalence on same graph
533: Author: Brouard
534:
1.153 brouard 535: Revision 1.152 2014/06/18 17:54:09 brouard
536: Summary: open browser, use gnuplot on same dir than imach if not found in the path
537:
1.152 brouard 538: Revision 1.151 2014/06/18 16:43:30 brouard
539: *** empty log message ***
540:
1.151 brouard 541: Revision 1.150 2014/06/18 16:42:35 brouard
542: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
543: Author: brouard
544:
1.150 brouard 545: Revision 1.149 2014/06/18 15:51:14 brouard
546: Summary: Some fixes in parameter files errors
547: Author: Nicolas Brouard
548:
1.149 brouard 549: Revision 1.148 2014/06/17 17:38:48 brouard
550: Summary: Nothing new
551: Author: Brouard
552:
553: Just a new packaging for OS/X version 0.98nS
554:
1.148 brouard 555: Revision 1.147 2014/06/16 10:33:11 brouard
556: *** empty log message ***
557:
1.147 brouard 558: Revision 1.146 2014/06/16 10:20:28 brouard
559: Summary: Merge
560: Author: Brouard
561:
562: Merge, before building revised version.
563:
1.146 brouard 564: Revision 1.145 2014/06/10 21:23:15 brouard
565: Summary: Debugging with valgrind
566: Author: Nicolas Brouard
567:
568: Lot of changes in order to output the results with some covariates
569: After the Edimburgh REVES conference 2014, it seems mandatory to
570: improve the code.
571: No more memory valgrind error but a lot has to be done in order to
572: continue the work of splitting the code into subroutines.
573: Also, decodemodel has been improved. Tricode is still not
574: optimal. nbcode should be improved. Documentation has been added in
575: the source code.
576:
1.144 brouard 577: Revision 1.143 2014/01/26 09:45:38 brouard
578: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
579:
580: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
581: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
582:
1.143 brouard 583: Revision 1.142 2014/01/26 03:57:36 brouard
584: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
585:
586: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
587:
1.142 brouard 588: Revision 1.141 2014/01/26 02:42:01 brouard
589: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
590:
1.141 brouard 591: Revision 1.140 2011/09/02 10:37:54 brouard
592: Summary: times.h is ok with mingw32 now.
593:
1.140 brouard 594: Revision 1.139 2010/06/14 07:50:17 brouard
595: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
596: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
597:
1.139 brouard 598: Revision 1.138 2010/04/30 18:19:40 brouard
599: *** empty log message ***
600:
1.138 brouard 601: Revision 1.137 2010/04/29 18:11:38 brouard
602: (Module): Checking covariates for more complex models
603: than V1+V2. A lot of change to be done. Unstable.
604:
1.137 brouard 605: Revision 1.136 2010/04/26 20:30:53 brouard
606: (Module): merging some libgsl code. Fixing computation
607: of likelione (using inter/intrapolation if mle = 0) in order to
608: get same likelihood as if mle=1.
609: Some cleaning of code and comments added.
610:
1.136 brouard 611: Revision 1.135 2009/10/29 15:33:14 brouard
612: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
613:
1.135 brouard 614: Revision 1.134 2009/10/29 13:18:53 brouard
615: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
616:
1.134 brouard 617: Revision 1.133 2009/07/06 10:21:25 brouard
618: just nforces
619:
1.133 brouard 620: Revision 1.132 2009/07/06 08:22:05 brouard
621: Many tings
622:
1.132 brouard 623: Revision 1.131 2009/06/20 16:22:47 brouard
624: Some dimensions resccaled
625:
1.131 brouard 626: Revision 1.130 2009/05/26 06:44:34 brouard
627: (Module): Max Covariate is now set to 20 instead of 8. A
628: lot of cleaning with variables initialized to 0. Trying to make
629: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
630:
1.130 brouard 631: Revision 1.129 2007/08/31 13:49:27 lievre
632: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
633:
1.129 lievre 634: Revision 1.128 2006/06/30 13:02:05 brouard
635: (Module): Clarifications on computing e.j
636:
1.128 brouard 637: Revision 1.127 2006/04/28 18:11:50 brouard
638: (Module): Yes the sum of survivors was wrong since
639: imach-114 because nhstepm was no more computed in the age
640: loop. Now we define nhstepma in the age loop.
641: (Module): In order to speed up (in case of numerous covariates) we
642: compute health expectancies (without variances) in a first step
643: and then all the health expectancies with variances or standard
644: deviation (needs data from the Hessian matrices) which slows the
645: computation.
646: In the future we should be able to stop the program is only health
647: expectancies and graph are needed without standard deviations.
648:
1.127 brouard 649: Revision 1.126 2006/04/28 17:23:28 brouard
650: (Module): Yes the sum of survivors was wrong since
651: imach-114 because nhstepm was no more computed in the age
652: loop. Now we define nhstepma in the age loop.
653: Version 0.98h
654:
1.126 brouard 655: Revision 1.125 2006/04/04 15:20:31 lievre
656: Errors in calculation of health expectancies. Age was not initialized.
657: Forecasting file added.
658:
659: Revision 1.124 2006/03/22 17:13:53 lievre
660: Parameters are printed with %lf instead of %f (more numbers after the comma).
661: The log-likelihood is printed in the log file
662:
663: Revision 1.123 2006/03/20 10:52:43 brouard
664: * imach.c (Module): <title> changed, corresponds to .htm file
665: name. <head> headers where missing.
666:
667: * imach.c (Module): Weights can have a decimal point as for
668: English (a comma might work with a correct LC_NUMERIC environment,
669: otherwise the weight is truncated).
670: Modification of warning when the covariates values are not 0 or
671: 1.
672: Version 0.98g
673:
674: Revision 1.122 2006/03/20 09:45:41 brouard
675: (Module): Weights can have a decimal point as for
676: English (a comma might work with a correct LC_NUMERIC environment,
677: otherwise the weight is truncated).
678: Modification of warning when the covariates values are not 0 or
679: 1.
680: Version 0.98g
681:
682: Revision 1.121 2006/03/16 17:45:01 lievre
683: * imach.c (Module): Comments concerning covariates added
684:
685: * imach.c (Module): refinements in the computation of lli if
686: status=-2 in order to have more reliable computation if stepm is
687: not 1 month. Version 0.98f
688:
689: Revision 1.120 2006/03/16 15:10:38 lievre
690: (Module): refinements in the computation of lli if
691: status=-2 in order to have more reliable computation if stepm is
692: not 1 month. Version 0.98f
693:
694: Revision 1.119 2006/03/15 17:42:26 brouard
695: (Module): Bug if status = -2, the loglikelihood was
696: computed as likelihood omitting the logarithm. Version O.98e
697:
698: Revision 1.118 2006/03/14 18:20:07 brouard
699: (Module): varevsij Comments added explaining the second
700: table of variances if popbased=1 .
701: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
702: (Module): Function pstamp added
703: (Module): Version 0.98d
704:
705: Revision 1.117 2006/03/14 17:16:22 brouard
706: (Module): varevsij Comments added explaining the second
707: table of variances if popbased=1 .
708: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
709: (Module): Function pstamp added
710: (Module): Version 0.98d
711:
712: Revision 1.116 2006/03/06 10:29:27 brouard
713: (Module): Variance-covariance wrong links and
714: varian-covariance of ej. is needed (Saito).
715:
716: Revision 1.115 2006/02/27 12:17:45 brouard
717: (Module): One freematrix added in mlikeli! 0.98c
718:
719: Revision 1.114 2006/02/26 12:57:58 brouard
720: (Module): Some improvements in processing parameter
721: filename with strsep.
722:
723: Revision 1.113 2006/02/24 14:20:24 brouard
724: (Module): Memory leaks checks with valgrind and:
725: datafile was not closed, some imatrix were not freed and on matrix
726: allocation too.
727:
728: Revision 1.112 2006/01/30 09:55:26 brouard
729: (Module): Back to gnuplot.exe instead of wgnuplot.exe
730:
731: Revision 1.111 2006/01/25 20:38:18 brouard
732: (Module): Lots of cleaning and bugs added (Gompertz)
733: (Module): Comments can be added in data file. Missing date values
734: can be a simple dot '.'.
735:
736: Revision 1.110 2006/01/25 00:51:50 brouard
737: (Module): Lots of cleaning and bugs added (Gompertz)
738:
739: Revision 1.109 2006/01/24 19:37:15 brouard
740: (Module): Comments (lines starting with a #) are allowed in data.
741:
742: Revision 1.108 2006/01/19 18:05:42 lievre
743: Gnuplot problem appeared...
744: To be fixed
745:
746: Revision 1.107 2006/01/19 16:20:37 brouard
747: Test existence of gnuplot in imach path
748:
749: Revision 1.106 2006/01/19 13:24:36 brouard
750: Some cleaning and links added in html output
751:
752: Revision 1.105 2006/01/05 20:23:19 lievre
753: *** empty log message ***
754:
755: Revision 1.104 2005/09/30 16:11:43 lievre
756: (Module): sump fixed, loop imx fixed, and simplifications.
757: (Module): If the status is missing at the last wave but we know
758: that the person is alive, then we can code his/her status as -2
759: (instead of missing=-1 in earlier versions) and his/her
760: contributions to the likelihood is 1 - Prob of dying from last
761: health status (= 1-p13= p11+p12 in the easiest case of somebody in
762: the healthy state at last known wave). Version is 0.98
763:
764: Revision 1.103 2005/09/30 15:54:49 lievre
765: (Module): sump fixed, loop imx fixed, and simplifications.
766:
767: Revision 1.102 2004/09/15 17:31:30 brouard
768: Add the possibility to read data file including tab characters.
769:
770: Revision 1.101 2004/09/15 10:38:38 brouard
771: Fix on curr_time
772:
773: Revision 1.100 2004/07/12 18:29:06 brouard
774: Add version for Mac OS X. Just define UNIX in Makefile
775:
776: Revision 1.99 2004/06/05 08:57:40 brouard
777: *** empty log message ***
778:
779: Revision 1.98 2004/05/16 15:05:56 brouard
780: New version 0.97 . First attempt to estimate force of mortality
781: directly from the data i.e. without the need of knowing the health
782: state at each age, but using a Gompertz model: log u =a + b*age .
783: This is the basic analysis of mortality and should be done before any
784: other analysis, in order to test if the mortality estimated from the
785: cross-longitudinal survey is different from the mortality estimated
786: from other sources like vital statistic data.
787:
788: The same imach parameter file can be used but the option for mle should be -3.
789:
1.133 brouard 790: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 791: former routines in order to include the new code within the former code.
792:
793: The output is very simple: only an estimate of the intercept and of
794: the slope with 95% confident intervals.
795:
796: Current limitations:
797: A) Even if you enter covariates, i.e. with the
798: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
799: B) There is no computation of Life Expectancy nor Life Table.
800:
801: Revision 1.97 2004/02/20 13:25:42 lievre
802: Version 0.96d. Population forecasting command line is (temporarily)
803: suppressed.
804:
805: Revision 1.96 2003/07/15 15:38:55 brouard
806: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
807: rewritten within the same printf. Workaround: many printfs.
808:
809: Revision 1.95 2003/07/08 07:54:34 brouard
810: * imach.c (Repository):
811: (Repository): Using imachwizard code to output a more meaningful covariance
812: matrix (cov(a12,c31) instead of numbers.
813:
814: Revision 1.94 2003/06/27 13:00:02 brouard
815: Just cleaning
816:
817: Revision 1.93 2003/06/25 16:33:55 brouard
818: (Module): On windows (cygwin) function asctime_r doesn't
819: exist so I changed back to asctime which exists.
820: (Module): Version 0.96b
821:
822: Revision 1.92 2003/06/25 16:30:45 brouard
823: (Module): On windows (cygwin) function asctime_r doesn't
824: exist so I changed back to asctime which exists.
825:
826: Revision 1.91 2003/06/25 15:30:29 brouard
827: * imach.c (Repository): Duplicated warning errors corrected.
828: (Repository): Elapsed time after each iteration is now output. It
829: helps to forecast when convergence will be reached. Elapsed time
830: is stamped in powell. We created a new html file for the graphs
831: concerning matrix of covariance. It has extension -cov.htm.
832:
833: Revision 1.90 2003/06/24 12:34:15 brouard
834: (Module): Some bugs corrected for windows. Also, when
835: mle=-1 a template is output in file "or"mypar.txt with the design
836: of the covariance matrix to be input.
837:
838: Revision 1.89 2003/06/24 12:30:52 brouard
839: (Module): Some bugs corrected for windows. Also, when
840: mle=-1 a template is output in file "or"mypar.txt with the design
841: of the covariance matrix to be input.
842:
843: Revision 1.88 2003/06/23 17:54:56 brouard
844: * 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.
845:
846: Revision 1.87 2003/06/18 12:26:01 brouard
847: Version 0.96
848:
849: Revision 1.86 2003/06/17 20:04:08 brouard
850: (Module): Change position of html and gnuplot routines and added
851: routine fileappend.
852:
853: Revision 1.85 2003/06/17 13:12:43 brouard
854: * imach.c (Repository): Check when date of death was earlier that
855: current date of interview. It may happen when the death was just
856: prior to the death. In this case, dh was negative and likelihood
857: was wrong (infinity). We still send an "Error" but patch by
858: assuming that the date of death was just one stepm after the
859: interview.
860: (Repository): Because some people have very long ID (first column)
861: we changed int to long in num[] and we added a new lvector for
862: memory allocation. But we also truncated to 8 characters (left
863: truncation)
864: (Repository): No more line truncation errors.
865:
866: Revision 1.84 2003/06/13 21:44:43 brouard
867: * imach.c (Repository): Replace "freqsummary" at a correct
868: place. It differs from routine "prevalence" which may be called
869: many times. Probs is memory consuming and must be used with
870: parcimony.
871: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
872:
873: Revision 1.83 2003/06/10 13:39:11 lievre
874: *** empty log message ***
875:
876: Revision 1.82 2003/06/05 15:57:20 brouard
877: Add log in imach.c and fullversion number is now printed.
878:
879: */
880: /*
881: Interpolated Markov Chain
882:
883: Short summary of the programme:
884:
1.227 brouard 885: This program computes Healthy Life Expectancies or State-specific
886: (if states aren't health statuses) Expectancies from
887: cross-longitudinal data. Cross-longitudinal data consist in:
888:
889: -1- a first survey ("cross") where individuals from different ages
890: are interviewed on their health status or degree of disability (in
891: the case of a health survey which is our main interest)
892:
893: -2- at least a second wave of interviews ("longitudinal") which
894: measure each change (if any) in individual health status. Health
895: expectancies are computed from the time spent in each health state
896: according to a model. More health states you consider, more time is
897: necessary to reach the Maximum Likelihood of the parameters involved
898: in the model. The simplest model is the multinomial logistic model
899: where pij is the probability to be observed in state j at the second
900: wave conditional to be observed in state i at the first
901: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
902: etc , where 'age' is age and 'sex' is a covariate. If you want to
903: have a more complex model than "constant and age", you should modify
904: the program where the markup *Covariates have to be included here
905: again* invites you to do it. More covariates you add, slower the
1.126 brouard 906: convergence.
907:
908: The advantage of this computer programme, compared to a simple
909: multinomial logistic model, is clear when the delay between waves is not
910: identical for each individual. Also, if a individual missed an
911: intermediate interview, the information is lost, but taken into
912: account using an interpolation or extrapolation.
913:
914: hPijx is the probability to be observed in state i at age x+h
915: conditional to the observed state i at age x. The delay 'h' can be
916: split into an exact number (nh*stepm) of unobserved intermediate
917: states. This elementary transition (by month, quarter,
918: semester or year) is modelled as a multinomial logistic. The hPx
919: matrix is simply the matrix product of nh*stepm elementary matrices
920: and the contribution of each individual to the likelihood is simply
921: hPijx.
922:
923: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 924: of the life expectancies. It also computes the period (stable) prevalence.
925:
926: Back prevalence and projections:
1.227 brouard 927:
928: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
929: double agemaxpar, double ftolpl, int *ncvyearp, double
930: dateprev1,double dateprev2, int firstpass, int lastpass, int
931: mobilavproj)
932:
933: Computes the back prevalence limit for any combination of
934: covariate values k at any age between ageminpar and agemaxpar and
935: returns it in **bprlim. In the loops,
936:
937: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
938: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
939:
940: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 941: Computes for any combination of covariates k and any age between bage and fage
942: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
943: oldm=oldms;savm=savms;
1.227 brouard 944:
1.267 brouard 945: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 946: Computes the transition matrix starting at age 'age' over
947: 'nhstepm*hstepm*stepm' months (i.e. until
948: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 949: nhstepm*hstepm matrices.
950:
951: Returns p3mat[i][j][h] after calling
952: p3mat[i][j][h]=matprod2(newm,
953: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
954: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
955: oldm);
1.226 brouard 956:
957: Important routines
958:
959: - func (or funcone), computes logit (pij) distinguishing
960: o fixed variables (single or product dummies or quantitative);
961: o varying variables by:
962: (1) wave (single, product dummies, quantitative),
963: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
964: % fixed dummy (treated) or quantitative (not done because time-consuming);
965: % varying dummy (not done) or quantitative (not done);
966: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
967: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
968: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
969: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
970: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 971:
1.226 brouard 972:
973:
1.133 brouard 974: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
975: Institut national d'études démographiques, Paris.
1.126 brouard 976: This software have been partly granted by Euro-REVES, a concerted action
977: from the European Union.
978: It is copyrighted identically to a GNU software product, ie programme and
979: software can be distributed freely for non commercial use. Latest version
980: can be accessed at http://euroreves.ined.fr/imach .
981:
982: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
983: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
984:
985: **********************************************************************/
986: /*
987: main
988: read parameterfile
989: read datafile
990: concatwav
991: freqsummary
992: if (mle >= 1)
993: mlikeli
994: print results files
995: if mle==1
996: computes hessian
997: read end of parameter file: agemin, agemax, bage, fage, estepm
998: begin-prev-date,...
999: open gnuplot file
1000: open html file
1.145 brouard 1001: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1002: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1003: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1004: freexexit2 possible for memory heap.
1005:
1006: h Pij x | pij_nom ficrestpij
1007: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1008: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1009: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1010:
1011: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1012: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1013: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1014: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1015: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1016:
1.126 brouard 1017: forecasting if prevfcast==1 prevforecast call prevalence()
1018: health expectancies
1019: Variance-covariance of DFLE
1020: prevalence()
1021: movingaverage()
1022: varevsij()
1023: if popbased==1 varevsij(,popbased)
1024: total life expectancies
1025: Variance of period (stable) prevalence
1026: end
1027: */
1028:
1.187 brouard 1029: /* #define DEBUG */
1030: /* #define DEBUGBRENT */
1.203 brouard 1031: /* #define DEBUGLINMIN */
1032: /* #define DEBUGHESS */
1033: #define DEBUGHESSIJ
1.224 brouard 1034: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1035: #define POWELL /* Instead of NLOPT */
1.224 brouard 1036: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1037: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1038: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1039:
1040: #include <math.h>
1041: #include <stdio.h>
1042: #include <stdlib.h>
1043: #include <string.h>
1.226 brouard 1044: #include <ctype.h>
1.159 brouard 1045:
1046: #ifdef _WIN32
1047: #include <io.h>
1.172 brouard 1048: #include <windows.h>
1049: #include <tchar.h>
1.159 brouard 1050: #else
1.126 brouard 1051: #include <unistd.h>
1.159 brouard 1052: #endif
1.126 brouard 1053:
1054: #include <limits.h>
1055: #include <sys/types.h>
1.171 brouard 1056:
1057: #if defined(__GNUC__)
1058: #include <sys/utsname.h> /* Doesn't work on Windows */
1059: #endif
1060:
1.126 brouard 1061: #include <sys/stat.h>
1062: #include <errno.h>
1.159 brouard 1063: /* extern int errno; */
1.126 brouard 1064:
1.157 brouard 1065: /* #ifdef LINUX */
1066: /* #include <time.h> */
1067: /* #include "timeval.h" */
1068: /* #else */
1069: /* #include <sys/time.h> */
1070: /* #endif */
1071:
1.126 brouard 1072: #include <time.h>
1073:
1.136 brouard 1074: #ifdef GSL
1075: #include <gsl/gsl_errno.h>
1076: #include <gsl/gsl_multimin.h>
1077: #endif
1078:
1.167 brouard 1079:
1.162 brouard 1080: #ifdef NLOPT
1081: #include <nlopt.h>
1082: typedef struct {
1083: double (* function)(double [] );
1084: } myfunc_data ;
1085: #endif
1086:
1.126 brouard 1087: /* #include <libintl.h> */
1088: /* #define _(String) gettext (String) */
1089:
1.251 brouard 1090: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1091:
1092: #define GNUPLOTPROGRAM "gnuplot"
1093: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1094: #define FILENAMELENGTH 132
1095:
1096: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1097: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1098:
1.144 brouard 1099: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1100: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1101:
1102: #define NINTERVMAX 8
1.144 brouard 1103: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1104: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1105: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1106: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1107: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1108: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1109: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1110: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1111: /* #define AGESUP 130 */
1.288 brouard 1112: /* #define AGESUP 150 */
1113: #define AGESUP 200
1.268 brouard 1114: #define AGEINF 0
1.218 brouard 1115: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1116: #define AGEBASE 40
1.194 brouard 1117: #define AGEOVERFLOW 1.e20
1.164 brouard 1118: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1119: #ifdef _WIN32
1120: #define DIRSEPARATOR '\\'
1121: #define CHARSEPARATOR "\\"
1122: #define ODIRSEPARATOR '/'
1123: #else
1.126 brouard 1124: #define DIRSEPARATOR '/'
1125: #define CHARSEPARATOR "/"
1126: #define ODIRSEPARATOR '\\'
1127: #endif
1128:
1.303 ! brouard 1129: /* $Id: imach.c,v 1.302 2020/02/22 21:00:05 brouard Exp $ */
1.126 brouard 1130: /* $State: Exp $ */
1.196 brouard 1131: #include "version.h"
1132: char version[]=__IMACH_VERSION__;
1.300 brouard 1133: char copyright[]="May 2019,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020";
1.303 ! brouard 1134: char fullversion[]="$Revision: 1.302 $ $Date: 2020/02/22 21:00:05 $";
1.126 brouard 1135: char strstart[80];
1136: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1137: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1138: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1139: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1140: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1141: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1142: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1143: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1144: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1145: int cptcovprodnoage=0; /**< Number of covariate products without age */
1146: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1147: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1148: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1149: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1150: int nsd=0; /**< Total number of single dummy variables (output) */
1151: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1152: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1153: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1154: int ntveff=0; /**< ntveff number of effective time varying variables */
1155: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1156: int cptcov=0; /* Working variable */
1.290 brouard 1157: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1158: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1159: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1160: int nlstate=2; /* Number of live states */
1161: int ndeath=1; /* Number of dead states */
1.130 brouard 1162: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1163: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1164: int popbased=0;
1165:
1166: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1167: int maxwav=0; /* Maxim number of waves */
1168: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1169: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1170: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1171: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1172: int mle=1, weightopt=0;
1.126 brouard 1173: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1174: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1175: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1176: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1177: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1178: int selected(int kvar); /* Is covariate kvar selected for printing results */
1179:
1.130 brouard 1180: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1181: double **matprod2(); /* test */
1.126 brouard 1182: double **oldm, **newm, **savm; /* Working pointers to matrices */
1183: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1184: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1185:
1.136 brouard 1186: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1187: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1188: FILE *ficlog, *ficrespow;
1.130 brouard 1189: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1190: double fretone; /* Only one call to likelihood */
1.130 brouard 1191: long ipmx=0; /* Number of contributions */
1.126 brouard 1192: double sw; /* Sum of weights */
1193: char filerespow[FILENAMELENGTH];
1194: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1195: FILE *ficresilk;
1196: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1197: FILE *ficresprobmorprev;
1198: FILE *fichtm, *fichtmcov; /* Html File */
1199: FILE *ficreseij;
1200: char filerese[FILENAMELENGTH];
1201: FILE *ficresstdeij;
1202: char fileresstde[FILENAMELENGTH];
1203: FILE *ficrescveij;
1204: char filerescve[FILENAMELENGTH];
1205: FILE *ficresvij;
1206: char fileresv[FILENAMELENGTH];
1.269 brouard 1207:
1.126 brouard 1208: char title[MAXLINE];
1.234 brouard 1209: char model[MAXLINE]; /**< The model line */
1.217 brouard 1210: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1211: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1212: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1213: char command[FILENAMELENGTH];
1214: int outcmd=0;
1215:
1.217 brouard 1216: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1217: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1218: char filelog[FILENAMELENGTH]; /* Log file */
1219: char filerest[FILENAMELENGTH];
1220: char fileregp[FILENAMELENGTH];
1221: char popfile[FILENAMELENGTH];
1222:
1223: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1224:
1.157 brouard 1225: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1226: /* struct timezone tzp; */
1227: /* extern int gettimeofday(); */
1228: struct tm tml, *gmtime(), *localtime();
1229:
1230: extern time_t time();
1231:
1232: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1233: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1234: struct tm tm;
1235:
1.126 brouard 1236: char strcurr[80], strfor[80];
1237:
1238: char *endptr;
1239: long lval;
1240: double dval;
1241:
1242: #define NR_END 1
1243: #define FREE_ARG char*
1244: #define FTOL 1.0e-10
1245:
1246: #define NRANSI
1.240 brouard 1247: #define ITMAX 200
1248: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1249:
1250: #define TOL 2.0e-4
1251:
1252: #define CGOLD 0.3819660
1253: #define ZEPS 1.0e-10
1254: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1255:
1256: #define GOLD 1.618034
1257: #define GLIMIT 100.0
1258: #define TINY 1.0e-20
1259:
1260: static double maxarg1,maxarg2;
1261: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1262: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1263:
1264: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1265: #define rint(a) floor(a+0.5)
1.166 brouard 1266: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1267: #define mytinydouble 1.0e-16
1.166 brouard 1268: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1269: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1270: /* static double dsqrarg; */
1271: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1272: static double sqrarg;
1273: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1274: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1275: int agegomp= AGEGOMP;
1276:
1277: int imx;
1278: int stepm=1;
1279: /* Stepm, step in month: minimum step interpolation*/
1280:
1281: int estepm;
1282: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1283:
1284: int m,nb;
1285: long *num;
1.197 brouard 1286: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1287: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1288: covariate for which somebody answered excluding
1289: undefined. Usually 2: 0 and 1. */
1290: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1291: covariate for which somebody answered including
1292: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1293: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1294: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1295: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1296: double *ageexmed,*agecens;
1297: double dateintmean=0;
1.296 brouard 1298: double anprojd, mprojd, jprojd; /* For eventual projections */
1299: double anprojf, mprojf, jprojf;
1.126 brouard 1300:
1.296 brouard 1301: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1302: double anbackf, mbackf, jbackf;
1303: double jintmean,mintmean,aintmean;
1.126 brouard 1304: double *weight;
1305: int **s; /* Status */
1.141 brouard 1306: double *agedc;
1.145 brouard 1307: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1308: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1309: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1310: double **coqvar; /* Fixed quantitative covariate nqv */
1311: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1312: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1313: double idx;
1314: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1315: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1316: /*k 1 2 3 4 5 6 7 8 9 */
1317: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1318: /* Tndvar[k] 1 2 3 4 5 */
1319: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1320: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1321: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1322: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1323: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1324: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1325: /* Tprod[i]=k 4 7 */
1326: /* Tage[i]=k 5 8 */
1327: /* */
1328: /* Type */
1329: /* V 1 2 3 4 5 */
1330: /* F F V V V */
1331: /* D Q D D Q */
1332: /* */
1333: int *TvarsD;
1334: int *TvarsDind;
1335: int *TvarsQ;
1336: int *TvarsQind;
1337:
1.235 brouard 1338: #define MAXRESULTLINES 10
1339: int nresult=0;
1.258 brouard 1340: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1341: int TKresult[MAXRESULTLINES];
1.237 brouard 1342: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1343: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1344: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1345: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1346: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1347: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1348:
1.234 brouard 1349: /* 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 1350: 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 */
1351: 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 */
1352: 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 */
1353: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1354: 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 */
1355: 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 1356: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1357: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1358: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1359: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1360: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1361: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1362: 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 */
1363: 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 */
1364:
1.230 brouard 1365: int *Tvarsel; /**< Selected covariates for output */
1366: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1367: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1368: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1369: 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 1370: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1371: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1372: int *Tage;
1.227 brouard 1373: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1374: 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 1375: 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*/
1376: 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 1377: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1378: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1379: int **Tvard;
1380: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1381: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1382: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1383: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1384: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1385: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1386: double *lsurv, *lpop, *tpop;
1387:
1.231 brouard 1388: #define FD 1; /* Fixed dummy covariate */
1389: #define FQ 2; /* Fixed quantitative covariate */
1390: #define FP 3; /* Fixed product covariate */
1391: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1392: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1393: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1394: #define VD 10; /* Varying dummy covariate */
1395: #define VQ 11; /* Varying quantitative covariate */
1396: #define VP 12; /* Varying product covariate */
1397: #define VPDD 13; /* Varying product dummy*dummy covariate */
1398: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1399: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1400: #define APFD 16; /* Age product * fixed dummy covariate */
1401: #define APFQ 17; /* Age product * fixed quantitative covariate */
1402: #define APVD 18; /* Age product * varying dummy covariate */
1403: #define APVQ 19; /* Age product * varying quantitative covariate */
1404:
1405: #define FTYPE 1; /* Fixed covariate */
1406: #define VTYPE 2; /* Varying covariate (loop in wave) */
1407: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1408:
1409: struct kmodel{
1410: int maintype; /* main type */
1411: int subtype; /* subtype */
1412: };
1413: struct kmodel modell[NCOVMAX];
1414:
1.143 brouard 1415: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1416: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1417:
1418: /**************** split *************************/
1419: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1420: {
1421: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1422: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1423: */
1424: char *ss; /* pointer */
1.186 brouard 1425: int l1=0, l2=0; /* length counters */
1.126 brouard 1426:
1427: l1 = strlen(path ); /* length of path */
1428: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1429: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1430: if ( ss == NULL ) { /* no directory, so determine current directory */
1431: strcpy( name, path ); /* we got the fullname name because no directory */
1432: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1433: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1434: /* get current working directory */
1435: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1436: #ifdef WIN32
1437: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1438: #else
1439: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1440: #endif
1.126 brouard 1441: return( GLOCK_ERROR_GETCWD );
1442: }
1443: /* got dirc from getcwd*/
1444: printf(" DIRC = %s \n",dirc);
1.205 brouard 1445: } else { /* strip directory from path */
1.126 brouard 1446: ss++; /* after this, the filename */
1447: l2 = strlen( ss ); /* length of filename */
1448: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1449: strcpy( name, ss ); /* save file name */
1450: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1451: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1452: printf(" DIRC2 = %s \n",dirc);
1453: }
1454: /* We add a separator at the end of dirc if not exists */
1455: l1 = strlen( dirc ); /* length of directory */
1456: if( dirc[l1-1] != DIRSEPARATOR ){
1457: dirc[l1] = DIRSEPARATOR;
1458: dirc[l1+1] = 0;
1459: printf(" DIRC3 = %s \n",dirc);
1460: }
1461: ss = strrchr( name, '.' ); /* find last / */
1462: if (ss >0){
1463: ss++;
1464: strcpy(ext,ss); /* save extension */
1465: l1= strlen( name);
1466: l2= strlen(ss)+1;
1467: strncpy( finame, name, l1-l2);
1468: finame[l1-l2]= 0;
1469: }
1470:
1471: return( 0 ); /* we're done */
1472: }
1473:
1474:
1475: /******************************************/
1476:
1477: void replace_back_to_slash(char *s, char*t)
1478: {
1479: int i;
1480: int lg=0;
1481: i=0;
1482: lg=strlen(t);
1483: for(i=0; i<= lg; i++) {
1484: (s[i] = t[i]);
1485: if (t[i]== '\\') s[i]='/';
1486: }
1487: }
1488:
1.132 brouard 1489: char *trimbb(char *out, char *in)
1.137 brouard 1490: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1491: char *s;
1492: s=out;
1493: while (*in != '\0'){
1.137 brouard 1494: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1495: in++;
1496: }
1497: *out++ = *in++;
1498: }
1499: *out='\0';
1500: return s;
1501: }
1502:
1.187 brouard 1503: /* char *substrchaine(char *out, char *in, char *chain) */
1504: /* { */
1505: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1506: /* char *s, *t; */
1507: /* t=in;s=out; */
1508: /* while ((*in != *chain) && (*in != '\0')){ */
1509: /* *out++ = *in++; */
1510: /* } */
1511:
1512: /* /\* *in matches *chain *\/ */
1513: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1514: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1515: /* } */
1516: /* in--; chain--; */
1517: /* while ( (*in != '\0')){ */
1518: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1519: /* *out++ = *in++; */
1520: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1521: /* } */
1522: /* *out='\0'; */
1523: /* out=s; */
1524: /* return out; */
1525: /* } */
1526: char *substrchaine(char *out, char *in, char *chain)
1527: {
1528: /* Substract chain 'chain' from 'in', return and output 'out' */
1529: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1530:
1531: char *strloc;
1532:
1533: strcpy (out, in);
1534: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1535: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1536: if(strloc != NULL){
1537: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1538: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1539: /* strcpy (strloc, strloc +strlen(chain));*/
1540: }
1541: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1542: return out;
1543: }
1544:
1545:
1.145 brouard 1546: char *cutl(char *blocc, char *alocc, char *in, char occ)
1547: {
1.187 brouard 1548: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1549: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1550: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1551: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1552: */
1.160 brouard 1553: char *s, *t;
1.145 brouard 1554: t=in;s=in;
1555: while ((*in != occ) && (*in != '\0')){
1556: *alocc++ = *in++;
1557: }
1558: if( *in == occ){
1559: *(alocc)='\0';
1560: s=++in;
1561: }
1562:
1563: if (s == t) {/* occ not found */
1564: *(alocc-(in-s))='\0';
1565: in=s;
1566: }
1567: while ( *in != '\0'){
1568: *blocc++ = *in++;
1569: }
1570:
1571: *blocc='\0';
1572: return t;
1573: }
1.137 brouard 1574: char *cutv(char *blocc, char *alocc, char *in, char occ)
1575: {
1.187 brouard 1576: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1577: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1578: gives blocc="abcdef2ghi" and alocc="j".
1579: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1580: */
1581: char *s, *t;
1582: t=in;s=in;
1583: while (*in != '\0'){
1584: while( *in == occ){
1585: *blocc++ = *in++;
1586: s=in;
1587: }
1588: *blocc++ = *in++;
1589: }
1590: if (s == t) /* occ not found */
1591: *(blocc-(in-s))='\0';
1592: else
1593: *(blocc-(in-s)-1)='\0';
1594: in=s;
1595: while ( *in != '\0'){
1596: *alocc++ = *in++;
1597: }
1598:
1599: *alocc='\0';
1600: return s;
1601: }
1602:
1.126 brouard 1603: int nbocc(char *s, char occ)
1604: {
1605: int i,j=0;
1606: int lg=20;
1607: i=0;
1608: lg=strlen(s);
1609: for(i=0; i<= lg; i++) {
1.234 brouard 1610: if (s[i] == occ ) j++;
1.126 brouard 1611: }
1612: return j;
1613: }
1614:
1.137 brouard 1615: /* void cutv(char *u,char *v, char*t, char occ) */
1616: /* { */
1617: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1618: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1619: /* gives u="abcdef2ghi" and v="j" *\/ */
1620: /* int i,lg,j,p=0; */
1621: /* i=0; */
1622: /* lg=strlen(t); */
1623: /* for(j=0; j<=lg-1; j++) { */
1624: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1625: /* } */
1.126 brouard 1626:
1.137 brouard 1627: /* for(j=0; j<p; j++) { */
1628: /* (u[j] = t[j]); */
1629: /* } */
1630: /* u[p]='\0'; */
1.126 brouard 1631:
1.137 brouard 1632: /* for(j=0; j<= lg; j++) { */
1633: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1634: /* } */
1635: /* } */
1.126 brouard 1636:
1.160 brouard 1637: #ifdef _WIN32
1638: char * strsep(char **pp, const char *delim)
1639: {
1640: char *p, *q;
1641:
1642: if ((p = *pp) == NULL)
1643: return 0;
1644: if ((q = strpbrk (p, delim)) != NULL)
1645: {
1646: *pp = q + 1;
1647: *q = '\0';
1648: }
1649: else
1650: *pp = 0;
1651: return p;
1652: }
1653: #endif
1654:
1.126 brouard 1655: /********************** nrerror ********************/
1656:
1657: void nrerror(char error_text[])
1658: {
1659: fprintf(stderr,"ERREUR ...\n");
1660: fprintf(stderr,"%s\n",error_text);
1661: exit(EXIT_FAILURE);
1662: }
1663: /*********************** vector *******************/
1664: double *vector(int nl, int nh)
1665: {
1666: double *v;
1667: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1668: if (!v) nrerror("allocation failure in vector");
1669: return v-nl+NR_END;
1670: }
1671:
1672: /************************ free vector ******************/
1673: void free_vector(double*v, int nl, int nh)
1674: {
1675: free((FREE_ARG)(v+nl-NR_END));
1676: }
1677:
1678: /************************ivector *******************************/
1679: int *ivector(long nl,long nh)
1680: {
1681: int *v;
1682: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1683: if (!v) nrerror("allocation failure in ivector");
1684: return v-nl+NR_END;
1685: }
1686:
1687: /******************free ivector **************************/
1688: void free_ivector(int *v, long nl, long nh)
1689: {
1690: free((FREE_ARG)(v+nl-NR_END));
1691: }
1692:
1693: /************************lvector *******************************/
1694: long *lvector(long nl,long nh)
1695: {
1696: long *v;
1697: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1698: if (!v) nrerror("allocation failure in ivector");
1699: return v-nl+NR_END;
1700: }
1701:
1702: /******************free lvector **************************/
1703: void free_lvector(long *v, long nl, long nh)
1704: {
1705: free((FREE_ARG)(v+nl-NR_END));
1706: }
1707:
1708: /******************* imatrix *******************************/
1709: int **imatrix(long nrl, long nrh, long ncl, long nch)
1710: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1711: {
1712: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1713: int **m;
1714:
1715: /* allocate pointers to rows */
1716: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1717: if (!m) nrerror("allocation failure 1 in matrix()");
1718: m += NR_END;
1719: m -= nrl;
1720:
1721:
1722: /* allocate rows and set pointers to them */
1723: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1724: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1725: m[nrl] += NR_END;
1726: m[nrl] -= ncl;
1727:
1728: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1729:
1730: /* return pointer to array of pointers to rows */
1731: return m;
1732: }
1733:
1734: /****************** free_imatrix *************************/
1735: void free_imatrix(m,nrl,nrh,ncl,nch)
1736: int **m;
1737: long nch,ncl,nrh,nrl;
1738: /* free an int matrix allocated by imatrix() */
1739: {
1740: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1741: free((FREE_ARG) (m+nrl-NR_END));
1742: }
1743:
1744: /******************* matrix *******************************/
1745: double **matrix(long nrl, long nrh, long ncl, long nch)
1746: {
1747: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1748: double **m;
1749:
1750: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1751: if (!m) nrerror("allocation failure 1 in matrix()");
1752: m += NR_END;
1753: m -= nrl;
1754:
1755: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1756: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1757: m[nrl] += NR_END;
1758: m[nrl] -= ncl;
1759:
1760: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1761: return m;
1.145 brouard 1762: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1763: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1764: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1765: */
1766: }
1767:
1768: /*************************free matrix ************************/
1769: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1770: {
1771: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1772: free((FREE_ARG)(m+nrl-NR_END));
1773: }
1774:
1775: /******************* ma3x *******************************/
1776: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1777: {
1778: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1779: double ***m;
1780:
1781: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1782: if (!m) nrerror("allocation failure 1 in matrix()");
1783: m += NR_END;
1784: m -= nrl;
1785:
1786: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1787: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1788: m[nrl] += NR_END;
1789: m[nrl] -= ncl;
1790:
1791: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1792:
1793: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1794: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1795: m[nrl][ncl] += NR_END;
1796: m[nrl][ncl] -= nll;
1797: for (j=ncl+1; j<=nch; j++)
1798: m[nrl][j]=m[nrl][j-1]+nlay;
1799:
1800: for (i=nrl+1; i<=nrh; i++) {
1801: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1802: for (j=ncl+1; j<=nch; j++)
1803: m[i][j]=m[i][j-1]+nlay;
1804: }
1805: return m;
1806: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1807: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1808: */
1809: }
1810:
1811: /*************************free ma3x ************************/
1812: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1813: {
1814: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1815: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1816: free((FREE_ARG)(m+nrl-NR_END));
1817: }
1818:
1819: /*************** function subdirf ***********/
1820: char *subdirf(char fileres[])
1821: {
1822: /* Caution optionfilefiname is hidden */
1823: strcpy(tmpout,optionfilefiname);
1824: strcat(tmpout,"/"); /* Add to the right */
1825: strcat(tmpout,fileres);
1826: return tmpout;
1827: }
1828:
1829: /*************** function subdirf2 ***********/
1830: char *subdirf2(char fileres[], char *preop)
1831: {
1832:
1833: /* Caution optionfilefiname is hidden */
1834: strcpy(tmpout,optionfilefiname);
1835: strcat(tmpout,"/");
1836: strcat(tmpout,preop);
1837: strcat(tmpout,fileres);
1838: return tmpout;
1839: }
1840:
1841: /*************** function subdirf3 ***********/
1842: char *subdirf3(char fileres[], char *preop, char *preop2)
1843: {
1844:
1845: /* Caution optionfilefiname is hidden */
1846: strcpy(tmpout,optionfilefiname);
1847: strcat(tmpout,"/");
1848: strcat(tmpout,preop);
1849: strcat(tmpout,preop2);
1850: strcat(tmpout,fileres);
1851: return tmpout;
1852: }
1.213 brouard 1853:
1854: /*************** function subdirfext ***********/
1855: char *subdirfext(char fileres[], char *preop, char *postop)
1856: {
1857:
1858: strcpy(tmpout,preop);
1859: strcat(tmpout,fileres);
1860: strcat(tmpout,postop);
1861: return tmpout;
1862: }
1.126 brouard 1863:
1.213 brouard 1864: /*************** function subdirfext3 ***********/
1865: char *subdirfext3(char fileres[], char *preop, char *postop)
1866: {
1867:
1868: /* Caution optionfilefiname is hidden */
1869: strcpy(tmpout,optionfilefiname);
1870: strcat(tmpout,"/");
1871: strcat(tmpout,preop);
1872: strcat(tmpout,fileres);
1873: strcat(tmpout,postop);
1874: return tmpout;
1875: }
1876:
1.162 brouard 1877: char *asc_diff_time(long time_sec, char ascdiff[])
1878: {
1879: long sec_left, days, hours, minutes;
1880: days = (time_sec) / (60*60*24);
1881: sec_left = (time_sec) % (60*60*24);
1882: hours = (sec_left) / (60*60) ;
1883: sec_left = (sec_left) %(60*60);
1884: minutes = (sec_left) /60;
1885: sec_left = (sec_left) % (60);
1886: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1887: return ascdiff;
1888: }
1889:
1.126 brouard 1890: /***************** f1dim *************************/
1891: extern int ncom;
1892: extern double *pcom,*xicom;
1893: extern double (*nrfunc)(double []);
1894:
1895: double f1dim(double x)
1896: {
1897: int j;
1898: double f;
1899: double *xt;
1900:
1901: xt=vector(1,ncom);
1902: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1903: f=(*nrfunc)(xt);
1904: free_vector(xt,1,ncom);
1905: return f;
1906: }
1907:
1908: /*****************brent *************************/
1909: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1910: {
1911: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1912: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1913: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1914: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1915: * returned function value.
