Annotation of imach/src/imach.c, revision 1.312
1.312 ! brouard 1: /* $Id: imach.c,v 1.311 2022/04/05 21:03:51 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.312 ! brouard 4: Revision 1.311 2022/04/05 21:03:51 brouard
! 5: Summary: Fixed quantitative covariates
! 6:
! 7: Fixed covariates (dummy or quantitative)
! 8: with missing values have never been allowed but are ERRORS and
! 9: program quits. Standard deviations of fixed covariates were
! 10: wrongly computed. Mean and standard deviations of time varying
! 11: covariates are still not computed.
! 12:
1.311 brouard 13: Revision 1.310 2022/03/17 08:45:53 brouard
14: Summary: 99r25
15:
16: Improving detection of errors: result lines should be compatible with
17: the model.
18:
1.310 brouard 19: Revision 1.309 2021/05/20 12:39:14 brouard
20: Summary: Version 0.99r24
21:
1.309 brouard 22: Revision 1.308 2021/03/31 13:11:57 brouard
23: Summary: Version 0.99r23
24:
25:
26: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
27:
1.308 brouard 28: Revision 1.307 2021/03/08 18:11:32 brouard
29: Summary: 0.99r22 fixed bug on result:
30:
1.307 brouard 31: Revision 1.306 2021/02/20 15:44:02 brouard
32: Summary: Version 0.99r21
33:
34: * imach.c (Module): Fix bug on quitting after result lines!
35: (Module): Version 0.99r21
36:
1.306 brouard 37: Revision 1.305 2021/02/20 15:28:30 brouard
38: * imach.c (Module): Fix bug on quitting after result lines!
39:
1.305 brouard 40: Revision 1.304 2021/02/12 11:34:20 brouard
41: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
42:
1.304 brouard 43: Revision 1.303 2021/02/11 19:50:15 brouard
44: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
45:
1.303 brouard 46: Revision 1.302 2020/02/22 21:00:05 brouard
47: * (Module): imach.c Update mle=-3 (for computing Life expectancy
48: and life table from the data without any state)
49:
1.302 brouard 50: Revision 1.301 2019/06/04 13:51:20 brouard
51: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
52:
1.301 brouard 53: Revision 1.300 2019/05/22 19:09:45 brouard
54: Summary: version 0.99r19 of May 2019
55:
1.300 brouard 56: Revision 1.299 2019/05/22 18:37:08 brouard
57: Summary: Cleaned 0.99r19
58:
1.299 brouard 59: Revision 1.298 2019/05/22 18:19:56 brouard
60: *** empty log message ***
61:
1.298 brouard 62: Revision 1.297 2019/05/22 17:56:10 brouard
63: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
64:
1.297 brouard 65: Revision 1.296 2019/05/20 13:03:18 brouard
66: Summary: Projection syntax simplified
67:
68:
69: We can now start projections, forward or backward, from the mean date
70: of inteviews up to or down to a number of years of projection:
71: prevforecast=1 yearsfproj=15.3 mobil_average=0
72: or
73: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
74: or
75: prevbackcast=1 yearsbproj=12.3 mobil_average=1
76: or
77: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
78:
1.296 brouard 79: Revision 1.295 2019/05/18 09:52:50 brouard
80: Summary: doxygen tex bug
81:
1.295 brouard 82: Revision 1.294 2019/05/16 14:54:33 brouard
83: Summary: There was some wrong lines added
84:
1.294 brouard 85: Revision 1.293 2019/05/09 15:17:34 brouard
86: *** empty log message ***
87:
1.293 brouard 88: Revision 1.292 2019/05/09 14:17:20 brouard
89: Summary: Some updates
90:
1.292 brouard 91: Revision 1.291 2019/05/09 13:44:18 brouard
92: Summary: Before ncovmax
93:
1.291 brouard 94: Revision 1.290 2019/05/09 13:39:37 brouard
95: Summary: 0.99r18 unlimited number of individuals
96:
97: 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.