1916: */
1.126 brouard 1917: int iter;
1918: double a,b,d,etemp;
1.159 brouard 1919: double fu=0,fv,fw,fx;
1.164 brouard 1920: double ftemp=0.;
1.126 brouard 1921: double p,q,r,tol1,tol2,u,v,w,x,xm;
1922: double e=0.0;
1923:
1924: a=(ax < cx ? ax : cx);
1925: b=(ax > cx ? ax : cx);
1926: x=w=v=bx;
1927: fw=fv=fx=(*f)(x);
1928: for (iter=1;iter<=ITMAX;iter++) {
1929: xm=0.5*(a+b);
1930: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1931: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1932: printf(".");fflush(stdout);
1933: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1934: #ifdef DEBUGBRENT
1.126 brouard 1935: 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);
1936: 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);
1937: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1938: #endif
1939: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1940: *xmin=x;
1941: return fx;
1942: }
1943: ftemp=fu;
1944: if (fabs(e) > tol1) {
1945: r=(x-w)*(fx-fv);
1946: q=(x-v)*(fx-fw);
1947: p=(x-v)*q-(x-w)*r;
1948: q=2.0*(q-r);
1949: if (q > 0.0) p = -p;
1950: q=fabs(q);
1951: etemp=e;
1952: e=d;
1953: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1954: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1955: else {
1.224 brouard 1956: d=p/q;
1957: u=x+d;
1958: if (u-a < tol2 || b-u < tol2)
1959: d=SIGN(tol1,xm-x);
1.126 brouard 1960: }
1961: } else {
1962: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1963: }
1964: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1965: fu=(*f)(u);
1966: if (fu <= fx) {
1967: if (u >= x) a=x; else b=x;
1968: SHFT(v,w,x,u)
1.183 brouard 1969: SHFT(fv,fw,fx,fu)
1970: } else {
1971: if (u < x) a=u; else b=u;
1972: if (fu <= fw || w == x) {
1.224 brouard 1973: v=w;
1974: w=u;
1975: fv=fw;
1976: fw=fu;
1.183 brouard 1977: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1978: v=u;
1979: fv=fu;
1.183 brouard 1980: }
1981: }
1.126 brouard 1982: }
1983: nrerror("Too many iterations in brent");
1984: *xmin=x;
1985: return fx;
1986: }
1987:
1988: /****************** mnbrak ***********************/
1989:
1990: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1991: double (*func)(double))
1.183 brouard 1992: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1993: the downhill direction (defined by the function as evaluated at the initial points) and returns
1994: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1995: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1996: */
1.126 brouard 1997: double ulim,u,r,q, dum;
1998: double fu;
1.187 brouard 1999:
2000: double scale=10.;
2001: int iterscale=0;
2002:
2003: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2004: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2005:
2006:
2007: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2008: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2009: /* *bx = *ax - (*ax - *bx)/scale; */
2010: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2011: /* } */
2012:
1.126 brouard 2013: if (*fb > *fa) {
2014: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2015: SHFT(dum,*fb,*fa,dum)
2016: }
1.126 brouard 2017: *cx=(*bx)+GOLD*(*bx-*ax);
2018: *fc=(*func)(*cx);
1.183 brouard 2019: #ifdef DEBUG
1.224 brouard 2020: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2021: 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 2022: #endif
1.224 brouard 2023: 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 2024: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2025: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2026: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2027: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2028: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2029: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2030: fu=(*func)(u);
1.163 brouard 2031: #ifdef DEBUG
2032: /* f(x)=A(x-u)**2+f(u) */
2033: double A, fparabu;
2034: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2035: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2036: 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);
2037: 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 2038: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2039: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2040: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2041: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2042: #endif
1.184 brouard 2043: #ifdef MNBRAKORIGINAL
1.183 brouard 2044: #else
1.191 brouard 2045: /* if (fu > *fc) { */
2046: /* #ifdef DEBUG */
2047: /* printf("mnbrak4 fu > fc \n"); */
2048: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2049: /* #endif */
2050: /* /\* 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 *\\/ *\/ */
2051: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2052: /* dum=u; /\* Shifting c and u *\/ */
2053: /* u = *cx; */
2054: /* *cx = dum; */
2055: /* dum = fu; */
2056: /* fu = *fc; */
2057: /* *fc =dum; */
2058: /* } else { /\* end *\/ */
2059: /* #ifdef DEBUG */
2060: /* printf("mnbrak3 fu < fc \n"); */
2061: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2062: /* #endif */
2063: /* dum=u; /\* Shifting c and u *\/ */
2064: /* u = *cx; */
2065: /* *cx = dum; */
2066: /* dum = fu; */
2067: /* fu = *fc; */
2068: /* *fc =dum; */
2069: /* } */
1.224 brouard 2070: #ifdef DEBUGMNBRAK
2071: double A, fparabu;
2072: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2073: fparabu= *fa - A*(*ax-u)*(*ax-u);
2074: 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);
2075: 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 2076: #endif
1.191 brouard 2077: dum=u; /* Shifting c and u */
2078: u = *cx;
2079: *cx = dum;
2080: dum = fu;
2081: fu = *fc;
2082: *fc =dum;
1.183 brouard 2083: #endif
1.162 brouard 2084: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2085: #ifdef DEBUG
1.224 brouard 2086: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2087: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2088: #endif
1.126 brouard 2089: fu=(*func)(u);
2090: if (fu < *fc) {
1.183 brouard 2091: #ifdef DEBUG
1.224 brouard 2092: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2093: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2094: #endif
2095: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2096: SHFT(*fb,*fc,fu,(*func)(u))
2097: #ifdef DEBUG
2098: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2099: #endif
2100: }
1.162 brouard 2101: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2102: #ifdef DEBUG
1.224 brouard 2103: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2104: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2105: #endif
1.126 brouard 2106: u=ulim;
2107: fu=(*func)(u);
1.183 brouard 2108: } else { /* u could be left to b (if r > q parabola has a maximum) */
2109: #ifdef DEBUG
1.224 brouard 2110: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2111: 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 2112: #endif
1.126 brouard 2113: u=(*cx)+GOLD*(*cx-*bx);
2114: fu=(*func)(u);
1.224 brouard 2115: #ifdef DEBUG
2116: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2117: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2118: #endif
1.183 brouard 2119: } /* end tests */
1.126 brouard 2120: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2121: SHFT(*fa,*fb,*fc,fu)
2122: #ifdef DEBUG
1.224 brouard 2123: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2124: 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 2125: #endif
2126: } /* 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 2127: }
2128:
2129: /*************** linmin ************************/
1.162 brouard 2130: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2131: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2132: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2133: the value of func at the returned location p . This is actually all accomplished by calling the
2134: routines mnbrak and brent .*/
1.126 brouard 2135: int ncom;
2136: double *pcom,*xicom;
2137: double (*nrfunc)(double []);
2138:
1.224 brouard 2139: #ifdef LINMINORIGINAL
1.126 brouard 2140: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2141: #else
2142: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2143: #endif
1.126 brouard 2144: {
2145: double brent(double ax, double bx, double cx,
2146: double (*f)(double), double tol, double *xmin);
2147: double f1dim(double x);
2148: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2149: double *fc, double (*func)(double));
2150: int j;
2151: double xx,xmin,bx,ax;
2152: double fx,fb,fa;
1.187 brouard 2153:
1.203 brouard 2154: #ifdef LINMINORIGINAL
2155: #else
2156: double scale=10., axs, xxs; /* Scale added for infinity */
2157: #endif
2158:
1.126 brouard 2159: ncom=n;
2160: pcom=vector(1,n);
2161: xicom=vector(1,n);
2162: nrfunc=func;
2163: for (j=1;j<=n;j++) {
2164: pcom[j]=p[j];
1.202 brouard 2165: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2166: }
1.187 brouard 2167:
1.203 brouard 2168: #ifdef LINMINORIGINAL
2169: xx=1.;
2170: #else
2171: axs=0.0;
2172: xxs=1.;
2173: do{
2174: xx= xxs;
2175: #endif
1.187 brouard 2176: ax=0.;
2177: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2178: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2179: /* 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)) */
2180: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2181: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2182: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2183: /* 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 2184: #ifdef LINMINORIGINAL
2185: #else
2186: if (fx != fx){
1.224 brouard 2187: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2188: printf("|");
2189: fprintf(ficlog,"|");
1.203 brouard 2190: #ifdef DEBUGLINMIN
1.224 brouard 2191: 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 2192: #endif
2193: }
1.224 brouard 2194: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2195: #endif
2196:
1.191 brouard 2197: #ifdef DEBUGLINMIN
2198: 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 2199: 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 2200: #endif
1.224 brouard 2201: #ifdef LINMINORIGINAL
2202: #else
2203: if(fb == fx){ /* Flat function in the direction */
2204: xmin=xx;
2205: *flat=1;
2206: }else{
2207: *flat=0;
2208: #endif
2209: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2210: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2211: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2212: /* fmin = f(p[j] + xmin * xi[j]) */
2213: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2214: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2215: #ifdef DEBUG
1.224 brouard 2216: 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);
2217: 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);
2218: #endif
2219: #ifdef LINMINORIGINAL
2220: #else
2221: }
1.126 brouard 2222: #endif
1.191 brouard 2223: #ifdef DEBUGLINMIN
2224: printf("linmin end ");
1.202 brouard 2225: fprintf(ficlog,"linmin end ");
1.191 brouard 2226: #endif
1.126 brouard 2227: for (j=1;j<=n;j++) {
1.203 brouard 2228: #ifdef LINMINORIGINAL
2229: xi[j] *= xmin;
2230: #else
2231: #ifdef DEBUGLINMIN
2232: if(xxs <1.0)
2233: printf(" before xi[%d]=%12.8f", j,xi[j]);
2234: #endif
2235: 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) */
2236: #ifdef DEBUGLINMIN
2237: if(xxs <1.0)
2238: 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 );
2239: #endif
2240: #endif
1.187 brouard 2241: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2242: }
1.191 brouard 2243: #ifdef DEBUGLINMIN
1.203 brouard 2244: printf("\n");
1.191 brouard 2245: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2246: 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 2247: for (j=1;j<=n;j++) {
1.202 brouard 2248: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2249: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2250: if(j % ncovmodel == 0){
1.191 brouard 2251: printf("\n");
1.202 brouard 2252: fprintf(ficlog,"\n");
2253: }
1.191 brouard 2254: }
1.203 brouard 2255: #else
1.191 brouard 2256: #endif
1.126 brouard 2257: free_vector(xicom,1,n);
2258: free_vector(pcom,1,n);
2259: }
2260:
2261:
2262: /*************** powell ************************/
1.162 brouard 2263: /*
2264: Minimization of a function func of n variables. Input consists of an initial starting point
2265: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2266: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2267: such that failure to decrease by more than this amount on one iteration signals doneness. On
2268: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2269: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2270: */
1.224 brouard 2271: #ifdef LINMINORIGINAL
2272: #else
2273: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2274: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2275: #endif
1.126 brouard 2276: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2277: double (*func)(double []))
2278: {
1.224 brouard 2279: #ifdef LINMINORIGINAL
2280: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2281: double (*func)(double []));
1.224 brouard 2282: #else
1.241 brouard 2283: void linmin(double p[], double xi[], int n, double *fret,
2284: double (*func)(double []),int *flat);
1.224 brouard 2285: #endif
1.239 brouard 2286: int i,ibig,j,jk,k;
1.126 brouard 2287: double del,t,*pt,*ptt,*xit;
1.181 brouard 2288: double directest;
1.126 brouard 2289: double fp,fptt;
2290: double *xits;
2291: int niterf, itmp;
1.224 brouard 2292: #ifdef LINMINORIGINAL
2293: #else
2294:
2295: flatdir=ivector(1,n);
2296: for (j=1;j<=n;j++) flatdir[j]=0;
2297: #endif
1.126 brouard 2298:
2299: pt=vector(1,n);
2300: ptt=vector(1,n);
2301: xit=vector(1,n);
2302: xits=vector(1,n);
2303: *fret=(*func)(p);
2304: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2305: rcurr_time = time(NULL);
1.126 brouard 2306: for (*iter=1;;++(*iter)) {
1.187 brouard 2307: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2308: ibig=0;
2309: del=0.0;
1.157 brouard 2310: rlast_time=rcurr_time;
2311: /* (void) gettimeofday(&curr_time,&tzp); */
2312: rcurr_time = time(NULL);
2313: curr_time = *localtime(&rcurr_time);
2314: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2315: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2316: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2317: for (i=1;i<=n;i++) {
1.126 brouard 2318: fprintf(ficrespow," %.12lf", p[i]);
2319: }
1.239 brouard 2320: fprintf(ficrespow,"\n");fflush(ficrespow);
2321: printf("\n#model= 1 + age ");
2322: fprintf(ficlog,"\n#model= 1 + age ");
2323: if(nagesqr==1){
1.241 brouard 2324: printf(" + age*age ");
2325: fprintf(ficlog," + age*age ");
1.239 brouard 2326: }
2327: for(j=1;j <=ncovmodel-2;j++){
2328: if(Typevar[j]==0) {
2329: printf(" + V%d ",Tvar[j]);
2330: fprintf(ficlog," + V%d ",Tvar[j]);
2331: }else if(Typevar[j]==1) {
2332: printf(" + V%d*age ",Tvar[j]);
2333: fprintf(ficlog," + V%d*age ",Tvar[j]);
2334: }else if(Typevar[j]==2) {
2335: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2336: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2337: }
2338: }
1.126 brouard 2339: printf("\n");
1.239 brouard 2340: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2341: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2342: fprintf(ficlog,"\n");
1.239 brouard 2343: for(i=1,jk=1; i <=nlstate; i++){
2344: for(k=1; k <=(nlstate+ndeath); k++){
2345: if (k != i) {
2346: printf("%d%d ",i,k);
2347: fprintf(ficlog,"%d%d ",i,k);
2348: for(j=1; j <=ncovmodel; j++){
2349: printf("%12.7f ",p[jk]);
2350: fprintf(ficlog,"%12.7f ",p[jk]);
2351: jk++;
2352: }
2353: printf("\n");
2354: fprintf(ficlog,"\n");
2355: }
2356: }
2357: }
1.241 brouard 2358: if(*iter <=3 && *iter >1){
1.157 brouard 2359: tml = *localtime(&rcurr_time);
2360: strcpy(strcurr,asctime(&tml));
2361: rforecast_time=rcurr_time;
1.126 brouard 2362: itmp = strlen(strcurr);
2363: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2364: strcurr[itmp-1]='\0';
1.162 brouard 2365: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2366: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2367: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2368: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2369: forecast_time = *localtime(&rforecast_time);
2370: strcpy(strfor,asctime(&forecast_time));
2371: itmp = strlen(strfor);
2372: if(strfor[itmp-1]=='\n')
2373: strfor[itmp-1]='\0';
2374: 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);
2375: 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 2376: }
2377: }
1.187 brouard 2378: for (i=1;i<=n;i++) { /* For each direction i */
2379: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2380: fptt=(*fret);
2381: #ifdef DEBUG
1.203 brouard 2382: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2383: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2384: #endif
1.203 brouard 2385: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2386: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2387: #ifdef LINMINORIGINAL
1.188 brouard 2388: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2389: #else
2390: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2391: flatdir[i]=flat; /* Function is vanishing in that direction i */
2392: #endif
2393: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2394: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2395: /* because that direction will be replaced unless the gain del is small */
2396: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2397: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2398: /* with the new direction. */
2399: del=fabs(fptt-(*fret));
2400: ibig=i;
1.126 brouard 2401: }
2402: #ifdef DEBUG
2403: printf("%d %.12e",i,(*fret));
2404: fprintf(ficlog,"%d %.12e",i,(*fret));
2405: for (j=1;j<=n;j++) {
1.224 brouard 2406: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2407: printf(" x(%d)=%.12e",j,xit[j]);
2408: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2409: }
2410: for(j=1;j<=n;j++) {
1.225 brouard 2411: printf(" p(%d)=%.12e",j,p[j]);
2412: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2413: }
2414: printf("\n");
2415: fprintf(ficlog,"\n");
2416: #endif
1.187 brouard 2417: } /* end loop on each direction i */
2418: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2419: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2420: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2421: for(j=1;j<=n;j++) {
1.302 brouard 2422: if(flatdir[j] >0){
2423: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2424: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2425: }
2426: /* printf("\n"); */
2427: /* fprintf(ficlog,"\n"); */
2428: }
1.243 brouard 2429: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2430: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2431: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2432: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2433: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2434: /* decreased of more than 3.84 */
2435: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2436: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2437: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2438:
1.188 brouard 2439: /* Starting the program with initial values given by a former maximization will simply change */
2440: /* the scales of the directions and the directions, because the are reset to canonical directions */
2441: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2442: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2443: #ifdef DEBUG
2444: int k[2],l;
2445: k[0]=1;
2446: k[1]=-1;
2447: printf("Max: %.12e",(*func)(p));
2448: fprintf(ficlog,"Max: %.12e",(*func)(p));
2449: for (j=1;j<=n;j++) {
2450: printf(" %.12e",p[j]);
2451: fprintf(ficlog," %.12e",p[j]);
2452: }
2453: printf("\n");
2454: fprintf(ficlog,"\n");
2455: for(l=0;l<=1;l++) {
2456: for (j=1;j<=n;j++) {
2457: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2458: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2459: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2460: }
2461: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2462: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2463: }
2464: #endif
2465:
1.224 brouard 2466: #ifdef LINMINORIGINAL
2467: #else
2468: free_ivector(flatdir,1,n);
2469: #endif
1.126 brouard 2470: free_vector(xit,1,n);
2471: free_vector(xits,1,n);
2472: free_vector(ptt,1,n);
2473: free_vector(pt,1,n);
2474: return;
1.192 brouard 2475: } /* enough precision */
1.240 brouard 2476: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2477: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2478: ptt[j]=2.0*p[j]-pt[j];
2479: xit[j]=p[j]-pt[j];
2480: pt[j]=p[j];
2481: }
1.181 brouard 2482: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2483: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2484: if (*iter <=4) {
1.225 brouard 2485: #else
2486: #endif
1.224 brouard 2487: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2488: #else
1.161 brouard 2489: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2490: #endif
1.162 brouard 2491: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2492: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2493: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2494: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2495: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2496: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2497: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2498: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2499: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2500: /* Even if f3 <f1, directest can be negative and t >0 */
2501: /* mu² and del² are equal when f3=f1 */
2502: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2503: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2504: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2505: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2506: #ifdef NRCORIGINAL
2507: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2508: #else
2509: 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 2510: t= t- del*SQR(fp-fptt);
1.183 brouard 2511: #endif
1.202 brouard 2512: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2513: #ifdef DEBUG
1.181 brouard 2514: 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);
2515: 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 2516: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2517: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2518: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2519: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2520: 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);
2521: 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);
2522: #endif
1.183 brouard 2523: #ifdef POWELLORIGINAL
2524: if (t < 0.0) { /* Then we use it for new direction */
2525: #else
1.182 brouard 2526: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2527: 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 2528: 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 2529: 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 2530: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2531: }
1.181 brouard 2532: if (directest < 0.0) { /* Then we use it for new direction */
2533: #endif
1.191 brouard 2534: #ifdef DEBUGLINMIN
1.234 brouard 2535: printf("Before linmin in direction P%d-P0\n",n);
2536: for (j=1;j<=n;j++) {
2537: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2538: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2539: if(j % ncovmodel == 0){
2540: printf("\n");
2541: fprintf(ficlog,"\n");
2542: }
2543: }
1.224 brouard 2544: #endif
2545: #ifdef LINMINORIGINAL
1.234 brouard 2546: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2547: #else
1.234 brouard 2548: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2549: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2550: #endif
1.234 brouard 2551:
1.191 brouard 2552: #ifdef DEBUGLINMIN
1.234 brouard 2553: for (j=1;j<=n;j++) {
2554: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2555: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2556: if(j % ncovmodel == 0){
2557: printf("\n");
2558: fprintf(ficlog,"\n");
2559: }
2560: }
1.224 brouard 2561: #endif
1.234 brouard 2562: for (j=1;j<=n;j++) {
2563: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2564: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2565: }
1.224 brouard 2566: #ifdef LINMINORIGINAL
2567: #else
1.234 brouard 2568: for (j=1, flatd=0;j<=n;j++) {
2569: if(flatdir[j]>0)
2570: flatd++;
2571: }
2572: if(flatd >0){
1.255 brouard 2573: printf("%d flat directions: ",flatd);
2574: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2575: for (j=1;j<=n;j++) {
2576: if(flatdir[j]>0){
2577: printf("%d ",j);
2578: fprintf(ficlog,"%d ",j);
2579: }
2580: }
2581: printf("\n");
2582: fprintf(ficlog,"\n");
2583: }
1.191 brouard 2584: #endif
1.234 brouard 2585: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2586: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2587:
1.126 brouard 2588: #ifdef DEBUG
1.234 brouard 2589: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2590: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2591: for(j=1;j<=n;j++){
2592: printf(" %lf",xit[j]);
2593: fprintf(ficlog," %lf",xit[j]);
2594: }
2595: printf("\n");
2596: fprintf(ficlog,"\n");
1.126 brouard 2597: #endif
1.192 brouard 2598: } /* end of t or directest negative */
1.224 brouard 2599: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2600: #else
1.234 brouard 2601: } /* end if (fptt < fp) */
1.192 brouard 2602: #endif
1.225 brouard 2603: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2604: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2605: #else
1.224 brouard 2606: #endif
1.234 brouard 2607: } /* loop iteration */
1.126 brouard 2608: }
1.234 brouard 2609:
1.126 brouard 2610: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2611:
1.235 brouard 2612: 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 2613: {
1.279 brouard 2614: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2615: * (and selected quantitative values in nres)
2616: * by left multiplying the unit
2617: * matrix by transitions matrix until convergence is reached with precision ftolpl
2618: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2619: * Wx is row vector: population in state 1, population in state 2, population dead
2620: * or prevalence in state 1, prevalence in state 2, 0
2621: * newm is the matrix after multiplications, its rows are identical at a factor.
2622: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2623: * Output is prlim.
2624: * Initial matrix pimij
2625: */
1.206 brouard 2626: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2627: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2628: /* 0, 0 , 1} */
2629: /*
2630: * and after some iteration: */
2631: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2632: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2633: /* 0, 0 , 1} */
2634: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2635: /* {0.51571254859325999, 0.4842874514067399, */
2636: /* 0.51326036147820708, 0.48673963852179264} */
2637: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2638:
1.126 brouard 2639: int i, ii,j,k;
1.209 brouard 2640: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2641: /* double **matprod2(); */ /* test */
1.218 brouard 2642: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2643: double **newm;
1.209 brouard 2644: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2645: int ncvloop=0;
1.288 brouard 2646: int first=0;
1.169 brouard 2647:
1.209 brouard 2648: min=vector(1,nlstate);
2649: max=vector(1,nlstate);
2650: meandiff=vector(1,nlstate);
2651:
1.218 brouard 2652: /* Starting with matrix unity */
1.126 brouard 2653: for (ii=1;ii<=nlstate+ndeath;ii++)
2654: for (j=1;j<=nlstate+ndeath;j++){
2655: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2656: }
1.169 brouard 2657:
2658: cov[1]=1.;
2659:
2660: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2661: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2662: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2663: ncvloop++;
1.126 brouard 2664: newm=savm;
2665: /* Covariates have to be included here again */
1.138 brouard 2666: cov[2]=agefin;
1.187 brouard 2667: if(nagesqr==1)
2668: cov[3]= agefin*agefin;;
1.234 brouard 2669: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2670: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2671: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2672: /* 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 2673: }
2674: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2675: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2676: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2677: /* 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 2678: }
1.237 brouard 2679: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2680: if(Dummy[Tvar[Tage[k]]]){
2681: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2682: } else{
1.235 brouard 2683: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2684: }
1.235 brouard 2685: /* 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 2686: }
1.237 brouard 2687: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2688: /* 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 2689: if(Dummy[Tvard[k][1]==0]){
2690: if(Dummy[Tvard[k][2]==0]){
2691: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2692: }else{
2693: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2694: }
2695: }else{
2696: if(Dummy[Tvard[k][2]==0]){
2697: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2698: }else{
2699: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2700: }
2701: }
1.234 brouard 2702: }
1.138 brouard 2703: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2704: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2705: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2706: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2707: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2708: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2709: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2710:
1.126 brouard 2711: savm=oldm;
2712: oldm=newm;
1.209 brouard 2713:
2714: for(j=1; j<=nlstate; j++){
2715: max[j]=0.;
2716: min[j]=1.;
2717: }
2718: for(i=1;i<=nlstate;i++){
2719: sumnew=0;
2720: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2721: for(j=1; j<=nlstate; j++){
2722: prlim[i][j]= newm[i][j]/(1-sumnew);
2723: max[j]=FMAX(max[j],prlim[i][j]);
2724: min[j]=FMIN(min[j],prlim[i][j]);
2725: }
2726: }
2727:
1.126 brouard 2728: maxmax=0.;
1.209 brouard 2729: for(j=1; j<=nlstate; j++){
2730: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2731: maxmax=FMAX(maxmax,meandiff[j]);
2732: /* 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 2733: } /* j loop */
1.203 brouard 2734: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2735: /* 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 2736: if(maxmax < ftolpl){
1.209 brouard 2737: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2738: free_vector(min,1,nlstate);
2739: free_vector(max,1,nlstate);
2740: free_vector(meandiff,1,nlstate);
1.126 brouard 2741: return prlim;
2742: }
1.288 brouard 2743: } /* agefin loop */
1.208 brouard 2744: /* After some age loop it doesn't converge */
1.288 brouard 2745: if(!first){
2746: first=1;
2747: 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);
2748: }
2749: 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);
2750:
1.209 brouard 2751: /* 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); */
2752: free_vector(min,1,nlstate);
2753: free_vector(max,1,nlstate);
2754: free_vector(meandiff,1,nlstate);
1.208 brouard 2755:
1.169 brouard 2756: return prlim; /* should not reach here */
1.126 brouard 2757: }
2758:
1.217 brouard 2759:
2760: /**** Back Prevalence limit (stable or period prevalence) ****************/
2761:
1.218 brouard 2762: /* 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) */
2763: /* 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 2764: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2765: {
1.264 brouard 2766: /* 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 2767: matrix by transitions matrix until convergence is reached with precision ftolpl */
2768: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2769: /* Wx is row vector: population in state 1, population in state 2, population dead */
2770: /* or prevalence in state 1, prevalence in state 2, 0 */
2771: /* newm is the matrix after multiplications, its rows are identical at a factor */
2772: /* Initial matrix pimij */
2773: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2774: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2775: /* 0, 0 , 1} */
2776: /*
2777: * and after some iteration: */
2778: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2779: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2780: /* 0, 0 , 1} */
2781: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2782: /* {0.51571254859325999, 0.4842874514067399, */
2783: /* 0.51326036147820708, 0.48673963852179264} */
2784: /* If we start from prlim again, prlim tends to a constant matrix */
2785:
2786: int i, ii,j,k;
1.247 brouard 2787: int first=0;
1.217 brouard 2788: double *min, *max, *meandiff, maxmax,sumnew=0.;
2789: /* double **matprod2(); */ /* test */
2790: double **out, cov[NCOVMAX+1], **bmij();
2791: double **newm;
1.218 brouard 2792: double **dnewm, **doldm, **dsavm; /* for use */
2793: double **oldm, **savm; /* for use */
2794:
1.217 brouard 2795: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2796: int ncvloop=0;
2797:
2798: min=vector(1,nlstate);
2799: max=vector(1,nlstate);
2800: meandiff=vector(1,nlstate);
2801:
1.266 brouard 2802: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2803: oldm=oldms; savm=savms;
2804:
2805: /* Starting with matrix unity */
2806: for (ii=1;ii<=nlstate+ndeath;ii++)
2807: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2808: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2809: }
2810:
2811: cov[1]=1.;
2812:
2813: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2814: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2815: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2816: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2817: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2818: ncvloop++;
1.218 brouard 2819: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2820: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2821: /* Covariates have to be included here again */
2822: cov[2]=agefin;
2823: if(nagesqr==1)
2824: cov[3]= agefin*agefin;;
1.242 brouard 2825: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2826: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2827: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2828: /* 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 2829: }
2830: /* for (k=1; k<=cptcovn;k++) { */
2831: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2832: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2833: /* /\* 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])]); *\/ */
2834: /* } */
2835: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2836: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2837: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2838: /* 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]); */
2839: }
2840: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2841: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2842: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2843: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2844: for (k=1; k<=cptcovage;k++){ /* For product with age */
2845: if(Dummy[Tvar[Tage[k]]]){
2846: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2847: } else{
2848: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2849: }
2850: /* 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]); */
2851: }
2852: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2853: /* 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]); */
2854: if(Dummy[Tvard[k][1]==0]){
2855: if(Dummy[Tvard[k][2]==0]){
2856: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2857: }else{
2858: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2859: }
2860: }else{
2861: if(Dummy[Tvard[k][2]==0]){
2862: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2863: }else{
2864: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2865: }
2866: }
1.217 brouard 2867: }
2868:
2869: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2870: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2871: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2872: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2873: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2874: /* ij should be linked to the correct index of cov */
2875: /* age and covariate values ij are in 'cov', but we need to pass
2876: * ij for the observed prevalence at age and status and covariate
2877: * number: prevacurrent[(int)agefin][ii][ij]
2878: */
2879: /* 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 *\/ */
2880: /* 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 *\/ */
2881: 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 2882: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2883: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2884: /* for(i=1; i<=nlstate+ndeath; i++) { */
2885: /* printf("%d newm= ",i); */
2886: /* for(j=1;j<=nlstate+ndeath;j++) { */
2887: /* printf("%f ",newm[i][j]); */
2888: /* } */
2889: /* printf("oldm * "); */
2890: /* for(j=1;j<=nlstate+ndeath;j++) { */
2891: /* printf("%f ",oldm[i][j]); */
2892: /* } */
1.268 brouard 2893: /* printf(" bmmij "); */
1.266 brouard 2894: /* for(j=1;j<=nlstate+ndeath;j++) { */
2895: /* printf("%f ",pmmij[i][j]); */
2896: /* } */
2897: /* printf("\n"); */
2898: /* } */
2899: /* } */
1.217 brouard 2900: savm=oldm;
2901: oldm=newm;
1.266 brouard 2902:
1.217 brouard 2903: for(j=1; j<=nlstate; j++){
2904: max[j]=0.;
2905: min[j]=1.;
2906: }
2907: for(j=1; j<=nlstate; j++){
2908: for(i=1;i<=nlstate;i++){
1.234 brouard 2909: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2910: bprlim[i][j]= newm[i][j];
2911: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2912: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2913: }
2914: }
1.218 brouard 2915:
1.217 brouard 2916: maxmax=0.;
2917: for(i=1; i<=nlstate; i++){
2918: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2919: maxmax=FMAX(maxmax,meandiff[i]);
2920: /* 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 2921: } /* i loop */
1.217 brouard 2922: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2923: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2924: if(maxmax < ftolpl){
1.220 brouard 2925: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2926: free_vector(min,1,nlstate);
2927: free_vector(max,1,nlstate);
2928: free_vector(meandiff,1,nlstate);
2929: return bprlim;
2930: }
1.288 brouard 2931: } /* agefin loop */
1.217 brouard 2932: /* After some age loop it doesn't converge */
1.288 brouard 2933: if(!first){
1.247 brouard 2934: first=1;
2935: 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\
2936: 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);
2937: }
2938: 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 2939: 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);
2940: /* 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); */
2941: free_vector(min,1,nlstate);
2942: free_vector(max,1,nlstate);
2943: free_vector(meandiff,1,nlstate);
2944:
2945: return bprlim; /* should not reach here */
2946: }
2947:
1.126 brouard 2948: /*************** transition probabilities ***************/
2949:
2950: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2951: {
1.138 brouard 2952: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2953: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2954: model to the ncovmodel covariates (including constant and age).
2955: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2956: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2957: ncth covariate in the global vector x is given by the formula:
2958: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2959: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2960: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2961: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2962: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2963: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2964: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2965: */
2966: double s1, lnpijopii;
1.126 brouard 2967: /*double t34;*/
1.164 brouard 2968: int i,j, nc, ii, jj;
1.126 brouard 2969:
1.223 brouard 2970: for(i=1; i<= nlstate; i++){
2971: for(j=1; j<i;j++){
2972: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2973: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2974: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2975: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2976: }
2977: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2978: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2979: }
2980: for(j=i+1; j<=nlstate+ndeath;j++){
2981: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2982: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2983: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2984: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2985: }
2986: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2987: }
2988: }
1.218 brouard 2989:
1.223 brouard 2990: for(i=1; i<= nlstate; i++){
2991: s1=0;
2992: for(j=1; j<i; j++){
2993: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2994: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2995: }
2996: for(j=i+1; j<=nlstate+ndeath; j++){
2997: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2998: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2999: }
3000: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3001: ps[i][i]=1./(s1+1.);
3002: /* Computing other pijs */
3003: for(j=1; j<i; j++)
3004: ps[i][j]= exp(ps[i][j])*ps[i][i];
3005: for(j=i+1; j<=nlstate+ndeath; j++)
3006: ps[i][j]= exp(ps[i][j])*ps[i][i];
3007: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3008: } /* end i */
1.218 brouard 3009:
1.223 brouard 3010: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3011: for(jj=1; jj<= nlstate+ndeath; jj++){
3012: ps[ii][jj]=0;
3013: ps[ii][ii]=1;
3014: }
3015: }
1.294 brouard 3016:
3017:
1.223 brouard 3018: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3019: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3020: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3021: /* } */
3022: /* printf("\n "); */
3023: /* } */
3024: /* printf("\n ");printf("%lf ",cov[2]);*/
3025: /*
3026: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3027: goto end;*/
1.266 brouard 3028: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3029: }
3030:
1.218 brouard 3031: /*************** backward transition probabilities ***************/
3032:
3033: /* 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 ) */
3034: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3035: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3036: {
1.302 brouard 3037: /* 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 3038: * 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 3039: */
1.218 brouard 3040: int i, ii, j,k;
1.222 brouard 3041:
3042: double **out, **pmij();
3043: double sumnew=0.;
1.218 brouard 3044: double agefin;
1.292 brouard 3045: 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 3046: double **dnewm, **dsavm, **doldm;
3047: double **bbmij;
3048:
1.218 brouard 3049: doldm=ddoldms; /* global pointers */
1.222 brouard 3050: dnewm=ddnewms;
3051: dsavm=ddsavms;
3052:
3053: agefin=cov[2];
1.268 brouard 3054: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3055: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3056: the observed prevalence (with this covariate ij) at beginning of transition */
3057: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3058:
3059: /* P_x */
1.266 brouard 3060: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3061: /* outputs pmmij which is a stochastic matrix in row */
3062:
3063: /* Diag(w_x) */
1.292 brouard 3064: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3065: sumnew=0.;
1.269 brouard 3066: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3067: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3068: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3069: sumnew+=prevacurrent[(int)agefin][ii][ij];
3070: }
3071: if(sumnew >0.01){ /* At least some value in the prevalence */
3072: for (ii=1;ii<=nlstate+ndeath;ii++){
3073: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3074: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3075: }
3076: }else{
3077: for (ii=1;ii<=nlstate+ndeath;ii++){
3078: for (j=1;j<=nlstate+ndeath;j++)
3079: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3080: }
3081: /* if(sumnew <0.9){ */
3082: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3083: /* } */
3084: }
3085: k3=0.0; /* We put the last diagonal to 0 */
3086: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3087: doldm[ii][ii]= k3;
3088: }
3089: /* End doldm, At the end doldm is diag[(w_i)] */
3090:
1.292 brouard 3091: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3092: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3093:
1.292 brouard 3094: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3095: /* 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 3096: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3097: sumnew=0.;
1.222 brouard 3098: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3099: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3100: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3101: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3102: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3103: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3104: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3105: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3106: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3107: /* }else */
1.268 brouard 3108: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3109: } /*End ii */
3110: } /* 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 */
3111:
1.292 brouard 3112: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3113: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3114: /* end bmij */
1.266 brouard 3115: return ps; /*pointer is unchanged */
1.218 brouard 3116: }
1.217 brouard 3117: /*************** transition probabilities ***************/
3118:
1.218 brouard 3119: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3120: {
3121: /* According to parameters values stored in x and the covariate's values stored in cov,
3122: computes the probability to be observed in state j being in state i by appying the
3123: model to the ncovmodel covariates (including constant and age).
3124: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3125: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3126: ncth covariate in the global vector x is given by the formula:
3127: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3128: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3129: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3130: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3131: Outputs ps[i][j] the probability to be observed in j being in j according to
3132: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3133: */
3134: double s1, lnpijopii;
3135: /*double t34;*/
3136: int i,j, nc, ii, jj;
3137:
1.234 brouard 3138: for(i=1; i<= nlstate; i++){
3139: for(j=1; j<i;j++){
3140: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3141: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3142: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3143: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3144: }
3145: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3146: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3147: }
3148: for(j=i+1; j<=nlstate+ndeath;j++){
3149: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3150: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3151: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3152: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3153: }
3154: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3155: }
3156: }
3157:
3158: for(i=1; i<= nlstate; i++){
3159: s1=0;
3160: for(j=1; j<i; j++){
3161: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3162: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3163: }
3164: for(j=i+1; j<=nlstate+ndeath; j++){
3165: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3166: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3167: }
3168: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3169: ps[i][i]=1./(s1+1.);
3170: /* Computing other pijs */
3171: for(j=1; j<i; j++)
3172: ps[i][j]= exp(ps[i][j])*ps[i][i];
3173: for(j=i+1; j<=nlstate+ndeath; j++)
3174: ps[i][j]= exp(ps[i][j])*ps[i][i];
3175: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3176: } /* end i */
3177:
3178: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3179: for(jj=1; jj<= nlstate+ndeath; jj++){
3180: ps[ii][jj]=0;
3181: ps[ii][ii]=1;
3182: }
3183: }
1.296 brouard 3184: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3185: for(jj=1; jj<= nlstate+ndeath; jj++){
3186: s1=0.;
3187: for(ii=1; ii<= nlstate+ndeath; ii++){
3188: s1+=ps[ii][jj];
3189: }
3190: for(ii=1; ii<= nlstate; ii++){
3191: ps[ii][jj]=ps[ii][jj]/s1;
3192: }
3193: }
3194: /* Transposition */
3195: for(jj=1; jj<= nlstate+ndeath; jj++){
3196: for(ii=jj; ii<= nlstate+ndeath; ii++){
3197: s1=ps[ii][jj];
3198: ps[ii][jj]=ps[jj][ii];
3199: ps[jj][ii]=s1;
3200: }
3201: }
3202: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3203: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3204: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3205: /* } */
3206: /* printf("\n "); */
3207: /* } */
3208: /* printf("\n ");printf("%lf ",cov[2]);*/
3209: /*
3210: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3211: goto end;*/
3212: return ps;
1.217 brouard 3213: }
3214:
3215:
1.126 brouard 3216: /**************** Product of 2 matrices ******************/
3217:
1.145 brouard 3218: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3219: {
3220: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3221: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3222: /* in, b, out are matrice of pointers which should have been initialized
3223: before: only the contents of out is modified. The function returns
3224: a pointer to pointers identical to out */
1.145 brouard 3225: int i, j, k;
1.126 brouard 3226: for(i=nrl; i<= nrh; i++)
1.145 brouard 3227: for(k=ncolol; k<=ncoloh; k++){
3228: out[i][k]=0.;
3229: for(j=ncl; j<=nch; j++)
3230: out[i][k] +=in[i][j]*b[j][k];
3231: }
1.126 brouard 3232: return out;
3233: }
3234:
3235:
3236: /************* Higher Matrix Product ***************/
3237:
1.235 brouard 3238: 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 3239: {
1.218 brouard 3240: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3241: 'nhstepm*hstepm*stepm' months (i.e. until
3242: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3243: nhstepm*hstepm matrices.
3244: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3245: (typically every 2 years instead of every month which is too big
3246: for the memory).
3247: Model is determined by parameters x and covariates have to be
3248: included manually here.
3249:
3250: */
3251:
3252: int i, j, d, h, k;
1.131 brouard 3253: double **out, cov[NCOVMAX+1];
1.126 brouard 3254: double **newm;
1.187 brouard 3255: double agexact;
1.214 brouard 3256: double agebegin, ageend;
1.126 brouard 3257:
3258: /* Hstepm could be zero and should return the unit matrix */
3259: for (i=1;i<=nlstate+ndeath;i++)
3260: for (j=1;j<=nlstate+ndeath;j++){
3261: oldm[i][j]=(i==j ? 1.0 : 0.0);
3262: po[i][j][0]=(i==j ? 1.0 : 0.0);
3263: }
3264: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3265: for(h=1; h <=nhstepm; h++){
3266: for(d=1; d <=hstepm; d++){
3267: newm=savm;
3268: /* Covariates have to be included here again */
3269: cov[1]=1.;
1.214 brouard 3270: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3271: cov[2]=agexact;
3272: if(nagesqr==1)
1.227 brouard 3273: cov[3]= agexact*agexact;
1.235 brouard 3274: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3275: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3276: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3277: /* 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)); */
3278: }
3279: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3280: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3281: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3282: /* 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]); */
3283: }
3284: for (k=1; k<=cptcovage;k++){
3285: if(Dummy[Tvar[Tage[k]]]){
3286: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3287: } else{
3288: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3289: }
3290: /* 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]); */
3291: }
3292: for (k=1; k<=cptcovprod;k++){ /* */
3293: /* 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]); */
3294: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3295: }
3296: /* for (k=1; k<=cptcovn;k++) */
3297: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3298: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3299: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3300: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3301: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3302:
3303:
1.126 brouard 3304: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3305: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3306: /* right multiplication of oldm by the current matrix */
1.126 brouard 3307: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3308: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3309: /* if((int)age == 70){ */
3310: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3311: /* for(i=1; i<=nlstate+ndeath; i++) { */
3312: /* printf("%d pmmij ",i); */
3313: /* for(j=1;j<=nlstate+ndeath;j++) { */
3314: /* printf("%f ",pmmij[i][j]); */
3315: /* } */
3316: /* printf(" oldm "); */
3317: /* for(j=1;j<=nlstate+ndeath;j++) { */
3318: /* printf("%f ",oldm[i][j]); */
3319: /* } */
3320: /* printf("\n"); */
3321: /* } */
3322: /* } */
1.126 brouard 3323: savm=oldm;
3324: oldm=newm;
3325: }
3326: for(i=1; i<=nlstate+ndeath; i++)
3327: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3328: po[i][j][h]=newm[i][j];
3329: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3330: }
1.128 brouard 3331: /*printf("h=%d ",h);*/
1.126 brouard 3332: } /* end h */
1.267 brouard 3333: /* printf("\n H=%d \n",h); */
1.126 brouard 3334: return po;
3335: }
3336:
1.217 brouard 3337: /************* Higher Back Matrix Product ***************/
1.218 brouard 3338: /* 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 3339: 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 3340: {
1.266 brouard 3341: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3342: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3343: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3344: nhstepm*hstepm matrices.
3345: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3346: (typically every 2 years instead of every month which is too big
1.217 brouard 3347: for the memory).
1.218 brouard 3348: Model is determined by parameters x and covariates have to be
1.266 brouard 3349: included manually here. Then we use a call to bmij(x and cov)
3350: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3351: */
1.217 brouard 3352:
3353: int i, j, d, h, k;
1.266 brouard 3354: double **out, cov[NCOVMAX+1], **bmij();
3355: double **newm, ***newmm;
1.217 brouard 3356: double agexact;
3357: double agebegin, ageend;
1.222 brouard 3358: double **oldm, **savm;
1.217 brouard 3359:
1.266 brouard 3360: newmm=po; /* To be saved */
3361: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3362: /* Hstepm could be zero and should return the unit matrix */
3363: for (i=1;i<=nlstate+ndeath;i++)
3364: for (j=1;j<=nlstate+ndeath;j++){
3365: oldm[i][j]=(i==j ? 1.0 : 0.0);
3366: po[i][j][0]=(i==j ? 1.0 : 0.0);
3367: }
3368: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3369: for(h=1; h <=nhstepm; h++){
3370: for(d=1; d <=hstepm; d++){
3371: newm=savm;
3372: /* Covariates have to be included here again */
3373: cov[1]=1.;
1.271 brouard 3374: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3375: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3376: cov[2]=agexact;
3377: if(nagesqr==1)
1.222 brouard 3378: cov[3]= agexact*agexact;
1.266 brouard 3379: for (k=1; k<=cptcovn;k++){
3380: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3381: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3382: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3383: /* 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)); */
3384: }
1.267 brouard 3385: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3386: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3387: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3388: /* 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]); */
3389: }
3390: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3391: if(Dummy[Tvar[Tage[k]]]){
3392: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3393: } else{
3394: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3395: }
3396: /* 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]); */
3397: }
3398: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3399: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3400: }
1.217 brouard 3401: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3402: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3403:
1.218 brouard 3404: /* Careful transposed matrix */
1.266 brouard 3405: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3406: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3407: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3408: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3409: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3410: /* if((int)age == 70){ */
3411: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3412: /* for(i=1; i<=nlstate+ndeath; i++) { */
3413: /* printf("%d pmmij ",i); */
3414: /* for(j=1;j<=nlstate+ndeath;j++) { */
3415: /* printf("%f ",pmmij[i][j]); */
3416: /* } */
3417: /* printf(" oldm "); */
3418: /* for(j=1;j<=nlstate+ndeath;j++) { */
3419: /* printf("%f ",oldm[i][j]); */
3420: /* } */
3421: /* printf("\n"); */
3422: /* } */
3423: /* } */
3424: savm=oldm;
3425: oldm=newm;
3426: }
3427: for(i=1; i<=nlstate+ndeath; i++)
3428: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3429: po[i][j][h]=newm[i][j];
1.268 brouard 3430: /* if(h==nhstepm) */
3431: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3432: }
1.268 brouard 3433: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3434: } /* end h */
1.268 brouard 3435: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3436: return po;
3437: }
3438:
3439:
1.162 brouard 3440: #ifdef NLOPT
3441: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3442: double fret;
3443: double *xt;
3444: int j;
3445: myfunc_data *d2 = (myfunc_data *) pd;
3446: /* xt = (p1-1); */
3447: xt=vector(1,n);
3448: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3449:
3450: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3451: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3452: printf("Function = %.12lf ",fret);
3453: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3454: printf("\n");
3455: free_vector(xt,1,n);
3456: return fret;
3457: }
3458: #endif
1.126 brouard 3459:
3460: /*************** log-likelihood *************/
3461: double func( double *x)
3462: {
1.226 brouard 3463: int i, ii, j, k, mi, d, kk;
3464: int ioffset=0;
3465: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3466: double **out;
3467: double lli; /* Individual log likelihood */
3468: int s1, s2;
1.228 brouard 3469: 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 3470: double bbh, survp;
3471: long ipmx;
3472: double agexact;
3473: /*extern weight */
3474: /* We are differentiating ll according to initial status */
3475: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3476: /*for(i=1;i<imx;i++)
3477: printf(" %d\n",s[4][i]);
3478: */
1.162 brouard 3479:
1.226 brouard 3480: ++countcallfunc;
1.162 brouard 3481:
1.226 brouard 3482: cov[1]=1.;
1.126 brouard 3483:
1.226 brouard 3484: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3485: ioffset=0;
1.226 brouard 3486: if(mle==1){
3487: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3488: /* Computes the values of the ncovmodel covariates of the model
3489: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3490: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3491: to be observed in j being in i according to the model.
3492: */
1.243 brouard 3493: ioffset=2+nagesqr ;
1.233 brouard 3494: /* Fixed */
1.234 brouard 3495: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3496: 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)*/
3497: }
1.226 brouard 3498: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3499: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3500: has been calculated etc */
3501: /* For an individual i, wav[i] gives the number of effective waves */
3502: /* We compute the contribution to Likelihood of each effective transition
3503: mw[mi][i] is real wave of the mi th effectve wave */
3504: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3505: s2=s[mw[mi+1][i]][i];
3506: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3507: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3508: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3509: */
3510: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3511: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3512: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3513: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3514: }
3515: for (ii=1;ii<=nlstate+ndeath;ii++)
3516: for (j=1;j<=nlstate+ndeath;j++){
3517: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3518: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3519: }
3520: for(d=0; d<dh[mi][i]; d++){
3521: newm=savm;
3522: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3523: cov[2]=agexact;
3524: if(nagesqr==1)
3525: cov[3]= agexact*agexact; /* Should be changed here */
3526: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3527: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3528: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3529: else
3530: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3531: }
3532: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3533: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3534: savm=oldm;
3535: oldm=newm;
3536: } /* end mult */
3537:
3538: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3539: /* But now since version 0.9 we anticipate for bias at large stepm.
3540: * If stepm is larger than one month (smallest stepm) and if the exact delay
3541: * (in months) between two waves is not a multiple of stepm, we rounded to
3542: * the nearest (and in case of equal distance, to the lowest) interval but now
3543: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3544: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3545: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3546: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3547: * -stepm/2 to stepm/2 .
3548: * For stepm=1 the results are the same as for previous versions of Imach.
3549: * For stepm > 1 the results are less biased than in previous versions.
3550: */
1.234 brouard 3551: s1=s[mw[mi][i]][i];
3552: s2=s[mw[mi+1][i]][i];
3553: bbh=(double)bh[mi][i]/(double)stepm;
3554: /* bias bh is positive if real duration
3555: * is higher than the multiple of stepm and negative otherwise.
3556: */
3557: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3558: if( s2 > nlstate){
3559: /* i.e. if s2 is a death state and if the date of death is known
3560: then the contribution to the likelihood is the probability to
3561: die between last step unit time and current step unit time,
3562: which is also equal to probability to die before dh
3563: minus probability to die before dh-stepm .
3564: In version up to 0.92 likelihood was computed
3565: as if date of death was unknown. Death was treated as any other
3566: health state: the date of the interview describes the actual state
3567: and not the date of a change in health state. The former idea was
3568: to consider that at each interview the state was recorded
3569: (healthy, disable or death) and IMaCh was corrected; but when we
3570: introduced the exact date of death then we should have modified
3571: the contribution of an exact death to the likelihood. This new
3572: contribution is smaller and very dependent of the step unit
3573: stepm. It is no more the probability to die between last interview
3574: and month of death but the probability to survive from last
3575: interview up to one month before death multiplied by the
3576: probability to die within a month. Thanks to Chris
3577: Jackson for correcting this bug. Former versions increased
3578: mortality artificially. The bad side is that we add another loop
3579: which slows down the processing. The difference can be up to 10%
3580: lower mortality.
3581: */
3582: /* If, at the beginning of the maximization mostly, the
3583: cumulative probability or probability to be dead is
3584: constant (ie = 1) over time d, the difference is equal to
3585: 0. out[s1][3] = savm[s1][3]: probability, being at state
3586: s1 at precedent wave, to be dead a month before current
3587: wave is equal to probability, being at state s1 at
3588: precedent wave, to be dead at mont of the current
3589: wave. Then the observed probability (that this person died)
3590: is null according to current estimated parameter. In fact,
3591: it should be very low but not zero otherwise the log go to
3592: infinity.
3593: */
1.183 brouard 3594: /* #ifdef INFINITYORIGINAL */
3595: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3596: /* #else */
3597: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3598: /* lli=log(mytinydouble); */
3599: /* else */
3600: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3601: /* #endif */
1.226 brouard 3602: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3603:
1.226 brouard 3604: } else if ( s2==-1 ) { /* alive */
3605: for (j=1,survp=0. ; j<=nlstate; j++)
3606: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3607: /*survp += out[s1][j]; */
3608: lli= log(survp);
3609: }
3610: else if (s2==-4) {
3611: for (j=3,survp=0. ; j<=nlstate; j++)
3612: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3613: lli= log(survp);
3614: }
3615: else if (s2==-5) {
3616: for (j=1,survp=0. ; j<=2; j++)
3617: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3618: lli= log(survp);
3619: }
3620: else{
3621: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3622: /* 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 */
3623: }
3624: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3625: /*if(lli ==000.0)*/
3626: /*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); */
3627: ipmx +=1;
3628: sw += weight[i];
3629: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3630: /* if (lli < log(mytinydouble)){ */
3631: /* 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); */
3632: /* 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]); */
3633: /* } */
3634: } /* end of wave */
3635: } /* end of individual */
3636: } else if(mle==2){
3637: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3638: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3639: for(mi=1; mi<= wav[i]-1; mi++){
3640: for (ii=1;ii<=nlstate+ndeath;ii++)
3641: for (j=1;j<=nlstate+ndeath;j++){
3642: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3643: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3644: }
3645: for(d=0; d<=dh[mi][i]; d++){
3646: newm=savm;
3647: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3648: cov[2]=agexact;
3649: if(nagesqr==1)
3650: cov[3]= agexact*agexact;
3651: for (kk=1; kk<=cptcovage;kk++) {
3652: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3653: }
3654: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3655: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3656: savm=oldm;
3657: oldm=newm;
3658: } /* end mult */
3659:
3660: s1=s[mw[mi][i]][i];
3661: s2=s[mw[mi+1][i]][i];
3662: bbh=(double)bh[mi][i]/(double)stepm;
3663: 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 */
3664: ipmx +=1;
3665: sw += weight[i];
3666: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3667: } /* end of wave */
3668: } /* end of individual */
3669: } else if(mle==3){ /* exponential inter-extrapolation */
3670: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3671: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3672: for(mi=1; mi<= wav[i]-1; mi++){
3673: for (ii=1;ii<=nlstate+ndeath;ii++)
3674: for (j=1;j<=nlstate+ndeath;j++){
3675: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3676: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3677: }
3678: for(d=0; d<dh[mi][i]; d++){
3679: newm=savm;
3680: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3681: cov[2]=agexact;
3682: if(nagesqr==1)
3683: cov[3]= agexact*agexact;
3684: for (kk=1; kk<=cptcovage;kk++) {
3685: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3686: }
3687: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3688: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3689: savm=oldm;
3690: oldm=newm;
3691: } /* end mult */
3692:
3693: s1=s[mw[mi][i]][i];
3694: s2=s[mw[mi+1][i]][i];
3695: bbh=(double)bh[mi][i]/(double)stepm;
3696: 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 */
3697: ipmx +=1;
3698: sw += weight[i];
3699: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3700: } /* end of wave */
3701: } /* end of individual */
3702: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3703: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3704: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3705: for(mi=1; mi<= wav[i]-1; mi++){
3706: for (ii=1;ii<=nlstate+ndeath;ii++)
3707: for (j=1;j<=nlstate+ndeath;j++){
3708: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3709: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3710: }
3711: for(d=0; d<dh[mi][i]; d++){
3712: newm=savm;
3713: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3714: cov[2]=agexact;
3715: if(nagesqr==1)
3716: cov[3]= agexact*agexact;
3717: for (kk=1; kk<=cptcovage;kk++) {
3718: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3719: }
1.126 brouard 3720:
1.226 brouard 3721: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3722: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3723: savm=oldm;
3724: oldm=newm;
3725: } /* end mult */
3726:
3727: s1=s[mw[mi][i]][i];
3728: s2=s[mw[mi+1][i]][i];
3729: if( s2 > nlstate){
3730: lli=log(out[s1][s2] - savm[s1][s2]);
3731: } else if ( s2==-1 ) { /* alive */
3732: for (j=1,survp=0. ; j<=nlstate; j++)
3733: survp += out[s1][j];
3734: lli= log(survp);
3735: }else{
3736: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3737: }
3738: ipmx +=1;
3739: sw += weight[i];
3740: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3741: /* 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 3742: } /* end of wave */
3743: } /* end of individual */
3744: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3745: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3746: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3747: for(mi=1; mi<= wav[i]-1; mi++){
3748: for (ii=1;ii<=nlstate+ndeath;ii++)
3749: for (j=1;j<=nlstate+ndeath;j++){
3750: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3751: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3752: }
3753: for(d=0; d<dh[mi][i]; d++){
3754: newm=savm;
3755: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3756: cov[2]=agexact;
3757: if(nagesqr==1)
3758: cov[3]= agexact*agexact;
3759: for (kk=1; kk<=cptcovage;kk++) {
3760: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3761: }
1.126 brouard 3762:
1.226 brouard 3763: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3764: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3765: savm=oldm;
3766: oldm=newm;
3767: } /* end mult */
3768:
3769: s1=s[mw[mi][i]][i];
3770: s2=s[mw[mi+1][i]][i];
3771: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3772: ipmx +=1;
3773: sw += weight[i];
3774: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3775: /*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]);*/
3776: } /* end of wave */
3777: } /* end of individual */
3778: } /* End of if */
3779: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3780: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3781: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3782: return -l;
1.126 brouard 3783: }
3784:
3785: /*************** log-likelihood *************/
3786: double funcone( double *x)
3787: {
1.228 brouard 3788: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3789: int i, ii, j, k, mi, d, kk;
1.228 brouard 3790: int ioffset=0;
1.131 brouard 3791: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3792: double **out;
3793: double lli; /* Individual log likelihood */
3794: double llt;
3795: int s1, s2;
1.228 brouard 3796: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3797:
1.126 brouard 3798: double bbh, survp;
1.187 brouard 3799: double agexact;
1.214 brouard 3800: double agebegin, ageend;
1.126 brouard 3801: /*extern weight */
3802: /* We are differentiating ll according to initial status */
3803: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3804: /*for(i=1;i<imx;i++)
3805: printf(" %d\n",s[4][i]);
3806: */
3807: cov[1]=1.;
3808:
3809: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3810: ioffset=0;
3811: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3812: /* ioffset=2+nagesqr+cptcovage; */
3813: ioffset=2+nagesqr;
1.232 brouard 3814: /* Fixed */
1.224 brouard 3815: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3816: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3817: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3818: 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)*/
3819: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3820: /* cov[2+6]=covar[Tvar[6]][i]; */
3821: /* cov[2+6]=covar[2][i]; V2 */
3822: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3823: /* cov[2+7]=covar[Tvar[7]][i]; */
3824: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3825: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3826: /* cov[2+9]=covar[Tvar[9]][i]; */
3827: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3828: }
1.232 brouard 3829: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3830: /* 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?)*\/ */
3831: /* } */
1.231 brouard 3832: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3833: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3834: /* } */
1.225 brouard 3835:
1.233 brouard 3836:
3837: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3838: /* Wave varying (but not age varying) */
3839: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3840: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3841: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3842: }
1.232 brouard 3843: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3844: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3845: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3846: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3847: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3848: /* 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 3849: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3850: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3851: /* /\* 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]); *\/ */
3852: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3853: /* } */
1.126 brouard 3854: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3855: for (j=1;j<=nlstate+ndeath;j++){
3856: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3857: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3858: }
1.214 brouard 3859:
3860: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3861: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3862: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3863: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3864: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3865: and mw[mi+1][i]. dh depends on stepm.*/
3866: newm=savm;
1.247 brouard 3867: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3868: cov[2]=agexact;
3869: if(nagesqr==1)
3870: cov[3]= agexact*agexact;
3871: for (kk=1; kk<=cptcovage;kk++) {
3872: if(!FixedV[Tvar[Tage[kk]]])
3873: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3874: else
3875: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3876: }
3877: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3878: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3879: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3880: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3881: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3882: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3883: savm=oldm;
3884: oldm=newm;
1.126 brouard 3885: } /* end mult */
3886:
3887: s1=s[mw[mi][i]][i];
3888: s2=s[mw[mi+1][i]][i];
1.217 brouard 3889: /* if(s2==-1){ */
1.268 brouard 3890: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3891: /* /\* exit(1); *\/ */
3892: /* } */
1.126 brouard 3893: bbh=(double)bh[mi][i]/(double)stepm;
3894: /* bias is positive if real duration
3895: * is higher than the multiple of stepm and negative otherwise.
3896: */
3897: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3898: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3899: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3900: for (j=1,survp=0. ; j<=nlstate; j++)
3901: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3902: lli= log(survp);
1.126 brouard 3903: }else if (mle==1){
1.242 brouard 3904: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3905: } else if(mle==2){
1.242 brouard 3906: 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 3907: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3908: 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 3909: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3910: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3911: } else{ /* mle=0 back to 1 */
1.242 brouard 3912: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3913: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3914: } /* End of if */
3915: ipmx +=1;
3916: sw += weight[i];
3917: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3918: /*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 3919: if(globpr){
1.246 brouard 3920: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3921: %11.6f %11.6f %11.6f ", \
1.242 brouard 3922: 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 3923: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3924: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3925: llt +=ll[k]*gipmx/gsw;
3926: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3927: }
3928: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3929: }
1.232 brouard 3930: } /* end of wave */
3931: } /* end of individual */
3932: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3933: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3934: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3935: if(globpr==0){ /* First time we count the contributions and weights */
3936: gipmx=ipmx;
3937: gsw=sw;
3938: }
3939: return -l;
1.126 brouard 3940: }
3941:
3942:
3943: /*************** function likelione ***********/
1.292 brouard 3944: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3945: {
3946: /* This routine should help understanding what is done with
3947: the selection of individuals/waves and
3948: to check the exact contribution to the likelihood.
3949: Plotting could be done.
3950: */
3951: int k;
3952:
3953: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3954: strcpy(fileresilk,"ILK_");
1.202 brouard 3955: strcat(fileresilk,fileresu);
1.126 brouard 3956: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3957: printf("Problem with resultfile: %s\n", fileresilk);
3958: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3959: }
1.214 brouard 3960: 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");
3961: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3962: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3963: for(k=1; k<=nlstate; k++)
3964: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3965: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3966: }
3967:
1.292 brouard 3968: *fretone=(*func)(p);
1.126 brouard 3969: if(*globpri !=0){
3970: fclose(ficresilk);
1.205 brouard 3971: if (mle ==0)
3972: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3973: else if(mle >=1)
3974: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3975: 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 3976: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 3977:
3978: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3979: 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 3980: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3981: }
1.207 brouard 3982: 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 3983: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3984: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3985: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3986: fflush(fichtm);
1.205 brouard 3987: }
1.126 brouard 3988: return;
3989: }
3990:
3991:
3992: /*********** Maximum Likelihood Estimation ***************/
3993:
3994: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3995: {
1.165 brouard 3996: int i,j, iter=0;
1.126 brouard 3997: double **xi;
3998: double fret;
3999: double fretone; /* Only one call to likelihood */
4000: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4001:
4002: #ifdef NLOPT
4003: int creturn;
4004: nlopt_opt opt;
4005: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4006: double *lb;
4007: double minf; /* the minimum objective value, upon return */
4008: double * p1; /* Shifted parameters from 0 instead of 1 */
4009: myfunc_data dinst, *d = &dinst;
4010: #endif
4011:
4012:
1.126 brouard 4013: xi=matrix(1,npar,1,npar);
4014: for (i=1;i<=npar;i++)
4015: for (j=1;j<=npar;j++)
4016: xi[i][j]=(i==j ? 1.0 : 0.0);
4017: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4018: strcpy(filerespow,"POW_");
1.126 brouard 4019: strcat(filerespow,fileres);
4020: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4021: printf("Problem with resultfile: %s\n", filerespow);
4022: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4023: }
4024: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4025: for (i=1;i<=nlstate;i++)
4026: for(j=1;j<=nlstate+ndeath;j++)
4027: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4028: fprintf(ficrespow,"\n");
1.162 brouard 4029: #ifdef POWELL
1.126 brouard 4030: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4031: #endif
1.126 brouard 4032:
1.162 brouard 4033: #ifdef NLOPT
4034: #ifdef NEWUOA
4035: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4036: #else
4037: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4038: #endif
4039: lb=vector(0,npar-1);
4040: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4041: nlopt_set_lower_bounds(opt, lb);
4042: nlopt_set_initial_step1(opt, 0.1);
4043:
4044: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4045: d->function = func;
4046: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4047: nlopt_set_min_objective(opt, myfunc, d);
4048: nlopt_set_xtol_rel(opt, ftol);
4049: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4050: printf("nlopt failed! %d\n",creturn);
4051: }
4052: else {
4053: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4054: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4055: iter=1; /* not equal */
4056: }
4057: nlopt_destroy(opt);
4058: #endif
1.126 brouard 4059: free_matrix(xi,1,npar,1,npar);
4060: fclose(ficrespow);
1.203 brouard 4061: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4062: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4063: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4064:
4065: }
4066:
4067: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4068: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4069: {
4070: double **a,**y,*x,pd;
1.203 brouard 4071: /* double **hess; */
1.164 brouard 4072: int i, j;
1.126 brouard 4073: int *indx;
4074:
4075: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4076: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4077: void lubksb(double **a, int npar, int *indx, double b[]) ;
4078: void ludcmp(double **a, int npar, int *indx, double *d) ;
4079: double gompertz(double p[]);
1.203 brouard 4080: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4081:
4082: printf("\nCalculation of the hessian matrix. Wait...\n");
4083: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4084: for (i=1;i<=npar;i++){
1.203 brouard 4085: printf("%d-",i);fflush(stdout);
4086: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4087:
4088: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4089:
4090: /* printf(" %f ",p[i]);
4091: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4092: }
4093:
4094: for (i=1;i<=npar;i++) {
4095: for (j=1;j<=npar;j++) {
4096: if (j>i) {
1.203 brouard 4097: printf(".%d-%d",i,j);fflush(stdout);
4098: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4099: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4100:
4101: hess[j][i]=hess[i][j];
4102: /*printf(" %lf ",hess[i][j]);*/
4103: }
4104: }
4105: }
4106: printf("\n");
4107: fprintf(ficlog,"\n");
4108:
4109: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4110: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4111:
4112: a=matrix(1,npar,1,npar);
4113: y=matrix(1,npar,1,npar);
4114: x=vector(1,npar);
4115: indx=ivector(1,npar);
4116: for (i=1;i<=npar;i++)
4117: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4118: ludcmp(a,npar,indx,&pd);
4119:
4120: for (j=1;j<=npar;j++) {
4121: for (i=1;i<=npar;i++) x[i]=0;
4122: x[j]=1;
4123: lubksb(a,npar,indx,x);
4124: for (i=1;i<=npar;i++){
4125: matcov[i][j]=x[i];
4126: }
4127: }
4128:
4129: printf("\n#Hessian matrix#\n");
4130: fprintf(ficlog,"\n#Hessian matrix#\n");
4131: for (i=1;i<=npar;i++) {
4132: for (j=1;j<=npar;j++) {
1.203 brouard 4133: printf("%.6e ",hess[i][j]);
4134: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4135: }
4136: printf("\n");
4137: fprintf(ficlog,"\n");
4138: }
4139:
1.203 brouard 4140: /* printf("\n#Covariance matrix#\n"); */
4141: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4142: /* for (i=1;i<=npar;i++) { */
4143: /* for (j=1;j<=npar;j++) { */
4144: /* printf("%.6e ",matcov[i][j]); */
4145: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4146: /* } */
4147: /* printf("\n"); */
4148: /* fprintf(ficlog,"\n"); */
4149: /* } */
4150:
1.126 brouard 4151: /* Recompute Inverse */
1.203 brouard 4152: /* for (i=1;i<=npar;i++) */
4153: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4154: /* ludcmp(a,npar,indx,&pd); */
4155:
4156: /* printf("\n#Hessian matrix recomputed#\n"); */
4157:
4158: /* for (j=1;j<=npar;j++) { */
4159: /* for (i=1;i<=npar;i++) x[i]=0; */
4160: /* x[j]=1; */
4161: /* lubksb(a,npar,indx,x); */
4162: /* for (i=1;i<=npar;i++){ */
4163: /* y[i][j]=x[i]; */
4164: /* printf("%.3e ",y[i][j]); */
4165: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4166: /* } */
4167: /* printf("\n"); */
4168: /* fprintf(ficlog,"\n"); */
4169: /* } */
4170:
4171: /* Verifying the inverse matrix */
4172: #ifdef DEBUGHESS
4173: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4174:
1.203 brouard 4175: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4176: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4177:
4178: for (j=1;j<=npar;j++) {
4179: for (i=1;i<=npar;i++){
1.203 brouard 4180: printf("%.2f ",y[i][j]);
4181: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4182: }
4183: printf("\n");
4184: fprintf(ficlog,"\n");
4185: }
1.203 brouard 4186: #endif
1.126 brouard 4187:
4188: free_matrix(a,1,npar,1,npar);
4189: free_matrix(y,1,npar,1,npar);
4190: free_vector(x,1,npar);
4191: free_ivector(indx,1,npar);
1.203 brouard 4192: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4193:
4194:
4195: }
4196:
4197: /*************** hessian matrix ****************/
4198: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4199: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4200: int i;
4201: int l=1, lmax=20;
1.203 brouard 4202: double k1,k2, res, fx;
1.132 brouard 4203: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4204: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4205: int k=0,kmax=10;
4206: double l1;
4207:
4208: fx=func(x);
4209: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4210: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4211: l1=pow(10,l);
4212: delts=delt;
4213: for(k=1 ; k <kmax; k=k+1){
4214: delt = delta*(l1*k);
4215: p2[theta]=x[theta] +delt;
1.145 brouard 4216: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4217: p2[theta]=x[theta]-delt;
4218: k2=func(p2)-fx;
4219: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4220: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4221:
1.203 brouard 4222: #ifdef DEBUGHESSII
1.126 brouard 4223: 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);
4224: 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);
4225: #endif
4226: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4227: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4228: k=kmax;
4229: }
4230: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4231: k=kmax; l=lmax*10;
1.126 brouard 4232: }
4233: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4234: delts=delt;
4235: }
1.203 brouard 4236: } /* End loop k */
1.126 brouard 4237: }
4238: delti[theta]=delts;
4239: return res;
4240:
4241: }
4242:
1.203 brouard 4243: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4244: {
4245: int i;
1.164 brouard 4246: int l=1, lmax=20;
1.126 brouard 4247: double k1,k2,k3,k4,res,fx;
1.132 brouard 4248: double p2[MAXPARM+1];
1.203 brouard 4249: int k, kmax=1;
4250: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4251:
4252: int firstime=0;
1.203 brouard 4253:
1.126 brouard 4254: fx=func(x);
1.203 brouard 4255: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4256: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4257: p2[thetai]=x[thetai]+delti[thetai]*k;
4258: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4259: k1=func(p2)-fx;
4260:
1.203 brouard 4261: p2[thetai]=x[thetai]+delti[thetai]*k;
4262: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4263: k2=func(p2)-fx;
4264:
1.203 brouard 4265: p2[thetai]=x[thetai]-delti[thetai]*k;
4266: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4267: k3=func(p2)-fx;
4268:
1.203 brouard 4269: p2[thetai]=x[thetai]-delti[thetai]*k;
4270: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4271: k4=func(p2)-fx;
1.203 brouard 4272: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4273: if(k1*k2*k3*k4 <0.){
1.208 brouard 4274: firstime=1;
1.203 brouard 4275: kmax=kmax+10;
1.208 brouard 4276: }
4277: if(kmax >=10 || firstime ==1){
1.246 brouard 4278: 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);
4279: 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 4280: 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);
4281: 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);
4282: }
4283: #ifdef DEBUGHESSIJ
4284: v1=hess[thetai][thetai];
4285: v2=hess[thetaj][thetaj];
4286: cv12=res;
4287: /* Computing eigen value of Hessian matrix */
4288: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4289: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4290: if ((lc2 <0) || (lc1 <0) ){
4291: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4292: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4293: 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);
4294: 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);
4295: }
1.126 brouard 4296: #endif
4297: }
4298: return res;
4299: }
4300:
1.203 brouard 4301: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4302: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4303: /* { */
4304: /* int i; */
4305: /* int l=1, lmax=20; */
4306: /* double k1,k2,k3,k4,res,fx; */
4307: /* double p2[MAXPARM+1]; */
4308: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4309: /* int k=0,kmax=10; */
4310: /* double l1; */
4311:
4312: /* fx=func(x); */
4313: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4314: /* l1=pow(10,l); */
4315: /* delts=delt; */
4316: /* for(k=1 ; k <kmax; k=k+1){ */
4317: /* delt = delti*(l1*k); */
4318: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4319: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4320: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4321: /* k1=func(p2)-fx; */
4322:
4323: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4324: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4325: /* k2=func(p2)-fx; */
4326:
4327: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4328: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4329: /* k3=func(p2)-fx; */
4330:
4331: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4332: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4333: /* k4=func(p2)-fx; */
4334: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4335: /* #ifdef DEBUGHESSIJ */
4336: /* 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); */
4337: /* 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); */
4338: /* #endif */
4339: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4340: /* k=kmax; */
4341: /* } */
4342: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4343: /* k=kmax; l=lmax*10; */
4344: /* } */
4345: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4346: /* delts=delt; */
4347: /* } */
4348: /* } /\* End loop k *\/ */
4349: /* } */
4350: /* delti[theta]=delts; */
4351: /* return res; */
4352: /* } */
4353:
4354:
1.126 brouard 4355: /************** Inverse of matrix **************/
4356: void ludcmp(double **a, int n, int *indx, double *d)
4357: {
4358: int i,imax,j,k;
4359: double big,dum,sum,temp;
4360: double *vv;
4361:
4362: vv=vector(1,n);
4363: *d=1.0;
4364: for (i=1;i<=n;i++) {
4365: big=0.0;
4366: for (j=1;j<=n;j++)
4367: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4368: if (big == 0.0){
4369: printf(" Singular Hessian matrix at row %d:\n",i);
4370: for (j=1;j<=n;j++) {
4371: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4372: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4373: }
4374: fflush(ficlog);
4375: fclose(ficlog);
4376: nrerror("Singular matrix in routine ludcmp");
4377: }
1.126 brouard 4378: vv[i]=1.0/big;
4379: }
4380: for (j=1;j<=n;j++) {
4381: for (i=1;i<j;i++) {
4382: sum=a[i][j];
4383: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4384: a[i][j]=sum;
4385: }
4386: big=0.0;
4387: for (i=j;i<=n;i++) {
4388: sum=a[i][j];
4389: for (k=1;k<j;k++)
4390: sum -= a[i][k]*a[k][j];
4391: a[i][j]=sum;
4392: if ( (dum=vv[i]*fabs(sum)) >= big) {
4393: big=dum;
4394: imax=i;
4395: }
4396: }
4397: if (j != imax) {
4398: for (k=1;k<=n;k++) {
4399: dum=a[imax][k];
4400: a[imax][k]=a[j][k];
4401: a[j][k]=dum;
4402: }
4403: *d = -(*d);
4404: vv[imax]=vv[j];
4405: }
4406: indx[j]=imax;
4407: if (a[j][j] == 0.0) a[j][j]=TINY;
4408: if (j != n) {
4409: dum=1.0/(a[j][j]);
4410: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4411: }
4412: }
4413: free_vector(vv,1,n); /* Doesn't work */
4414: ;
4415: }
4416:
4417: void lubksb(double **a, int n, int *indx, double b[])
4418: {
4419: int i,ii=0,ip,j;
4420: double sum;
4421:
4422: for (i=1;i<=n;i++) {
4423: ip=indx[i];
4424: sum=b[ip];
4425: b[ip]=b[i];
4426: if (ii)
4427: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4428: else if (sum) ii=i;
4429: b[i]=sum;
4430: }
4431: for (i=n;i>=1;i--) {
4432: sum=b[i];
4433: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4434: b[i]=sum/a[i][i];
4435: }
4436: }
4437:
4438: void pstamp(FILE *fichier)
4439: {
1.196 brouard 4440: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4441: }
4442:
1.297 brouard 4443: void date2dmy(double date,double *day, double *month, double *year){
4444: double yp=0., yp1=0., yp2=0.;
4445:
4446: yp1=modf(date,&yp);/* extracts integral of date in yp and
4447: fractional in yp1 */
4448: *year=yp;
4449: yp2=modf((yp1*12),&yp);
4450: *month=yp;
4451: yp1=modf((yp2*30.5),&yp);
4452: *day=yp;
4453: if(*day==0) *day=1;
4454: if(*month==0) *month=1;
4455: }
4456:
1.253 brouard 4457:
4458:
1.126 brouard 4459: /************ Frequencies ********************/
1.251 brouard 4460: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4461: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4462: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4463: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4464:
1.265 brouard 4465: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4466: int iind=0, iage=0;
4467: int mi; /* Effective wave */
4468: int first;
4469: double ***freq; /* Frequencies */
1.268 brouard 4470: 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 */
4471: 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 4472: double *meanq, *stdq, *idq;
1.226 brouard 4473: double **meanqt;
4474: double *pp, **prop, *posprop, *pospropt;
4475: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4476: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4477: double agebegin, ageend;
4478:
4479: pp=vector(1,nlstate);
1.251 brouard 4480: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4481: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4482: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4483: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4484: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4485: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4486: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4487: meanqt=matrix(1,lastpass,1,nqtveff);
4488: strcpy(fileresp,"P_");
4489: strcat(fileresp,fileresu);
4490: /*strcat(fileresphtm,fileresu);*/
4491: if((ficresp=fopen(fileresp,"w"))==NULL) {
4492: printf("Problem with prevalence resultfile: %s\n", fileresp);
4493: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4494: exit(0);
4495: }
1.240 brouard 4496:
1.226 brouard 4497: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4498: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4499: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4500: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4501: fflush(ficlog);
4502: exit(70);
4503: }
4504: else{
4505: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4506: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4507: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4508: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4509: }
1.237 brouard 4510: fprintf(ficresphtm,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Frequencies and prevalence by age at begin of transition and dummy covariate value at beginning of transition</h4>\n",fileresphtm, fileresphtm);
1.240 brouard 4511:
1.226 brouard 4512: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4513: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4514: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4515: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4516: fflush(ficlog);
4517: exit(70);
1.240 brouard 4518: } else{
1.226 brouard 4519: fprintf(ficresphtmfr,"<html><head>\n<title>IMaCh PHTM_Frequency table %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4520: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4521: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4522: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4523: }
1.240 brouard 4524: fprintf(ficresphtmfr,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>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);
4525:
1.253 brouard 4526: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4527: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4528: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4529: j1=0;
1.126 brouard 4530:
1.227 brouard 4531: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4532: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4533: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4534:
4535:
1.226 brouard 4536: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4537: reference=low_education V1=0,V2=0
4538: med_educ V1=1 V2=0,
4539: high_educ V1=0 V2=1
4540: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4541: */
1.249 brouard 4542: dateintsum=0;
4543: k2cpt=0;
4544:
1.253 brouard 4545: if(cptcoveff == 0 )
1.265 brouard 4546: nl=1; /* Constant and age model only */
1.253 brouard 4547: else
4548: nl=2;
1.265 brouard 4549:
4550: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4551: /* Loop on nj=1 or 2 if dummy covariates j!=0
4552: * Loop on j1(1 to 2**cptcoveff) covariate combination
4553: * freq[s1][s2][iage] =0.
4554: * Loop on iind
4555: * ++freq[s1][s2][iage] weighted
4556: * end iind
4557: * if covariate and j!0
4558: * headers Variable on one line
4559: * endif cov j!=0
4560: * header of frequency table by age
4561: * Loop on age
4562: * pp[s1]+=freq[s1][s2][iage] weighted
4563: * pos+=freq[s1][s2][iage] weighted
4564: * Loop on s1 initial state
4565: * fprintf(ficresp
4566: * end s1
4567: * end age
4568: * if j!=0 computes starting values
4569: * end compute starting values
4570: * end j1
4571: * end nl
4572: */
1.253 brouard 4573: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4574: if(nj==1)
4575: j=0; /* First pass for the constant */
1.265 brouard 4576: else{
1.253 brouard 4577: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4578: }
1.251 brouard 4579: first=1;
1.265 brouard 4580: 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 4581: posproptt=0.;
4582: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4583: scanf("%d", i);*/
4584: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4585: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4586: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4587: freq[i][s2][m]=0;
1.251 brouard 4588:
4589: for (i=1; i<=nlstate; i++) {
1.240 brouard 4590: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4591: prop[i][m]=0;
4592: posprop[i]=0;
4593: pospropt[i]=0;
4594: }
1.283 brouard 4595: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4596: idq[z1]=0.;
4597: meanq[z1]=0.;
4598: stdq[z1]=0.;
1.283 brouard 4599: }
4600: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4601: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4602: /* meanqt[m][z1]=0.; */
4603: /* } */
4604: /* } */
1.251 brouard 4605: /* dateintsum=0; */
4606: /* k2cpt=0; */
4607:
1.265 brouard 4608: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4609: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4610: bool=1;
4611: if(j !=0){
4612: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4613: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4614: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4615: /* if(Tvaraff[z1] ==-20){ */
4616: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4617: /* }else if(Tvaraff[z1] ==-10){ */
4618: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4619: /* }else */
4620: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4621: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4622: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4623: /* 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",
4624: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4625: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4626: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4627: } /* Onlyf fixed */
4628: } /* end z1 */
4629: } /* cptcovn > 0 */
4630: } /* end any */
4631: }/* end j==0 */
1.265 brouard 4632: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4633: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4634: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4635: m=mw[mi][iind];
4636: if(j!=0){
4637: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4638: for (z1=1; z1<=cptcoveff; z1++) {
4639: if( Fixed[Tmodelind[z1]]==1){
4640: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4641: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4642: value is -1, we don't select. It differs from the
4643: constant and age model which counts them. */
4644: bool=0; /* not selected */
4645: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4646: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4647: bool=0;
4648: }
4649: }
4650: }
4651: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4652: } /* end j==0 */
4653: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4654: if(bool==1){ /*Selected */
1.251 brouard 4655: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4656: and mw[mi+1][iind]. dh depends on stepm. */
4657: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4658: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4659: if(m >=firstpass && m <=lastpass){
4660: k2=anint[m][iind]+(mint[m][iind]/12.);
4661: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4662: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4663: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4664: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4665: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4666: if (m<lastpass) {
4667: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4668: /* 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]); */
4669: if(s[m][iind]==-1)
4670: 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.));
4671: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.284 brouard 4672: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean */
4673: idq[z1]=idq[z1]+weight[iind];
4674: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4675: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4676: }
1.251 brouard 4677: /* if((int)agev[m][iind] == 55) */
4678: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4679: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4680: 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 4681: }
1.251 brouard 4682: } /* end if between passes */
4683: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4684: dateintsum=dateintsum+k2; /* on all covariates ?*/
4685: k2cpt++;
4686: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4687: }
1.251 brouard 4688: }else{
4689: bool=1;
4690: }/* end bool 2 */
4691: } /* end m */
1.284 brouard 4692: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4693: /* idq[z1]=idq[z1]+weight[iind]; */
4694: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4695: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4696: /* } */
1.251 brouard 4697: } /* end bool */
4698: } /* end iind = 1 to imx */
4699: /* prop[s][age] is feeded for any initial and valid live state as well as
4700: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4701:
4702:
4703: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4704: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4705: pstamp(ficresp);
1.251 brouard 4706: if (cptcoveff>0 && j!=0){
1.265 brouard 4707: pstamp(ficresp);
1.251 brouard 4708: printf( "\n#********** Variable ");
4709: fprintf(ficresp, "\n#********** Variable ");
4710: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4711: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4712: fprintf(ficlog, "\n#********** Variable ");
4713: for (z1=1; z1<=cptcoveff; z1++){
4714: if(!FixedV[Tvaraff[z1]]){
4715: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4716: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4717: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4718: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4719: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4720: }else{
1.251 brouard 4721: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4722: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4723: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4724: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4725: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4726: }
4727: }
4728: printf( "**********\n#");
4729: fprintf(ficresp, "**********\n#");
4730: fprintf(ficresphtm, "**********</h3>\n");
4731: fprintf(ficresphtmfr, "**********</h3>\n");
4732: fprintf(ficlog, "**********\n");
4733: }
1.284 brouard 4734: /*
4735: Printing means of quantitative variables if any
4736: */
4737: for (z1=1; z1<= nqfveff; z1++) {
1.285 brouard 4738: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.0f individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.284 brouard 4739: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
4740: if(weightopt==1){
4741: printf(" Weighted mean and standard deviation of");
4742: fprintf(ficlog," Weighted mean and standard deviation of");
4743: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4744: }
1.285 brouard 4745: printf(" fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4746: fprintf(ficlog," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
4747: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.0f representatives of the population : %6.3g (%6.3g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt((stdq[z1]-meanq[z1]*meanq[z1]/idq[z1])/idq[z1]));
1.284 brouard 4748: }
4749: /* for (z1=1; z1<= nqtveff; z1++) { */
4750: /* for(m=1;m<=lastpass;m++){ */
4751: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4752: /* } */
4753: /* } */
1.283 brouard 4754:
1.251 brouard 4755: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4756: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4757: fprintf(ficresp, " Age");
4758: 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 4759: for(i=1; i<=nlstate;i++) {
1.265 brouard 4760: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4761: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4762: }
1.265 brouard 4763: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4764: fprintf(ficresphtm, "\n");
4765:
4766: /* Header of frequency table by age */
4767: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4768: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4769: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4770: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4771: if(s2!=0 && m!=0)
4772: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4773: }
1.226 brouard 4774: }
1.251 brouard 4775: fprintf(ficresphtmfr, "\n");
4776:
4777: /* For each age */
4778: for(iage=iagemin; iage <= iagemax+3; iage++){
4779: fprintf(ficresphtm,"<tr>");
4780: if(iage==iagemax+1){
4781: fprintf(ficlog,"1");
4782: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4783: }else if(iage==iagemax+2){
4784: fprintf(ficlog,"0");
4785: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4786: }else if(iage==iagemax+3){
4787: fprintf(ficlog,"Total");
4788: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4789: }else{
1.240 brouard 4790: if(first==1){
1.251 brouard 4791: first=0;
4792: printf("See log file for details...\n");
4793: }
4794: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4795: fprintf(ficlog,"Age %d", iage);
4796: }
1.265 brouard 4797: for(s1=1; s1 <=nlstate ; s1++){
4798: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4799: pp[s1] += freq[s1][m][iage];
1.251 brouard 4800: }
1.265 brouard 4801: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4802: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4803: pos += freq[s1][m][iage];
4804: if(pp[s1]>=1.e-10){
1.251 brouard 4805: if(first==1){
1.265 brouard 4806: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4807: }
1.265 brouard 4808: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4809: }else{
4810: if(first==1)
1.265 brouard 4811: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4812: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4813: }
4814: }
4815:
1.265 brouard 4816: for(s1=1; s1 <=nlstate ; s1++){
4817: /* posprop[s1]=0; */
4818: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4819: pp[s1] += freq[s1][m][iage];
4820: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4821:
4822: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4823: pos += pp[s1]; /* pos is the total number of transitions until this age */
4824: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4825: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4826: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4827: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4828: }
4829:
4830: /* Writing ficresp */
4831: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4832: if( iage <= iagemax){
4833: fprintf(ficresp," %d",iage);
4834: }
4835: }else if( nj==2){
4836: if( iage <= iagemax){
4837: fprintf(ficresp," %d",iage);
4838: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4839: }
1.240 brouard 4840: }
1.265 brouard 4841: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4842: if(pos>=1.e-5){
1.251 brouard 4843: if(first==1)
1.265 brouard 4844: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4845: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4846: }else{
4847: if(first==1)
1.265 brouard 4848: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4849: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4850: }
4851: if( iage <= iagemax){
4852: if(pos>=1.e-5){
1.265 brouard 4853: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4854: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4855: }else if( nj==2){
4856: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4857: }
4858: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4859: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4860: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4861: } else{
4862: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4863: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4864: }
1.240 brouard 4865: }
1.265 brouard 4866: pospropt[s1] +=posprop[s1];
4867: } /* end loop s1 */
1.251 brouard 4868: /* pospropt=0.; */
1.265 brouard 4869: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4870: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4871: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4872: if(first==1){
1.265 brouard 4873: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4874: }
1.265 brouard 4875: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4876: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4877: }
1.265 brouard 4878: if(s1!=0 && m!=0)
4879: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4880: }
1.265 brouard 4881: } /* end loop s1 */
1.251 brouard 4882: posproptt=0.;
1.265 brouard 4883: for(s1=1; s1 <=nlstate; s1++){
4884: posproptt += pospropt[s1];
1.251 brouard 4885: }
4886: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4887: fprintf(ficresphtm,"</tr>\n");
4888: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4889: if(iage <= iagemax)
4890: fprintf(ficresp,"\n");
1.240 brouard 4891: }
1.251 brouard 4892: if(first==1)
4893: printf("Others in log...\n");
4894: fprintf(ficlog,"\n");
4895: } /* end loop age iage */
1.265 brouard 4896:
1.251 brouard 4897: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4898: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4899: if(posproptt < 1.e-5){
1.265 brouard 4900: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4901: }else{
1.265 brouard 4902: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4903: }
1.226 brouard 4904: }
1.251 brouard 4905: fprintf(ficresphtm,"</tr>\n");
4906: fprintf(ficresphtm,"</table>\n");
4907: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4908: if(posproptt < 1.e-5){
1.251 brouard 4909: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4910: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4911: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4912: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4913: invalidvarcomb[j1]=1;
1.226 brouard 4914: }else{
1.251 brouard 4915: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4916: invalidvarcomb[j1]=0;
1.226 brouard 4917: }
1.251 brouard 4918: fprintf(ficresphtmfr,"</table>\n");
4919: fprintf(ficlog,"\n");
4920: if(j!=0){
4921: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4922: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4923: for(k=1; k <=(nlstate+ndeath); k++){
4924: if (k != i) {
1.265 brouard 4925: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4926: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4927: if(j1==1){ /* All dummy covariates to zero */
4928: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4929: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4930: printf("%d%d ",i,k);
4931: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4932: 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]));
4933: 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]));
4934: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4935: }
1.253 brouard 4936: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4937: for(iage=iagemin; iage <= iagemax+3; iage++){
4938: x[iage]= (double)iage;
4939: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4940: /* 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 4941: }
1.268 brouard 4942: /* Some are not finite, but linreg will ignore these ages */
4943: no=0;
1.253 brouard 4944: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4945: pstart[s1]=b;
4946: pstart[s1-1]=a;
1.252 brouard 4947: }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 */
4948: 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]);
4949: 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 4950: 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 4951: printf("%d%d ",i,k);
4952: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4953: 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 4954: }else{ /* Other cases, like quantitative fixed or varying covariates */
4955: ;
4956: }
4957: /* printf("%12.7f )", param[i][jj][k]); */
4958: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4959: s1++;
1.251 brouard 4960: } /* end jj */
4961: } /* end k!= i */
4962: } /* end k */
1.265 brouard 4963: } /* end i, s1 */
1.251 brouard 4964: } /* end j !=0 */
4965: } /* end selected combination of covariate j1 */
4966: if(j==0){ /* We can estimate starting values from the occurences in each case */
4967: printf("#Freqsummary: Starting values for the constants:\n");
4968: fprintf(ficlog,"\n");
1.265 brouard 4969: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4970: for(k=1; k <=(nlstate+ndeath); k++){
4971: if (k != i) {
4972: printf("%d%d ",i,k);
4973: fprintf(ficlog,"%d%d ",i,k);
4974: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4975: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4976: if(jj==1){ /* Age has to be done */
1.265 brouard 4977: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4978: 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]));
4979: 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 4980: }
4981: /* printf("%12.7f )", param[i][jj][k]); */
4982: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4983: s1++;
1.250 brouard 4984: }
1.251 brouard 4985: printf("\n");
4986: fprintf(ficlog,"\n");
1.250 brouard 4987: }
4988: }
1.284 brouard 4989: } /* end of state i */
1.251 brouard 4990: printf("#Freqsummary\n");
4991: fprintf(ficlog,"\n");
1.265 brouard 4992: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4993: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4994: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4995: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4996: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4997: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4998: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4999: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5000: /* } */
5001: }
1.265 brouard 5002: } /* end loop s1 */
1.251 brouard 5003:
5004: printf("\n");
5005: fprintf(ficlog,"\n");
5006: } /* end j=0 */
1.249 brouard 5007: } /* end j */
1.252 brouard 5008:
1.253 brouard 5009: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5010: for(i=1, jk=1; i <=nlstate; i++){
5011: for(j=1; j <=nlstate+ndeath; j++){
5012: if(j!=i){
5013: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5014: printf("%1d%1d",i,j);
5015: fprintf(ficparo,"%1d%1d",i,j);
5016: for(k=1; k<=ncovmodel;k++){
5017: /* printf(" %lf",param[i][j][k]); */
5018: /* fprintf(ficparo," %lf",param[i][j][k]); */
5019: p[jk]=pstart[jk];
5020: printf(" %f ",pstart[jk]);
5021: fprintf(ficparo," %f ",pstart[jk]);
5022: jk++;
5023: }
5024: printf("\n");
5025: fprintf(ficparo,"\n");
5026: }
5027: }
5028: }
5029: } /* end mle=-2 */
1.226 brouard 5030: dateintmean=dateintsum/k2cpt;
1.296 brouard 5031: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5032:
1.226 brouard 5033: fclose(ficresp);
5034: fclose(ficresphtm);
5035: fclose(ficresphtmfr);
1.283 brouard 5036: free_vector(idq,1,nqfveff);
1.226 brouard 5037: free_vector(meanq,1,nqfveff);
1.284 brouard 5038: free_vector(stdq,1,nqfveff);
1.226 brouard 5039: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5040: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5041: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5042: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5043: free_vector(pospropt,1,nlstate);
5044: free_vector(posprop,1,nlstate);
1.251 brouard 5045: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5046: free_vector(pp,1,nlstate);
5047: /* End of freqsummary */
5048: }
1.126 brouard 5049:
1.268 brouard 5050: /* Simple linear regression */
5051: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5052:
5053: /* y=a+bx regression */
5054: double sumx = 0.0; /* sum of x */
5055: double sumx2 = 0.0; /* sum of x**2 */
5056: double sumxy = 0.0; /* sum of x * y */
5057: double sumy = 0.0; /* sum of y */
5058: double sumy2 = 0.0; /* sum of y**2 */
5059: double sume2 = 0.0; /* sum of square or residuals */
5060: double yhat;
5061:
5062: double denom=0;
5063: int i;
5064: int ne=*no;
5065:
5066: for ( i=ifi, ne=0;i<=ila;i++) {
5067: if(!isfinite(x[i]) || !isfinite(y[i])){
5068: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5069: continue;
5070: }
5071: ne=ne+1;
5072: sumx += x[i];
5073: sumx2 += x[i]*x[i];
5074: sumxy += x[i] * y[i];
5075: sumy += y[i];
5076: sumy2 += y[i]*y[i];
5077: denom = (ne * sumx2 - sumx*sumx);
5078: /* 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); */
5079: }
5080:
5081: denom = (ne * sumx2 - sumx*sumx);
5082: if (denom == 0) {
5083: // vertical, slope m is infinity
5084: *b = INFINITY;
5085: *a = 0;
5086: if (r) *r = 0;
5087: return 1;
5088: }
5089:
5090: *b = (ne * sumxy - sumx * sumy) / denom;
5091: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5092: if (r!=NULL) {
5093: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5094: sqrt((sumx2 - sumx*sumx/ne) *
5095: (sumy2 - sumy*sumy/ne));
5096: }
5097: *no=ne;
5098: for ( i=ifi, ne=0;i<=ila;i++) {
5099: if(!isfinite(x[i]) || !isfinite(y[i])){
5100: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5101: continue;
5102: }
5103: ne=ne+1;
5104: yhat = y[i] - *a -*b* x[i];
5105: sume2 += yhat * yhat ;
5106:
5107: denom = (ne * sumx2 - sumx*sumx);
5108: /* 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); */
5109: }
5110: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5111: *sa= *sb * sqrt(sumx2/ne);
5112:
5113: return 0;
5114: }
5115:
1.126 brouard 5116: /************ Prevalence ********************/
1.227 brouard 5117: 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)
5118: {
5119: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5120: in each health status at the date of interview (if between dateprev1 and dateprev2).