98:
1.290 brouard 99: Revision 1.289 2018/12/13 09:16:26 brouard
100: Summary: Bug for young ages (<-30) will be in r17
101:
1.289 brouard 102: Revision 1.288 2018/05/02 20:58:27 brouard
103: Summary: Some bugs fixed
104:
1.288 brouard 105: Revision 1.287 2018/05/01 17:57:25 brouard
106: Summary: Bug fixed by providing frequencies only for non missing covariates
107:
1.287 brouard 108: Revision 1.286 2018/04/27 14:27:04 brouard
109: Summary: some minor bugs
110:
1.286 brouard 111: Revision 1.285 2018/04/21 21:02:16 brouard
112: Summary: Some bugs fixed, valgrind tested
113:
1.285 brouard 114: Revision 1.284 2018/04/20 05:22:13 brouard
115: Summary: Computing mean and stdeviation of fixed quantitative variables
116:
1.284 brouard 117: Revision 1.283 2018/04/19 14:49:16 brouard
118: Summary: Some minor bugs fixed
119:
1.283 brouard 120: Revision 1.282 2018/02/27 22:50:02 brouard
121: *** empty log message ***
122:
1.282 brouard 123: Revision 1.281 2018/02/27 19:25:23 brouard
124: Summary: Adding second argument for quitting
125:
1.281 brouard 126: Revision 1.280 2018/02/21 07:58:13 brouard
127: Summary: 0.99r15
128:
129: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
130:
1.280 brouard 131: Revision 1.279 2017/07/20 13:35:01 brouard
132: Summary: temporary working
133:
1.279 brouard 134: Revision 1.278 2017/07/19 14:09:02 brouard
135: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
136:
1.278 brouard 137: Revision 1.277 2017/07/17 08:53:49 brouard
138: Summary: BOM files can be read now
139:
1.277 brouard 140: Revision 1.276 2017/06/30 15:48:31 brouard
141: Summary: Graphs improvements
142:
1.276 brouard 143: Revision 1.275 2017/06/30 13:39:33 brouard
144: Summary: Saito's color
145:
1.275 brouard 146: Revision 1.274 2017/06/29 09:47:08 brouard
147: Summary: Version 0.99r14
148:
1.274 brouard 149: Revision 1.273 2017/06/27 11:06:02 brouard
150: Summary: More documentation on projections
151:
1.273 brouard 152: Revision 1.272 2017/06/27 10:22:40 brouard
153: Summary: Color of backprojection changed from 6 to 5(yellow)
154:
1.272 brouard 155: Revision 1.271 2017/06/27 10:17:50 brouard
156: Summary: Some bug with rint
157:
1.271 brouard 158: Revision 1.270 2017/05/24 05:45:29 brouard
159: *** empty log message ***
160:
1.270 brouard 161: Revision 1.269 2017/05/23 08:39:25 brouard
162: Summary: Code into subroutine, cleanings
163:
1.269 brouard 164: Revision 1.268 2017/05/18 20:09:32 brouard
165: Summary: backprojection and confidence intervals of backprevalence
166:
1.268 brouard 167: Revision 1.267 2017/05/13 10:25:05 brouard
168: Summary: temporary save for backprojection
169:
1.267 brouard 170: Revision 1.266 2017/05/13 07:26:12 brouard
171: Summary: Version 0.99r13 (improvements and bugs fixed)
172:
1.266 brouard 173: Revision 1.265 2017/04/26 16:22:11 brouard
174: Summary: imach 0.99r13 Some bugs fixed
175:
1.265 brouard 176: Revision 1.264 2017/04/26 06:01:29 brouard
177: Summary: Labels in graphs
178:
1.264 brouard 179: Revision 1.263 2017/04/24 15:23:15 brouard
180: Summary: to save
181:
1.263 brouard 182: Revision 1.262 2017/04/18 16:48:12 brouard
183: *** empty log message ***
184:
1.262 brouard 185: Revision 1.261 2017/04/05 10:14:09 brouard
186: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
187:
1.261 brouard 188: Revision 1.260 2017/04/04 17:46:59 brouard
189: Summary: Gnuplot indexations fixed (humm)
190:
1.260 brouard 191: Revision 1.259 2017/04/04 13:01:16 brouard
192: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
193:
1.259 brouard 194: Revision 1.258 2017/04/03 10:17:47 brouard
195: Summary: Version 0.99r12
196:
197: Some cleanings, conformed with updated documentation.