5121: We still use firstpass and lastpass as another selection.
5122: */
1.126 brouard 5123:
1.227 brouard 5124: int i, m, jk, j1, bool, z1,j, iv;
5125: int mi; /* Effective wave */
5126: int iage;
5127: double agebegin, ageend;
5128:
5129: double **prop;
5130: double posprop;
5131: double y2; /* in fractional years */
5132: int iagemin, iagemax;
5133: int first; /** to stop verbosity which is redirected to log file */
5134:
5135: iagemin= (int) agemin;
5136: iagemax= (int) agemax;
5137: /*pp=vector(1,nlstate);*/
1.251 brouard 5138: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5139: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5140: j1=0;
1.222 brouard 5141:
1.227 brouard 5142: /*j=cptcoveff;*/
5143: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5144:
1.288 brouard 5145: first=0;
1.227 brouard 5146: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5147: for (i=1; i<=nlstate; i++)
1.251 brouard 5148: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5149: prop[i][iage]=0.0;
5150: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5151: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5152: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5153:
5154: for (i=1; i<=imx; i++) { /* Each individual */
5155: bool=1;
5156: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5157: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5158: m=mw[mi][i];
5159: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5160: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5161: for (z1=1; z1<=cptcoveff; z1++){
5162: if( Fixed[Tmodelind[z1]]==1){
5163: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5164: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5165: bool=0;
5166: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5167: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5168: bool=0;
5169: }
5170: }
5171: if(bool==1){ /* Otherwise we skip that wave/person */
5172: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5173: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5174: if(m >=firstpass && m <=lastpass){
5175: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5176: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5177: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5178: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5179: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5180: 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);
5181: exit(1);
5182: }
5183: if (s[m][i]>0 && s[m][i]<=nlstate) {
5184: /*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]]);*/
5185: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5186: prop[s[m][i]][iagemax+3] += weight[i];
5187: } /* end valid statuses */
5188: } /* end selection of dates */
5189: } /* end selection of waves */
5190: } /* end bool */
5191: } /* end wave */
5192: } /* end individual */
5193: for(i=iagemin; i <= iagemax+3; i++){
5194: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5195: posprop += prop[jk][i];
5196: }
5197:
5198: for(jk=1; jk <=nlstate ; jk++){
5199: if( i <= iagemax){
5200: if(posprop>=1.e-5){
5201: probs[i][jk][j1]= prop[jk][i]/posprop;
5202: } else{
1.288 brouard 5203: if(!first){
5204: first=1;
1.266 brouard 5205: 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]);
5206: }else{
1.288 brouard 5207: 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 5208: }
5209: }
5210: }
5211: }/* end jk */
5212: }/* end i */
1.222 brouard 5213: /*} *//* end i1 */
1.227 brouard 5214: } /* end j1 */
1.222 brouard 5215:
1.227 brouard 5216: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5217: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5218: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5219: } /* End of prevalence */
1.126 brouard 5220:
5221: /************* Waves Concatenation ***************/
5222:
5223: 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)
5224: {
1.298 brouard 5225: /* 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 5226: Death is a valid wave (if date is known).
5227: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5228: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5229: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5230: */
1.126 brouard 5231:
1.224 brouard 5232: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5233: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5234: double sum=0., jmean=0.;*/
1.224 brouard 5235: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5236: int j, k=0,jk, ju, jl;
5237: double sum=0.;
5238: first=0;
1.214 brouard 5239: firstwo=0;
1.217 brouard 5240: firsthree=0;
1.218 brouard 5241: firstfour=0;
1.164 brouard 5242: jmin=100000;
1.126 brouard 5243: jmax=-1;
5244: jmean=0.;
1.224 brouard 5245:
5246: /* Treating live states */
1.214 brouard 5247: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5248: mi=0; /* First valid wave */
1.227 brouard 5249: mli=0; /* Last valid wave */
1.126 brouard 5250: m=firstpass;
1.214 brouard 5251: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5252: 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 */
5253: mli=m-1;/* mw[++mi][i]=m-1; */
5254: }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 */
5255: mw[++mi][i]=m;
5256: mli=m;
1.224 brouard 5257: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5258: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5259: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5260: }
1.227 brouard 5261: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5262: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5263: break;
1.224 brouard 5264: #else
1.227 brouard 5265: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5266: if(firsthree == 0){
1.302 brouard 5267: 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 5268: firsthree=1;
5269: }
1.302 brouard 5270: 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);
1.227 brouard 5271: mw[++mi][i]=m;
5272: mli=m;
5273: }
5274: if(s[m][i]==-2){ /* Vital status is really unknown */
5275: nbwarn++;
5276: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5277: 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);
5278: 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);
5279: }
5280: break;
5281: }
5282: break;
1.224 brouard 5283: #endif
1.227 brouard 5284: }/* End m >= lastpass */
1.126 brouard 5285: }/* end while */
1.224 brouard 5286:
1.227 brouard 5287: /* 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 5288: /* After last pass */
1.224 brouard 5289: /* Treating death states */
1.214 brouard 5290: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5291: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5292: /* } */
1.126 brouard 5293: mi++; /* Death is another wave */
5294: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5295: /* Only death is a correct wave */
1.126 brouard 5296: mw[mi][i]=m;
1.257 brouard 5297: } /* else not in a death state */
1.224 brouard 5298: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5299: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5300: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5301: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* death occured before last wave and status should have been death instead of -1 */
5302: nbwarn++;
5303: if(firstfiv==0){
5304: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %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 );
5305: firstfiv=1;
5306: }else{
5307: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %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 );
5308: }
5309: }else{ /* Death occured afer last wave potential bias */
5310: nberr++;
5311: if(firstwo==0){
1.257 brouard 5312: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). 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], i,m );
1.227 brouard 5313: firstwo=1;
5314: }
1.257 brouard 5315: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.227 brouard 5316: }
1.257 brouard 5317: }else{ /* if date of interview is unknown */
1.227 brouard 5318: /* death is known but not confirmed by death status at any wave */
5319: if(firstfour==0){
5320: 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. 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], i,m );
5321: firstfour=1;
5322: }
5323: 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. 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], i,m );
1.214 brouard 5324: }
1.224 brouard 5325: } /* end if date of death is known */
5326: #endif
5327: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5328: /* wav[i]=mw[mi][i]; */
1.126 brouard 5329: if(mi==0){
5330: nbwarn++;
5331: if(first==0){
1.227 brouard 5332: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5333: first=1;
1.126 brouard 5334: }
5335: if(first==1){
1.227 brouard 5336: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5337: }
5338: } /* end mi==0 */
5339: } /* End individuals */
1.214 brouard 5340: /* wav and mw are no more changed */
1.223 brouard 5341:
1.214 brouard 5342:
1.126 brouard 5343: for(i=1; i<=imx; i++){
5344: for(mi=1; mi<wav[i];mi++){
5345: if (stepm <=0)
1.227 brouard 5346: dh[mi][i]=1;
1.126 brouard 5347: else{
1.260 brouard 5348: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5349: if (agedc[i] < 2*AGESUP) {
5350: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5351: if(j==0) j=1; /* Survives at least one month after exam */
5352: else if(j<0){
5353: nberr++;
5354: 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]);
5355: j=1; /* Temporary Dangerous patch */
5356: 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);
5357: 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]);
5358: 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);
5359: }
5360: k=k+1;
5361: if (j >= jmax){
5362: jmax=j;
5363: ijmax=i;
5364: }
5365: if (j <= jmin){
5366: jmin=j;
5367: ijmin=i;
5368: }
5369: sum=sum+j;
5370: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5371: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5372: }
5373: }
5374: else{
5375: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5376: /* 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 5377:
1.227 brouard 5378: k=k+1;
5379: if (j >= jmax) {
5380: jmax=j;
5381: ijmax=i;
5382: }
5383: else if (j <= jmin){
5384: jmin=j;
5385: ijmin=i;
5386: }
5387: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5388: /*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]);*/
5389: if(j<0){
5390: nberr++;
5391: 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]);
5392: 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]);
5393: }
5394: sum=sum+j;
5395: }
5396: jk= j/stepm;
5397: jl= j -jk*stepm;
5398: ju= j -(jk+1)*stepm;
5399: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5400: if(jl==0){
5401: dh[mi][i]=jk;
5402: bh[mi][i]=0;
5403: }else{ /* We want a negative bias in order to only have interpolation ie
5404: * to avoid the price of an extra matrix product in likelihood */
5405: dh[mi][i]=jk+1;
5406: bh[mi][i]=ju;
5407: }
5408: }else{
5409: if(jl <= -ju){
5410: dh[mi][i]=jk;
5411: bh[mi][i]=jl; /* bias is positive if real duration
5412: * is higher than the multiple of stepm and negative otherwise.
5413: */
5414: }
5415: else{
5416: dh[mi][i]=jk+1;
5417: bh[mi][i]=ju;
5418: }
5419: if(dh[mi][i]==0){
5420: dh[mi][i]=1; /* At least one step */
5421: bh[mi][i]=ju; /* At least one step */
5422: /* 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);*/
5423: }
5424: } /* end if mle */
1.126 brouard 5425: }
5426: } /* end wave */
5427: }
5428: jmean=sum/k;
5429: 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 5430: 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 5431: }
1.126 brouard 5432:
5433: /*********** Tricode ****************************/
1.220 brouard 5434: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5435: {
5436: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5437: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5438: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5439: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5440: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5441: */
1.130 brouard 5442:
1.242 brouard 5443: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5444: int modmaxcovj=0; /* Modality max of covariates j */
5445: int cptcode=0; /* Modality max of covariates j */
5446: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5447:
5448:
1.242 brouard 5449: /* cptcoveff=0; */
5450: /* *cptcov=0; */
1.126 brouard 5451:
1.242 brouard 5452: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5453: for (k=1; k <= maxncov; k++)
5454: for(j=1; j<=2; j++)
5455: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5456:
1.242 brouard 5457: /* Loop on covariates without age and products and no quantitative variable */
5458: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5459: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5460: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5461: switch(Fixed[k]) {
5462: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5463: 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*/
5464: ij=(int)(covar[Tvar[k]][i]);
5465: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5466: * If product of Vn*Vm, still boolean *:
5467: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5468: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5469: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5470: modality of the nth covariate of individual i. */
5471: if (ij > modmaxcovj)
5472: modmaxcovj=ij;
5473: else if (ij < modmincovj)
5474: modmincovj=ij;
1.287 brouard 5475: if (ij <0 || ij >1 ){
5476: printf("Information, IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5477: fprintf(ficlog,"Information, currently IMaCh doesn't treat covariate with missing values (-1), individual %d will be skipped.\n",i);
5478: }
5479: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5480: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5481: exit(1);
5482: }else
5483: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5484: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5485: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5486: /* getting the maximum value of the modality of the covariate
5487: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5488: female ies 1, then modmaxcovj=1.
5489: */
5490: } /* end for loop on individuals i */
5491: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5492: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5493: cptcode=modmaxcovj;
5494: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5495: /*for (i=0; i<=cptcode; i++) {*/
5496: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5497: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5498: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5499: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5500: if( j != -1){
5501: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5502: covariate for which somebody answered excluding
5503: undefined. Usually 2: 0 and 1. */
5504: }
5505: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5506: covariate for which somebody answered including
5507: undefined. Usually 3: -1, 0 and 1. */
5508: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5509: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5510: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5511:
1.242 brouard 5512: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5513: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5514: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5515: /* modmincovj=3; modmaxcovj = 7; */
5516: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5517: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5518: /* defining two dummy variables: variables V1_1 and V1_2.*/
5519: /* nbcode[Tvar[j]][ij]=k; */
5520: /* nbcode[Tvar[j]][1]=0; */
5521: /* nbcode[Tvar[j]][2]=1; */
5522: /* nbcode[Tvar[j]][3]=2; */
5523: /* To be continued (not working yet). */
5524: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5525:
5526: /* 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*/
5527: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5528: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5529: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5530: /*, could be restored in the future */
5531: 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 5532: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5533: break;
5534: }
5535: ij++;
1.287 brouard 5536: 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 5537: cptcode = ij; /* New max modality for covar j */
5538: } /* end of loop on modality i=-1 to 1 or more */
5539: break;
5540: case 1: /* Testing on varying covariate, could be simple and
5541: * should look at waves or product of fixed *
5542: * varying. No time to test -1, assuming 0 and 1 only */
5543: ij=0;
5544: for(i=0; i<=1;i++){
5545: nbcode[Tvar[k]][++ij]=i;
5546: }
5547: break;
5548: default:
5549: break;
5550: } /* end switch */
5551: } /* end dummy test */
1.287 brouard 5552: } /* 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 5553:
5554: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5555: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5556: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5557: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5558: 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 */
5559: 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 */
5560: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5561: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5562:
5563: ij=0;
5564: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5565: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5566: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5567: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5568: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5569: /* If product not in single variable we don't print results */
5570: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5571: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5572: 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*/
5573: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5574: 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 */
5575: if(Fixed[k]!=0)
5576: anyvaryingduminmodel=1;
5577: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5578: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5579: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5580: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5581: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5582: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5583: }
5584: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5585: /* ij--; */
5586: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5587: *cptcov=ij; /*Number of total real effective covariates: effective
5588: * because they can be excluded from the model and real
5589: * if in the model but excluded because missing values, but how to get k from ij?*/
5590: for(j=ij+1; j<= cptcovt; j++){
5591: Tvaraff[j]=0;
5592: Tmodelind[j]=0;
5593: }
5594: for(j=ntveff+1; j<= cptcovt; j++){
5595: TmodelInvind[j]=0;
5596: }
5597: /* To be sorted */
5598: ;
5599: }
1.126 brouard 5600:
1.145 brouard 5601:
1.126 brouard 5602: /*********** Health Expectancies ****************/
5603:
1.235 brouard 5604: 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 5605:
5606: {
5607: /* Health expectancies, no variances */
1.164 brouard 5608: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5609: int nhstepma, nstepma; /* Decreasing with age */
5610: double age, agelim, hf;
5611: double ***p3mat;
5612: double eip;
5613:
1.238 brouard 5614: /* pstamp(ficreseij); */
1.126 brouard 5615: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5616: fprintf(ficreseij,"# Age");
5617: for(i=1; i<=nlstate;i++){
5618: for(j=1; j<=nlstate;j++){
5619: fprintf(ficreseij," e%1d%1d ",i,j);
5620: }
5621: fprintf(ficreseij," e%1d. ",i);
5622: }
5623: fprintf(ficreseij,"\n");
5624:
5625:
5626: if(estepm < stepm){
5627: printf ("Problem %d lower than %d\n",estepm, stepm);
5628: }
5629: else hstepm=estepm;
5630: /* We compute the life expectancy from trapezoids spaced every estepm months
5631: * This is mainly to measure the difference between two models: for example
5632: * if stepm=24 months pijx are given only every 2 years and by summing them
5633: * we are calculating an estimate of the Life Expectancy assuming a linear
5634: * progression in between and thus overestimating or underestimating according
5635: * to the curvature of the survival function. If, for the same date, we
5636: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5637: * to compare the new estimate of Life expectancy with the same linear
5638: * hypothesis. A more precise result, taking into account a more precise
5639: * curvature will be obtained if estepm is as small as stepm. */
5640:
5641: /* For example we decided to compute the life expectancy with the smallest unit */
5642: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5643: nhstepm is the number of hstepm from age to agelim
5644: nstepm is the number of stepm from age to agelin.
1.270 brouard 5645: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5646: and note for a fixed period like estepm months */
5647: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5648: survival function given by stepm (the optimization length). Unfortunately it
5649: means that if the survival funtion is printed only each two years of age and if
5650: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5651: results. So we changed our mind and took the option of the best precision.
5652: */
5653: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5654:
5655: agelim=AGESUP;
5656: /* If stepm=6 months */
5657: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5658: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5659:
5660: /* nhstepm age range expressed in number of stepm */
5661: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5662: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5663: /* if (stepm >= YEARM) hstepm=1;*/
5664: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5665: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5666:
5667: for (age=bage; age<=fage; age ++){
5668: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5669: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5670: /* if (stepm >= YEARM) hstepm=1;*/
5671: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5672:
5673: /* If stepm=6 months */
5674: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5675: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5676:
1.235 brouard 5677: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5678:
5679: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5680:
5681: printf("%d|",(int)age);fflush(stdout);
5682: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5683:
5684: /* Computing expectancies */
5685: for(i=1; i<=nlstate;i++)
5686: for(j=1; j<=nlstate;j++)
5687: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5688: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5689:
5690: /* 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]);*/
5691:
5692: }
5693:
5694: fprintf(ficreseij,"%3.0f",age );
5695: for(i=1; i<=nlstate;i++){
5696: eip=0;
5697: for(j=1; j<=nlstate;j++){
5698: eip +=eij[i][j][(int)age];
5699: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5700: }
5701: fprintf(ficreseij,"%9.4f", eip );
5702: }
5703: fprintf(ficreseij,"\n");
5704:
5705: }
5706: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5707: printf("\n");
5708: fprintf(ficlog,"\n");
5709:
5710: }
5711:
1.235 brouard 5712: 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 5713:
5714: {
5715: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5716: to initial status i, ei. .
1.126 brouard 5717: */
5718: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5719: int nhstepma, nstepma; /* Decreasing with age */
5720: double age, agelim, hf;
5721: double ***p3matp, ***p3matm, ***varhe;
5722: double **dnewm,**doldm;
5723: double *xp, *xm;
5724: double **gp, **gm;
5725: double ***gradg, ***trgradg;
5726: int theta;
5727:
5728: double eip, vip;
5729:
5730: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5731: xp=vector(1,npar);
5732: xm=vector(1,npar);
5733: dnewm=matrix(1,nlstate*nlstate,1,npar);
5734: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5735:
5736: pstamp(ficresstdeij);
5737: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5738: fprintf(ficresstdeij,"# Age");
5739: for(i=1; i<=nlstate;i++){
5740: for(j=1; j<=nlstate;j++)
5741: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5742: fprintf(ficresstdeij," e%1d. ",i);
5743: }
5744: fprintf(ficresstdeij,"\n");
5745:
5746: pstamp(ficrescveij);
5747: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5748: fprintf(ficrescveij,"# Age");
5749: for(i=1; i<=nlstate;i++)
5750: for(j=1; j<=nlstate;j++){
5751: cptj= (j-1)*nlstate+i;
5752: for(i2=1; i2<=nlstate;i2++)
5753: for(j2=1; j2<=nlstate;j2++){
5754: cptj2= (j2-1)*nlstate+i2;
5755: if(cptj2 <= cptj)
5756: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5757: }
5758: }
5759: fprintf(ficrescveij,"\n");
5760:
5761: if(estepm < stepm){
5762: printf ("Problem %d lower than %d\n",estepm, stepm);
5763: }
5764: else hstepm=estepm;
5765: /* We compute the life expectancy from trapezoids spaced every estepm months
5766: * This is mainly to measure the difference between two models: for example
5767: * if stepm=24 months pijx are given only every 2 years and by summing them
5768: * we are calculating an estimate of the Life Expectancy assuming a linear
5769: * progression in between and thus overestimating or underestimating according
5770: * to the curvature of the survival function. If, for the same date, we
5771: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5772: * to compare the new estimate of Life expectancy with the same linear
5773: * hypothesis. A more precise result, taking into account a more precise
5774: * curvature will be obtained if estepm is as small as stepm. */
5775:
5776: /* For example we decided to compute the life expectancy with the smallest unit */
5777: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5778: nhstepm is the number of hstepm from age to agelim
5779: nstepm is the number of stepm from age to agelin.
5780: Look at hpijx to understand the reason of that which relies in memory size
5781: and note for a fixed period like estepm months */
5782: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5783: survival function given by stepm (the optimization length). Unfortunately it
5784: means that if the survival funtion is printed only each two years of age and if
5785: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5786: results. So we changed our mind and took the option of the best precision.
5787: */
5788: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5789:
5790: /* If stepm=6 months */
5791: /* nhstepm age range expressed in number of stepm */
5792: agelim=AGESUP;
5793: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5794: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5795: /* if (stepm >= YEARM) hstepm=1;*/
5796: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5797:
5798: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5799: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5800: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5801: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5802: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5803: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5804:
5805: for (age=bage; age<=fage; age ++){
5806: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5807: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5808: /* if (stepm >= YEARM) hstepm=1;*/
5809: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5810:
1.126 brouard 5811: /* If stepm=6 months */
5812: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5813: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5814:
5815: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5816:
1.126 brouard 5817: /* Computing Variances of health expectancies */
5818: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5819: decrease memory allocation */
5820: for(theta=1; theta <=npar; theta++){
5821: for(i=1; i<=npar; i++){
1.222 brouard 5822: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5823: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5824: }
1.235 brouard 5825: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5826: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5827:
1.126 brouard 5828: for(j=1; j<= nlstate; j++){
1.222 brouard 5829: for(i=1; i<=nlstate; i++){
5830: for(h=0; h<=nhstepm-1; h++){
5831: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5832: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5833: }
5834: }
1.126 brouard 5835: }
1.218 brouard 5836:
1.126 brouard 5837: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5838: for(h=0; h<=nhstepm-1; h++){
5839: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5840: }
1.126 brouard 5841: }/* End theta */
5842:
5843:
5844: for(h=0; h<=nhstepm-1; h++)
5845: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5846: for(theta=1; theta <=npar; theta++)
5847: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5848:
1.218 brouard 5849:
1.222 brouard 5850: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5851: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5852: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5853:
1.222 brouard 5854: printf("%d|",(int)age);fflush(stdout);
5855: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5856: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5857: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5858: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5859: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5860: for(ij=1;ij<=nlstate*nlstate;ij++)
5861: for(ji=1;ji<=nlstate*nlstate;ji++)
5862: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5863: }
5864: }
1.218 brouard 5865:
1.126 brouard 5866: /* Computing expectancies */
1.235 brouard 5867: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5868: for(i=1; i<=nlstate;i++)
5869: for(j=1; j<=nlstate;j++)
1.222 brouard 5870: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5871: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5872:
1.222 brouard 5873: /* 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 5874:
1.222 brouard 5875: }
1.269 brouard 5876:
5877: /* Standard deviation of expectancies ij */
1.126 brouard 5878: fprintf(ficresstdeij,"%3.0f",age );
5879: for(i=1; i<=nlstate;i++){
5880: eip=0.;
5881: vip=0.;
5882: for(j=1; j<=nlstate;j++){
1.222 brouard 5883: eip += eij[i][j][(int)age];
5884: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5885: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5886: 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 5887: }
5888: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5889: }
5890: fprintf(ficresstdeij,"\n");
1.218 brouard 5891:
1.269 brouard 5892: /* Variance of expectancies ij */
1.126 brouard 5893: fprintf(ficrescveij,"%3.0f",age );
5894: for(i=1; i<=nlstate;i++)
5895: for(j=1; j<=nlstate;j++){
1.222 brouard 5896: cptj= (j-1)*nlstate+i;
5897: for(i2=1; i2<=nlstate;i2++)
5898: for(j2=1; j2<=nlstate;j2++){
5899: cptj2= (j2-1)*nlstate+i2;
5900: if(cptj2 <= cptj)
5901: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5902: }
1.126 brouard 5903: }
5904: fprintf(ficrescveij,"\n");
1.218 brouard 5905:
1.126 brouard 5906: }
5907: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5908: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5909: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5910: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5911: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5912: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5913: printf("\n");
5914: fprintf(ficlog,"\n");
1.218 brouard 5915:
1.126 brouard 5916: free_vector(xm,1,npar);
5917: free_vector(xp,1,npar);
5918: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5919: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5920: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5921: }
1.218 brouard 5922:
1.126 brouard 5923: /************ Variance ******************/
1.235 brouard 5924: 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 5925: {
1.279 brouard 5926: /** Variance of health expectancies
5927: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5928: * double **newm;
5929: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5930: */
1.218 brouard 5931:
5932: /* int movingaverage(); */
5933: double **dnewm,**doldm;
5934: double **dnewmp,**doldmp;
5935: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5936: int first=0;
1.218 brouard 5937: int k;
5938: double *xp;
1.279 brouard 5939: double **gp, **gm; /**< for var eij */
5940: double ***gradg, ***trgradg; /**< for var eij */
5941: double **gradgp, **trgradgp; /**< for var p point j */
5942: double *gpp, *gmp; /**< for var p point j */
5943: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 5944: double ***p3mat;
5945: double age,agelim, hf;
5946: /* double ***mobaverage; */
5947: int theta;
5948: char digit[4];
5949: char digitp[25];
5950:
5951: char fileresprobmorprev[FILENAMELENGTH];
5952:
5953: if(popbased==1){
5954: if(mobilav!=0)
5955: strcpy(digitp,"-POPULBASED-MOBILAV_");
5956: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5957: }
5958: else
5959: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5960:
1.218 brouard 5961: /* if (mobilav!=0) { */
5962: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5963: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5964: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5965: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5966: /* } */
5967: /* } */
5968:
5969: strcpy(fileresprobmorprev,"PRMORPREV-");
5970: sprintf(digit,"%-d",ij);
5971: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5972: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5973: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5974: strcat(fileresprobmorprev,fileresu);
5975: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5976: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5977: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5978: }
5979: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5980: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5981: pstamp(ficresprobmorprev);
5982: 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 5983: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5984: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5985: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5986: }
5987: for(j=1;j<=cptcoveff;j++)
5988: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5989: fprintf(ficresprobmorprev,"\n");
5990:
1.218 brouard 5991: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5992: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5993: fprintf(ficresprobmorprev," p.%-d SE",j);
5994: for(i=1; i<=nlstate;i++)
5995: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5996: }
5997: fprintf(ficresprobmorprev,"\n");
5998:
5999: fprintf(ficgp,"\n# Routine varevsij");
6000: fprintf(ficgp,"\nunset title \n");
6001: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6002: 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");
6003: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6004:
1.218 brouard 6005: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6006: pstamp(ficresvij);
6007: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6008: if(popbased==1)
6009: 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);
6010: else
6011: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6012: fprintf(ficresvij,"# Age");
6013: for(i=1; i<=nlstate;i++)
6014: for(j=1; j<=nlstate;j++)
6015: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6016: fprintf(ficresvij,"\n");
6017:
6018: xp=vector(1,npar);
6019: dnewm=matrix(1,nlstate,1,npar);
6020: doldm=matrix(1,nlstate,1,nlstate);
6021: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6022: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6023:
6024: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6025: gpp=vector(nlstate+1,nlstate+ndeath);
6026: gmp=vector(nlstate+1,nlstate+ndeath);
6027: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6028:
1.218 brouard 6029: if(estepm < stepm){
6030: printf ("Problem %d lower than %d\n",estepm, stepm);
6031: }
6032: else hstepm=estepm;
6033: /* For example we decided to compute the life expectancy with the smallest unit */
6034: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6035: nhstepm is the number of hstepm from age to agelim
6036: nstepm is the number of stepm from age to agelim.
6037: Look at function hpijx to understand why because of memory size limitations,
6038: we decided (b) to get a life expectancy respecting the most precise curvature of the
6039: survival function given by stepm (the optimization length). Unfortunately it
6040: means that if the survival funtion is printed every two years of age and if
6041: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6042: results. So we changed our mind and took the option of the best precision.
6043: */
6044: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6045: agelim = AGESUP;
6046: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6047: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6048: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6049: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6050: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6051: gp=matrix(0,nhstepm,1,nlstate);
6052: gm=matrix(0,nhstepm,1,nlstate);
6053:
6054:
6055: for(theta=1; theta <=npar; theta++){
6056: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6057: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6058: }
1.279 brouard 6059: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6060: * returns into prlim .
1.288 brouard 6061: */
1.242 brouard 6062: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6063:
6064: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6065: if (popbased==1) {
6066: if(mobilav ==0){
6067: for(i=1; i<=nlstate;i++)
6068: prlim[i][i]=probs[(int)age][i][ij];
6069: }else{ /* mobilav */
6070: for(i=1; i<=nlstate;i++)
6071: prlim[i][i]=mobaverage[(int)age][i][ij];
6072: }
6073: }
1.295 brouard 6074: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6075: */
6076: 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 6077: /**< 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 6078: * at horizon h in state j including mortality.
6079: */
1.218 brouard 6080: for(j=1; j<= nlstate; j++){
6081: for(h=0; h<=nhstepm; h++){
6082: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6083: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6084: }
6085: }
1.279 brouard 6086: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6087: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6088: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6089: */
6090: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6091: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6092: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6093: }
6094:
6095: /* Again with minus shift */
1.218 brouard 6096:
6097: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6098: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6099:
1.242 brouard 6100: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6101:
6102: if (popbased==1) {
6103: if(mobilav ==0){
6104: for(i=1; i<=nlstate;i++)
6105: prlim[i][i]=probs[(int)age][i][ij];
6106: }else{ /* mobilav */
6107: for(i=1; i<=nlstate;i++)
6108: prlim[i][i]=mobaverage[(int)age][i][ij];
6109: }
6110: }
6111:
1.235 brouard 6112: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6113:
6114: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6115: for(h=0; h<=nhstepm; h++){
6116: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6117: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6118: }
6119: }
6120: /* This for computing probability of death (h=1 means
6121: computed over hstepm matrices product = hstepm*stepm months)
6122: as a weighted average of prlim.
6123: */
6124: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6125: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6126: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6127: }
1.279 brouard 6128: /* end shifting computations */
6129:
6130: /**< Computing gradient matrix at horizon h
6131: */
1.218 brouard 6132: for(j=1; j<= nlstate; j++) /* vareij */
6133: for(h=0; h<=nhstepm; h++){
6134: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6135: }
1.279 brouard 6136: /**< Gradient of overall mortality p.3 (or p.j)
6137: */
6138: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6139: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6140: }
6141:
6142: } /* End theta */
1.279 brouard 6143:
6144: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6145: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6146:
6147: for(h=0; h<=nhstepm; h++) /* veij */
6148: for(j=1; j<=nlstate;j++)
6149: for(theta=1; theta <=npar; theta++)
6150: trgradg[h][j][theta]=gradg[h][theta][j];
6151:
6152: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6153: for(theta=1; theta <=npar; theta++)
6154: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6155: /**< as well as its transposed matrix
6156: */
1.218 brouard 6157:
6158: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6159: for(i=1;i<=nlstate;i++)
6160: for(j=1;j<=nlstate;j++)
6161: vareij[i][j][(int)age] =0.;
1.279 brouard 6162:
6163: /* Computing trgradg by matcov by gradg at age and summing over h
6164: * and k (nhstepm) formula 15 of article
6165: * Lievre-Brouard-Heathcote
6166: */
6167:
1.218 brouard 6168: for(h=0;h<=nhstepm;h++){
6169: for(k=0;k<=nhstepm;k++){
6170: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6171: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6172: for(i=1;i<=nlstate;i++)
6173: for(j=1;j<=nlstate;j++)
6174: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6175: }
6176: }
6177:
1.279 brouard 6178: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6179: * p.j overall mortality formula 49 but computed directly because
6180: * we compute the grad (wix pijx) instead of grad (pijx),even if
6181: * wix is independent of theta.
6182: */
1.218 brouard 6183: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6184: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6185: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6186: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6187: varppt[j][i]=doldmp[j][i];
6188: /* end ppptj */
6189: /* x centered again */
6190:
1.242 brouard 6191: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6192:
6193: if (popbased==1) {
6194: if(mobilav ==0){
6195: for(i=1; i<=nlstate;i++)
6196: prlim[i][i]=probs[(int)age][i][ij];
6197: }else{ /* mobilav */
6198: for(i=1; i<=nlstate;i++)
6199: prlim[i][i]=mobaverage[(int)age][i][ij];
6200: }
6201: }
6202:
6203: /* This for computing probability of death (h=1 means
6204: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6205: as a weighted average of prlim.
6206: */
1.235 brouard 6207: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6208: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6209: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6210: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6211: }
6212: /* end probability of death */
6213:
6214: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6215: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6216: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6217: for(i=1; i<=nlstate;i++){
6218: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6219: }
6220: }
6221: fprintf(ficresprobmorprev,"\n");
6222:
6223: fprintf(ficresvij,"%.0f ",age );
6224: for(i=1; i<=nlstate;i++)
6225: for(j=1; j<=nlstate;j++){
6226: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6227: }
6228: fprintf(ficresvij,"\n");
6229: free_matrix(gp,0,nhstepm,1,nlstate);
6230: free_matrix(gm,0,nhstepm,1,nlstate);
6231: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6232: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6233: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6234: } /* End age */
6235: free_vector(gpp,nlstate+1,nlstate+ndeath);
6236: free_vector(gmp,nlstate+1,nlstate+ndeath);
6237: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6238: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6239: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6240: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6241: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6242: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6243: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6244: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6245: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6246: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6247: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6248: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6249: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6250: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6251: 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);
6252: /* 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 6253: */
1.218 brouard 6254: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6255: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6256:
1.218 brouard 6257: free_vector(xp,1,npar);
6258: free_matrix(doldm,1,nlstate,1,nlstate);
6259: free_matrix(dnewm,1,nlstate,1,npar);
6260: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6261: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6262: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6263: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6264: fclose(ficresprobmorprev);
6265: fflush(ficgp);
6266: fflush(fichtm);
6267: } /* end varevsij */
1.126 brouard 6268:
6269: /************ Variance of prevlim ******************/
1.269 brouard 6270: 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 6271: {
1.205 brouard 6272: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6273: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6274:
1.268 brouard 6275: double **dnewmpar,**doldm;
1.126 brouard 6276: int i, j, nhstepm, hstepm;
6277: double *xp;
6278: double *gp, *gm;
6279: double **gradg, **trgradg;
1.208 brouard 6280: double **mgm, **mgp;
1.126 brouard 6281: double age,agelim;
6282: int theta;
6283:
6284: pstamp(ficresvpl);
1.288 brouard 6285: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6286: fprintf(ficresvpl,"# Age ");
6287: if(nresult >=1)
6288: fprintf(ficresvpl," Result# ");
1.126 brouard 6289: for(i=1; i<=nlstate;i++)
6290: fprintf(ficresvpl," %1d-%1d",i,i);
6291: fprintf(ficresvpl,"\n");
6292:
6293: xp=vector(1,npar);
1.268 brouard 6294: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6295: doldm=matrix(1,nlstate,1,nlstate);
6296:
6297: hstepm=1*YEARM; /* Every year of age */
6298: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6299: agelim = AGESUP;
6300: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6301: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6302: if (stepm >= YEARM) hstepm=1;
6303: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6304: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6305: mgp=matrix(1,npar,1,nlstate);
6306: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6307: gp=vector(1,nlstate);
6308: gm=vector(1,nlstate);
6309:
6310: for(theta=1; theta <=npar; theta++){
6311: for(i=1; i<=npar; i++){ /* Computes gradient */
6312: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6313: }
1.288 brouard 6314: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6315: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6316: /* else */
6317: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6318: for(i=1;i<=nlstate;i++){
1.126 brouard 6319: gp[i] = prlim[i][i];
1.208 brouard 6320: mgp[theta][i] = prlim[i][i];
6321: }
1.126 brouard 6322: for(i=1; i<=npar; i++) /* Computes gradient */
6323: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6324: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6325: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6326: /* else */
6327: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6328: for(i=1;i<=nlstate;i++){
1.126 brouard 6329: gm[i] = prlim[i][i];
1.208 brouard 6330: mgm[theta][i] = prlim[i][i];
6331: }
1.126 brouard 6332: for(i=1;i<=nlstate;i++)
6333: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6334: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6335: } /* End theta */
6336:
6337: trgradg =matrix(1,nlstate,1,npar);
6338:
6339: for(j=1; j<=nlstate;j++)
6340: for(theta=1; theta <=npar; theta++)
6341: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6342: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6343: /* printf("\nmgm mgp %d ",(int)age); */
6344: /* for(j=1; j<=nlstate;j++){ */
6345: /* printf(" %d ",j); */
6346: /* for(theta=1; theta <=npar; theta++) */
6347: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6348: /* printf("\n "); */
6349: /* } */
6350: /* } */
6351: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6352: /* printf("\n gradg %d ",(int)age); */
6353: /* for(j=1; j<=nlstate;j++){ */
6354: /* printf("%d ",j); */
6355: /* for(theta=1; theta <=npar; theta++) */
6356: /* printf("%d %lf ",theta,gradg[theta][j]); */
6357: /* printf("\n "); */
6358: /* } */
6359: /* } */
1.126 brouard 6360:
6361: for(i=1;i<=nlstate;i++)
6362: varpl[i][(int)age] =0.;
1.209 brouard 6363: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6364: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6365: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6366: }else{
1.268 brouard 6367: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6368: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6369: }
1.126 brouard 6370: for(i=1;i<=nlstate;i++)
6371: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6372:
6373: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6374: if(nresult >=1)
6375: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6376: for(i=1; i<=nlstate;i++){
1.126 brouard 6377: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6378: /* for(j=1;j<=nlstate;j++) */
6379: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6380: }
1.126 brouard 6381: fprintf(ficresvpl,"\n");
6382: free_vector(gp,1,nlstate);
6383: free_vector(gm,1,nlstate);
1.208 brouard 6384: free_matrix(mgm,1,npar,1,nlstate);
6385: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6386: free_matrix(gradg,1,npar,1,nlstate);
6387: free_matrix(trgradg,1,nlstate,1,npar);
6388: } /* End age */
6389:
6390: free_vector(xp,1,npar);
6391: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6392: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6393:
6394: }
6395:
6396:
6397: /************ Variance of backprevalence limit ******************/
1.269 brouard 6398: 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 6399: {
6400: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6401: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6402:
6403: double **dnewmpar,**doldm;
6404: int i, j, nhstepm, hstepm;
6405: double *xp;
6406: double *gp, *gm;
6407: double **gradg, **trgradg;
6408: double **mgm, **mgp;
6409: double age,agelim;
6410: int theta;
6411:
6412: pstamp(ficresvbl);
6413: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6414: fprintf(ficresvbl,"# Age ");
6415: if(nresult >=1)
6416: fprintf(ficresvbl," Result# ");
6417: for(i=1; i<=nlstate;i++)
6418: fprintf(ficresvbl," %1d-%1d",i,i);
6419: fprintf(ficresvbl,"\n");
6420:
6421: xp=vector(1,npar);
6422: dnewmpar=matrix(1,nlstate,1,npar);
6423: doldm=matrix(1,nlstate,1,nlstate);
6424:
6425: hstepm=1*YEARM; /* Every year of age */
6426: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6427: agelim = AGEINF;
6428: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6429: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6430: if (stepm >= YEARM) hstepm=1;
6431: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6432: gradg=matrix(1,npar,1,nlstate);
6433: mgp=matrix(1,npar,1,nlstate);
6434: mgm=matrix(1,npar,1,nlstate);
6435: gp=vector(1,nlstate);
6436: gm=vector(1,nlstate);
6437:
6438: for(theta=1; theta <=npar; theta++){
6439: for(i=1; i<=npar; i++){ /* Computes gradient */
6440: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6441: }
6442: if(mobilavproj > 0 )
6443: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6444: else
6445: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6446: for(i=1;i<=nlstate;i++){
6447: gp[i] = bprlim[i][i];
6448: mgp[theta][i] = bprlim[i][i];
6449: }
6450: for(i=1; i<=npar; i++) /* Computes gradient */
6451: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6452: if(mobilavproj > 0 )
6453: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6454: else
6455: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6456: for(i=1;i<=nlstate;i++){
6457: gm[i] = bprlim[i][i];
6458: mgm[theta][i] = bprlim[i][i];
6459: }
6460: for(i=1;i<=nlstate;i++)
6461: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6462: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6463: } /* End theta */
6464:
6465: trgradg =matrix(1,nlstate,1,npar);
6466:
6467: for(j=1; j<=nlstate;j++)
6468: for(theta=1; theta <=npar; theta++)
6469: trgradg[j][theta]=gradg[theta][j];
6470: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6471: /* printf("\nmgm mgp %d ",(int)age); */
6472: /* for(j=1; j<=nlstate;j++){ */
6473: /* printf(" %d ",j); */
6474: /* for(theta=1; theta <=npar; theta++) */
6475: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6476: /* printf("\n "); */
6477: /* } */
6478: /* } */
6479: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6480: /* printf("\n gradg %d ",(int)age); */
6481: /* for(j=1; j<=nlstate;j++){ */
6482: /* printf("%d ",j); */
6483: /* for(theta=1; theta <=npar; theta++) */
6484: /* printf("%d %lf ",theta,gradg[theta][j]); */
6485: /* printf("\n "); */
6486: /* } */
6487: /* } */
6488:
6489: for(i=1;i<=nlstate;i++)
6490: varbpl[i][(int)age] =0.;
6491: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6492: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6493: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6494: }else{
6495: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6496: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6497: }
6498: for(i=1;i<=nlstate;i++)
6499: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6500:
6501: fprintf(ficresvbl,"%.0f ",age );
6502: if(nresult >=1)
6503: fprintf(ficresvbl,"%d ",nres );
6504: for(i=1; i<=nlstate;i++)
6505: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6506: fprintf(ficresvbl,"\n");
6507: free_vector(gp,1,nlstate);
6508: free_vector(gm,1,nlstate);
6509: free_matrix(mgm,1,npar,1,nlstate);
6510: free_matrix(mgp,1,npar,1,nlstate);
6511: free_matrix(gradg,1,npar,1,nlstate);
6512: free_matrix(trgradg,1,nlstate,1,npar);
6513: } /* End age */
6514:
6515: free_vector(xp,1,npar);
6516: free_matrix(doldm,1,nlstate,1,npar);
6517: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6518:
6519: }
6520:
6521: /************ Variance of one-step probabilities ******************/
6522: 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 6523: {
6524: int i, j=0, k1, l1, tj;
6525: int k2, l2, j1, z1;
6526: int k=0, l;
6527: int first=1, first1, first2;
6528: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6529: double **dnewm,**doldm;
6530: double *xp;
6531: double *gp, *gm;
6532: double **gradg, **trgradg;
6533: double **mu;
6534: double age, cov[NCOVMAX+1];
6535: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6536: int theta;
6537: char fileresprob[FILENAMELENGTH];
6538: char fileresprobcov[FILENAMELENGTH];
6539: char fileresprobcor[FILENAMELENGTH];
6540: double ***varpij;
6541:
6542: strcpy(fileresprob,"PROB_");
6543: strcat(fileresprob,fileres);
6544: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6545: printf("Problem with resultfile: %s\n", fileresprob);
6546: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6547: }
6548: strcpy(fileresprobcov,"PROBCOV_");
6549: strcat(fileresprobcov,fileresu);
6550: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6551: printf("Problem with resultfile: %s\n", fileresprobcov);
6552: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6553: }
6554: strcpy(fileresprobcor,"PROBCOR_");
6555: strcat(fileresprobcor,fileresu);
6556: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6557: printf("Problem with resultfile: %s\n", fileresprobcor);
6558: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6559: }
6560: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6561: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6562: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6563: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6564: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6565: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6566: pstamp(ficresprob);
6567: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6568: fprintf(ficresprob,"# Age");
6569: pstamp(ficresprobcov);
6570: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6571: fprintf(ficresprobcov,"# Age");
6572: pstamp(ficresprobcor);
6573: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6574: fprintf(ficresprobcor,"# Age");
1.126 brouard 6575:
6576:
1.222 brouard 6577: for(i=1; i<=nlstate;i++)
6578: for(j=1; j<=(nlstate+ndeath);j++){
6579: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6580: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6581: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6582: }
6583: /* fprintf(ficresprob,"\n");
6584: fprintf(ficresprobcov,"\n");
6585: fprintf(ficresprobcor,"\n");
6586: */
6587: xp=vector(1,npar);
6588: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6589: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6590: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6591: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6592: first=1;
6593: fprintf(ficgp,"\n# Routine varprob");
6594: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6595: fprintf(fichtm,"\n");
6596:
1.288 brouard 6597: 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 6598: 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);
6599: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6600: and drawn. It helps understanding how is the covariance between two incidences.\
6601: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6602: 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 6603: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6604: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6605: standard deviations wide on each axis. <br>\
6606: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6607: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6608: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6609:
1.222 brouard 6610: cov[1]=1;
6611: /* tj=cptcoveff; */
1.225 brouard 6612: tj = (int) pow(2,cptcoveff);
1.222 brouard 6613: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6614: j1=0;
1.224 brouard 6615: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6616: if (cptcovn>0) {
6617: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6618: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6619: fprintf(ficresprob, "**********\n#\n");
6620: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6621: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6622: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6623:
1.222 brouard 6624: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6625: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6626: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6627:
6628:
1.222 brouard 6629: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6630: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6631: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6632:
1.222 brouard 6633: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6634: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6635: fprintf(ficresprobcor, "**********\n#");
6636: if(invalidvarcomb[j1]){
6637: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6638: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6639: continue;
6640: }
6641: }
6642: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6643: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6644: gp=vector(1,(nlstate)*(nlstate+ndeath));
6645: gm=vector(1,(nlstate)*(nlstate+ndeath));
6646: for (age=bage; age<=fage; age ++){
6647: cov[2]=age;
6648: if(nagesqr==1)
6649: cov[3]= age*age;
6650: for (k=1; k<=cptcovn;k++) {
6651: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6652: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6653: * 1 1 1 1 1
6654: * 2 2 1 1 1
6655: * 3 1 2 1 1
6656: */
6657: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6658: }
6659: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6660: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6661: for (k=1; k<=cptcovprod;k++)
6662: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6663:
6664:
1.222 brouard 6665: for(theta=1; theta <=npar; theta++){
6666: for(i=1; i<=npar; i++)
6667: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6668:
1.222 brouard 6669: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6670:
1.222 brouard 6671: k=0;
6672: for(i=1; i<= (nlstate); i++){
6673: for(j=1; j<=(nlstate+ndeath);j++){
6674: k=k+1;
6675: gp[k]=pmmij[i][j];
6676: }
6677: }
1.220 brouard 6678:
1.222 brouard 6679: for(i=1; i<=npar; i++)
6680: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6681:
1.222 brouard 6682: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6683: k=0;
6684: for(i=1; i<=(nlstate); i++){
6685: for(j=1; j<=(nlstate+ndeath);j++){
6686: k=k+1;
6687: gm[k]=pmmij[i][j];
6688: }
6689: }
1.220 brouard 6690:
1.222 brouard 6691: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6692: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6693: }
1.126 brouard 6694:
1.222 brouard 6695: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6696: for(theta=1; theta <=npar; theta++)
6697: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6698:
1.222 brouard 6699: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6700: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6701:
1.222 brouard 6702: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6703:
1.222 brouard 6704: k=0;
6705: for(i=1; i<=(nlstate); i++){
6706: for(j=1; j<=(nlstate+ndeath);j++){
6707: k=k+1;
6708: mu[k][(int) age]=pmmij[i][j];
6709: }
6710: }
6711: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6712: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6713: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6714:
1.222 brouard 6715: /*printf("\n%d ",(int)age);
6716: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6717: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6718: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6719: }*/
1.220 brouard 6720:
1.222 brouard 6721: fprintf(ficresprob,"\n%d ",(int)age);
6722: fprintf(ficresprobcov,"\n%d ",(int)age);
6723: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6724:
1.222 brouard 6725: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6726: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6727: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6728: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6729: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6730: }
6731: i=0;
6732: for (k=1; k<=(nlstate);k++){
6733: for (l=1; l<=(nlstate+ndeath);l++){
6734: i++;
6735: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6736: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6737: for (j=1; j<=i;j++){
6738: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6739: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6740: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6741: }
6742: }
6743: }/* end of loop for state */
6744: } /* end of loop for age */
6745: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6746: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6747: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6748: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6749:
6750: /* Confidence intervalle of pij */
6751: /*
6752: fprintf(ficgp,"\nunset parametric;unset label");
6753: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6754: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6755: 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);
6756: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6757: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6758: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6759: */
6760:
6761: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6762: first1=1;first2=2;
6763: for (k2=1; k2<=(nlstate);k2++){
6764: for (l2=1; l2<=(nlstate+ndeath);l2++){
6765: if(l2==k2) continue;
6766: j=(k2-1)*(nlstate+ndeath)+l2;
6767: for (k1=1; k1<=(nlstate);k1++){
6768: for (l1=1; l1<=(nlstate+ndeath);l1++){
6769: if(l1==k1) continue;
6770: i=(k1-1)*(nlstate+ndeath)+l1;
6771: if(i<=j) continue;
6772: for (age=bage; age<=fage; age ++){
6773: if ((int)age %5==0){
6774: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6775: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6776: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6777: mu1=mu[i][(int) age]/stepm*YEARM ;
6778: mu2=mu[j][(int) age]/stepm*YEARM;
6779: c12=cv12/sqrt(v1*v2);
6780: /* Computing eigen value of matrix of covariance */
6781: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6782: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6783: if ((lc2 <0) || (lc1 <0) ){
6784: if(first2==1){
6785: first1=0;
6786: 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);
6787: }
6788: 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);
6789: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6790: /* lc2=fabs(lc2); */
6791: }
1.220 brouard 6792:
1.222 brouard 6793: /* Eigen vectors */
1.280 brouard 6794: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6795: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6796: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6797: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6798: }else
6799: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6800: /*v21=sqrt(1.-v11*v11); *//* error */
6801: v21=(lc1-v1)/cv12*v11;
6802: v12=-v21;
6803: v22=v11;
6804: tnalp=v21/v11;
6805: if(first1==1){
6806: first1=0;
6807: 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);
6808: }
6809: 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);
6810: /*printf(fignu*/
6811: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6812: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6813: if(first==1){
6814: first=0;
6815: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6816: fprintf(ficgp,"\nset parametric;unset label");
6817: 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);
6818: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6819: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6820: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6821: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6822: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6823: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6824: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6825: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6826: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6827: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6828: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6829: 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 6830: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6831: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6832: }else{
6833: first=0;
6834: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6835: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6836: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6837: 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 6838: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6839: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6840: }/* if first */
6841: } /* age mod 5 */
6842: } /* end loop age */
6843: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6844: first=1;
6845: } /*l12 */
6846: } /* k12 */
6847: } /*l1 */
6848: }/* k1 */
6849: } /* loop on combination of covariates j1 */
6850: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6851: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6852: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6853: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6854: free_vector(xp,1,npar);
6855: fclose(ficresprob);
6856: fclose(ficresprobcov);
6857: fclose(ficresprobcor);
6858: fflush(ficgp);
6859: fflush(fichtmcov);
6860: }
1.126 brouard 6861:
6862:
6863: /******************* Printing html file ***********/
1.201 brouard 6864: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6865: int lastpass, int stepm, int weightopt, char model[],\
6866: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6867: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6868: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6869: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6870: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6871:
6872: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6873: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6874: </ul>");
1.237 brouard 6875: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6876: </ul>", model);
1.214 brouard 6877: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6878: 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",
6879: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6880: 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 6881: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6882: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6883: fprintf(fichtm,"\
6884: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6885: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6886: fprintf(fichtm,"\
1.217 brouard 6887: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6888: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6889: fprintf(fichtm,"\
1.288 brouard 6890: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6891: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6892: fprintf(fichtm,"\
1.288 brouard 6893: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6894: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6895: fprintf(fichtm,"\
1.211 brouard 6896: - (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 6897: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6898: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6899: if(prevfcast==1){
6900: fprintf(fichtm,"\
6901: - Prevalence projections by age and states: \
1.201 brouard 6902: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6903: }
1.126 brouard 6904:
6905:
1.225 brouard 6906: m=pow(2,cptcoveff);
1.222 brouard 6907: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6908:
1.264 brouard 6909: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6910:
6911: jj1=0;
6912:
6913: fprintf(fichtm," \n<ul>");
6914: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6915: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6916: if(m != 1 && TKresult[nres]!= k1)
6917: continue;
6918: jj1++;
6919: if (cptcovn > 0) {
6920: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6921: for (cpt=1; cpt<=cptcoveff;cpt++){
6922: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6923: }
6924: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6925: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6926: }
6927: fprintf(fichtm,"\">");
6928:
6929: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6930: fprintf(fichtm,"************ Results for covariates");
6931: for (cpt=1; cpt<=cptcoveff;cpt++){
6932: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6933: }
6934: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6935: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6936: }
6937: if(invalidvarcomb[k1]){
6938: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6939: continue;
6940: }
6941: fprintf(fichtm,"</a></li>");
6942: } /* cptcovn >0 */
6943: }
6944: fprintf(fichtm," \n</ul>");
6945:
1.222 brouard 6946: jj1=0;
1.237 brouard 6947:
6948: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6949: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6950: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6951: continue;
1.220 brouard 6952:
1.222 brouard 6953: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6954: jj1++;
6955: if (cptcovn > 0) {
1.264 brouard 6956: fprintf(fichtm,"\n<p><a name=\"rescov");
6957: for (cpt=1; cpt<=cptcoveff;cpt++){
6958: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6959: }
6960: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6961: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6962: }
6963: fprintf(fichtm,"\"</a>");
6964:
1.222 brouard 6965: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6966: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6967: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6968: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6969: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6970: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6971: }
1.237 brouard 6972: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6973: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6974: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6975: }
6976:
1.230 brouard 6977: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6978: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6979: if(invalidvarcomb[k1]){
6980: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6981: printf("\nCombination (%d) ignored because no cases \n",k1);
6982: continue;
6983: }
6984: }
6985: /* aij, bij */
1.259 brouard 6986: 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 6987: <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 6988: /* Pij */
1.241 brouard 6989: 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> \
6990: <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 6991: /* Quasi-incidences */
6992: 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 6993: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6994: 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 6995: 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> \
6996: <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 6997: /* Survival functions (period) in state j */
6998: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 6999: 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 7000: <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 7001: }
7002: /* State specific survival functions (period) */
7003: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7004: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7005: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7006: <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 7007: }
1.288 brouard 7008: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7009: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7010: 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> \
7011: <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 7012: }
1.296 brouard 7013: if(prevbcast==1){
1.288 brouard 7014: /* Backward prevalence in each health state */
1.222 brouard 7015: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7016: 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 7017: <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 7018: }
1.217 brouard 7019: }
1.222 brouard 7020: if(prevfcast==1){
1.288 brouard 7021: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7022: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7023: 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><br> \
1.296 brouard 7024: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7025: }
7026: }
1.296 brouard 7027: if(prevbcast==1){
1.268 brouard 7028: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7029: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7030: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7031: 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 \
7032: 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) \
7033: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7034: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7035: }
7036: }
1.220 brouard 7037:
1.222 brouard 7038: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7039: 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> <br> \
7040: <img src=\"%s_%d-%d-%d.svg\">",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.222 brouard 7041: }
7042: /* } /\* end i1 *\/ */
7043: }/* End k1 */
7044: fprintf(fichtm,"</ul>");
1.126 brouard 7045:
1.222 brouard 7046: fprintf(fichtm,"\
1.126 brouard 7047: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7048: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7049: - 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 7050: But because parameters are usually highly correlated (a higher incidence of disability \
7051: and a higher incidence of recovery can give very close observed transition) it might \
7052: be very useful to look not only at linear confidence intervals estimated from the \
7053: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7054: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7055: covariance matrix of the one-step probabilities. \
7056: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7057:
1.222 brouard 7058: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7059: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7060: fprintf(fichtm,"\
1.126 brouard 7061: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7062: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7063:
1.222 brouard 7064: fprintf(fichtm,"\
1.126 brouard 7065: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7066: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7067: fprintf(fichtm,"\
1.126 brouard 7068: - 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): \
7069: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7070: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7071: fprintf(fichtm,"\
1.126 brouard 7072: - (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): \
7073: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7074: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7075: fprintf(fichtm,"\
1.288 brouard 7076: - 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 7077: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7078: fprintf(fichtm,"\
1.128 brouard 7079: - 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 7080: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7081: fprintf(fichtm,"\
1.288 brouard 7082: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7083: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7084:
7085: /* if(popforecast==1) fprintf(fichtm,"\n */
7086: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7087: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7088: /* <br>",fileres,fileres,fileres,fileres); */
7089: /* else */
7090: /* 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 7091: fflush(fichtm);
7092: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7093:
1.225 brouard 7094: m=pow(2,cptcoveff);
1.222 brouard 7095: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7096:
1.222 brouard 7097: jj1=0;
1.237 brouard 7098:
1.241 brouard 7099: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7100: for(k1=1; k1<=m;k1++){
1.253 brouard 7101: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7102: continue;
1.222 brouard 7103: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7104: jj1++;
1.126 brouard 7105: if (cptcovn > 0) {
7106: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7107: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7108: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7109: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7110: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7111: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7112: }
7113:
1.126 brouard 7114: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7115:
1.222 brouard 7116: if(invalidvarcomb[k1]){
7117: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7118: continue;
7119: }
1.126 brouard 7120: }
7121: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7122: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7123: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
1.258 brouard 7124: <img src=\"%s_%d-%d-%d.svg\">",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 7125: }
7126: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7127: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7128: true period expectancies (those weighted with period prevalences are also\
7129: drawn in addition to the population based expectancies computed using\
1.241 brouard 7130: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7131: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7132: /* } /\* end i1 *\/ */
7133: }/* End k1 */
1.241 brouard 7134: }/* End nres */
1.222 brouard 7135: fprintf(fichtm,"</ul>");
7136: fflush(fichtm);
1.126 brouard 7137: }
7138:
7139: /******************* Gnuplot file **************/
1.296 brouard 7140: 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 7141:
7142: char dirfileres[132],optfileres[132];
1.264 brouard 7143: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7144: 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 7145: int lv=0, vlv=0, kl=0;
1.130 brouard 7146: int ng=0;
1.201 brouard 7147: int vpopbased;
1.223 brouard 7148: int ioffset; /* variable offset for columns */
1.270 brouard 7149: int iyearc=1; /* variable column for year of projection */
7150: int iagec=1; /* variable column for age of projection */
1.235 brouard 7151: int nres=0; /* Index of resultline */
1.266 brouard 7152: int istart=1; /* For starting graphs in projections */
1.219 brouard 7153:
1.126 brouard 7154: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7155: /* printf("Problem with file %s",optionfilegnuplot); */
7156: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7157: /* } */
7158:
7159: /*#ifdef windows */
7160: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7161: /*#endif */
1.225 brouard 7162: m=pow(2,cptcoveff);
1.126 brouard 7163:
1.274 brouard 7164: /* diagram of the model */
7165: fprintf(ficgp,"\n#Diagram of the model \n");
7166: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7167: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7168: 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);
7169:
7170: 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);
7171: fprintf(ficgp,"\n#show arrow\nunset label\n");
7172: 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);
7173: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7174: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7175: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7176: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7177:
1.202 brouard 7178: /* Contribution to likelihood */
7179: /* Plot the probability implied in the likelihood */
1.223 brouard 7180: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7181: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7182: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7183: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7184: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7185: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7186: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7187: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7188: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7189: 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));
7190: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7191: 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));
7192: for (i=1; i<= nlstate ; i ++) {
7193: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7194: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7195: 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);
7196: for (j=2; j<= nlstate+ndeath ; j ++) {
7197: 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);
7198: }
7199: fprintf(ficgp,";\nset out; unset ylabel;\n");
7200: }
7201: /* 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 */
7202: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7203: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7204: fprintf(ficgp,"\nset out;unset log\n");
7205: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7206:
1.126 brouard 7207: strcpy(dirfileres,optionfilefiname);
7208: strcpy(optfileres,"vpl");
1.223 brouard 7209: /* 1eme*/
1.238 brouard 7210: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7211: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7212: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7213: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7214: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7215: continue;
7216: /* We are interested in selected combination by the resultline */
1.246 brouard 7217: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7218: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7219: strcpy(gplotlabel,"(");
1.238 brouard 7220: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7221: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7222: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7223: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7224: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7225: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7226: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7227: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7228: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7229: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7230: }
7231: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7232: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7233: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7234: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7235: }
7236: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7237: /* printf("\n#\n"); */
1.238 brouard 7238: fprintf(ficgp,"\n#\n");
7239: if(invalidvarcomb[k1]){
1.260 brouard 7240: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7241: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7242: continue;
7243: }
1.235 brouard 7244:
1.241 brouard 7245: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7246: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7247: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7248: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7249: 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);
7250: /* 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); */
7251: /* k1-1 error should be nres-1*/
1.238 brouard 7252: for (i=1; i<= nlstate ; i ++) {
7253: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7254: else fprintf(ficgp," %%*lf (%%*lf)");
7255: }
1.288 brouard 7256: 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 7257: for (i=1; i<= nlstate ; i ++) {
7258: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7259: else fprintf(ficgp," %%*lf (%%*lf)");
7260: }
1.260 brouard 7261: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7262: for (i=1; i<= nlstate ; i ++) {
7263: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7264: else fprintf(ficgp," %%*lf (%%*lf)");
7265: }
1.265 brouard 7266: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7267:
7268: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7269: if(cptcoveff ==0){
1.271 brouard 7270: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7271: }else{
7272: kl=0;
7273: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7274: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7275: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7276: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7277: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7278: vlv= nbcode[Tvaraff[k]][lv];
7279: kl++;
7280: /* 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 *\/ */
7281: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7282: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7283: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7284: if(k==cptcoveff){
7285: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7286: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7287: }else{
7288: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7289: kl++;
7290: }
7291: } /* end covariate */
7292: } /* end if no covariate */
7293:
1.296 brouard 7294: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7295: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7296: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7297: if(cptcoveff ==0){
1.245 brouard 7298: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7299: }else{
7300: kl=0;
7301: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7302: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7303: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7304: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7305: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7306: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7307: kl++;
1.238 brouard 7308: /* 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 *\/ */
7309: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7310: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7311: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7312: if(k==cptcoveff){
1.245 brouard 7313: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7314: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7315: }else{
7316: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7317: kl++;
7318: }
7319: } /* end covariate */
7320: } /* end if no covariate */
1.296 brouard 7321: if(prevbcast == 1){
1.268 brouard 7322: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7323: /* k1-1 error should be nres-1*/
7324: for (i=1; i<= nlstate ; i ++) {
7325: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7326: else fprintf(ficgp," %%*lf (%%*lf)");
7327: }
1.271 brouard 7328: 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 7329: for (i=1; i<= nlstate ; i ++) {
7330: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7331: else fprintf(ficgp," %%*lf (%%*lf)");
7332: }
1.276 brouard 7333: 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 7334: for (i=1; i<= nlstate ; i ++) {
7335: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7336: else fprintf(ficgp," %%*lf (%%*lf)");
7337: }
1.274 brouard 7338: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7339: } /* end if backprojcast */
1.296 brouard 7340: } /* end if prevbcast */
1.276 brouard 7341: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7342: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7343: } /* nres */
1.201 brouard 7344: } /* k1 */
7345: } /* cpt */
1.235 brouard 7346:
7347:
1.126 brouard 7348: /*2 eme*/
1.238 brouard 7349: for (k1=1; k1<= m ; k1 ++){
7350: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7351: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7352: continue;
7353: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7354: strcpy(gplotlabel,"(");
1.238 brouard 7355: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7356: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7357: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7358: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7359: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7360: vlv= nbcode[Tvaraff[k]][lv];
7361: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7362: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7363: }
1.237 brouard 7364: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7365: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7366: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7367: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7368: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7369: }
1.264 brouard 7370: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7371: fprintf(ficgp,"\n#\n");
1.223 brouard 7372: if(invalidvarcomb[k1]){
7373: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7374: continue;
7375: }
1.219 brouard 7376:
1.241 brouard 7377: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7378: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7379: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7380: if(vpopbased==0){
1.238 brouard 7381: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7382: }else
1.238 brouard 7383: fprintf(ficgp,"\nreplot ");
7384: for (i=1; i<= nlstate+1 ; i ++) {
7385: k=2*i;
1.261 brouard 7386: 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 7387: for (j=1; j<= nlstate+1 ; j ++) {
7388: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7389: else fprintf(ficgp," %%*lf (%%*lf)");
7390: }
7391: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7392: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7393: 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 7394: for (j=1; j<= nlstate+1 ; j ++) {
7395: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7396: else fprintf(ficgp," %%*lf (%%*lf)");
7397: }
7398: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7399: 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 7400: for (j=1; j<= nlstate+1 ; j ++) {
7401: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7402: else fprintf(ficgp," %%*lf (%%*lf)");
7403: }
7404: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7405: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7406: } /* state */
7407: } /* vpopbased */
1.264 brouard 7408: 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 7409: } /* end nres */
7410: } /* k1 end 2 eme*/
7411:
7412:
7413: /*3eme*/
7414: for (k1=1; k1<= m ; k1 ++){
7415: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7416: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7417: continue;
7418:
7419: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7420: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7421: strcpy(gplotlabel,"(");
1.238 brouard 7422: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7423: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7424: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7425: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7426: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7427: vlv= nbcode[Tvaraff[k]][lv];
7428: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7429: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7430: }
7431: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7432: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7433: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7434: }
1.264 brouard 7435: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7436: fprintf(ficgp,"\n#\n");
7437: if(invalidvarcomb[k1]){
7438: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7439: continue;
7440: }
7441:
7442: /* k=2+nlstate*(2*cpt-2); */
7443: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7444: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7445: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7446: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7447: 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 7448: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7449: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7450: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7451: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7452: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7453: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7454:
1.238 brouard 7455: */
7456: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7457: 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 7458: /* 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 7459:
1.238 brouard 7460: }
1.261 brouard 7461: 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 7462: }
1.264 brouard 7463: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7464: } /* end nres */
7465: } /* end kl 3eme */
1.126 brouard 7466:
1.223 brouard 7467: /* 4eme */
1.201 brouard 7468: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7469: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7470: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7471: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7472: continue;
1.238 brouard 7473: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7474: strcpy(gplotlabel,"(");
1.238 brouard 7475: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7476: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7477: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7478: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7479: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7480: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7481: vlv= nbcode[Tvaraff[k]][lv];
7482: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7483: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7484: }
7485: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7486: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7487: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7488: }
1.264 brouard 7489: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7490: fprintf(ficgp,"\n#\n");
7491: if(invalidvarcomb[k1]){
7492: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7493: continue;
1.223 brouard 7494: }
1.238 brouard 7495:
1.241 brouard 7496: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7497: 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 7498: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7499: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7500: k=3;
7501: for (i=1; i<= nlstate ; i ++){
7502: if(i==1){
7503: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7504: }else{
7505: fprintf(ficgp,", '' ");
7506: }
7507: l=(nlstate+ndeath)*(i-1)+1;
7508: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7509: for (j=2; j<= nlstate+ndeath ; j ++)
7510: fprintf(ficgp,"+$%d",k+l+j-1);
7511: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7512: } /* nlstate */
1.264 brouard 7513: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7514: } /* end cpt state*/
7515: } /* end nres */
7516: } /* end covariate k1 */
7517:
1.220 brouard 7518: /* 5eme */
1.201 brouard 7519: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7520: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7521: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7522: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7523: continue;
1.238 brouard 7524: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7525: strcpy(gplotlabel,"(");
1.238 brouard 7526: 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);
7527: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7528: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7529: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7530: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7531: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7532: vlv= nbcode[Tvaraff[k]][lv];
7533: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7534: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7535: }
7536: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7537: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7538: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7539: }
1.264 brouard 7540: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7541: fprintf(ficgp,"\n#\n");
7542: if(invalidvarcomb[k1]){
7543: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7544: continue;
7545: }
1.227 brouard 7546:
1.241 brouard 7547: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7548: 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 7549: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7550: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7551: k=3;
7552: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7553: if(j==1)
7554: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7555: else
7556: fprintf(ficgp,", '' ");
7557: l=(nlstate+ndeath)*(cpt-1) +j;
7558: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7559: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7560: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7561: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7562: } /* nlstate */
7563: fprintf(ficgp,", '' ");
7564: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7565: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7566: l=(nlstate+ndeath)*(cpt-1) +j;
7567: if(j < nlstate)
7568: fprintf(ficgp,"$%d +",k+l);
7569: else
7570: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7571: }
1.264 brouard 7572: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7573: } /* end cpt state*/
7574: } /* end covariate */
7575: } /* end nres */
1.227 brouard 7576:
1.220 brouard 7577: /* 6eme */
1.202 brouard 7578: /* CV preval stable (period) for each covariate */
1.237 brouard 7579: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7580: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7581: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7582: continue;
1.255 brouard 7583: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7584: strcpy(gplotlabel,"(");
1.288 brouard 7585: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7586: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7587: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7588: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7589: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7590: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7591: vlv= nbcode[Tvaraff[k]][lv];
7592: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7593: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7594: }
1.237 brouard 7595: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7596: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7597: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7598: }
1.264 brouard 7599: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7600: fprintf(ficgp,"\n#\n");
1.223 brouard 7601: if(invalidvarcomb[k1]){
1.227 brouard 7602: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7603: continue;
1.223 brouard 7604: }
1.227 brouard 7605:
1.241 brouard 7606: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7607: 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 7608: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7609: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7610: k=3; /* Offset */
1.255 brouard 7611: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7612: if(i==1)
7613: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7614: else
7615: fprintf(ficgp,", '' ");
1.255 brouard 7616: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7617: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7618: for (j=2; j<= nlstate ; j ++)
7619: fprintf(ficgp,"+$%d",k+l+j-1);
7620: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7621: } /* nlstate */
1.264 brouard 7622: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7623: } /* end cpt state*/
7624: } /* end covariate */
1.227 brouard 7625:
7626:
1.220 brouard 7627: /* 7eme */
1.296 brouard 7628: if(prevbcast == 1){
1.288 brouard 7629: /* CV backward prevalence for each covariate */
1.237 brouard 7630: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7631: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7632: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7633: continue;
1.268 brouard 7634: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7635: strcpy(gplotlabel,"(");
1.288 brouard 7636: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7637: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7638: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7639: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7640: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7641: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7642: vlv= nbcode[Tvaraff[k]][lv];
7643: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7644: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7645: }
1.237 brouard 7646: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7647: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7648: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7649: }
1.264 brouard 7650: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7651: fprintf(ficgp,"\n#\n");
7652: if(invalidvarcomb[k1]){
7653: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7654: continue;
7655: }
7656:
1.241 brouard 7657: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7658: 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 7659: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7660: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7661: k=3; /* Offset */
1.268 brouard 7662: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7663: if(i==1)
7664: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7665: else
7666: fprintf(ficgp,", '' ");
7667: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7668: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7669: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7670: /* 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 7671: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7672: /* for (j=2; j<= nlstate ; j ++) */
7673: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7674: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7675: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7676: } /* nlstate */
1.264 brouard 7677: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7678: } /* end cpt state*/
7679: } /* end covariate */
1.296 brouard 7680: } /* End if prevbcast */
1.218 brouard 7681:
1.223 brouard 7682: /* 8eme */
1.218 brouard 7683: if(prevfcast==1){
1.288 brouard 7684: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7685:
1.237 brouard 7686: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7687: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7688: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7689: continue;
1.211 brouard 7690: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7691: strcpy(gplotlabel,"(");
1.288 brouard 7692: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7693: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7694: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7695: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7696: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7697: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7698: vlv= nbcode[Tvaraff[k]][lv];
7699: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7700: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7701: }
1.237 brouard 7702: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7703: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7704: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7705: }
1.264 brouard 7706: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7707: fprintf(ficgp,"\n#\n");
7708: if(invalidvarcomb[k1]){
7709: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7710: continue;
7711: }
7712:
7713: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7714: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7715: 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 7716: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7717: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7718:
7719: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7720: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7721: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7722: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7723: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7724: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7725: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7726: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7727: if(i==istart){
1.227 brouard 7728: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7729: }else{
7730: fprintf(ficgp,",\\\n '' ");
7731: }
7732: if(cptcoveff ==0){ /* No covariate */
7733: ioffset=2; /* Age is in 2 */
7734: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7735: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7736: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7737: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7738: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7739: if(i==nlstate+1){
1.270 brouard 7740: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7741: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7742: fprintf(ficgp,",\\\n '' ");
7743: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7744: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7745: offyear, \
1.268 brouard 7746: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7747: }else
1.227 brouard 7748: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7749: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7750: }else{ /* more than 2 covariates */
1.270 brouard 7751: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7752: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7753: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7754: iyearc=ioffset-1;
7755: iagec=ioffset;
1.227 brouard 7756: fprintf(ficgp," u %d:(",ioffset);
7757: kl=0;
7758: strcpy(gplotcondition,"(");
7759: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7760: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7761: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7762: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7763: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7764: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7765: kl++;
7766: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7767: kl++;
7768: if(k <cptcoveff && cptcoveff>1)
7769: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7770: }
7771: strcpy(gplotcondition+strlen(gplotcondition),")");
7772: /* 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 *\/ */
7773: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7774: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7775: /* '' 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*/
7776: if(i==nlstate+1){
1.270 brouard 7777: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7778: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7779: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7780: fprintf(ficgp," u %d:(",iagec);
7781: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7782: iyearc, iagec, offyear, \
7783: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7784: /* '' 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 7785: }else{
7786: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7787: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7788: }
7789: } /* end if covariate */
7790: } /* nlstate */
1.264 brouard 7791: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7792: } /* end cpt state*/
7793: } /* end covariate */
7794: } /* End if prevfcast */
1.227 brouard 7795:
1.296 brouard 7796: if(prevbcast==1){
1.268 brouard 7797: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7798:
7799: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7800: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7801: if(m != 1 && TKresult[nres]!= k1)
7802: continue;
7803: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7804: strcpy(gplotlabel,"(");
7805: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7806: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7807: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7808: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7809: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7810: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7811: vlv= nbcode[Tvaraff[k]][lv];
7812: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7813: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7814: }
7815: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7816: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7817: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7818: }
7819: strcpy(gplotlabel+strlen(gplotlabel),")");
7820: fprintf(ficgp,"\n#\n");
7821: if(invalidvarcomb[k1]){
7822: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7823: continue;
7824: }
7825:
7826: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7827: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7828: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7829: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7830: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7831:
7832: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7833: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7834: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7835: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7836: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7837: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7838: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7839: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7840: if(i==istart){
7841: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7842: }else{
7843: fprintf(ficgp,",\\\n '' ");
7844: }
7845: if(cptcoveff ==0){ /* No covariate */
7846: ioffset=2; /* Age is in 2 */
7847: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7848: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7849: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7850: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7851: fprintf(ficgp," u %d:(", ioffset);
7852: if(i==nlstate+1){
1.270 brouard 7853: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7854: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7855: fprintf(ficgp,",\\\n '' ");
7856: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7857: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7858: offbyear, \
7859: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7860: }else
7861: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7862: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7863: }else{ /* more than 2 covariates */
1.270 brouard 7864: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7865: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7866: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7867: iyearc=ioffset-1;
7868: iagec=ioffset;
1.268 brouard 7869: fprintf(ficgp," u %d:(",ioffset);
7870: kl=0;
7871: strcpy(gplotcondition,"(");
7872: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7873: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7874: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7875: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7876: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7877: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7878: kl++;
7879: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7880: kl++;
7881: if(k <cptcoveff && cptcoveff>1)
7882: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7883: }
7884: strcpy(gplotcondition+strlen(gplotcondition),")");
7885: /* 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 *\/ */
7886: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7887: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7888: /* '' 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*/
7889: if(i==nlstate+1){
1.270 brouard 7890: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7891: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7892: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7893: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7894: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7895: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7896: iyearc,iagec,offbyear, \
7897: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7898: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7899: }else{
7900: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7901: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7902: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7903: }
7904: } /* end if covariate */
7905: } /* nlstate */
7906: fprintf(ficgp,"\nset out; unset label;\n");
7907: } /* end cpt state*/
7908: } /* end covariate */
1.296 brouard 7909: } /* End if prevbcast */
1.268 brouard 7910:
1.227 brouard 7911:
1.238 brouard 7912: /* 9eme writing MLE parameters */
7913: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7914: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7915: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7916: for(k=1; k <=(nlstate+ndeath); k++){
7917: if (k != i) {
1.227 brouard 7918: fprintf(ficgp,"# current state %d\n",k);
7919: for(j=1; j <=ncovmodel; j++){
7920: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7921: jk++;
7922: }
7923: fprintf(ficgp,"\n");
1.126 brouard 7924: }
7925: }
1.223 brouard 7926: }
1.187 brouard 7927: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7928:
1.145 brouard 7929: /*goto avoid;*/
1.238 brouard 7930: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7931: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7932: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7933: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7934: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7935: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7936: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7937: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7938: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7939: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7940: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7941: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7942: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7943: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7944: fprintf(ficgp,"#\n");
1.223 brouard 7945: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7946: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7947: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7948: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7949: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7950: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7951: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7952: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7953: continue;
1.264 brouard 7954: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7955: strcpy(gplotlabel,"(");
1.276 brouard 7956: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 7957: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7958: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7959: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7960: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7961: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7962: vlv= nbcode[Tvaraff[k]][lv];
7963: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7964: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7965: }
1.237 brouard 7966: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7967: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7968: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7969: }
1.264 brouard 7970: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7971: fprintf(ficgp,"\n#\n");
1.264 brouard 7972: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 7973: fprintf(ficgp,"\nset key outside ");
7974: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
7975: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7976: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7977: if (ng==1){
7978: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7979: fprintf(ficgp,"\nunset log y");
7980: }else if (ng==2){
7981: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7982: fprintf(ficgp,"\nset log y");
7983: }else if (ng==3){
7984: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7985: fprintf(ficgp,"\nset log y");
7986: }else
7987: fprintf(ficgp,"\nunset title ");
7988: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7989: i=1;
7990: for(k2=1; k2<=nlstate; k2++) {
7991: k3=i;
7992: for(k=1; k<=(nlstate+ndeath); k++) {
7993: if (k != k2){
7994: switch( ng) {
7995: case 1:
7996: if(nagesqr==0)
7997: fprintf(ficgp," p%d+p%d*x",i,i+1);
7998: else /* nagesqr =1 */
7999: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8000: break;
8001: case 2: /* ng=2 */
8002: if(nagesqr==0)
8003: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8004: else /* nagesqr =1 */
8005: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8006: break;
8007: case 3:
8008: if(nagesqr==0)
8009: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8010: else /* nagesqr =1 */
8011: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8012: break;
8013: }
8014: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8015: ijp=1; /* product no age */
8016: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8017: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8018: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8019: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8020: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8021: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8022: if(DummyV[j]==0){
8023: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8024: }else{ /* quantitative */
8025: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8026: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8027: }
8028: ij++;
1.237 brouard 8029: }
1.268 brouard 8030: }
8031: }else if(cptcovprod >0){
8032: if(j==Tprod[ijp]) { /* */
8033: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8034: if(ijp <=cptcovprod) { /* Product */
8035: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8036: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8037: /* 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)]); */
8038: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8039: }else{ /* Vn is dummy and Vm is quanti */
8040: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8041: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8042: }
8043: }else{ /* Vn*Vm Vn is quanti */
8044: if(DummyV[Tvard[ijp][2]]==0){
8045: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8046: }else{ /* Both quanti */
8047: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8048: }
1.237 brouard 8049: }
1.268 brouard 8050: ijp++;
1.237 brouard 8051: }
1.268 brouard 8052: } /* end Tprod */
1.237 brouard 8053: } else{ /* simple covariate */
1.264 brouard 8054: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8055: if(Dummy[j]==0){
8056: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8057: }else{ /* quantitative */
8058: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8059: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8060: }
1.237 brouard 8061: } /* end simple */
8062: } /* end j */
1.223 brouard 8063: }else{
8064: i=i-ncovmodel;
8065: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8066: fprintf(ficgp," (1.");
8067: }
1.227 brouard 8068:
1.223 brouard 8069: if(ng != 1){
8070: fprintf(ficgp,")/(1");
1.227 brouard 8071:
1.264 brouard 8072: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8073: if(nagesqr==0)
1.264 brouard 8074: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8075: else /* nagesqr =1 */
1.264 brouard 8076: 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 8077:
1.223 brouard 8078: ij=1;
8079: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8080: if(cptcovage >0){
8081: if((j-2)==Tage[ij]) { /* Bug valgrind */
8082: if(ij <=cptcovage) { /* Bug valgrind */
8083: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8084: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8085: ij++;
8086: }
8087: }
8088: }else
8089: 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 8090: }
8091: fprintf(ficgp,")");
8092: }
8093: fprintf(ficgp,")");
8094: if(ng ==2)
1.276 brouard 8095: 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 8096: else /* ng= 3 */
1.276 brouard 8097: 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 8098: }else{ /* end ng <> 1 */
8099: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8100: 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 8101: }
8102: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8103: fprintf(ficgp,",");
8104: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8105: fprintf(ficgp,",");
8106: i=i+ncovmodel;
8107: } /* end k */
8108: } /* end k2 */
1.276 brouard 8109: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8110: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8111: } /* end k1 */
1.223 brouard 8112: } /* end ng */
8113: /* avoid: */
8114: fflush(ficgp);
1.126 brouard 8115: } /* end gnuplot */
8116:
8117:
8118: /*************** Moving average **************/
1.219 brouard 8119: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8120: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8121:
1.222 brouard 8122: int i, cpt, cptcod;
8123: int modcovmax =1;
8124: int mobilavrange, mob;
8125: int iage=0;
1.288 brouard 8126: int firstA1=0, firstA2=0;
1.222 brouard 8127:
1.266 brouard 8128: double sum=0., sumr=0.;
1.222 brouard 8129: double age;
1.266 brouard 8130: double *sumnewp, *sumnewm, *sumnewmr;
8131: double *agemingood, *agemaxgood;
8132: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8133:
8134:
1.278 brouard 8135: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8136: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8137:
8138: sumnewp = vector(1,ncovcombmax);
8139: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8140: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8141: agemingood = vector(1,ncovcombmax);
1.266 brouard 8142: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8143: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8144: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8145:
8146: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8147: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8148: sumnewp[cptcod]=0.;
1.266 brouard 8149: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8150: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8151: }
8152: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8153:
1.266 brouard 8154: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8155: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8156: else mobilavrange=mobilav;
8157: for (age=bage; age<=fage; age++)
8158: for (i=1; i<=nlstate;i++)
8159: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8160: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8161: /* We keep the original values on the extreme ages bage, fage and for
8162: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8163: we use a 5 terms etc. until the borders are no more concerned.
8164: */
8165: for (mob=3;mob <=mobilavrange;mob=mob+2){
8166: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8167: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8168: sumnewm[cptcod]=0.;
8169: for (i=1; i<=nlstate;i++){
1.222 brouard 8170: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8171: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8172: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8173: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8174: }
8175: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8176: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8177: } /* end i */
8178: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8179: } /* end cptcod */
1.222 brouard 8180: }/* end age */
8181: }/* end mob */
1.266 brouard 8182: }else{
8183: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8184: return -1;
1.266 brouard 8185: }
8186:
8187: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8188: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8189: if(invalidvarcomb[cptcod]){
8190: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8191: continue;
8192: }
1.219 brouard 8193:
1.266 brouard 8194: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8195: sumnewm[cptcod]=0.;
8196: sumnewmr[cptcod]=0.;
8197: for (i=1; i<=nlstate;i++){
8198: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8199: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8200: }
8201: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8202: agemingoodr[cptcod]=age;
8203: }
8204: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8205: agemingood[cptcod]=age;
8206: }
8207: } /* age */
8208: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8209: sumnewm[cptcod]=0.;
1.266 brouard 8210: sumnewmr[cptcod]=0.;
1.222 brouard 8211: for (i=1; i<=nlstate;i++){
8212: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8213: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8214: }
8215: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8216: agemaxgoodr[cptcod]=age;
1.222 brouard 8217: }
8218: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8219: agemaxgood[cptcod]=age;
8220: }
8221: } /* age */
8222: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8223: /* but they will change */
1.288 brouard 8224: firstA1=0;firstA2=0;
1.266 brouard 8225: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8226: sumnewm[cptcod]=0.;
8227: sumnewmr[cptcod]=0.;
8228: for (i=1; i<=nlstate;i++){
8229: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8230: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8231: }
8232: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8233: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8234: agemaxgoodr[cptcod]=age; /* age min */
8235: for (i=1; i<=nlstate;i++)
8236: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8237: }else{ /* bad we change the value with the values of good ages */
8238: for (i=1; i<=nlstate;i++){
8239: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8240: } /* i */
8241: } /* end bad */
8242: }else{
8243: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8244: agemaxgood[cptcod]=age;
8245: }else{ /* bad we change the value with the values of good ages */
8246: for (i=1; i<=nlstate;i++){
8247: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8248: } /* i */
8249: } /* end bad */
8250: }/* end else */
8251: sum=0.;sumr=0.;
8252: for (i=1; i<=nlstate;i++){
8253: sum+=mobaverage[(int)age][i][cptcod];
8254: sumr+=probs[(int)age][i][cptcod];
8255: }
8256: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8257: if(!firstA1){
8258: firstA1=1;
8259: 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);
8260: }
8261: 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 8262: } /* end bad */
8263: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8264: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8265: if(!firstA2){
8266: firstA2=1;
8267: 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);
8268: }
8269: 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 8270: } /* end bad */
8271: }/* age */
1.266 brouard 8272:
8273: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8274: sumnewm[cptcod]=0.;
1.266 brouard 8275: sumnewmr[cptcod]=0.;
1.222 brouard 8276: for (i=1; i<=nlstate;i++){
8277: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8278: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8279: }
8280: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8281: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8282: agemingoodr[cptcod]=age;
8283: for (i=1; i<=nlstate;i++)
8284: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8285: }else{ /* bad we change the value with the values of good ages */
8286: for (i=1; i<=nlstate;i++){
8287: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8288: } /* i */
8289: } /* end bad */
8290: }else{
8291: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8292: agemingood[cptcod]=age;
8293: }else{ /* bad */
8294: for (i=1; i<=nlstate;i++){
8295: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8296: } /* i */
8297: } /* end bad */
8298: }/* end else */
8299: sum=0.;sumr=0.;
8300: for (i=1; i<=nlstate;i++){
8301: sum+=mobaverage[(int)age][i][cptcod];
8302: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8303: }
1.266 brouard 8304: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8305: 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 8306: } /* end bad */
8307: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8308: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8309: 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 8310: } /* end bad */
8311: }/* age */
1.266 brouard 8312:
1.222 brouard 8313:
8314: for (age=bage; age<=fage; age++){
1.235 brouard 8315: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8316: sumnewp[cptcod]=0.;
8317: sumnewm[cptcod]=0.;
8318: for (i=1; i<=nlstate;i++){
8319: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8320: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8321: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8322: }
8323: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8324: }
8325: /* printf("\n"); */
8326: /* } */
1.266 brouard 8327:
1.222 brouard 8328: /* brutal averaging */
1.266 brouard 8329: /* for (i=1; i<=nlstate;i++){ */
8330: /* for (age=1; age<=bage; age++){ */
8331: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8332: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8333: /* } */
8334: /* for (age=fage; age<=AGESUP; age++){ */
8335: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8336: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8337: /* } */
8338: /* } /\* end i status *\/ */
8339: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8340: /* for (age=1; age<=AGESUP; age++){ */
8341: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8342: /* mobaverage[(int)age][i][cptcod]=0.; */
8343: /* } */
8344: /* } */
1.222 brouard 8345: }/* end cptcod */
1.266 brouard 8346: free_vector(agemaxgoodr,1, ncovcombmax);
8347: free_vector(agemaxgood,1, ncovcombmax);
8348: free_vector(agemingood,1, ncovcombmax);
8349: free_vector(agemingoodr,1, ncovcombmax);
8350: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8351: free_vector(sumnewm,1, ncovcombmax);
8352: free_vector(sumnewp,1, ncovcombmax);
8353: return 0;
8354: }/* End movingaverage */
1.218 brouard 8355:
1.126 brouard 8356:
1.296 brouard 8357:
1.126 brouard 8358: /************** Forecasting ******************/
1.296 brouard 8359: /* 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)*/
8360: 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){
8361: /* dateintemean, mean date of interviews
8362: dateprojd, year, month, day of starting projection
8363: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8364: agemin, agemax range of age
8365: dateprev1 dateprev2 range of dates during which prevalence is computed
8366: */
1.296 brouard 8367: /* double anprojd, mprojd, jprojd; */
8368: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8369: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8370: double agec; /* generic age */
1.296 brouard 8371: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8372: double *popeffectif,*popcount;
8373: double ***p3mat;
1.218 brouard 8374: /* double ***mobaverage; */
1.126 brouard 8375: char fileresf[FILENAMELENGTH];
8376:
8377: agelim=AGESUP;
1.211 brouard 8378: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8379: in each health status at the date of interview (if between dateprev1 and dateprev2).
8380: We still use firstpass and lastpass as another selection.
8381: */
1.214 brouard 8382: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8383: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8384:
1.201 brouard 8385: strcpy(fileresf,"F_");
8386: strcat(fileresf,fileresu);
1.126 brouard 8387: if((ficresf=fopen(fileresf,"w"))==NULL) {
8388: printf("Problem with forecast resultfile: %s\n", fileresf);
8389: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8390: }
1.235 brouard 8391: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8392: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8393:
1.225 brouard 8394: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8395:
8396:
8397: stepsize=(int) (stepm+YEARM-1)/YEARM;
8398: if (stepm<=12) stepsize=1;
8399: if(estepm < stepm){
8400: printf ("Problem %d lower than %d\n",estepm, stepm);
8401: }
1.270 brouard 8402: else{
8403: hstepm=estepm;
8404: }
8405: if(estepm > stepm){ /* Yes every two year */
8406: stepsize=2;
8407: }
1.296 brouard 8408: hstepm=hstepm/stepm;
1.126 brouard 8409:
1.296 brouard 8410:
8411: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8412: /* fractional in yp1 *\/ */
8413: /* aintmean=yp; */
8414: /* yp2=modf((yp1*12),&yp); */
8415: /* mintmean=yp; */
8416: /* yp1=modf((yp2*30.5),&yp); */
8417: /* jintmean=yp; */
8418: /* if(jintmean==0) jintmean=1; */
8419: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8420:
1.296 brouard 8421:
8422: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8423: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8424: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8425: i1=pow(2,cptcoveff);
1.126 brouard 8426: if (cptcovn < 1){i1=1;}
8427:
1.296 brouard 8428: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8429:
8430: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8431:
1.126 brouard 8432: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8433: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8434: for(k=1; k<=i1;k++){
1.253 brouard 8435: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8436: continue;
1.227 brouard 8437: if(invalidvarcomb[k]){
8438: printf("\nCombination (%d) projection ignored because no cases \n",k);
8439: continue;
8440: }
8441: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8442: for(j=1;j<=cptcoveff;j++) {
8443: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8444: }
1.235 brouard 8445: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8446: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8447: }
1.227 brouard 8448: fprintf(ficresf," yearproj age");
8449: for(j=1; j<=nlstate+ndeath;j++){
8450: for(i=1; i<=nlstate;i++)
8451: fprintf(ficresf," p%d%d",i,j);
8452: fprintf(ficresf," wp.%d",j);
8453: }
1.296 brouard 8454: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8455: fprintf(ficresf,"\n");
1.296 brouard 8456: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8457: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8458: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8459: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8460: nhstepm = nhstepm/hstepm;
8461: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8462: oldm=oldms;savm=savms;
1.268 brouard 8463: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8464: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8465: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8466: for (h=0; h<=nhstepm; h++){
8467: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8468: break;
8469: }
8470: }
8471: fprintf(ficresf,"\n");
8472: for(j=1;j<=cptcoveff;j++)
8473: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8474: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8475:
8476: for(j=1; j<=nlstate+ndeath;j++) {
8477: ppij=0.;
8478: for(i=1; i<=nlstate;i++) {
1.278 brouard 8479: if (mobilav>=1)
8480: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8481: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8482: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8483: }
1.268 brouard 8484: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8485: } /* end i */
8486: fprintf(ficresf," %.3f", ppij);
8487: }/* end j */
1.227 brouard 8488: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8489: } /* end agec */
1.266 brouard 8490: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8491: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8492: } /* end yearp */
8493: } /* end k */
1.219 brouard 8494:
1.126 brouard 8495: fclose(ficresf);
1.215 brouard 8496: printf("End of Computing forecasting \n");
8497: fprintf(ficlog,"End of Computing forecasting\n");
8498:
1.126 brouard 8499: }
8500:
1.269 brouard 8501: /************** Back Forecasting ******************/
1.296 brouard 8502: /* 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){ */
8503: 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){
8504: /* back1, year, month, day of starting backprojection
1.267 brouard 8505: agemin, agemax range of age
8506: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8507: anback2 year of end of backprojection (same day and month as back1).
8508: prevacurrent and prev are prevalences.
1.267 brouard 8509: */
8510: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8511: double agec; /* generic age */
1.302 brouard 8512: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8513: double *popeffectif,*popcount;
8514: double ***p3mat;
8515: /* double ***mobaverage; */
8516: char fileresfb[FILENAMELENGTH];
8517:
1.268 brouard 8518: agelim=AGEINF;
1.267 brouard 8519: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8520: in each health status at the date of interview (if between dateprev1 and dateprev2).
8521: We still use firstpass and lastpass as another selection.
8522: */
8523: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8524: /* firstpass, lastpass, stepm, weightopt, model); */
8525:
8526: /*Do we need to compute prevalence again?*/
8527:
8528: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8529:
8530: strcpy(fileresfb,"FB_");
8531: strcat(fileresfb,fileresu);
8532: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8533: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8534: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8535: }
8536: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8537: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8538:
8539: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8540:
8541:
8542: stepsize=(int) (stepm+YEARM-1)/YEARM;
8543: if (stepm<=12) stepsize=1;
8544: if(estepm < stepm){
8545: printf ("Problem %d lower than %d\n",estepm, stepm);
8546: }
1.270 brouard 8547: else{
8548: hstepm=estepm;
8549: }
8550: if(estepm >= stepm){ /* Yes every two year */
8551: stepsize=2;
8552: }
1.267 brouard 8553:
8554: hstepm=hstepm/stepm;
1.296 brouard 8555: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8556: /* fractional in yp1 *\/ */
8557: /* aintmean=yp; */
8558: /* yp2=modf((yp1*12),&yp); */
8559: /* mintmean=yp; */
8560: /* yp1=modf((yp2*30.5),&yp); */
8561: /* jintmean=yp; */
8562: /* if(jintmean==0) jintmean=1; */
8563: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8564:
8565: i1=pow(2,cptcoveff);
8566: if (cptcovn < 1){i1=1;}
8567:
1.296 brouard 8568: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8569: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8570:
8571: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8572:
8573: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8574: for(k=1; k<=i1;k++){
8575: if(i1 != 1 && TKresult[nres]!= k)
8576: continue;
8577: if(invalidvarcomb[k]){
8578: printf("\nCombination (%d) projection ignored because no cases \n",k);
8579: continue;
8580: }
1.268 brouard 8581: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8582: for(j=1;j<=cptcoveff;j++) {
8583: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8584: }
8585: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8586: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8587: }
8588: fprintf(ficresfb," yearbproj age");
8589: for(j=1; j<=nlstate+ndeath;j++){
8590: for(i=1; i<=nlstate;i++)
1.268 brouard 8591: fprintf(ficresfb," b%d%d",i,j);
8592: fprintf(ficresfb," b.%d",j);
1.267 brouard 8593: }
1.296 brouard 8594: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8595: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8596: fprintf(ficresfb,"\n");
1.296 brouard 8597: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8598: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8599: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8600: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8601: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8602: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8603: nhstepm = nhstepm/hstepm;
8604: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8605: oldm=oldms;savm=savms;
1.268 brouard 8606: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8607: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8608: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8609: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8610: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8611: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8612: for (h=0; h<=nhstepm; h++){
1.268 brouard 8613: if (h*hstepm/YEARM*stepm ==-yearp) {
8614: break;
8615: }
8616: }
8617: fprintf(ficresfb,"\n");
8618: for(j=1;j<=cptcoveff;j++)
8619: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8620: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8621: for(i=1; i<=nlstate+ndeath;i++) {
8622: ppij=0.;ppi=0.;
8623: for(j=1; j<=nlstate;j++) {
8624: /* if (mobilav==1) */
1.269 brouard 8625: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8626: ppi=ppi+prevacurrent[(int)agec][j][k];
8627: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8628: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8629: /* else { */
8630: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8631: /* } */
1.268 brouard 8632: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8633: } /* end j */
8634: if(ppi <0.99){
8635: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8636: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8637: }
8638: fprintf(ficresfb," %.3f", ppij);
8639: }/* end j */
1.267 brouard 8640: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8641: } /* end agec */
8642: } /* end yearp */
8643: } /* end k */
1.217 brouard 8644:
1.267 brouard 8645: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8646:
1.267 brouard 8647: fclose(ficresfb);
8648: printf("End of Computing Back forecasting \n");
8649: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8650:
1.267 brouard 8651: }
1.217 brouard 8652:
1.269 brouard 8653: /* Variance of prevalence limit: varprlim */
8654: 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 8655: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8656:
8657: char fileresvpl[FILENAMELENGTH];
8658: FILE *ficresvpl;
8659: double **oldm, **savm;
8660: double **varpl; /* Variances of prevalence limits by age */
8661: int i1, k, nres, j ;
8662:
8663: strcpy(fileresvpl,"VPL_");
8664: strcat(fileresvpl,fileresu);
8665: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8666: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8667: exit(0);
8668: }
1.288 brouard 8669: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8670: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8671:
8672: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8673: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8674:
8675: i1=pow(2,cptcoveff);
8676: if (cptcovn < 1){i1=1;}
8677:
8678: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8679: for(k=1; k<=i1;k++){
8680: if(i1 != 1 && TKresult[nres]!= k)
8681: continue;
8682: fprintf(ficresvpl,"\n#****** ");
8683: printf("\n#****** ");
8684: fprintf(ficlog,"\n#****** ");
8685: for(j=1;j<=cptcoveff;j++) {
8686: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8687: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8688: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8689: }
8690: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8691: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8692: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8693: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8694: }
8695: fprintf(ficresvpl,"******\n");
8696: printf("******\n");
8697: fprintf(ficlog,"******\n");
8698:
8699: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8700: oldm=oldms;savm=savms;
8701: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8702: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8703: /*}*/
8704: }
8705:
8706: fclose(ficresvpl);
1.288 brouard 8707: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8708: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8709:
8710: }
8711: /* Variance of back prevalence: varbprlim */
8712: 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){
8713: /*------- Variance of back (stable) prevalence------*/
8714:
8715: char fileresvbl[FILENAMELENGTH];
8716: FILE *ficresvbl;
8717:
8718: double **oldm, **savm;
8719: double **varbpl; /* Variances of back prevalence limits by age */
8720: int i1, k, nres, j ;
8721:
8722: strcpy(fileresvbl,"VBL_");
8723: strcat(fileresvbl,fileresu);
8724: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8725: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8726: exit(0);
8727: }
8728: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8729: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8730:
8731:
8732: i1=pow(2,cptcoveff);
8733: if (cptcovn < 1){i1=1;}
8734:
8735: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8736: for(k=1; k<=i1;k++){
8737: if(i1 != 1 && TKresult[nres]!= k)
8738: continue;
8739: fprintf(ficresvbl,"\n#****** ");
8740: printf("\n#****** ");
8741: fprintf(ficlog,"\n#****** ");
8742: for(j=1;j<=cptcoveff;j++) {
8743: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8744: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8745: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8746: }
8747: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8748: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8749: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8750: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8751: }
8752: fprintf(ficresvbl,"******\n");
8753: printf("******\n");
8754: fprintf(ficlog,"******\n");
8755:
8756: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8757: oldm=oldms;savm=savms;
8758:
8759: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8760: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8761: /*}*/
8762: }
8763:
8764: fclose(ficresvbl);
8765: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8766: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8767:
8768: } /* End of varbprlim */
8769:
1.126 brouard 8770: /************** Forecasting *****not tested NB*************/
1.227 brouard 8771: /* 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 8772:
1.227 brouard 8773: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8774: /* int *popage; */
8775: /* double calagedatem, agelim, kk1, kk2; */
8776: /* double *popeffectif,*popcount; */
8777: /* double ***p3mat,***tabpop,***tabpopprev; */
8778: /* /\* double ***mobaverage; *\/ */
8779: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8780:
1.227 brouard 8781: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8782: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8783: /* agelim=AGESUP; */
8784: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8785:
1.227 brouard 8786: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8787:
8788:
1.227 brouard 8789: /* strcpy(filerespop,"POP_"); */
8790: /* strcat(filerespop,fileresu); */
8791: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8792: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8793: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8794: /* } */
8795: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8796: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8797:
1.227 brouard 8798: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8799:
1.227 brouard 8800: /* /\* if (mobilav!=0) { *\/ */
8801: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8802: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8803: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8804: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8805: /* /\* } *\/ */
8806: /* /\* } *\/ */
1.126 brouard 8807:
1.227 brouard 8808: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8809: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8810:
1.227 brouard 8811: /* agelim=AGESUP; */
1.126 brouard 8812:
1.227 brouard 8813: /* hstepm=1; */
8814: /* hstepm=hstepm/stepm; */
1.218 brouard 8815:
1.227 brouard 8816: /* if (popforecast==1) { */
8817: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8818: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8819: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8820: /* } */
8821: /* popage=ivector(0,AGESUP); */
8822: /* popeffectif=vector(0,AGESUP); */
8823: /* popcount=vector(0,AGESUP); */
1.126 brouard 8824:
1.227 brouard 8825: /* i=1; */
8826: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8827:
1.227 brouard 8828: /* imx=i; */
8829: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8830: /* } */
1.218 brouard 8831:
1.227 brouard 8832: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8833: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8834: /* k=k+1; */
8835: /* fprintf(ficrespop,"\n#******"); */
8836: /* for(j=1;j<=cptcoveff;j++) { */
8837: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8838: /* } */
8839: /* fprintf(ficrespop,"******\n"); */
8840: /* fprintf(ficrespop,"# Age"); */
8841: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8842: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8843:
1.227 brouard 8844: /* for (cpt=0; cpt<=0;cpt++) { */
8845: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8846:
1.227 brouard 8847: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8848: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8849: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8850:
1.227 brouard 8851: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8852: /* oldm=oldms;savm=savms; */
8853: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8854:
1.227 brouard 8855: /* for (h=0; h<=nhstepm; h++){ */
8856: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8857: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8858: /* } */
8859: /* for(j=1; j<=nlstate+ndeath;j++) { */
8860: /* kk1=0.;kk2=0; */
8861: /* for(i=1; i<=nlstate;i++) { */
8862: /* if (mobilav==1) */
8863: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8864: /* else { */
8865: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8866: /* } */
8867: /* } */
8868: /* if (h==(int)(calagedatem+12*cpt)){ */
8869: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8870: /* /\*fprintf(ficrespop," %.3f", kk1); */
8871: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8872: /* } */
8873: /* } */
8874: /* for(i=1; i<=nlstate;i++){ */
8875: /* kk1=0.; */
8876: /* for(j=1; j<=nlstate;j++){ */
8877: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8878: /* } */
8879: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8880: /* } */
1.218 brouard 8881:
1.227 brouard 8882: /* if (h==(int)(calagedatem+12*cpt)) */
8883: /* for(j=1; j<=nlstate;j++) */
8884: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8885: /* } */
8886: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8887: /* } */
8888: /* } */
1.218 brouard 8889:
1.227 brouard 8890: /* /\******\/ */
1.218 brouard 8891:
1.227 brouard 8892: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8893: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8894: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8895: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8896: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8897:
1.227 brouard 8898: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8899: /* oldm=oldms;savm=savms; */
8900: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8901: /* for (h=0; h<=nhstepm; h++){ */
8902: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8903: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8904: /* } */
8905: /* for(j=1; j<=nlstate+ndeath;j++) { */
8906: /* kk1=0.;kk2=0; */
8907: /* for(i=1; i<=nlstate;i++) { */
8908: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8909: /* } */
8910: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8911: /* } */
8912: /* } */
8913: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8914: /* } */
8915: /* } */
8916: /* } */
8917: /* } */
1.218 brouard 8918:
1.227 brouard 8919: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8920:
1.227 brouard 8921: /* if (popforecast==1) { */
8922: /* free_ivector(popage,0,AGESUP); */
8923: /* free_vector(popeffectif,0,AGESUP); */
8924: /* free_vector(popcount,0,AGESUP); */
8925: /* } */
8926: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8927: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8928: /* fclose(ficrespop); */
8929: /* } /\* End of popforecast *\/ */
1.218 brouard 8930:
1.126 brouard 8931: int fileappend(FILE *fichier, char *optionfich)
8932: {
8933: if((fichier=fopen(optionfich,"a"))==NULL) {
8934: printf("Problem with file: %s\n", optionfich);
8935: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8936: return (0);
8937: }
8938: fflush(fichier);
8939: return (1);
8940: }
8941:
8942:
8943: /**************** function prwizard **********************/
8944: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8945: {
8946:
8947: /* Wizard to print covariance matrix template */
8948:
1.164 brouard 8949: char ca[32], cb[32];
8950: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8951: int numlinepar;
8952:
8953: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8954: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8955: for(i=1; i <=nlstate; i++){
8956: jj=0;
8957: for(j=1; j <=nlstate+ndeath; j++){
8958: if(j==i) continue;
8959: jj++;
8960: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8961: printf("%1d%1d",i,j);
8962: fprintf(ficparo,"%1d%1d",i,j);
8963: for(k=1; k<=ncovmodel;k++){
8964: /* printf(" %lf",param[i][j][k]); */
8965: /* fprintf(ficparo," %lf",param[i][j][k]); */
8966: printf(" 0.");
8967: fprintf(ficparo," 0.");
8968: }
8969: printf("\n");
8970: fprintf(ficparo,"\n");
8971: }
8972: }
8973: printf("# Scales (for hessian or gradient estimation)\n");
8974: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8975: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8976: for(i=1; i <=nlstate; i++){
8977: jj=0;
8978: for(j=1; j <=nlstate+ndeath; j++){
8979: if(j==i) continue;
8980: jj++;
8981: fprintf(ficparo,"%1d%1d",i,j);
8982: printf("%1d%1d",i,j);
8983: fflush(stdout);
8984: for(k=1; k<=ncovmodel;k++){
8985: /* printf(" %le",delti3[i][j][k]); */
8986: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8987: printf(" 0.");
8988: fprintf(ficparo," 0.");
8989: }
8990: numlinepar++;
8991: printf("\n");
8992: fprintf(ficparo,"\n");
8993: }
8994: }
8995: printf("# Covariance matrix\n");
8996: /* # 121 Var(a12)\n\ */
8997: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8998: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8999: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9000: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9001: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9002: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9003: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9004: fflush(stdout);
9005: fprintf(ficparo,"# Covariance matrix\n");
9006: /* # 121 Var(a12)\n\ */
9007: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9008: /* # ...\n\ */
9009: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9010:
9011: for(itimes=1;itimes<=2;itimes++){
9012: jj=0;
9013: for(i=1; i <=nlstate; i++){
9014: for(j=1; j <=nlstate+ndeath; j++){
9015: if(j==i) continue;
9016: for(k=1; k<=ncovmodel;k++){
9017: jj++;
9018: ca[0]= k+'a'-1;ca[1]='\0';
9019: if(itimes==1){
9020: printf("#%1d%1d%d",i,j,k);
9021: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9022: }else{
9023: printf("%1d%1d%d",i,j,k);
9024: fprintf(ficparo,"%1d%1d%d",i,j,k);
9025: /* printf(" %.5le",matcov[i][j]); */
9026: }
9027: ll=0;
9028: for(li=1;li <=nlstate; li++){
9029: for(lj=1;lj <=nlstate+ndeath; lj++){
9030: if(lj==li) continue;
9031: for(lk=1;lk<=ncovmodel;lk++){
9032: ll++;
9033: if(ll<=jj){
9034: cb[0]= lk +'a'-1;cb[1]='\0';
9035: if(ll<jj){
9036: if(itimes==1){
9037: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9038: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9039: }else{
9040: printf(" 0.");
9041: fprintf(ficparo," 0.");
9042: }
9043: }else{
9044: if(itimes==1){
9045: printf(" Var(%s%1d%1d)",ca,i,j);
9046: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9047: }else{
9048: printf(" 0.");
9049: fprintf(ficparo," 0.");
9050: }
9051: }
9052: }
9053: } /* end lk */
9054: } /* end lj */
9055: } /* end li */
9056: printf("\n");
9057: fprintf(ficparo,"\n");
9058: numlinepar++;
9059: } /* end k*/
9060: } /*end j */
9061: } /* end i */
9062: } /* end itimes */
9063:
9064: } /* end of prwizard */
9065: /******************* Gompertz Likelihood ******************************/
9066: double gompertz(double x[])
9067: {
1.302 brouard 9068: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9069: int i,n=0; /* n is the size of the sample */
9070:
1.220 brouard 9071: for (i=1;i<=imx ; i++) {
1.126 brouard 9072: sump=sump+weight[i];
9073: /* sump=sump+1;*/
9074: num=num+1;
9075: }
1.302 brouard 9076: L=0.0;
9077: /* agegomp=AGEGOMP; */
1.126 brouard 9078: /* for (i=0; i<=imx; i++)
9079: 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]);*/
9080:
1.302 brouard 9081: for (i=1;i<=imx ; i++) {
9082: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9083: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9084: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9085: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9086: * +
9087: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9088: */
9089: if (wav[i] > 1 || agedc[i] < AGESUP) {
9090: if (cens[i] == 1){
9091: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9092: } else if (cens[i] == 0){
1.126 brouard 9093: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9094: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9095: } else
9096: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9097: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9098: L=L+A*weight[i];
1.126 brouard 9099: /* 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 9100: }
9101: }
1.126 brouard 9102:
1.302 brouard 9103: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9104:
9105: return -2*L*num/sump;
9106: }
9107:
1.136 brouard 9108: #ifdef GSL
9109: /******************* Gompertz_f Likelihood ******************************/
9110: double gompertz_f(const gsl_vector *v, void *params)
9111: {
1.302 brouard 9112: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9113: double *x= (double *) v->data;
9114: int i,n=0; /* n is the size of the sample */
9115:
9116: for (i=0;i<=imx-1 ; i++) {
9117: sump=sump+weight[i];
9118: /* sump=sump+1;*/
9119: num=num+1;
9120: }
9121:
9122:
9123: /* for (i=0; i<=imx; i++)
9124: 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]);*/
9125: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9126: for (i=1;i<=imx ; i++)
9127: {
9128: if (cens[i] == 1 && wav[i]>1)
9129: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9130:
9131: if (cens[i] == 0 && wav[i]>1)
9132: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9133: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9134:
9135: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9136: if (wav[i] > 1 ) { /* ??? */
9137: LL=LL+A*weight[i];
9138: /* 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]);*/
9139: }
9140: }
9141:
9142: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9143: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9144:
9145: return -2*LL*num/sump;
9146: }
9147: #endif
9148:
1.126 brouard 9149: /******************* Printing html file ***********/
1.201 brouard 9150: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9151: int lastpass, int stepm, int weightopt, char model[],\
9152: int imx, double p[],double **matcov,double agemortsup){
9153: int i,k;
9154:
9155: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9156: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9157: for (i=1;i<=2;i++)
9158: 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 9159: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9160: fprintf(fichtm,"</ul>");
9161:
9162: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9163:
9164: 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>");
9165:
9166: for (k=agegomp;k<(agemortsup-2);k++)
9167: 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]);
9168:
9169:
9170: fflush(fichtm);
9171: }
9172:
9173: /******************* Gnuplot file **************/
1.201 brouard 9174: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9175:
9176: char dirfileres[132],optfileres[132];
1.164 brouard 9177:
1.126 brouard 9178: int ng;
9179:
9180:
9181: /*#ifdef windows */
9182: fprintf(ficgp,"cd \"%s\" \n",pathc);
9183: /*#endif */
9184:
9185:
9186: strcpy(dirfileres,optionfilefiname);
9187: strcpy(optfileres,"vpl");
1.199 brouard 9188: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9189: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9190: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9191: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9192: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9193:
9194: }
9195:
1.136 brouard 9196: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9197: {
1.126 brouard 9198:
1.136 brouard 9199: /*-------- data file ----------*/
9200: FILE *fic;
9201: char dummy[]=" ";
1.240 brouard 9202: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9203: int lstra;
1.136 brouard 9204: int linei, month, year,iout;
1.302 brouard 9205: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9206: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9207: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9208: char *stratrunc;
1.223 brouard 9209:
1.240 brouard 9210: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9211: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9212:
1.240 brouard 9213: for(v=1; v <=ncovcol;v++){
9214: DummyV[v]=0;
9215: FixedV[v]=0;
9216: }
9217: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9218: DummyV[v]=1;
9219: FixedV[v]=0;
9220: }
9221: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9222: DummyV[v]=0;
9223: FixedV[v]=1;
9224: }
9225: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9226: DummyV[v]=1;
9227: FixedV[v]=1;
9228: }
9229: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9230: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9231: 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]);
9232: }
1.126 brouard 9233:
1.136 brouard 9234: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9235: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9236: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9237: }
1.126 brouard 9238:
1.302 brouard 9239: /* Is it a BOM UTF-8 Windows file? */
9240: /* First data line */
9241: linei=0;
9242: while(fgets(line, MAXLINE, fic)) {
9243: noffset=0;
9244: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9245: {
9246: noffset=noffset+3;
9247: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9248: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9249: fflush(ficlog); return 1;
9250: }
9251: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9252: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9253: {
9254: noffset=noffset+2;
9255: printf("# 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);
9256: fprintf(ficlog,"# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
9257: fflush(ficlog); return 1;
9258: }
9259: else if( line[0] == 0 && line[1] == 0)
9260: {
9261: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9262: noffset=noffset+4;
9263: printf("# 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);
9264: fprintf(ficlog,"# Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
9265: fflush(ficlog); return 1;
9266: }
9267: } else{
9268: ;/*printf(" Not a BOM file\n");*/
9269: }
9270: /* If line starts with a # it is a comment */
9271: if (line[noffset] == '#') {
9272: linei=linei+1;
9273: break;
9274: }else{
9275: break;
9276: }
9277: }
9278: fclose(fic);
9279: if((fic=fopen(datafile,"r"))==NULL) {
9280: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9281: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9282: }
9283: /* Not a Bom file */
9284:
1.136 brouard 9285: i=1;
9286: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9287: linei=linei+1;
9288: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9289: if(line[j] == '\t')
9290: line[j] = ' ';
9291: }
9292: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9293: ;
9294: };
9295: line[j+1]=0; /* Trims blanks at end of line */
9296: if(line[0]=='#'){
9297: fprintf(ficlog,"Comment line\n%s\n",line);
9298: printf("Comment line\n%s\n",line);
9299: continue;
9300: }
9301: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9302: strcpy(line, linetmp);
1.223 brouard 9303:
9304: /* Loops on waves */
9305: for (j=maxwav;j>=1;j--){
9306: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9307: cutv(stra, strb, line, ' ');
9308: if(strb[0]=='.') { /* Missing value */
9309: lval=-1;
9310: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9311: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9312: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9313: 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);
9314: 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);
9315: return 1;
9316: }
9317: }else{
9318: errno=0;
9319: /* what_kind_of_number(strb); */
9320: dval=strtod(strb,&endptr);
9321: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9322: /* if(strb != endptr && *endptr == '\0') */
9323: /* dval=dlval; */
9324: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9325: if( strb[0]=='\0' || (*endptr != '\0')){
9326: 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);
9327: 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);
9328: return 1;
9329: }
9330: cotqvar[j][iv][i]=dval;
9331: cotvar[j][ntv+iv][i]=dval;
9332: }
9333: strcpy(line,stra);
1.223 brouard 9334: }/* end loop ntqv */
1.225 brouard 9335:
1.223 brouard 9336: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9337: cutv(stra, strb, line, ' ');
9338: if(strb[0]=='.') { /* Missing value */
9339: lval=-1;
9340: }else{
9341: errno=0;
9342: lval=strtol(strb,&endptr,10);
9343: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9344: if( strb[0]=='\0' || (*endptr != '\0')){
9345: 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);
9346: 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);
9347: return 1;
9348: }
9349: }
9350: if(lval <-1 || lval >1){
9351: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9352: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9353: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9354: For example, for multinomial values like 1, 2 and 3,\n \
9355: build V1=0 V2=0 for the reference value (1),\n \
9356: V1=1 V2=0 for (2) \n \
1.223 brouard 9357: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9358: output of IMaCh is often meaningless.\n \
1.223 brouard 9359: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9360: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9361: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9362: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9363: For example, for multinomial values like 1, 2 and 3,\n \
9364: build V1=0 V2=0 for the reference value (1),\n \
9365: V1=1 V2=0 for (2) \n \
1.223 brouard 9366: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9367: output of IMaCh is often meaningless.\n \
1.223 brouard 9368: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9369: return 1;
9370: }
9371: cotvar[j][iv][i]=(double)(lval);
9372: strcpy(line,stra);
1.223 brouard 9373: }/* end loop ntv */
1.225 brouard 9374:
1.223 brouard 9375: /* Statuses at wave */
1.137 brouard 9376: cutv(stra, strb, line, ' ');
1.223 brouard 9377: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9378: lval=-1;
1.136 brouard 9379: }else{
1.238 brouard 9380: errno=0;
9381: lval=strtol(strb,&endptr,10);
9382: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9383: if( strb[0]=='\0' || (*endptr != '\0')){
9384: 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);
9385: 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);
9386: return 1;
9387: }
1.136 brouard 9388: }
1.225 brouard 9389:
1.136 brouard 9390: s[j][i]=lval;
1.225 brouard 9391:
1.223 brouard 9392: /* Date of Interview */
1.136 brouard 9393: strcpy(line,stra);
9394: cutv(stra, strb,line,' ');
1.169 brouard 9395: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9396: }
1.169 brouard 9397: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9398: month=99;
9399: year=9999;
1.136 brouard 9400: }else{
1.225 brouard 9401: 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);
9402: 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);
9403: return 1;
1.136 brouard 9404: }
9405: anint[j][i]= (double) year;
1.302 brouard 9406: mint[j][i]= (double)month;
9407: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9408: /* 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]); */
9409: /* 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]); */
9410: /* } */
1.136 brouard 9411: strcpy(line,stra);
1.223 brouard 9412: } /* End loop on waves */
1.225 brouard 9413:
1.223 brouard 9414: /* Date of death */
1.136 brouard 9415: cutv(stra, strb,line,' ');
1.169 brouard 9416: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9417: }
1.169 brouard 9418: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9419: month=99;
9420: year=9999;
9421: }else{
1.141 brouard 9422: 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 9423: 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);
9424: return 1;
1.136 brouard 9425: }
9426: andc[i]=(double) year;
9427: moisdc[i]=(double) month;
9428: strcpy(line,stra);
9429:
1.223 brouard 9430: /* Date of birth */
1.136 brouard 9431: cutv(stra, strb,line,' ');
1.169 brouard 9432: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9433: }
1.169 brouard 9434: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9435: month=99;
9436: year=9999;
9437: }else{
1.141 brouard 9438: 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);
9439: 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 9440: return 1;
1.136 brouard 9441: }
9442: if (year==9999) {
1.141 brouard 9443: 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);
9444: 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 9445: return 1;
9446:
1.136 brouard 9447: }
9448: annais[i]=(double)(year);
1.302 brouard 9449: moisnais[i]=(double)(month);
9450: for (j=1;j<=maxwav;j++){
9451: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9452: 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]);
9453: 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]);
9454: }
9455: }
9456:
1.136 brouard 9457: strcpy(line,stra);
1.225 brouard 9458:
1.223 brouard 9459: /* Sample weight */
1.136 brouard 9460: cutv(stra, strb,line,' ');
9461: errno=0;
9462: dval=strtod(strb,&endptr);
9463: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9464: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9465: 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 9466: fflush(ficlog);
9467: return 1;
9468: }
9469: weight[i]=dval;
9470: strcpy(line,stra);
1.225 brouard 9471:
1.223 brouard 9472: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9473: cutv(stra, strb, line, ' ');
9474: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9475: lval=-1;
1.223 brouard 9476: }else{
1.225 brouard 9477: errno=0;
9478: /* what_kind_of_number(strb); */
9479: dval=strtod(strb,&endptr);
9480: /* if(strb != endptr && *endptr == '\0') */
9481: /* dval=dlval; */
9482: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9483: if( strb[0]=='\0' || (*endptr != '\0')){
9484: 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);
9485: 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);
9486: return 1;
9487: }
9488: coqvar[iv][i]=dval;
1.226 brouard 9489: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9490: }
9491: strcpy(line,stra);
9492: }/* end loop nqv */
1.136 brouard 9493:
1.223 brouard 9494: /* Covariate values */
1.136 brouard 9495: for (j=ncovcol;j>=1;j--){
9496: cutv(stra, strb,line,' ');
1.223 brouard 9497: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9498: lval=-1;
1.136 brouard 9499: }else{
1.225 brouard 9500: errno=0;
9501: lval=strtol(strb,&endptr,10);
9502: if( strb[0]=='\0' || (*endptr != '\0')){
9503: 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);
9504: 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);
9505: return 1;
9506: }
1.136 brouard 9507: }
9508: if(lval <-1 || lval >1){
1.225 brouard 9509: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9510: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9511: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9512: For example, for multinomial values like 1, 2 and 3,\n \
9513: build V1=0 V2=0 for the reference value (1),\n \
9514: V1=1 V2=0 for (2) \n \
1.136 brouard 9515: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9516: output of IMaCh is often meaningless.\n \
1.136 brouard 9517: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9518: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9519: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9520: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9521: For example, for multinomial values like 1, 2 and 3,\n \
9522: build V1=0 V2=0 for the reference value (1),\n \
9523: V1=1 V2=0 for (2) \n \
1.136 brouard 9524: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9525: output of IMaCh is often meaningless.\n \
1.136 brouard 9526: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9527: return 1;
1.136 brouard 9528: }
9529: covar[j][i]=(double)(lval);
9530: strcpy(line,stra);
9531: }
9532: lstra=strlen(stra);
1.225 brouard 9533:
1.136 brouard 9534: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9535: stratrunc = &(stra[lstra-9]);
9536: num[i]=atol(stratrunc);
9537: }
9538: else
9539: num[i]=atol(stra);
9540: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9541: 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;}*/
9542:
9543: i=i+1;
9544: } /* End loop reading data */
1.225 brouard 9545:
1.136 brouard 9546: *imax=i-1; /* Number of individuals */
9547: fclose(fic);
1.225 brouard 9548:
1.136 brouard 9549: return (0);
1.164 brouard 9550: /* endread: */
1.225 brouard 9551: printf("Exiting readdata: ");
9552: fclose(fic);
9553: return (1);
1.223 brouard 9554: }
1.126 brouard 9555:
1.234 brouard 9556: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9557: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9558: while (*p2 == ' ')
1.234 brouard 9559: p2++;
9560: /* while ((*p1++ = *p2++) !=0) */
9561: /* ; */
9562: /* do */
9563: /* while (*p2 == ' ') */
9564: /* p2++; */
9565: /* while (*p1++ == *p2++); */
9566: *stri=p2;
1.145 brouard 9567: }
9568:
1.235 brouard 9569: int decoderesult ( char resultline[], int nres)
1.230 brouard 9570: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9571: {
1.235 brouard 9572: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9573: char resultsav[MAXLINE];
1.234 brouard 9574: int resultmodel[MAXLINE];
9575: int modelresult[MAXLINE];
1.230 brouard 9576: char stra[80], strb[80], strc[80], strd[80],stre[80];
9577:
1.234 brouard 9578: removefirstspace(&resultline);
1.233 brouard 9579: printf("decoderesult:%s\n",resultline);
1.230 brouard 9580:
9581: if (strstr(resultline,"v") !=0){
9582: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9583: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9584: return 1;
9585: }
9586: trimbb(resultsav, resultline);
9587: if (strlen(resultsav) >1){
9588: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9589: }
1.253 brouard 9590: if(j == 0){ /* Resultline but no = */
9591: TKresult[nres]=0; /* Combination for the nresult and the model */
9592: return (0);
9593: }
9594:
1.234 brouard 9595: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
9596: printf("ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
9597: fprintf(ficlog,"ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
9598: }
9599: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9600: if(nbocc(resultsav,'=') >1){
9601: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
9602: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
9603: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9604: }else
9605: cutl(strc,strd,resultsav,'=');
1.230 brouard 9606: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9607:
1.230 brouard 9608: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9609: Tvarsel[k]=atoi(strc);
9610: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9611: /* cptcovsel++; */
9612: if (nbocc(stra,'=') >0)
9613: strcpy(resultsav,stra); /* and analyzes it */
9614: }
1.235 brouard 9615: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9616: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9617: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9618: match=0;
1.236 brouard 9619: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9620: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9621: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9622: match=1;
9623: break;
9624: }
9625: }
9626: if(match == 0){
9627: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9628: }
9629: }
9630: }
1.235 brouard 9631: /* Checking for missing or useless values in comparison of current model needs */
9632: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9633: match=0;
1.235 brouard 9634: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9635: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9636: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9637: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9638: ++match;
9639: }
9640: }
9641: }
9642: if(match == 0){
9643: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9644: }else if(match > 1){
9645: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9646: }
9647: }
1.235 brouard 9648:
1.234 brouard 9649: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9650: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9651: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9652: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9653: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9654: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9655: /* 1 0 0 0 */
9656: /* 2 1 0 0 */
9657: /* 3 0 1 0 */
9658: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9659: /* 5 0 0 1 */
9660: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9661: /* 7 0 1 1 */
9662: /* 8 1 1 1 */
1.237 brouard 9663: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9664: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9665: /* V5*age V5 known which value for nres? */
9666: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9667: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9668: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9669: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9670: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9671: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9672: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9673: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9674: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9675: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9676: k4++;;
9677: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9678: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9679: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9680: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9681: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9682: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9683: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9684: k4q++;;
9685: }
9686: }
1.234 brouard 9687:
1.235 brouard 9688: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9689: return (0);
9690: }
1.235 brouard 9691:
1.230 brouard 9692: int decodemodel( char model[], int lastobs)
9693: /**< This routine decodes the model and returns:
1.224 brouard 9694: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9695: * - nagesqr = 1 if age*age in the model, otherwise 0.
9696: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9697: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9698: * - cptcovage number of covariates with age*products =2
9699: * - cptcovs number of simple covariates
9700: * - 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
9701: * which is a new column after the 9 (ncovcol) variables.
9702: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9703: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9704: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9705: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9706: */
1.136 brouard 9707: {
1.238 brouard 9708: int i, j, k, ks, v;
1.227 brouard 9709: int j1, k1, k2, k3, k4;
1.136 brouard 9710: char modelsav[80];
1.145 brouard 9711: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9712: char *strpt;
1.136 brouard 9713:
1.145 brouard 9714: /*removespace(model);*/
1.136 brouard 9715: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9716: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9717: if (strstr(model,"AGE") !=0){
1.192 brouard 9718: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9719: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9720: return 1;
9721: }
1.141 brouard 9722: if (strstr(model,"v") !=0){
9723: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9724: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9725: return 1;
9726: }
1.187 brouard 9727: strcpy(modelsav,model);
9728: if ((strpt=strstr(model,"age*age")) !=0){
9729: printf(" strpt=%s, model=%s\n",strpt, model);
9730: if(strpt != model){
1.234 brouard 9731: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9732: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9733: corresponding column of parameters.\n",model);
1.234 brouard 9734: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9735: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9736: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9737: return 1;
1.225 brouard 9738: }
1.187 brouard 9739: nagesqr=1;
9740: if (strstr(model,"+age*age") !=0)
1.234 brouard 9741: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9742: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9743: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9744: else
1.234 brouard 9745: substrchaine(modelsav, model, "age*age");
1.187 brouard 9746: }else
9747: nagesqr=0;
9748: if (strlen(modelsav) >1){
9749: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9750: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9751: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9752: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9753: * cst, age and age*age
9754: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9755: /* including age products which are counted in cptcovage.
9756: * but the covariates which are products must be treated
9757: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9758: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9759: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9760:
9761:
1.187 brouard 9762: /* Design
9763: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9764: * < ncovcol=8 >
9765: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9766: * k= 1 2 3 4 5 6 7 8
9767: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9768: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9769: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9770: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9771: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9772: * Tage[++cptcovage]=k
9773: * if products, new covar are created after ncovcol with k1
9774: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9775: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9776: * 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
9777: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9778: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9779: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9780: * < ncovcol=8 >
9781: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9782: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9783: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9784: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9785: * p Tprod[1]@2={ 6, 5}
9786: *p Tvard[1][1]@4= {7, 8, 5, 6}
9787: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9788: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9789: *How to reorganize?
9790: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9791: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9792: * {2, 1, 4, 8, 5, 6, 3, 7}
9793: * Struct []
9794: */
1.225 brouard 9795:
1.187 brouard 9796: /* This loop fills the array Tvar from the string 'model'.*/
9797: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9798: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9799: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9800: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9801: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9802: /* k=1 Tvar[1]=2 (from V2) */
9803: /* k=5 Tvar[5] */
9804: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9805: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9806: /* } */
1.198 brouard 9807: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9808: /*
9809: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9810: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9811: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9812: }
1.187 brouard 9813: cptcovage=0;
9814: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9815: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9816: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9817: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9818: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9819: /*scanf("%d",i);*/
9820: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9821: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9822: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9823: /* covar is not filled and then is empty */
9824: cptcovprod--;
9825: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9826: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9827: Typevar[k]=1; /* 1 for age product */
9828: cptcovage++; /* Sums the number of covariates which include age as a product */
9829: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9830: /*printf("stre=%s ", stre);*/
9831: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9832: cptcovprod--;
9833: cutl(stre,strb,strc,'V');
9834: Tvar[k]=atoi(stre);
9835: Typevar[k]=1; /* 1 for age product */
9836: cptcovage++;
9837: Tage[cptcovage]=k;
9838: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9839: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9840: cptcovn++;
9841: cptcovprodnoage++;k1++;
9842: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9843: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9844: because this model-covariate is a construction we invent a new column
9845: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9846: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9847: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9848: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9849: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9850: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9851: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9852: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9853: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9854: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9855: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9856: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9857: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9858: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9859: for (i=1; i<=lastobs;i++){
9860: /* Computes the new covariate which is a product of
9861: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9862: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9863: }
9864: } /* End age is not in the model */
9865: } /* End if model includes a product */
9866: else { /* no more sum */
9867: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9868: /* scanf("%d",i);*/
9869: cutl(strd,strc,strb,'V');
9870: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9871: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9872: Tvar[k]=atoi(strd);
9873: Typevar[k]=0; /* 0 for simple covariates */
9874: }
9875: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9876: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9877: scanf("%d",i);*/
1.187 brouard 9878: } /* end of loop + on total covariates */
9879: } /* end if strlen(modelsave == 0) age*age might exist */
9880: } /* end if strlen(model == 0) */
1.136 brouard 9881:
9882: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9883: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9884:
1.136 brouard 9885: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9886: printf("cptcovprod=%d ", cptcovprod);
9887: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9888: scanf("%d ",i);*/
9889:
9890:
1.230 brouard 9891: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9892: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9893: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9894: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9895: k = 1 2 3 4 5 6 7 8 9
9896: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9897: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9898: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9899: Dummy[k] 1 0 0 0 3 1 1 2 3
9900: Tmodelind[combination of covar]=k;
1.225 brouard 9901: */
9902: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9903: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9904: /* 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 9905: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9906: printf("Model=%s\n\
9907: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9908: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9909: 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);
9910: fprintf(ficlog,"Model=%s\n\
9911: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9912: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9913: 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 9914: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9915: 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 */
9916: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9917: Fixed[k]= 0;
9918: Dummy[k]= 0;
1.225 brouard 9919: ncoveff++;
1.232 brouard 9920: ncovf++;
1.234 brouard 9921: nsd++;
9922: modell[k].maintype= FTYPE;
9923: TvarsD[nsd]=Tvar[k];
9924: TvarsDind[nsd]=k;
9925: TvarF[ncovf]=Tvar[k];
9926: TvarFind[ncovf]=k;
9927: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9928: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9929: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9930: Fixed[k]= 0;
9931: Dummy[k]= 0;
9932: ncoveff++;
9933: ncovf++;
9934: modell[k].maintype= FTYPE;
9935: TvarF[ncovf]=Tvar[k];
9936: TvarFind[ncovf]=k;
1.230 brouard 9937: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9938: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9939: }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 9940: Fixed[k]= 0;
9941: Dummy[k]= 1;
1.230 brouard 9942: nqfveff++;
1.234 brouard 9943: modell[k].maintype= FTYPE;
9944: modell[k].subtype= FQ;
9945: nsq++;
9946: TvarsQ[nsq]=Tvar[k];
9947: TvarsQind[nsq]=k;
1.232 brouard 9948: ncovf++;
1.234 brouard 9949: TvarF[ncovf]=Tvar[k];
9950: TvarFind[ncovf]=k;
1.231 brouard 9951: 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 9952: 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 9953: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9954: Fixed[k]= 1;
9955: Dummy[k]= 0;
1.225 brouard 9956: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9957: modell[k].maintype= VTYPE;
9958: modell[k].subtype= VD;
9959: nsd++;
9960: TvarsD[nsd]=Tvar[k];
9961: TvarsDind[nsd]=k;
9962: ncovv++; /* Only simple time varying variables */
9963: TvarV[ncovv]=Tvar[k];
1.242 brouard 9964: 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 9965: 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 */
9966: 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 9967: 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);
9968: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9969: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9970: Fixed[k]= 1;
9971: Dummy[k]= 1;
9972: nqtveff++;
9973: modell[k].maintype= VTYPE;
9974: modell[k].subtype= VQ;
9975: ncovv++; /* Only simple time varying variables */
9976: nsq++;
9977: TvarsQ[nsq]=Tvar[k];
9978: TvarsQind[nsq]=k;
9979: TvarV[ncovv]=Tvar[k];
1.242 brouard 9980: 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 9981: 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 */
9982: 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 9983: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9984: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9985: 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 9986: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9987: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9988: ncova++;
9989: TvarA[ncova]=Tvar[k];
9990: TvarAind[ncova]=k;
1.231 brouard 9991: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9992: Fixed[k]= 2;
9993: Dummy[k]= 2;
9994: modell[k].maintype= ATYPE;
9995: modell[k].subtype= APFD;
9996: /* ncoveff++; */
1.227 brouard 9997: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9998: Fixed[k]= 2;
9999: Dummy[k]= 3;
10000: modell[k].maintype= ATYPE;
10001: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10002: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10003: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10004: Fixed[k]= 3;
10005: Dummy[k]= 2;
10006: modell[k].maintype= ATYPE;
10007: modell[k].subtype= APVD; /* Product age * varying dummy */
10008: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10009: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10010: Fixed[k]= 3;
10011: Dummy[k]= 3;
10012: modell[k].maintype= ATYPE;
10013: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10014: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10015: }
10016: }else if (Typevar[k] == 2) { /* product without age */
10017: k1=Tposprod[k];
10018: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10019: if(Tvard[k1][2] <=ncovcol){
10020: Fixed[k]= 1;
10021: Dummy[k]= 0;
10022: modell[k].maintype= FTYPE;
10023: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10024: ncovf++; /* Fixed variables without age */
10025: TvarF[ncovf]=Tvar[k];
10026: TvarFind[ncovf]=k;
10027: }else if(Tvard[k1][2] <=ncovcol+nqv){
10028: Fixed[k]= 0; /* or 2 ?*/
10029: Dummy[k]= 1;
10030: modell[k].maintype= FTYPE;
10031: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10032: ncovf++; /* Varying variables without age */
10033: TvarF[ncovf]=Tvar[k];
10034: TvarFind[ncovf]=k;
10035: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10036: Fixed[k]= 1;
10037: Dummy[k]= 0;
10038: modell[k].maintype= VTYPE;
10039: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10040: ncovv++; /* Varying variables without age */
10041: TvarV[ncovv]=Tvar[k];
10042: TvarVind[ncovv]=k;
10043: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10044: Fixed[k]= 1;
10045: Dummy[k]= 1;
10046: modell[k].maintype= VTYPE;
10047: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10048: ncovv++; /* Varying variables without age */
10049: TvarV[ncovv]=Tvar[k];
10050: TvarVind[ncovv]=k;
10051: }
1.227 brouard 10052: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10053: if(Tvard[k1][2] <=ncovcol){
10054: Fixed[k]= 0; /* or 2 ?*/
10055: Dummy[k]= 1;
10056: modell[k].maintype= FTYPE;
10057: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10058: ncovf++; /* Fixed variables without age */
10059: TvarF[ncovf]=Tvar[k];
10060: TvarFind[ncovf]=k;
10061: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10062: Fixed[k]= 1;
10063: Dummy[k]= 1;
10064: modell[k].maintype= VTYPE;
10065: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10066: ncovv++; /* Varying variables without age */
10067: TvarV[ncovv]=Tvar[k];
10068: TvarVind[ncovv]=k;
10069: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10070: Fixed[k]= 1;
10071: Dummy[k]= 1;
10072: modell[k].maintype= VTYPE;
10073: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10074: ncovv++; /* Varying variables without age */
10075: TvarV[ncovv]=Tvar[k];
10076: TvarVind[ncovv]=k;
10077: ncovv++; /* Varying variables without age */
10078: TvarV[ncovv]=Tvar[k];
10079: TvarVind[ncovv]=k;
10080: }
1.227 brouard 10081: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10082: if(Tvard[k1][2] <=ncovcol){
10083: Fixed[k]= 1;
10084: Dummy[k]= 1;
10085: modell[k].maintype= VTYPE;
10086: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10087: ncovv++; /* Varying variables without age */
10088: TvarV[ncovv]=Tvar[k];
10089: TvarVind[ncovv]=k;
10090: }else if(Tvard[k1][2] <=ncovcol+nqv){
10091: Fixed[k]= 1;
10092: Dummy[k]= 1;
10093: modell[k].maintype= VTYPE;
10094: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10095: ncovv++; /* Varying variables without age */
10096: TvarV[ncovv]=Tvar[k];
10097: TvarVind[ncovv]=k;
10098: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10099: Fixed[k]= 1;
10100: Dummy[k]= 0;
10101: modell[k].maintype= VTYPE;
10102: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10103: ncovv++; /* Varying variables without age */
10104: TvarV[ncovv]=Tvar[k];
10105: TvarVind[ncovv]=k;
10106: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10107: Fixed[k]= 1;
10108: Dummy[k]= 1;
10109: modell[k].maintype= VTYPE;
10110: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10111: ncovv++; /* Varying variables without age */
10112: TvarV[ncovv]=Tvar[k];
10113: TvarVind[ncovv]=k;
10114: }
1.227 brouard 10115: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10116: if(Tvard[k1][2] <=ncovcol){
10117: Fixed[k]= 1;
10118: Dummy[k]= 1;
10119: modell[k].maintype= VTYPE;
10120: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10121: ncovv++; /* Varying variables without age */
10122: TvarV[ncovv]=Tvar[k];
10123: TvarVind[ncovv]=k;
10124: }else if(Tvard[k1][2] <=ncovcol+nqv){
10125: Fixed[k]= 1;
10126: Dummy[k]= 1;
10127: modell[k].maintype= VTYPE;
10128: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10129: ncovv++; /* Varying variables without age */
10130: TvarV[ncovv]=Tvar[k];
10131: TvarVind[ncovv]=k;
10132: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10133: Fixed[k]= 1;
10134: Dummy[k]= 1;
10135: modell[k].maintype= VTYPE;
10136: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10137: ncovv++; /* Varying variables without age */
10138: TvarV[ncovv]=Tvar[k];
10139: TvarVind[ncovv]=k;
10140: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10141: Fixed[k]= 1;
10142: Dummy[k]= 1;
10143: modell[k].maintype= VTYPE;
10144: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10145: ncovv++; /* Varying variables without age */
10146: TvarV[ncovv]=Tvar[k];
10147: TvarVind[ncovv]=k;
10148: }
1.227 brouard 10149: }else{
1.240 brouard 10150: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10151: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10152: } /*end k1*/
1.225 brouard 10153: }else{
1.226 brouard 10154: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10155: 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 10156: }
1.227 brouard 10157: 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 10158: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10159: 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]);
10160: }
10161: /* Searching for doublons in the model */
10162: for(k1=1; k1<= cptcovt;k1++){
10163: for(k2=1; k2 <k1;k2++){
1.285 brouard 10164: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10165: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10166: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10167: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10168: 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]);
10169: 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 10170: return(1);
10171: }
10172: }else if (Typevar[k1] ==2){
10173: k3=Tposprod[k1];
10174: k4=Tposprod[k2];
10175: 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])) ){
10176: 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]]);
10177: 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);
10178: return(1);
10179: }
10180: }
1.227 brouard 10181: }
10182: }
1.225 brouard 10183: }
10184: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10185: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10186: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10187: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10188: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10189: /*endread:*/
1.225 brouard 10190: printf("Exiting decodemodel: ");
10191: return (1);
1.136 brouard 10192: }
10193:
1.169 brouard 10194: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10195: {/* Check ages at death */
1.136 brouard 10196: int i, m;
1.218 brouard 10197: int firstone=0;
10198:
1.136 brouard 10199: for (i=1; i<=imx; i++) {
10200: for(m=2; (m<= maxwav); m++) {
10201: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10202: anint[m][i]=9999;
1.216 brouard 10203: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10204: s[m][i]=-1;
1.136 brouard 10205: }
10206: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10207: *nberr = *nberr + 1;
1.218 brouard 10208: if(firstone == 0){
10209: firstone=1;
1.260 brouard 10210: 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 10211: }
1.262 brouard 10212: 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 10213: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10214: }
10215: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10216: (*nberr)++;
1.259 brouard 10217: 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 10218: 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 10219: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10220: }
10221: }
10222: }
10223:
10224: for (i=1; i<=imx; i++) {
10225: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10226: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10227: 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 10228: if (s[m][i] >= nlstate+1) {
1.169 brouard 10229: if(agedc[i]>0){
10230: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10231: agev[m][i]=agedc[i];
1.214 brouard 10232: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10233: }else {
1.136 brouard 10234: if ((int)andc[i]!=9999){
10235: nbwarn++;
10236: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10237: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10238: agev[m][i]=-1;
10239: }
10240: }
1.169 brouard 10241: } /* agedc > 0 */
1.214 brouard 10242: } /* end if */
1.136 brouard 10243: else if(s[m][i] !=9){ /* Standard case, age in fractional
10244: years but with the precision of a month */
10245: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10246: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10247: agev[m][i]=1;
10248: else if(agev[m][i] < *agemin){
10249: *agemin=agev[m][i];
10250: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10251: }
10252: else if(agev[m][i] >*agemax){
10253: *agemax=agev[m][i];
1.156 brouard 10254: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10255: }
10256: /*agev[m][i]=anint[m][i]-annais[i];*/
10257: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10258: } /* en if 9*/
1.136 brouard 10259: else { /* =9 */
1.214 brouard 10260: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10261: agev[m][i]=1;
10262: s[m][i]=-1;
10263: }
10264: }
1.214 brouard 10265: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10266: agev[m][i]=1;
1.214 brouard 10267: else{
10268: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10269: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10270: agev[m][i]=0;
10271: }
10272: } /* End for lastpass */
10273: }
1.136 brouard 10274:
10275: for (i=1; i<=imx; i++) {
10276: for(m=firstpass; (m<=lastpass); m++){
10277: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10278: (*nberr)++;
1.136 brouard 10279: 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);
10280: 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);
10281: return 1;
10282: }
10283: }
10284: }
10285:
10286: /*for (i=1; i<=imx; i++){
10287: for (m=firstpass; (m<lastpass); m++){
10288: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10289: }
10290:
10291: }*/
10292:
10293:
1.139 brouard 10294: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10295: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10296:
10297: return (0);
1.164 brouard 10298: /* endread:*/
1.136 brouard 10299: printf("Exiting calandcheckages: ");
10300: return (1);
10301: }
10302:
1.172 brouard 10303: #if defined(_MSC_VER)
10304: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10305: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10306: //#include "stdafx.h"
10307: //#include <stdio.h>
10308: //#include <tchar.h>
10309: //#include <windows.h>
10310: //#include <iostream>
10311: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10312:
10313: LPFN_ISWOW64PROCESS fnIsWow64Process;
10314:
10315: BOOL IsWow64()
10316: {
10317: BOOL bIsWow64 = FALSE;
10318:
10319: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10320: // (HANDLE, PBOOL);
10321:
10322: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10323:
10324: HMODULE module = GetModuleHandle(_T("kernel32"));
10325: const char funcName[] = "IsWow64Process";
10326: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10327: GetProcAddress(module, funcName);
10328:
10329: if (NULL != fnIsWow64Process)
10330: {
10331: if (!fnIsWow64Process(GetCurrentProcess(),
10332: &bIsWow64))
10333: //throw std::exception("Unknown error");
10334: printf("Unknown error\n");
10335: }
10336: return bIsWow64 != FALSE;
10337: }
10338: #endif
1.177 brouard 10339:
1.191 brouard 10340: void syscompilerinfo(int logged)
1.292 brouard 10341: {
10342: #include <stdint.h>
10343:
10344: /* #include "syscompilerinfo.h"*/
1.185 brouard 10345: /* command line Intel compiler 32bit windows, XP compatible:*/
10346: /* /GS /W3 /Gy
10347: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10348: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10349: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10350: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10351: */
10352: /* 64 bits */
1.185 brouard 10353: /*
10354: /GS /W3 /Gy
10355: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10356: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10357: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10358: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10359: /* Optimization are useless and O3 is slower than O2 */
10360: /*
10361: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10362: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10363: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10364: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10365: */
1.186 brouard 10366: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10367: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10368: /PDB:"visual studio
10369: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10370: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10371: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10372: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10373: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10374: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10375: uiAccess='false'"
10376: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10377: /NOLOGO /TLBID:1
10378: */
1.292 brouard 10379:
10380:
1.177 brouard 10381: #if defined __INTEL_COMPILER
1.178 brouard 10382: #if defined(__GNUC__)
10383: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10384: #endif
1.177 brouard 10385: #elif defined(__GNUC__)
1.179 brouard 10386: #ifndef __APPLE__
1.174 brouard 10387: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10388: #endif
1.177 brouard 10389: struct utsname sysInfo;
1.178 brouard 10390: int cross = CROSS;
10391: if (cross){
10392: printf("Cross-");
1.191 brouard 10393: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10394: }
1.174 brouard 10395: #endif
10396:
1.191 brouard 10397: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10398: #if defined(__clang__)
1.191 brouard 10399: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10400: #endif
10401: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10402: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10403: #endif
10404: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10405: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10406: #endif
10407: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10408: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10409: #endif
10410: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10411: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10412: #endif
10413: #if defined(_MSC_VER)
1.191 brouard 10414: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10415: #endif
10416: #if defined(__PGI)
1.191 brouard 10417: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10418: #endif
10419: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10420: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10421: #endif
1.191 brouard 10422: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10423:
1.167 brouard 10424: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10425: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10426: // Windows (x64 and x86)
1.191 brouard 10427: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10428: #elif __unix__ // all unices, not all compilers
10429: // Unix
1.191 brouard 10430: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10431: #elif __linux__
10432: // linux
1.191 brouard 10433: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10434: #elif __APPLE__
1.174 brouard 10435: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10436: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10437: #endif
10438:
10439: /* __MINGW32__ */
10440: /* __CYGWIN__ */
10441: /* __MINGW64__ */
10442: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10443: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10444: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10445: /* _WIN64 // Defined for applications for Win64. */
10446: /* _M_X64 // Defined for compilations that target x64 processors. */
10447: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10448:
1.167 brouard 10449: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10450: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10451: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10452: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10453: #else
1.191 brouard 10454: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10455: #endif
10456:
1.169 brouard 10457: #if defined(__GNUC__)
10458: # if defined(__GNUC_PATCHLEVEL__)
10459: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10460: + __GNUC_MINOR__ * 100 \
10461: + __GNUC_PATCHLEVEL__)
10462: # else
10463: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10464: + __GNUC_MINOR__ * 100)
10465: # endif
1.174 brouard 10466: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10467: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10468:
10469: if (uname(&sysInfo) != -1) {
10470: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10471: 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 10472: }
10473: else
10474: perror("uname() error");
1.179 brouard 10475: //#ifndef __INTEL_COMPILER
10476: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10477: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10478: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10479: #endif
1.169 brouard 10480: #endif
1.172 brouard 10481:
1.286 brouard 10482: // void main ()
1.172 brouard 10483: // {
1.169 brouard 10484: #if defined(_MSC_VER)
1.174 brouard 10485: if (IsWow64()){
1.191 brouard 10486: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10487: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10488: }
10489: else{
1.191 brouard 10490: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10491: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10492: }
1.172 brouard 10493: // printf("\nPress Enter to continue...");
10494: // getchar();
10495: // }
10496:
1.169 brouard 10497: #endif
10498:
1.167 brouard 10499:
1.219 brouard 10500: }
1.136 brouard 10501:
1.219 brouard 10502: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10503: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10504: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10505: /* double ftolpl = 1.e-10; */
1.180 brouard 10506: double age, agebase, agelim;
1.203 brouard 10507: double tot;
1.180 brouard 10508:
1.202 brouard 10509: strcpy(filerespl,"PL_");
10510: strcat(filerespl,fileresu);
10511: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10512: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10513: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10514: }
1.288 brouard 10515: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10516: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10517: pstamp(ficrespl);
1.288 brouard 10518: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10519: fprintf(ficrespl,"#Age ");
10520: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10521: fprintf(ficrespl,"\n");
1.180 brouard 10522:
1.219 brouard 10523: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10524:
1.219 brouard 10525: agebase=ageminpar;
10526: agelim=agemaxpar;
1.180 brouard 10527:
1.227 brouard 10528: /* i1=pow(2,ncoveff); */
1.234 brouard 10529: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10530: if (cptcovn < 1){i1=1;}
1.180 brouard 10531:
1.238 brouard 10532: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10533: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10534: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10535: continue;
1.235 brouard 10536:
1.238 brouard 10537: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10538: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10539: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10540: /* k=k+1; */
10541: /* to clean */
10542: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10543: fprintf(ficrespl,"#******");
10544: printf("#******");
10545: fprintf(ficlog,"#******");
10546: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10547: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10548: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10549: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10550: }
10551: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10552: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10553: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10554: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10555: }
10556: fprintf(ficrespl,"******\n");
10557: printf("******\n");
10558: fprintf(ficlog,"******\n");
10559: if(invalidvarcomb[k]){
10560: printf("\nCombination (%d) ignored because no case \n",k);
10561: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10562: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10563: continue;
10564: }
1.219 brouard 10565:
1.238 brouard 10566: fprintf(ficrespl,"#Age ");
10567: for(j=1;j<=cptcoveff;j++) {
10568: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10569: }
10570: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10571: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10572:
1.238 brouard 10573: for (age=agebase; age<=agelim; age++){
10574: /* for (age=agebase; age<=agebase; age++){ */
10575: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10576: fprintf(ficrespl,"%.0f ",age );
10577: for(j=1;j<=cptcoveff;j++)
10578: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10579: tot=0.;
10580: for(i=1; i<=nlstate;i++){
10581: tot += prlim[i][i];
10582: fprintf(ficrespl," %.5f", prlim[i][i]);
10583: }
10584: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10585: } /* Age */
10586: /* was end of cptcod */
10587: } /* cptcov */
10588: } /* nres */
1.219 brouard 10589: return 0;
1.180 brouard 10590: }
10591:
1.218 brouard 10592: 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 10593: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10594:
10595: /* Computes the back prevalence limit for any combination of covariate values
10596: * at any age between ageminpar and agemaxpar
10597: */
1.235 brouard 10598: int i, j, k, i1, nres=0 ;
1.217 brouard 10599: /* double ftolpl = 1.e-10; */
10600: double age, agebase, agelim;
10601: double tot;
1.218 brouard 10602: /* double ***mobaverage; */
10603: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10604:
10605: strcpy(fileresplb,"PLB_");
10606: strcat(fileresplb,fileresu);
10607: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10608: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10609: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10610: }
1.288 brouard 10611: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10612: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10613: pstamp(ficresplb);
1.288 brouard 10614: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10615: fprintf(ficresplb,"#Age ");
10616: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10617: fprintf(ficresplb,"\n");
10618:
1.218 brouard 10619:
10620: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10621:
10622: agebase=ageminpar;
10623: agelim=agemaxpar;
10624:
10625:
1.227 brouard 10626: i1=pow(2,cptcoveff);
1.218 brouard 10627: if (cptcovn < 1){i1=1;}
1.227 brouard 10628:
1.238 brouard 10629: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10630: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10631: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10632: continue;
10633: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10634: fprintf(ficresplb,"#******");
10635: printf("#******");
10636: fprintf(ficlog,"#******");
10637: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10638: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10639: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10640: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10641: }
10642: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10643: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10644: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10645: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10646: }
10647: fprintf(ficresplb,"******\n");
10648: printf("******\n");
10649: fprintf(ficlog,"******\n");
10650: if(invalidvarcomb[k]){
10651: printf("\nCombination (%d) ignored because no cases \n",k);
10652: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10653: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10654: continue;
10655: }
1.218 brouard 10656:
1.238 brouard 10657: fprintf(ficresplb,"#Age ");
10658: for(j=1;j<=cptcoveff;j++) {
10659: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10660: }
10661: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10662: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10663:
10664:
1.238 brouard 10665: for (age=agebase; age<=agelim; age++){
10666: /* for (age=agebase; age<=agebase; age++){ */
10667: if(mobilavproj > 0){
10668: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10669: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10670: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10671: }else if (mobilavproj == 0){
10672: 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);
10673: 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);
10674: exit(1);
10675: }else{
10676: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10677: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10678: /* printf("TOTOT\n"); */
10679: /* exit(1); */
1.238 brouard 10680: }
10681: fprintf(ficresplb,"%.0f ",age );
10682: for(j=1;j<=cptcoveff;j++)
10683: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10684: tot=0.;
10685: for(i=1; i<=nlstate;i++){
10686: tot += bprlim[i][i];
10687: fprintf(ficresplb," %.5f", bprlim[i][i]);
10688: }
10689: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10690: } /* Age */
10691: /* was end of cptcod */
1.255 brouard 10692: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10693: } /* end of any combination */
10694: } /* end of nres */
1.218 brouard 10695: /* hBijx(p, bage, fage); */
10696: /* fclose(ficrespijb); */
10697:
10698: return 0;
1.217 brouard 10699: }
1.218 brouard 10700:
1.180 brouard 10701: int hPijx(double *p, int bage, int fage){
10702: /*------------- h Pij x at various ages ------------*/
10703:
10704: int stepsize;
10705: int agelim;
10706: int hstepm;
10707: int nhstepm;
1.235 brouard 10708: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10709:
10710: double agedeb;
10711: double ***p3mat;
10712:
1.201 brouard 10713: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10714: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10715: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10716: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10717: }
10718: printf("Computing pij: result on file '%s' \n", filerespij);
10719: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10720:
10721: stepsize=(int) (stepm+YEARM-1)/YEARM;
10722: /*if (stepm<=24) stepsize=2;*/
10723:
10724: agelim=AGESUP;
10725: hstepm=stepsize*YEARM; /* Every year of age */
10726: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10727:
1.180 brouard 10728: /* hstepm=1; aff par mois*/
10729: pstamp(ficrespij);
10730: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10731: i1= pow(2,cptcoveff);
1.218 brouard 10732: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10733: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10734: /* k=k+1; */
1.235 brouard 10735: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10736: for(k=1; k<=i1;k++){
1.253 brouard 10737: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10738: continue;
1.183 brouard 10739: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10740: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10741: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10742: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10743: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10744: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10745: }
1.183 brouard 10746: fprintf(ficrespij,"******\n");
10747:
10748: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10749: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10750: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10751:
10752: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10753:
1.183 brouard 10754: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10755: oldm=oldms;savm=savms;
1.235 brouard 10756: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10757: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10758: for(i=1; i<=nlstate;i++)
10759: for(j=1; j<=nlstate+ndeath;j++)
10760: fprintf(ficrespij," %1d-%1d",i,j);
10761: fprintf(ficrespij,"\n");
10762: for (h=0; h<=nhstepm; h++){
10763: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10764: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10765: for(i=1; i<=nlstate;i++)
10766: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10767: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10768: fprintf(ficrespij,"\n");
10769: }
1.183 brouard 10770: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10771: fprintf(ficrespij,"\n");
10772: }
1.180 brouard 10773: /*}*/
10774: }
1.218 brouard 10775: return 0;
1.180 brouard 10776: }
1.218 brouard 10777:
10778: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10779: /*------------- h Bij x at various ages ------------*/
10780:
10781: int stepsize;
1.218 brouard 10782: /* int agelim; */
10783: int ageminl;
1.217 brouard 10784: int hstepm;
10785: int nhstepm;
1.238 brouard 10786: int h, i, i1, j, k, nres;
1.218 brouard 10787:
1.217 brouard 10788: double agedeb;
10789: double ***p3mat;
1.218 brouard 10790:
10791: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10792: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10793: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10794: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10795: }
10796: printf("Computing pij back: result on file '%s' \n", filerespijb);
10797: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10798:
10799: stepsize=(int) (stepm+YEARM-1)/YEARM;
10800: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10801:
1.218 brouard 10802: /* agelim=AGESUP; */
1.289 brouard 10803: ageminl=AGEINF; /* was 30 */
1.218 brouard 10804: hstepm=stepsize*YEARM; /* Every year of age */
10805: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10806:
10807: /* hstepm=1; aff par mois*/
10808: pstamp(ficrespijb);
1.255 brouard 10809: 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 10810: i1= pow(2,cptcoveff);
1.218 brouard 10811: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10812: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10813: /* k=k+1; */
1.238 brouard 10814: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10815: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10816: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10817: continue;
10818: fprintf(ficrespijb,"\n#****** ");
10819: for(j=1;j<=cptcoveff;j++)
10820: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10821: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10822: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10823: }
10824: fprintf(ficrespijb,"******\n");
1.264 brouard 10825: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10826: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10827: continue;
10828: }
10829:
10830: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10831: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10832: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10833: 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 */
10834: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10835:
10836: /* nhstepm=nhstepm*YEARM; aff par mois*/
10837:
1.266 brouard 10838: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10839: /* and memory limitations if stepm is small */
10840:
1.238 brouard 10841: /* oldm=oldms;savm=savms; */
10842: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10843: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10844: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10845: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10846: for(i=1; i<=nlstate;i++)
10847: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10848: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10849: fprintf(ficrespijb,"\n");
1.238 brouard 10850: for (h=0; h<=nhstepm; h++){
10851: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10852: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10853: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10854: for(i=1; i<=nlstate;i++)
10855: for(j=1; j<=nlstate+ndeath;j++)
10856: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10857: fprintf(ficrespijb,"\n");
10858: }
10859: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10860: fprintf(ficrespijb,"\n");
10861: } /* end age deb */
10862: } /* end combination */
10863: } /* end nres */
1.218 brouard 10864: return 0;
10865: } /* hBijx */
1.217 brouard 10866:
1.180 brouard 10867:
1.136 brouard 10868: /***********************************************/
10869: /**************** Main Program *****************/
10870: /***********************************************/
10871:
10872: int main(int argc, char *argv[])
10873: {
10874: #ifdef GSL
10875: const gsl_multimin_fminimizer_type *T;
10876: size_t iteri = 0, it;
10877: int rval = GSL_CONTINUE;
10878: int status = GSL_SUCCESS;
10879: double ssval;
10880: #endif
10881: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10882: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10883: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10884: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10885: int jj, ll, li, lj, lk;
1.136 brouard 10886: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10887: int num_filled;
1.136 brouard 10888: int itimes;
10889: int NDIM=2;
10890: int vpopbased=0;
1.235 brouard 10891: int nres=0;
1.258 brouard 10892: int endishere=0;
1.277 brouard 10893: int noffset=0;
1.274 brouard 10894: int ncurrv=0; /* Temporary variable */
10895:
1.164 brouard 10896: char ca[32], cb[32];
1.136 brouard 10897: /* FILE *fichtm; *//* Html File */
10898: /* FILE *ficgp;*/ /*Gnuplot File */
10899: struct stat info;
1.191 brouard 10900: double agedeb=0.;
1.194 brouard 10901:
10902: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10903: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10904:
1.165 brouard 10905: double fret;
1.191 brouard 10906: double dum=0.; /* Dummy variable */
1.136 brouard 10907: double ***p3mat;
1.218 brouard 10908: /* double ***mobaverage; */
1.164 brouard 10909:
10910: char line[MAXLINE];
1.197 brouard 10911: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10912:
1.234 brouard 10913: char modeltemp[MAXLINE];
1.230 brouard 10914: char resultline[MAXLINE];
10915:
1.136 brouard 10916: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10917: char *tok, *val; /* pathtot */
1.290 brouard 10918: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10919: int c, h , cpt, c2;
1.191 brouard 10920: int jl=0;
10921: int i1, j1, jk, stepsize=0;
1.194 brouard 10922: int count=0;
10923:
1.164 brouard 10924: int *tab;
1.136 brouard 10925: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10926: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10927: /* double anprojf, mprojf, jprojf; */
10928: /* double jintmean,mintmean,aintmean; */
10929: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10930: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10931: double yrfproj= 10.0; /* Number of years of forward projections */
10932: double yrbproj= 10.0; /* Number of years of backward projections */
10933: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 10934: int mobilav=0,popforecast=0;
1.191 brouard 10935: int hstepm=0, nhstepm=0;
1.136 brouard 10936: int agemortsup;
10937: float sumlpop=0.;
10938: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10939: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10940:
1.191 brouard 10941: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10942: double ftolpl=FTOL;
10943: double **prlim;
1.217 brouard 10944: double **bprlim;
1.136 brouard 10945: double ***param; /* Matrix of parameters */
1.251 brouard 10946: double ***paramstart; /* Matrix of starting parameter values */
10947: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10948: double **matcov; /* Matrix of covariance */
1.203 brouard 10949: double **hess; /* Hessian matrix */
1.136 brouard 10950: double ***delti3; /* Scale */
10951: double *delti; /* Scale */
10952: double ***eij, ***vareij;
10953: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 10954:
1.136 brouard 10955: double *epj, vepp;
1.164 brouard 10956:
1.273 brouard 10957: double dateprev1, dateprev2;
1.296 brouard 10958: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
10959: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
10960:
1.217 brouard 10961:
1.136 brouard 10962: double **ximort;
1.145 brouard 10963: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10964: int *dcwave;
10965:
1.164 brouard 10966: char z[1]="c";
1.136 brouard 10967:
10968: /*char *strt;*/
10969: char strtend[80];
1.126 brouard 10970:
1.164 brouard 10971:
1.126 brouard 10972: /* setlocale (LC_ALL, ""); */
10973: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10974: /* textdomain (PACKAGE); */
10975: /* setlocale (LC_CTYPE, ""); */
10976: /* setlocale (LC_MESSAGES, ""); */
10977:
10978: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10979: rstart_time = time(NULL);
10980: /* (void) gettimeofday(&start_time,&tzp);*/
10981: start_time = *localtime(&rstart_time);
1.126 brouard 10982: curr_time=start_time;
1.157 brouard 10983: /*tml = *localtime(&start_time.tm_sec);*/
10984: /* strcpy(strstart,asctime(&tml)); */
10985: strcpy(strstart,asctime(&start_time));
1.126 brouard 10986:
10987: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10988: /* tp.tm_sec = tp.tm_sec +86400; */
10989: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10990: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10991: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10992: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10993: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10994: /* strt=asctime(&tmg); */
10995: /* printf("Time(after) =%s",strstart); */
10996: /* (void) time (&time_value);
10997: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10998: * tm = *localtime(&time_value);
10999: * strstart=asctime(&tm);
11000: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11001: */
11002:
11003: nberr=0; /* Number of errors and warnings */
11004: nbwarn=0;
1.184 brouard 11005: #ifdef WIN32
11006: _getcwd(pathcd, size);
11007: #else
1.126 brouard 11008: getcwd(pathcd, size);
1.184 brouard 11009: #endif
1.191 brouard 11010: syscompilerinfo(0);
1.196 brouard 11011: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11012: if(argc <=1){
11013: printf("\nEnter the parameter file name: ");
1.205 brouard 11014: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11015: printf("ERROR Empty parameter file name\n");
11016: goto end;
11017: }
1.126 brouard 11018: i=strlen(pathr);
11019: if(pathr[i-1]=='\n')
11020: pathr[i-1]='\0';
1.156 brouard 11021: i=strlen(pathr);
1.205 brouard 11022: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11023: pathr[i-1]='\0';
1.205 brouard 11024: }
11025: i=strlen(pathr);
11026: if( i==0 ){
11027: printf("ERROR Empty parameter file name\n");
11028: goto end;
11029: }
11030: for (tok = pathr; tok != NULL; ){
1.126 brouard 11031: printf("Pathr |%s|\n",pathr);
11032: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11033: printf("val= |%s| pathr=%s\n",val,pathr);
11034: strcpy (pathtot, val);
11035: if(pathr[0] == '\0') break; /* Dirty */
11036: }
11037: }
1.281 brouard 11038: else if (argc<=2){
11039: strcpy(pathtot,argv[1]);
11040: }
1.126 brouard 11041: else{
11042: strcpy(pathtot,argv[1]);
1.281 brouard 11043: strcpy(z,argv[2]);
11044: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11045: }
11046: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11047: /*cygwin_split_path(pathtot,path,optionfile);
11048: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11049: /* cutv(path,optionfile,pathtot,'\\');*/
11050:
11051: /* Split argv[0], imach program to get pathimach */
11052: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11053: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11054: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11055: /* strcpy(pathimach,argv[0]); */
11056: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11057: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11058: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11059: #ifdef WIN32
11060: _chdir(path); /* Can be a relative path */
11061: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11062: #else
1.126 brouard 11063: chdir(path); /* Can be a relative path */
1.184 brouard 11064: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11065: #endif
11066: printf("Current directory %s!\n",pathcd);
1.126 brouard 11067: strcpy(command,"mkdir ");
11068: strcat(command,optionfilefiname);
11069: if((outcmd=system(command)) != 0){
1.169 brouard 11070: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11071: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11072: /* fclose(ficlog); */
11073: /* exit(1); */
11074: }
11075: /* if((imk=mkdir(optionfilefiname))<0){ */
11076: /* perror("mkdir"); */
11077: /* } */
11078:
11079: /*-------- arguments in the command line --------*/
11080:
1.186 brouard 11081: /* Main Log file */
1.126 brouard 11082: strcat(filelog, optionfilefiname);
11083: strcat(filelog,".log"); /* */
11084: if((ficlog=fopen(filelog,"w"))==NULL) {
11085: printf("Problem with logfile %s\n",filelog);
11086: goto end;
11087: }
11088: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11089: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11090: fprintf(ficlog,"\nEnter the parameter file name: \n");
11091: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11092: path=%s \n\
11093: optionfile=%s\n\
11094: optionfilext=%s\n\
1.156 brouard 11095: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11096:
1.197 brouard 11097: syscompilerinfo(1);
1.167 brouard 11098:
1.126 brouard 11099: printf("Local time (at start):%s",strstart);
11100: fprintf(ficlog,"Local time (at start): %s",strstart);
11101: fflush(ficlog);
11102: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11103: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11104:
11105: /* */
11106: strcpy(fileres,"r");
11107: strcat(fileres, optionfilefiname);
1.201 brouard 11108: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11109: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11110: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11111:
1.186 brouard 11112: /* Main ---------arguments file --------*/
1.126 brouard 11113:
11114: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11115: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11116: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11117: fflush(ficlog);
1.149 brouard 11118: /* goto end; */
11119: exit(70);
1.126 brouard 11120: }
11121:
11122: strcpy(filereso,"o");
1.201 brouard 11123: strcat(filereso,fileresu);
1.126 brouard 11124: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11125: printf("Problem with Output resultfile: %s\n", filereso);
11126: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11127: fflush(ficlog);
11128: goto end;
11129: }
1.278 brouard 11130: /*-------- Rewriting parameter file ----------*/
11131: strcpy(rfileres,"r"); /* "Rparameterfile */
11132: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11133: strcat(rfileres,"."); /* */
11134: strcat(rfileres,optionfilext); /* Other files have txt extension */
11135: if((ficres =fopen(rfileres,"w"))==NULL) {
11136: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11137: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11138: fflush(ficlog);
11139: goto end;
11140: }
11141: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11142:
1.278 brouard 11143:
1.126 brouard 11144: /* Reads comments: lines beginning with '#' */
11145: numlinepar=0;
1.277 brouard 11146: /* Is it a BOM UTF-8 Windows file? */
11147: /* First parameter line */
1.197 brouard 11148: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11149: noffset=0;
11150: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11151: {
11152: noffset=noffset+3;
11153: printf("# File is an UTF8 Bom.\n"); // 0xBF
11154: }
1.302 brouard 11155: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11156: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11157: {
11158: noffset=noffset+2;
11159: printf("# File is an UTF16BE BOM file\n");
11160: }
11161: else if( line[0] == 0 && line[1] == 0)
11162: {
11163: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11164: noffset=noffset+4;
11165: printf("# File is an UTF16BE BOM file\n");
11166: }
11167: } else{
11168: ;/*printf(" Not a BOM file\n");*/
11169: }
11170:
1.197 brouard 11171: /* If line starts with a # it is a comment */
1.277 brouard 11172: if (line[noffset] == '#') {
1.197 brouard 11173: numlinepar++;
11174: fputs(line,stdout);
11175: fputs(line,ficparo);
1.278 brouard 11176: fputs(line,ficres);
1.197 brouard 11177: fputs(line,ficlog);
11178: continue;
11179: }else
11180: break;
11181: }
11182: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11183: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11184: if (num_filled != 5) {
11185: printf("Should be 5 parameters\n");
1.283 brouard 11186: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11187: }
1.126 brouard 11188: numlinepar++;
1.197 brouard 11189: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11190: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11191: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11192: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11193: }
11194: /* Second parameter line */
11195: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11196: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11197: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11198: if (line[0] == '#') {
11199: numlinepar++;
1.283 brouard 11200: printf("%s",line);
11201: fprintf(ficres,"%s",line);
11202: fprintf(ficparo,"%s",line);
11203: fprintf(ficlog,"%s",line);
1.197 brouard 11204: continue;
11205: }else
11206: break;
11207: }
1.223 brouard 11208: 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", \
11209: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11210: if (num_filled != 11) {
11211: 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 11212: printf("but line=%s\n",line);
1.283 brouard 11213: 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");
11214: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11215: }
1.286 brouard 11216: if( lastpass > maxwav){
11217: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11218: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11219: fflush(ficlog);
11220: goto end;
11221: }
11222: 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 11223: 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 11224: 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 11225: 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 11226: }
1.203 brouard 11227: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11228: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11229: /* Third parameter line */
11230: while(fgets(line, MAXLINE, ficpar)) {
11231: /* If line starts with a # it is a comment */
11232: if (line[0] == '#') {
11233: numlinepar++;
1.283 brouard 11234: printf("%s",line);
11235: fprintf(ficres,"%s",line);
11236: fprintf(ficparo,"%s",line);
11237: fprintf(ficlog,"%s",line);
1.197 brouard 11238: continue;
11239: }else
11240: break;
11241: }
1.201 brouard 11242: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11243: if (num_filled != 1){
1.302 brouard 11244: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11245: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11246: model[0]='\0';
11247: goto end;
11248: }
11249: else{
11250: if (model[0]=='+'){
11251: for(i=1; i<=strlen(model);i++)
11252: modeltemp[i-1]=model[i];
1.201 brouard 11253: strcpy(model,modeltemp);
1.197 brouard 11254: }
11255: }
1.199 brouard 11256: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11257: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11258: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11259: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11260: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11261: }
11262: /* 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); */
11263: /* numlinepar=numlinepar+3; /\* In general *\/ */
11264: /* 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 11265: /* 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); */
11266: /* 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 11267: fflush(ficlog);
1.190 brouard 11268: /* if(model[0]=='#'|| model[0]== '\0'){ */
11269: if(model[0]=='#'){
1.279 brouard 11270: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11271: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11272: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11273: if(mle != -1){
1.279 brouard 11274: 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 11275: exit(1);
11276: }
11277: }
1.126 brouard 11278: while((c=getc(ficpar))=='#' && c!= EOF){
11279: ungetc(c,ficpar);
11280: fgets(line, MAXLINE, ficpar);
11281: numlinepar++;
1.195 brouard 11282: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11283: z[0]=line[1];
11284: }
11285: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11286: fputs(line, stdout);
11287: //puts(line);
1.126 brouard 11288: fputs(line,ficparo);
11289: fputs(line,ficlog);
11290: }
11291: ungetc(c,ficpar);
11292:
11293:
1.290 brouard 11294: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11295: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11296: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11297: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11298: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11299: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11300: v1+v2*age+v2*v3 makes cptcovn = 3
11301: */
11302: if (strlen(model)>1)
1.187 brouard 11303: 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 11304: else
1.187 brouard 11305: ncovmodel=2; /* Constant and age */
1.133 brouard 11306: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11307: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11308: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11309: 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);
11310: 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);
11311: fflush(stdout);
11312: fclose (ficlog);
11313: goto end;
11314: }
1.126 brouard 11315: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11316: delti=delti3[1][1];
11317: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11318: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11319: /* We could also provide initial parameters values giving by simple logistic regression
11320: * only one way, that is without matrix product. We will have nlstate maximizations */
11321: /* for(i=1;i<nlstate;i++){ */
11322: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11323: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11324: /* } */
1.126 brouard 11325: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11326: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11327: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11328: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11329: fclose (ficparo);
11330: fclose (ficlog);
11331: goto end;
11332: exit(0);
1.220 brouard 11333: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11334: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11335: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11336: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11337: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11338: matcov=matrix(1,npar,1,npar);
1.203 brouard 11339: hess=matrix(1,npar,1,npar);
1.220 brouard 11340: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11341: /* Read guessed parameters */
1.126 brouard 11342: /* Reads comments: lines beginning with '#' */
11343: while((c=getc(ficpar))=='#' && c!= EOF){
11344: ungetc(c,ficpar);
11345: fgets(line, MAXLINE, ficpar);
11346: numlinepar++;
1.141 brouard 11347: fputs(line,stdout);
1.126 brouard 11348: fputs(line,ficparo);
11349: fputs(line,ficlog);
11350: }
11351: ungetc(c,ficpar);
11352:
11353: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11354: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11355: for(i=1; i <=nlstate; i++){
1.234 brouard 11356: j=0;
1.126 brouard 11357: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11358: if(jj==i) continue;
11359: j++;
1.292 brouard 11360: while((c=getc(ficpar))=='#' && c!= EOF){
11361: ungetc(c,ficpar);
11362: fgets(line, MAXLINE, ficpar);
11363: numlinepar++;
11364: fputs(line,stdout);
11365: fputs(line,ficparo);
11366: fputs(line,ficlog);
11367: }
11368: ungetc(c,ficpar);
1.234 brouard 11369: fscanf(ficpar,"%1d%1d",&i1,&j1);
11370: if ((i1 != i) || (j1 != jj)){
11371: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11372: It might be a problem of design; if ncovcol and the model are correct\n \
11373: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11374: exit(1);
11375: }
11376: fprintf(ficparo,"%1d%1d",i1,j1);
11377: if(mle==1)
11378: printf("%1d%1d",i,jj);
11379: fprintf(ficlog,"%1d%1d",i,jj);
11380: for(k=1; k<=ncovmodel;k++){
11381: fscanf(ficpar," %lf",¶m[i][j][k]);
11382: if(mle==1){
11383: printf(" %lf",param[i][j][k]);
11384: fprintf(ficlog," %lf",param[i][j][k]);
11385: }
11386: else
11387: fprintf(ficlog," %lf",param[i][j][k]);
11388: fprintf(ficparo," %lf",param[i][j][k]);
11389: }
11390: fscanf(ficpar,"\n");
11391: numlinepar++;
11392: if(mle==1)
11393: printf("\n");
11394: fprintf(ficlog,"\n");
11395: fprintf(ficparo,"\n");
1.126 brouard 11396: }
11397: }
11398: fflush(ficlog);
1.234 brouard 11399:
1.251 brouard 11400: /* Reads parameters values */
1.126 brouard 11401: p=param[1][1];
1.251 brouard 11402: pstart=paramstart[1][1];
1.126 brouard 11403:
11404: /* Reads comments: lines beginning with '#' */
11405: while((c=getc(ficpar))=='#' && c!= EOF){
11406: ungetc(c,ficpar);
11407: fgets(line, MAXLINE, ficpar);
11408: numlinepar++;
1.141 brouard 11409: fputs(line,stdout);
1.126 brouard 11410: fputs(line,ficparo);
11411: fputs(line,ficlog);
11412: }
11413: ungetc(c,ficpar);
11414:
11415: for(i=1; i <=nlstate; i++){
11416: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11417: fscanf(ficpar,"%1d%1d",&i1,&j1);
11418: if ( (i1-i) * (j1-j) != 0){
11419: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11420: exit(1);
11421: }
11422: printf("%1d%1d",i,j);
11423: fprintf(ficparo,"%1d%1d",i1,j1);
11424: fprintf(ficlog,"%1d%1d",i1,j1);
11425: for(k=1; k<=ncovmodel;k++){
11426: fscanf(ficpar,"%le",&delti3[i][j][k]);
11427: printf(" %le",delti3[i][j][k]);
11428: fprintf(ficparo," %le",delti3[i][j][k]);
11429: fprintf(ficlog," %le",delti3[i][j][k]);
11430: }
11431: fscanf(ficpar,"\n");
11432: numlinepar++;
11433: printf("\n");
11434: fprintf(ficparo,"\n");
11435: fprintf(ficlog,"\n");
1.126 brouard 11436: }
11437: }
11438: fflush(ficlog);
1.234 brouard 11439:
1.145 brouard 11440: /* Reads covariance matrix */
1.126 brouard 11441: delti=delti3[1][1];
1.220 brouard 11442:
11443:
1.126 brouard 11444: /* 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 11445:
1.126 brouard 11446: /* Reads comments: lines beginning with '#' */
11447: while((c=getc(ficpar))=='#' && c!= EOF){
11448: ungetc(c,ficpar);
11449: fgets(line, MAXLINE, ficpar);
11450: numlinepar++;
1.141 brouard 11451: fputs(line,stdout);
1.126 brouard 11452: fputs(line,ficparo);
11453: fputs(line,ficlog);
11454: }
11455: ungetc(c,ficpar);
1.220 brouard 11456:
1.126 brouard 11457: matcov=matrix(1,npar,1,npar);
1.203 brouard 11458: hess=matrix(1,npar,1,npar);
1.131 brouard 11459: for(i=1; i <=npar; i++)
11460: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11461:
1.194 brouard 11462: /* Scans npar lines */
1.126 brouard 11463: for(i=1; i <=npar; i++){
1.226 brouard 11464: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11465: if(count != 3){
1.226 brouard 11466: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11467: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11468: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11469: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11470: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11471: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11472: exit(1);
1.220 brouard 11473: }else{
1.226 brouard 11474: if(mle==1)
11475: printf("%1d%1d%d",i1,j1,jk);
11476: }
11477: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11478: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11479: for(j=1; j <=i; j++){
1.226 brouard 11480: fscanf(ficpar," %le",&matcov[i][j]);
11481: if(mle==1){
11482: printf(" %.5le",matcov[i][j]);
11483: }
11484: fprintf(ficlog," %.5le",matcov[i][j]);
11485: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11486: }
11487: fscanf(ficpar,"\n");
11488: numlinepar++;
11489: if(mle==1)
1.220 brouard 11490: printf("\n");
1.126 brouard 11491: fprintf(ficlog,"\n");
11492: fprintf(ficparo,"\n");
11493: }
1.194 brouard 11494: /* End of read covariance matrix npar lines */
1.126 brouard 11495: for(i=1; i <=npar; i++)
11496: for(j=i+1;j<=npar;j++)
1.226 brouard 11497: matcov[i][j]=matcov[j][i];
1.126 brouard 11498:
11499: if(mle==1)
11500: printf("\n");
11501: fprintf(ficlog,"\n");
11502:
11503: fflush(ficlog);
11504:
11505: } /* End of mle != -3 */
1.218 brouard 11506:
1.186 brouard 11507: /* Main data
11508: */
1.290 brouard 11509: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11510: /* num=lvector(1,n); */
11511: /* moisnais=vector(1,n); */
11512: /* annais=vector(1,n); */
11513: /* moisdc=vector(1,n); */
11514: /* andc=vector(1,n); */
11515: /* weight=vector(1,n); */
11516: /* agedc=vector(1,n); */
11517: /* cod=ivector(1,n); */
11518: /* for(i=1;i<=n;i++){ */
11519: num=lvector(firstobs,lastobs);
11520: moisnais=vector(firstobs,lastobs);
11521: annais=vector(firstobs,lastobs);
11522: moisdc=vector(firstobs,lastobs);
11523: andc=vector(firstobs,lastobs);
11524: weight=vector(firstobs,lastobs);
11525: agedc=vector(firstobs,lastobs);
11526: cod=ivector(firstobs,lastobs);
11527: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11528: num[i]=0;
11529: moisnais[i]=0;
11530: annais[i]=0;
11531: moisdc[i]=0;
11532: andc[i]=0;
11533: agedc[i]=0;
11534: cod[i]=0;
11535: weight[i]=1.0; /* Equal weights, 1 by default */
11536: }
1.290 brouard 11537: mint=matrix(1,maxwav,firstobs,lastobs);
11538: anint=matrix(1,maxwav,firstobs,lastobs);
11539: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11540: tab=ivector(1,NCOVMAX);
1.144 brouard 11541: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11542: 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 11543:
1.136 brouard 11544: /* Reads data from file datafile */
11545: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11546: goto end;
11547:
11548: /* Calculation of the number of parameters from char model */
1.234 brouard 11549: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11550: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11551: k=3 V4 Tvar[k=3]= 4 (from V4)
11552: k=2 V1 Tvar[k=2]= 1 (from V1)
11553: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11554: */
11555:
11556: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11557: TvarsDind=ivector(1,NCOVMAX); /* */
11558: TvarsD=ivector(1,NCOVMAX); /* */
11559: TvarsQind=ivector(1,NCOVMAX); /* */
11560: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11561: TvarF=ivector(1,NCOVMAX); /* */
11562: TvarFind=ivector(1,NCOVMAX); /* */
11563: TvarV=ivector(1,NCOVMAX); /* */
11564: TvarVind=ivector(1,NCOVMAX); /* */
11565: TvarA=ivector(1,NCOVMAX); /* */
11566: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11567: TvarFD=ivector(1,NCOVMAX); /* */
11568: TvarFDind=ivector(1,NCOVMAX); /* */
11569: TvarFQ=ivector(1,NCOVMAX); /* */
11570: TvarFQind=ivector(1,NCOVMAX); /* */
11571: TvarVD=ivector(1,NCOVMAX); /* */
11572: TvarVDind=ivector(1,NCOVMAX); /* */
11573: TvarVQ=ivector(1,NCOVMAX); /* */
11574: TvarVQind=ivector(1,NCOVMAX); /* */
11575:
1.230 brouard 11576: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11577: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11578: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11579: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11580: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11581: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11582: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11583: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11584: */
11585: /* For model-covariate k tells which data-covariate to use but
11586: because this model-covariate is a construction we invent a new column
11587: ncovcol + k1
11588: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11589: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11590: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11591: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11592: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11593: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11594: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11595: */
1.145 brouard 11596: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11597: 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 11598: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11599: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11600: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11601: 4 covariates (3 plus signs)
11602: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11603: */
1.230 brouard 11604: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11605: * individual dummy, fixed or varying:
11606: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11607: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11608: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11609: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11610: * Tmodelind[1]@9={9,0,3,2,}*/
11611: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11612: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11613: * individual quantitative, fixed or varying:
11614: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11615: * 3, 1, 0, 0, 0, 0, 0, 0},
11616: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11617: /* Main decodemodel */
11618:
1.187 brouard 11619:
1.223 brouard 11620: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11621: goto end;
11622:
1.137 brouard 11623: if((double)(lastobs-imx)/(double)imx > 1.10){
11624: nbwarn++;
11625: 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);
11626: 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);
11627: }
1.136 brouard 11628: /* if(mle==1){*/
1.137 brouard 11629: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11630: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11631: }
11632:
11633: /*-calculation of age at interview from date of interview and age at death -*/
11634: agev=matrix(1,maxwav,1,imx);
11635:
11636: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11637: goto end;
11638:
1.126 brouard 11639:
1.136 brouard 11640: agegomp=(int)agemin;
1.290 brouard 11641: free_vector(moisnais,firstobs,lastobs);
11642: free_vector(annais,firstobs,lastobs);
1.126 brouard 11643: /* free_matrix(mint,1,maxwav,1,n);
11644: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11645: /* free_vector(moisdc,1,n); */
11646: /* free_vector(andc,1,n); */
1.145 brouard 11647: /* */
11648:
1.126 brouard 11649: wav=ivector(1,imx);
1.214 brouard 11650: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11651: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11652: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11653: 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.*/
11654: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11655: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11656:
11657: /* Concatenates waves */
1.214 brouard 11658: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11659: Death is a valid wave (if date is known).
11660: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11661: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11662: and mw[mi+1][i]. dh depends on stepm.
11663: */
11664:
1.126 brouard 11665: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11666: /* Concatenates waves */
1.145 brouard 11667:
1.290 brouard 11668: free_vector(moisdc,firstobs,lastobs);
11669: free_vector(andc,firstobs,lastobs);
1.215 brouard 11670:
1.126 brouard 11671: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11672: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11673: ncodemax[1]=1;
1.145 brouard 11674: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11675: cptcoveff=0;
1.220 brouard 11676: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11677: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11678: }
11679:
11680: ncovcombmax=pow(2,cptcoveff);
11681: invalidvarcomb=ivector(1, ncovcombmax);
11682: for(i=1;i<ncovcombmax;i++)
11683: invalidvarcomb[i]=0;
11684:
1.211 brouard 11685: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11686: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11687: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11688:
1.200 brouard 11689: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11690: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11691: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11692: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11693: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11694: * (currently 0 or 1) in the data.
11695: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11696: * corresponding modality (h,j).
11697: */
11698:
1.145 brouard 11699: h=0;
11700: /*if (cptcovn > 0) */
1.126 brouard 11701: m=pow(2,cptcoveff);
11702:
1.144 brouard 11703: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11704: * For k=4 covariates, h goes from 1 to m=2**k
11705: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11706: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11707: * h\k 1 2 3 4
1.143 brouard 11708: *______________________________
11709: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11710: * 2 2 1 1 1
11711: * 3 i=2 1 2 1 1
11712: * 4 2 2 1 1
11713: * 5 i=3 1 i=2 1 2 1
11714: * 6 2 1 2 1
11715: * 7 i=4 1 2 2 1
11716: * 8 2 2 2 1
1.197 brouard 11717: * 9 i=5 1 i=3 1 i=2 1 2
11718: * 10 2 1 1 2
11719: * 11 i=6 1 2 1 2
11720: * 12 2 2 1 2
11721: * 13 i=7 1 i=4 1 2 2
11722: * 14 2 1 2 2
11723: * 15 i=8 1 2 2 2
11724: * 16 2 2 2 2
1.143 brouard 11725: */
1.212 brouard 11726: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11727: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11728: * and the value of each covariate?
11729: * V1=1, V2=1, V3=2, V4=1 ?
11730: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11731: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11732: * In order to get the real value in the data, we use nbcode
11733: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11734: * We are keeping this crazy system in order to be able (in the future?)
11735: * to have more than 2 values (0 or 1) for a covariate.
11736: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11737: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11738: * bbbbbbbb
11739: * 76543210
11740: * h-1 00000101 (6-1=5)
1.219 brouard 11741: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11742: * &
11743: * 1 00000001 (1)
1.219 brouard 11744: * 00000000 = 1 & ((h-1) >> (k-1))
11745: * +1= 00000001 =1
1.211 brouard 11746: *
11747: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11748: * h' 1101 =2^3+2^2+0x2^1+2^0
11749: * >>k' 11
11750: * & 00000001
11751: * = 00000001
11752: * +1 = 00000010=2 = codtabm(14,3)
11753: * Reverse h=6 and m=16?
11754: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11755: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11756: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11757: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11758: * V3=decodtabm(14,3,2**4)=2
11759: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11760: *(h-1) >> (j-1) 0011 =13 >> 2
11761: * &1 000000001
11762: * = 000000001
11763: * +1= 000000010 =2
11764: * 2211
11765: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11766: * V3=2
1.220 brouard 11767: * codtabm and decodtabm are identical
1.211 brouard 11768: */
11769:
1.145 brouard 11770:
11771: free_ivector(Ndum,-1,NCOVMAX);
11772:
11773:
1.126 brouard 11774:
1.186 brouard 11775: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11776: strcpy(optionfilegnuplot,optionfilefiname);
11777: if(mle==-3)
1.201 brouard 11778: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11779: strcat(optionfilegnuplot,".gp");
11780:
11781: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11782: printf("Problem with file %s",optionfilegnuplot);
11783: }
11784: else{
1.204 brouard 11785: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11786: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11787: //fprintf(ficgp,"set missing 'NaNq'\n");
11788: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11789: }
11790: /* fclose(ficgp);*/
1.186 brouard 11791:
11792:
11793: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11794:
11795: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11796: if(mle==-3)
1.201 brouard 11797: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11798: strcat(optionfilehtm,".htm");
11799: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11800: printf("Problem with %s \n",optionfilehtm);
11801: exit(0);
1.126 brouard 11802: }
11803:
11804: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11805: strcat(optionfilehtmcov,"-cov.htm");
11806: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11807: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11808: }
11809: else{
11810: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11811: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11812: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11813: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11814: }
11815:
1.213 brouard 11816: 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 11817: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11818: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11819: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11820: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11821: \n\
11822: <hr size=\"2\" color=\"#EC5E5E\">\
11823: <ul><li><h4>Parameter files</h4>\n\
11824: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11825: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11826: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11827: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11828: - Date and time at start: %s</ul>\n",\
11829: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11830: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11831: fileres,fileres,\
11832: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11833: fflush(fichtm);
11834:
11835: strcpy(pathr,path);
11836: strcat(pathr,optionfilefiname);
1.184 brouard 11837: #ifdef WIN32
11838: _chdir(optionfilefiname); /* Move to directory named optionfile */
11839: #else
1.126 brouard 11840: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11841: #endif
11842:
1.126 brouard 11843:
1.220 brouard 11844: /* Calculates basic frequencies. Computes observed prevalence at single age
11845: and for any valid combination of covariates
1.126 brouard 11846: and prints on file fileres'p'. */
1.251 brouard 11847: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11848: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11849:
11850: fprintf(fichtm,"\n");
1.286 brouard 11851: 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 11852: ftol, stepm);
11853: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11854: ncurrv=1;
11855: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11856: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11857: ncurrv=i;
11858: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11859: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11860: ncurrv=i;
11861: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11862: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11863: ncurrv=i;
11864: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11865: 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", \
11866: nlstate, ndeath, maxwav, mle, weightopt);
11867:
11868: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11869: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11870:
11871:
11872: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11873: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11874: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11875: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11876: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11877: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11878: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11879: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11880: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11881:
1.126 brouard 11882: /* For Powell, parameters are in a vector p[] starting at p[1]
11883: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11884: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11885:
11886: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11887: /* For mortality only */
1.126 brouard 11888: if (mle==-3){
1.136 brouard 11889: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11890: for(i=1;i<=NDIM;i++)
11891: for(j=1;j<=NDIM;j++)
11892: ximort[i][j]=0.;
1.186 brouard 11893: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11894: cens=ivector(firstobs,lastobs);
11895: ageexmed=vector(firstobs,lastobs);
11896: agecens=vector(firstobs,lastobs);
11897: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11898:
1.126 brouard 11899: for (i=1; i<=imx; i++){
11900: dcwave[i]=-1;
11901: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11902: if (s[m][i]>nlstate) {
11903: dcwave[i]=m;
11904: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11905: break;
11906: }
1.126 brouard 11907: }
1.226 brouard 11908:
1.126 brouard 11909: for (i=1; i<=imx; i++) {
11910: if (wav[i]>0){
1.226 brouard 11911: ageexmed[i]=agev[mw[1][i]][i];
11912: j=wav[i];
11913: agecens[i]=1.;
11914:
11915: if (ageexmed[i]> 1 && wav[i] > 0){
11916: agecens[i]=agev[mw[j][i]][i];
11917: cens[i]= 1;
11918: }else if (ageexmed[i]< 1)
11919: cens[i]= -1;
11920: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11921: cens[i]=0 ;
1.126 brouard 11922: }
11923: else cens[i]=-1;
11924: }
11925:
11926: for (i=1;i<=NDIM;i++) {
11927: for (j=1;j<=NDIM;j++)
1.226 brouard 11928: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11929: }
11930:
1.302 brouard 11931: p[1]=0.0268; p[NDIM]=0.083;
11932: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 11933:
11934:
1.136 brouard 11935: #ifdef GSL
11936: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11937: #else
1.126 brouard 11938: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11939: #endif
1.201 brouard 11940: strcpy(filerespow,"POW-MORT_");
11941: strcat(filerespow,fileresu);
1.126 brouard 11942: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11943: printf("Problem with resultfile: %s\n", filerespow);
11944: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11945: }
1.136 brouard 11946: #ifdef GSL
11947: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11948: #else
1.126 brouard 11949: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11950: #endif
1.126 brouard 11951: /* for (i=1;i<=nlstate;i++)
11952: for(j=1;j<=nlstate+ndeath;j++)
11953: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11954: */
11955: fprintf(ficrespow,"\n");
1.136 brouard 11956: #ifdef GSL
11957: /* gsl starts here */
11958: T = gsl_multimin_fminimizer_nmsimplex;
11959: gsl_multimin_fminimizer *sfm = NULL;
11960: gsl_vector *ss, *x;
11961: gsl_multimin_function minex_func;
11962:
11963: /* Initial vertex size vector */
11964: ss = gsl_vector_alloc (NDIM);
11965:
11966: if (ss == NULL){
11967: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11968: }
11969: /* Set all step sizes to 1 */
11970: gsl_vector_set_all (ss, 0.001);
11971:
11972: /* Starting point */
1.126 brouard 11973:
1.136 brouard 11974: x = gsl_vector_alloc (NDIM);
11975:
11976: if (x == NULL){
11977: gsl_vector_free(ss);
11978: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11979: }
11980:
11981: /* Initialize method and iterate */
11982: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11983: /* gsl_vector_set(x, 0, 0.0268); */
11984: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11985: gsl_vector_set(x, 0, p[1]);
11986: gsl_vector_set(x, 1, p[2]);
11987:
11988: minex_func.f = &gompertz_f;
11989: minex_func.n = NDIM;
11990: minex_func.params = (void *)&p; /* ??? */
11991:
11992: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11993: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11994:
11995: printf("Iterations beginning .....\n\n");
11996: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11997:
11998: iteri=0;
11999: while (rval == GSL_CONTINUE){
12000: iteri++;
12001: status = gsl_multimin_fminimizer_iterate(sfm);
12002:
12003: if (status) printf("error: %s\n", gsl_strerror (status));
12004: fflush(0);
12005:
12006: if (status)
12007: break;
12008:
12009: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12010: ssval = gsl_multimin_fminimizer_size (sfm);
12011:
12012: if (rval == GSL_SUCCESS)
12013: printf ("converged to a local maximum at\n");
12014:
12015: printf("%5d ", iteri);
12016: for (it = 0; it < NDIM; it++){
12017: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12018: }
12019: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12020: }
12021:
12022: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12023:
12024: gsl_vector_free(x); /* initial values */
12025: gsl_vector_free(ss); /* inital step size */
12026: for (it=0; it<NDIM; it++){
12027: p[it+1]=gsl_vector_get(sfm->x,it);
12028: fprintf(ficrespow," %.12lf", p[it]);
12029: }
12030: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12031: #endif
12032: #ifdef POWELL
12033: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12034: #endif
1.126 brouard 12035: fclose(ficrespow);
12036:
1.203 brouard 12037: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12038:
12039: for(i=1; i <=NDIM; i++)
12040: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12041: matcov[i][j]=matcov[j][i];
1.126 brouard 12042:
12043: printf("\nCovariance matrix\n ");
1.203 brouard 12044: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12045: for(i=1; i <=NDIM; i++) {
12046: for(j=1;j<=NDIM;j++){
1.220 brouard 12047: printf("%f ",matcov[i][j]);
12048: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12049: }
1.203 brouard 12050: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12051: }
12052:
12053: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12054: for (i=1;i<=NDIM;i++) {
1.126 brouard 12055: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12056: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12057: }
1.302 brouard 12058: lsurv=vector(agegomp,AGESUP);
12059: lpop=vector(agegomp,AGESUP);
12060: tpop=vector(agegomp,AGESUP);
1.126 brouard 12061: lsurv[agegomp]=100000;
12062:
12063: for (k=agegomp;k<=AGESUP;k++) {
12064: agemortsup=k;
12065: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12066: }
12067:
12068: for (k=agegomp;k<agemortsup;k++)
12069: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12070:
12071: for (k=agegomp;k<agemortsup;k++){
12072: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12073: sumlpop=sumlpop+lpop[k];
12074: }
12075:
12076: tpop[agegomp]=sumlpop;
12077: for (k=agegomp;k<(agemortsup-3);k++){
12078: /* tpop[k+1]=2;*/
12079: tpop[k+1]=tpop[k]-lpop[k];
12080: }
12081:
12082:
12083: printf("\nAge lx qx dx Lx Tx e(x)\n");
12084: for (k=agegomp;k<(agemortsup-2);k++)
12085: 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]);
12086:
12087:
12088: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12089: ageminpar=50;
12090: agemaxpar=100;
1.194 brouard 12091: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12092: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12093: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12094: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12095: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12096: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12097: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12098: }else{
12099: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12100: 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 12101: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12102: }
1.201 brouard 12103: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12104: stepm, weightopt,\
12105: model,imx,p,matcov,agemortsup);
12106:
1.302 brouard 12107: free_vector(lsurv,agegomp,AGESUP);
12108: free_vector(lpop,agegomp,AGESUP);
12109: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12110: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12111: free_ivector(dcwave,firstobs,lastobs);
12112: free_vector(agecens,firstobs,lastobs);
12113: free_vector(ageexmed,firstobs,lastobs);
12114: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12115: #ifdef GSL
1.136 brouard 12116: #endif
1.186 brouard 12117: } /* Endof if mle==-3 mortality only */
1.205 brouard 12118: /* Standard */
12119: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12120: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12121: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12122: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12123: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12124: for (k=1; k<=npar;k++)
12125: printf(" %d %8.5f",k,p[k]);
12126: printf("\n");
1.205 brouard 12127: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12128: /* mlikeli uses func not funcone */
1.247 brouard 12129: /* for(i=1;i<nlstate;i++){ */
12130: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12131: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12132: /* } */
1.205 brouard 12133: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12134: }
12135: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12136: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12137: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12138: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12139: }
12140: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12141: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12142: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12143: for (k=1; k<=npar;k++)
12144: printf(" %d %8.5f",k,p[k]);
12145: printf("\n");
12146:
12147: /*--------- results files --------------*/
1.283 brouard 12148: /* 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 12149:
12150:
12151: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12152: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12153: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12154: for(i=1,jk=1; i <=nlstate; i++){
12155: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12156: if (k != i) {
12157: printf("%d%d ",i,k);
12158: fprintf(ficlog,"%d%d ",i,k);
12159: fprintf(ficres,"%1d%1d ",i,k);
12160: for(j=1; j <=ncovmodel; j++){
12161: printf("%12.7f ",p[jk]);
12162: fprintf(ficlog,"%12.7f ",p[jk]);
12163: fprintf(ficres,"%12.7f ",p[jk]);
12164: jk++;
12165: }
12166: printf("\n");
12167: fprintf(ficlog,"\n");
12168: fprintf(ficres,"\n");
12169: }
1.126 brouard 12170: }
12171: }
1.203 brouard 12172: if(mle != 0){
12173: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12174: ftolhess=ftol; /* Usually correct */
1.203 brouard 12175: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12176: 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");
12177: 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");
12178: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12179: for(k=1; k <=(nlstate+ndeath); k++){
12180: if (k != i) {
12181: printf("%d%d ",i,k);
12182: fprintf(ficlog,"%d%d ",i,k);
12183: for(j=1; j <=ncovmodel; j++){
12184: printf("%12.7f W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk], p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
12185: fprintf(ficlog,"%12.7f W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk], p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
12186: jk++;
12187: }
12188: printf("\n");
12189: fprintf(ficlog,"\n");
12190: }
12191: }
1.193 brouard 12192: }
1.203 brouard 12193: } /* end of hesscov and Wald tests */
1.225 brouard 12194:
1.203 brouard 12195: /* */
1.126 brouard 12196: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12197: printf("# Scales (for hessian or gradient estimation)\n");
12198: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12199: for(i=1,jk=1; i <=nlstate; i++){
12200: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12201: if (j!=i) {
12202: fprintf(ficres,"%1d%1d",i,j);
12203: printf("%1d%1d",i,j);
12204: fprintf(ficlog,"%1d%1d",i,j);
12205: for(k=1; k<=ncovmodel;k++){
12206: printf(" %.5e",delti[jk]);
12207: fprintf(ficlog," %.5e",delti[jk]);
12208: fprintf(ficres," %.5e",delti[jk]);
12209: jk++;
12210: }
12211: printf("\n");
12212: fprintf(ficlog,"\n");
12213: fprintf(ficres,"\n");
12214: }
1.126 brouard 12215: }
12216: }
12217:
12218: 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 12219: if(mle >= 1) /* To big for the screen */
1.126 brouard 12220: 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");
12221: 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");
12222: /* # 121 Var(a12)\n\ */
12223: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12224: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12225: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12226: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12227: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12228: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12229: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12230:
12231:
12232: /* Just to have a covariance matrix which will be more understandable
12233: even is we still don't want to manage dictionary of variables
12234: */
12235: for(itimes=1;itimes<=2;itimes++){
12236: jj=0;
12237: for(i=1; i <=nlstate; i++){
1.225 brouard 12238: for(j=1; j <=nlstate+ndeath; j++){
12239: if(j==i) continue;
12240: for(k=1; k<=ncovmodel;k++){
12241: jj++;
12242: ca[0]= k+'a'-1;ca[1]='\0';
12243: if(itimes==1){
12244: if(mle>=1)
12245: printf("#%1d%1d%d",i,j,k);
12246: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12247: fprintf(ficres,"#%1d%1d%d",i,j,k);
12248: }else{
12249: if(mle>=1)
12250: printf("%1d%1d%d",i,j,k);
12251: fprintf(ficlog,"%1d%1d%d",i,j,k);
12252: fprintf(ficres,"%1d%1d%d",i,j,k);
12253: }
12254: ll=0;
12255: for(li=1;li <=nlstate; li++){
12256: for(lj=1;lj <=nlstate+ndeath; lj++){
12257: if(lj==li) continue;
12258: for(lk=1;lk<=ncovmodel;lk++){
12259: ll++;
12260: if(ll<=jj){
12261: cb[0]= lk +'a'-1;cb[1]='\0';
12262: if(ll<jj){
12263: if(itimes==1){
12264: if(mle>=1)
12265: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12266: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12267: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12268: }else{
12269: if(mle>=1)
12270: printf(" %.5e",matcov[jj][ll]);
12271: fprintf(ficlog," %.5e",matcov[jj][ll]);
12272: fprintf(ficres," %.5e",matcov[jj][ll]);
12273: }
12274: }else{
12275: if(itimes==1){
12276: if(mle>=1)
12277: printf(" Var(%s%1d%1d)",ca,i,j);
12278: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12279: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12280: }else{
12281: if(mle>=1)
12282: printf(" %.7e",matcov[jj][ll]);
12283: fprintf(ficlog," %.7e",matcov[jj][ll]);
12284: fprintf(ficres," %.7e",matcov[jj][ll]);
12285: }
12286: }
12287: }
12288: } /* end lk */
12289: } /* end lj */
12290: } /* end li */
12291: if(mle>=1)
12292: printf("\n");
12293: fprintf(ficlog,"\n");
12294: fprintf(ficres,"\n");
12295: numlinepar++;
12296: } /* end k*/
12297: } /*end j */
1.126 brouard 12298: } /* end i */
12299: } /* end itimes */
12300:
12301: fflush(ficlog);
12302: fflush(ficres);
1.225 brouard 12303: while(fgets(line, MAXLINE, ficpar)) {
12304: /* If line starts with a # it is a comment */
12305: if (line[0] == '#') {
12306: numlinepar++;
12307: fputs(line,stdout);
12308: fputs(line,ficparo);
12309: fputs(line,ficlog);
1.299 brouard 12310: fputs(line,ficres);
1.225 brouard 12311: continue;
12312: }else
12313: break;
12314: }
12315:
1.209 brouard 12316: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12317: /* ungetc(c,ficpar); */
12318: /* fgets(line, MAXLINE, ficpar); */
12319: /* fputs(line,stdout); */
12320: /* fputs(line,ficparo); */
12321: /* } */
12322: /* ungetc(c,ficpar); */
1.126 brouard 12323:
12324: estepm=0;
1.209 brouard 12325: 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 12326:
12327: if (num_filled != 6) {
12328: 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);
12329: 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);
12330: goto end;
12331: }
12332: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12333: }
12334: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12335: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12336:
1.209 brouard 12337: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12338: if (estepm==0 || estepm < stepm) estepm=stepm;
12339: if (fage <= 2) {
12340: bage = ageminpar;
12341: fage = agemaxpar;
12342: }
12343:
12344: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12345: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12346: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12347:
1.186 brouard 12348: /* Other stuffs, more or less useful */
1.254 brouard 12349: while(fgets(line, MAXLINE, ficpar)) {
12350: /* If line starts with a # it is a comment */
12351: if (line[0] == '#') {
12352: numlinepar++;
12353: fputs(line,stdout);
12354: fputs(line,ficparo);
12355: fputs(line,ficlog);
1.299 brouard 12356: fputs(line,ficres);
1.254 brouard 12357: continue;
12358: }else
12359: break;
12360: }
12361:
12362: 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){
12363:
12364: if (num_filled != 7) {
12365: 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);
12366: 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);
12367: goto end;
12368: }
12369: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12370: 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);
12371: 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);
12372: 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 12373: }
1.254 brouard 12374:
12375: while(fgets(line, MAXLINE, ficpar)) {
12376: /* If line starts with a # it is a comment */
12377: if (line[0] == '#') {
12378: numlinepar++;
12379: fputs(line,stdout);
12380: fputs(line,ficparo);
12381: fputs(line,ficlog);
1.299 brouard 12382: fputs(line,ficres);
1.254 brouard 12383: continue;
12384: }else
12385: break;
1.126 brouard 12386: }
12387:
12388:
12389: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12390: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12391:
1.254 brouard 12392: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12393: if (num_filled != 1) {
12394: 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);
12395: 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);
12396: goto end;
12397: }
12398: printf("pop_based=%d\n",popbased);
12399: fprintf(ficlog,"pop_based=%d\n",popbased);
12400: fprintf(ficparo,"pop_based=%d\n",popbased);
12401: fprintf(ficres,"pop_based=%d\n",popbased);
12402: }
12403:
1.258 brouard 12404: /* Results */
12405: nresult=0;
12406: do{
12407: if(!fgets(line, MAXLINE, ficpar)){
12408: endishere=1;
12409: parameterline=14;
12410: }else if (line[0] == '#') {
12411: /* If line starts with a # it is a comment */
1.254 brouard 12412: numlinepar++;
12413: fputs(line,stdout);
12414: fputs(line,ficparo);
12415: fputs(line,ficlog);
1.299 brouard 12416: fputs(line,ficres);
1.254 brouard 12417: continue;
1.258 brouard 12418: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12419: parameterline=11;
1.296 brouard 12420: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12421: parameterline=12;
12422: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
12423: parameterline=13;
12424: else{
12425: parameterline=14;
1.254 brouard 12426: }
1.258 brouard 12427: switch (parameterline){
12428: case 11:
1.296 brouard 12429: 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)){
12430: 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 12431: 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);
12432: 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);
12433: 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);
12434: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12435: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12436: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12437: prvforecast = 1;
12438: }
12439: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.302 brouard 12440: printf("prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12441: fprintf(ficlog,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12442: fprintf(ficres,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12443: prvforecast = 2;
12444: }
12445: else {
12446: 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);
12447: 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);
12448: goto end;
1.258 brouard 12449: }
1.254 brouard 12450: break;
1.258 brouard 12451: case 12:
1.296 brouard 12452: 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)){
12453: 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);
12454: 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);
12455: 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);
12456: 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);
12457: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12458: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12459: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12460: prvbackcast = 1;
12461: }
12462: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.302 brouard 12463: printf("prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12464: fprintf(ficlog,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12465: fprintf(ficres,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12466: prvbackcast = 2;
12467: }
12468: else {
12469: 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);
12470: 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);
12471: goto end;
1.258 brouard 12472: }
1.230 brouard 12473: break;
1.258 brouard 12474: case 13:
12475: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
12476: if (num_filled == 0){
12477: resultline[0]='\0';
12478: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12479: fprintf(ficlog,"Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
12480: break;
12481: } else if (num_filled != 1){
12482: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12483: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
12484: }
12485: nresult++; /* Sum of resultlines */
12486: printf("Result %d: result=%s\n",nresult, resultline);
12487: if(nresult > MAXRESULTLINES){
12488: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12489: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
12490: goto end;
12491: }
12492: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
12493: fprintf(ficparo,"result: %s\n",resultline);
12494: fprintf(ficres,"result: %s\n",resultline);
12495: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 12496: break;
1.303 ! brouard 12497: case 14:
! 12498: printf("Error: Unknown command '%s'\n",line);
! 12499: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
! 12500: if(ncovmodel >=2 && nresult==0 ){
1.259 brouard 12501: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.303 ! brouard 12502: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12503: }
1.303 ! brouard 12504: goto end;
1.259 brouard 12505: break;
1.258 brouard 12506: default:
12507: nresult=1;
12508: decoderesult(".",nresult ); /* No covariate */
12509: }
12510: } /* End switch parameterline */
12511: }while(endishere==0); /* End do */
1.126 brouard 12512:
1.230 brouard 12513: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12514: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12515:
12516: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12517: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12518: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12519: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12520: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12521: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12522: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12523: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12524: }else{
1.270 brouard 12525: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12526: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12527: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12528: if(prvforecast==1){
12529: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12530: jprojd=jproj1;
12531: mprojd=mproj1;
12532: anprojd=anproj1;
12533: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12534: jprojf=jproj2;
12535: mprojf=mproj2;
12536: anprojf=anproj2;
12537: } else if(prvforecast == 2){
12538: dateprojd=dateintmean;
12539: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12540: dateprojf=dateintmean+yrfproj;
12541: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12542: }
12543: if(prvbackcast==1){
12544: datebackd=(jback1+12*mback1+365*anback1)/365;
12545: jbackd=jback1;
12546: mbackd=mback1;
12547: anbackd=anback1;
12548: datebackf=(jback2+12*mback2+365*anback2)/365;
12549: jbackf=jback2;
12550: mbackf=mback2;
12551: anbackf=anback2;
12552: } else if(prvbackcast == 2){
12553: datebackd=dateintmean;
12554: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12555: datebackf=dateintmean-yrbproj;
12556: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12557: }
12558:
12559: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12560: }
12561: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12562: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12563: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12564:
1.225 brouard 12565: /*------------ free_vector -------------*/
12566: /* chdir(path); */
1.220 brouard 12567:
1.215 brouard 12568: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12569: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12570: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12571: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12572: free_lvector(num,firstobs,lastobs);
12573: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12574: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12575: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12576: fclose(ficparo);
12577: fclose(ficres);
1.220 brouard 12578:
12579:
1.186 brouard 12580: /* Other results (useful)*/
1.220 brouard 12581:
12582:
1.126 brouard 12583: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12584: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12585: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12586: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12587: fclose(ficrespl);
12588:
12589: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12590: /*#include "hpijx.h"*/
12591: hPijx(p, bage, fage);
1.145 brouard 12592: fclose(ficrespij);
1.227 brouard 12593:
1.220 brouard 12594: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12595: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12596: k=1;
1.126 brouard 12597: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12598:
1.269 brouard 12599: /* Prevalence for each covariate combination in probs[age][status][cov] */
12600: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12601: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12602: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12603: for(k=1;k<=ncovcombmax;k++)
12604: probs[i][j][k]=0.;
1.269 brouard 12605: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12606: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12607: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12608: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12609: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12610: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12611: for(k=1;k<=ncovcombmax;k++)
12612: mobaverages[i][j][k]=0.;
1.219 brouard 12613: mobaverage=mobaverages;
12614: if (mobilav!=0) {
1.235 brouard 12615: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12616: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12617: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12618: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12619: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12620: }
1.269 brouard 12621: } else if (mobilavproj !=0) {
1.235 brouard 12622: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12623: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12624: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12625: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12626: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12627: }
1.269 brouard 12628: }else{
12629: printf("Internal error moving average\n");
12630: fflush(stdout);
12631: exit(1);
1.219 brouard 12632: }
12633: }/* end if moving average */
1.227 brouard 12634:
1.126 brouard 12635: /*---------- Forecasting ------------------*/
1.296 brouard 12636: if(prevfcast==1){
12637: /* /\* if(stepm ==1){*\/ */
12638: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12639: /*This done previously after freqsummary.*/
12640: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12641: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12642:
12643: /* } else if (prvforecast==2){ */
12644: /* /\* if(stepm ==1){*\/ */
12645: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12646: /* } */
12647: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12648: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12649: }
1.269 brouard 12650:
1.296 brouard 12651: /* Prevbcasting */
12652: if(prevbcast==1){
1.219 brouard 12653: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12654: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12655: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12656:
12657: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12658:
12659: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12660:
1.219 brouard 12661: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12662: fclose(ficresplb);
12663:
1.222 brouard 12664: hBijx(p, bage, fage, mobaverage);
12665: fclose(ficrespijb);
1.219 brouard 12666:
1.296 brouard 12667: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12668: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12669: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12670: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12671: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12672: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12673:
12674:
1.269 brouard 12675: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12676:
12677:
1.269 brouard 12678: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12679: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12680: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12681: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12682: } /* end Prevbcasting */
1.268 brouard 12683:
1.186 brouard 12684:
12685: /* ------ Other prevalence ratios------------ */
1.126 brouard 12686:
1.215 brouard 12687: free_ivector(wav,1,imx);
12688: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12689: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12690: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12691:
12692:
1.127 brouard 12693: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12694:
1.201 brouard 12695: strcpy(filerese,"E_");
12696: strcat(filerese,fileresu);
1.126 brouard 12697: if((ficreseij=fopen(filerese,"w"))==NULL) {
12698: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12699: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12700: }
1.208 brouard 12701: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12702: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12703:
12704: pstamp(ficreseij);
1.219 brouard 12705:
1.235 brouard 12706: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12707: if (cptcovn < 1){i1=1;}
12708:
12709: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12710: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12711: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12712: continue;
1.219 brouard 12713: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12714: printf("\n#****** ");
1.225 brouard 12715: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12716: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12717: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12718: }
12719: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12720: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12721: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12722: }
12723: fprintf(ficreseij,"******\n");
1.235 brouard 12724: printf("******\n");
1.219 brouard 12725:
12726: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12727: oldm=oldms;savm=savms;
1.235 brouard 12728: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12729:
1.219 brouard 12730: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12731: }
12732: fclose(ficreseij);
1.208 brouard 12733: printf("done evsij\n");fflush(stdout);
12734: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12735:
1.218 brouard 12736:
1.227 brouard 12737: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12738:
1.201 brouard 12739: strcpy(filerest,"T_");
12740: strcat(filerest,fileresu);
1.127 brouard 12741: if((ficrest=fopen(filerest,"w"))==NULL) {
12742: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12743: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12744: }
1.208 brouard 12745: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12746: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12747: strcpy(fileresstde,"STDE_");
12748: strcat(fileresstde,fileresu);
1.126 brouard 12749: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12750: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12751: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12752: }
1.227 brouard 12753: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12754: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12755:
1.201 brouard 12756: strcpy(filerescve,"CVE_");
12757: strcat(filerescve,fileresu);
1.126 brouard 12758: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12759: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12760: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12761: }
1.227 brouard 12762: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12763: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12764:
1.201 brouard 12765: strcpy(fileresv,"V_");
12766: strcat(fileresv,fileresu);
1.126 brouard 12767: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12768: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12769: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12770: }
1.227 brouard 12771: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12772: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12773:
1.235 brouard 12774: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12775: if (cptcovn < 1){i1=1;}
12776:
12777: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12778: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12779: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12780: continue;
1.242 brouard 12781: printf("\n#****** Result for:");
12782: fprintf(ficrest,"\n#****** Result for:");
12783: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12784: for(j=1;j<=cptcoveff;j++){
12785: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12786: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12787: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12788: }
1.235 brouard 12789: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12790: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12791: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12792: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12793: }
1.208 brouard 12794: fprintf(ficrest,"******\n");
1.227 brouard 12795: fprintf(ficlog,"******\n");
12796: printf("******\n");
1.208 brouard 12797:
12798: fprintf(ficresstdeij,"\n#****** ");
12799: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12800: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12801: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12802: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12803: }
1.235 brouard 12804: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12805: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12806: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12807: }
1.208 brouard 12808: fprintf(ficresstdeij,"******\n");
12809: fprintf(ficrescveij,"******\n");
12810:
12811: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12812: /* pstamp(ficresvij); */
1.225 brouard 12813: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12814: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12815: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12816: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12817: }
1.208 brouard 12818: fprintf(ficresvij,"******\n");
12819:
12820: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12821: oldm=oldms;savm=savms;
1.235 brouard 12822: printf(" cvevsij ");
12823: fprintf(ficlog, " cvevsij ");
12824: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12825: printf(" end cvevsij \n ");
12826: fprintf(ficlog, " end cvevsij \n ");
12827:
12828: /*
12829: */
12830: /* goto endfree; */
12831:
12832: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12833: pstamp(ficrest);
12834:
1.269 brouard 12835: epj=vector(1,nlstate+1);
1.208 brouard 12836: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12837: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12838: cptcod= 0; /* To be deleted */
12839: printf("varevsij vpopbased=%d \n",vpopbased);
12840: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12841: 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 12842: 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 ");
12843: if(vpopbased==1)
12844: 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);
12845: else
1.288 brouard 12846: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12847: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12848: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12849: fprintf(ficrest,"\n");
12850: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12851: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12852: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12853: for(age=bage; age <=fage ;age++){
1.235 brouard 12854: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12855: if (vpopbased==1) {
12856: if(mobilav ==0){
12857: for(i=1; i<=nlstate;i++)
12858: prlim[i][i]=probs[(int)age][i][k];
12859: }else{ /* mobilav */
12860: for(i=1; i<=nlstate;i++)
12861: prlim[i][i]=mobaverage[(int)age][i][k];
12862: }
12863: }
1.219 brouard 12864:
1.227 brouard 12865: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12866: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12867: /* printf(" age %4.0f ",age); */
12868: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12869: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12870: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12871: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12872: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12873: }
12874: epj[nlstate+1] +=epj[j];
12875: }
12876: /* printf(" age %4.0f \n",age); */
1.219 brouard 12877:
1.227 brouard 12878: for(i=1, vepp=0.;i <=nlstate;i++)
12879: for(j=1;j <=nlstate;j++)
12880: vepp += vareij[i][j][(int)age];
12881: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12882: for(j=1;j <=nlstate;j++){
12883: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12884: }
12885: fprintf(ficrest,"\n");
12886: }
1.208 brouard 12887: } /* End vpopbased */
1.269 brouard 12888: free_vector(epj,1,nlstate+1);
1.208 brouard 12889: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12890: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12891: printf("done selection\n");fflush(stdout);
12892: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12893:
1.235 brouard 12894: } /* End k selection */
1.227 brouard 12895:
12896: printf("done State-specific expectancies\n");fflush(stdout);
12897: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12898:
1.288 brouard 12899: /* variance-covariance of forward period prevalence*/
1.269 brouard 12900: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12901:
1.227 brouard 12902:
1.290 brouard 12903: free_vector(weight,firstobs,lastobs);
1.227 brouard 12904: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12905: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12906: free_matrix(anint,1,maxwav,firstobs,lastobs);
12907: free_matrix(mint,1,maxwav,firstobs,lastobs);
12908: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12909: free_ivector(tab,1,NCOVMAX);
12910: fclose(ficresstdeij);
12911: fclose(ficrescveij);
12912: fclose(ficresvij);
12913: fclose(ficrest);
12914: fclose(ficpar);
12915:
12916:
1.126 brouard 12917: /*---------- End : free ----------------*/
1.219 brouard 12918: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12919: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12920: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12921: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12922: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12923: } /* mle==-3 arrives here for freeing */
1.227 brouard 12924: /* endfree:*/
12925: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12926: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12927: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12928: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12929: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12930: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12931: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 12932: free_matrix(matcov,1,npar,1,npar);
12933: free_matrix(hess,1,npar,1,npar);
12934: /*free_vector(delti,1,npar);*/
12935: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12936: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 12937: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 12938: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12939:
12940: free_ivector(ncodemax,1,NCOVMAX);
12941: free_ivector(ncodemaxwundef,1,NCOVMAX);
12942: free_ivector(Dummy,-1,NCOVMAX);
12943: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12944: free_ivector(DummyV,1,NCOVMAX);
12945: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12946: free_ivector(Typevar,-1,NCOVMAX);
12947: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12948: free_ivector(TvarsQ,1,NCOVMAX);
12949: free_ivector(TvarsQind,1,NCOVMAX);
12950: free_ivector(TvarsD,1,NCOVMAX);
12951: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12952: free_ivector(TvarFD,1,NCOVMAX);
12953: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12954: free_ivector(TvarF,1,NCOVMAX);
12955: free_ivector(TvarFind,1,NCOVMAX);
12956: free_ivector(TvarV,1,NCOVMAX);
12957: free_ivector(TvarVind,1,NCOVMAX);
12958: free_ivector(TvarA,1,NCOVMAX);
12959: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12960: free_ivector(TvarFQ,1,NCOVMAX);
12961: free_ivector(TvarFQind,1,NCOVMAX);
12962: free_ivector(TvarVD,1,NCOVMAX);
12963: free_ivector(TvarVDind,1,NCOVMAX);
12964: free_ivector(TvarVQ,1,NCOVMAX);
12965: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12966: free_ivector(Tvarsel,1,NCOVMAX);
12967: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12968: free_ivector(Tposprod,1,NCOVMAX);
12969: free_ivector(Tprod,1,NCOVMAX);
12970: free_ivector(Tvaraff,1,NCOVMAX);
12971: free_ivector(invalidvarcomb,1,ncovcombmax);
12972: free_ivector(Tage,1,NCOVMAX);
12973: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12974: free_ivector(TmodelInvind,1,NCOVMAX);
12975: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12976:
12977: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12978: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12979: fflush(fichtm);
12980: fflush(ficgp);
12981:
1.227 brouard 12982:
1.126 brouard 12983: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12984: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12985: 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 12986: }else{
12987: printf("End of Imach\n");
12988: fprintf(ficlog,"End of Imach\n");
12989: }
12990: printf("See log file on %s\n",filelog);
12991: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12992: /*(void) gettimeofday(&end_time,&tzp);*/
12993: rend_time = time(NULL);
12994: end_time = *localtime(&rend_time);
12995: /* tml = *localtime(&end_time.tm_sec); */
12996: strcpy(strtend,asctime(&end_time));
1.126 brouard 12997: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12998: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12999: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13000:
1.157 brouard 13001: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13002: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13003: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13004: /* printf("Total time was %d uSec.\n", total_usecs);*/
13005: /* if(fileappend(fichtm,optionfilehtm)){ */
13006: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13007: fclose(fichtm);
13008: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13009: fclose(fichtmcov);
13010: fclose(ficgp);
13011: fclose(ficlog);
13012: /*------ End -----------*/
1.227 brouard 13013:
1.281 brouard 13014:
13015: /* Executes gnuplot */
1.227 brouard 13016:
13017: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13018: #ifdef WIN32
1.227 brouard 13019: if (_chdir(pathcd) != 0)
13020: printf("Can't move to directory %s!\n",path);
13021: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13022: #else
1.227 brouard 13023: if(chdir(pathcd) != 0)
13024: printf("Can't move to directory %s!\n", path);
13025: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13026: #endif
1.126 brouard 13027: printf("Current directory %s!\n",pathcd);
13028: /*strcat(plotcmd,CHARSEPARATOR);*/
13029: sprintf(plotcmd,"gnuplot");
1.157 brouard 13030: #ifdef _WIN32
1.126 brouard 13031: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13032: #endif
13033: if(!stat(plotcmd,&info)){
1.158 brouard 13034: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13035: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13036: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13037: }else
13038: strcpy(pplotcmd,plotcmd);
1.157 brouard 13039: #ifdef __unix
1.126 brouard 13040: strcpy(plotcmd,GNUPLOTPROGRAM);
13041: if(!stat(plotcmd,&info)){
1.158 brouard 13042: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13043: }else
13044: strcpy(pplotcmd,plotcmd);
13045: #endif
13046: }else
13047: strcpy(pplotcmd,plotcmd);
13048:
13049: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13050: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13051: strcpy(pplotcmd,plotcmd);
1.227 brouard 13052:
1.126 brouard 13053: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13054: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13055: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13056: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13057: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13058: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13059: strcpy(plotcmd,pplotcmd);
13060: }
1.126 brouard 13061: }
1.158 brouard 13062: printf(" Successful, please wait...");
1.126 brouard 13063: while (z[0] != 'q') {
13064: /* chdir(path); */
1.154 brouard 13065: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13066: scanf("%s",z);
13067: /* if (z[0] == 'c') system("./imach"); */
13068: if (z[0] == 'e') {
1.158 brouard 13069: #ifdef __APPLE__
1.152 brouard 13070: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13071: #elif __linux
13072: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13073: #else
1.152 brouard 13074: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13075: #endif
13076: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13077: system(pplotcmd);
1.126 brouard 13078: }
13079: else if (z[0] == 'g') system(plotcmd);
13080: else if (z[0] == 'q') exit(0);
13081: }
1.227 brouard 13082: end:
1.126 brouard 13083: while (z[0] != 'q') {
1.195 brouard 13084: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13085: scanf("%s",z);
13086: }
1.283 brouard 13087: printf("End\n");
1.282 brouard 13088: exit(0);
1.126 brouard 13089: }
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