198:
1.258 brouard 199: Revision 1.257 2017/03/29 16:53:30 brouard
200: Summary: Temp
201:
1.257 brouard 202: Revision 1.256 2017/03/27 05:50:23 brouard
203: Summary: Temporary
204:
1.256 brouard 205: Revision 1.255 2017/03/08 16:02:28 brouard
206: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
207:
1.255 brouard 208: Revision 1.254 2017/03/08 07:13:00 brouard
209: Summary: Fixing data parameter line
210:
1.254 brouard 211: Revision 1.253 2016/12/15 11:59:41 brouard
212: Summary: 0.99 in progress
213:
1.253 brouard 214: Revision 1.252 2016/09/15 21:15:37 brouard
215: *** empty log message ***
216:
1.252 brouard 217: Revision 1.251 2016/09/15 15:01:13 brouard
218: Summary: not working
219:
1.251 brouard 220: Revision 1.250 2016/09/08 16:07:27 brouard
221: Summary: continue
222:
1.250 brouard 223: Revision 1.249 2016/09/07 17:14:18 brouard
224: Summary: Starting values from frequencies
225:
1.249 brouard 226: Revision 1.248 2016/09/07 14:10:18 brouard
227: *** empty log message ***
228:
1.248 brouard 229: Revision 1.247 2016/09/02 11:11:21 brouard
230: *** empty log message ***
231:
1.247 brouard 232: Revision 1.246 2016/09/02 08:49:22 brouard
233: *** empty log message ***
234:
1.246 brouard 235: Revision 1.245 2016/09/02 07:25:01 brouard
236: *** empty log message ***
237:
1.245 brouard 238: Revision 1.244 2016/09/02 07:17:34 brouard
239: *** empty log message ***
240:
1.244 brouard 241: Revision 1.243 2016/09/02 06:45:35 brouard
242: *** empty log message ***
243:
1.243 brouard 244: Revision 1.242 2016/08/30 15:01:20 brouard
245: Summary: Fixing a lots
246:
1.242 brouard 247: Revision 1.241 2016/08/29 17:17:25 brouard
248: Summary: gnuplot problem in Back projection to fix
249:
1.241 brouard 250: Revision 1.240 2016/08/29 07:53:18 brouard
251: Summary: Better
252:
1.240 brouard 253: Revision 1.239 2016/08/26 15:51:03 brouard
254: Summary: Improvement in Powell output in order to copy and paste
255:
256: Author:
257:
1.239 brouard 258: Revision 1.238 2016/08/26 14:23:35 brouard
259: Summary: Starting tests of 0.99
260:
1.238 brouard 261: Revision 1.237 2016/08/26 09:20:19 brouard
262: Summary: to valgrind
263:
1.237 brouard 264: Revision 1.236 2016/08/25 10:50:18 brouard
265: *** empty log message ***
266:
1.236 brouard 267: Revision 1.235 2016/08/25 06:59:23 brouard
268: *** empty log message ***
269:
1.235 brouard 270: Revision 1.234 2016/08/23 16:51:20 brouard
271: *** empty log message ***
272:
1.234 brouard 273: Revision 1.233 2016/08/23 07:40:50 brouard
274: Summary: not working
275:
1.233 brouard 276: Revision 1.232 2016/08/22 14:20:21 brouard
277: Summary: not working
278:
1.232 brouard 279: Revision 1.231 2016/08/22 07:17:15 brouard
280: Summary: not working
281:
1.231 brouard 282: Revision 1.230 2016/08/22 06:55:53 brouard
283: Summary: Not working
284:
1.230 brouard 285: Revision 1.229 2016/07/23 09:45:53 brouard
286: Summary: Completing for func too
287:
1.229 brouard 288: Revision 1.228 2016/07/22 17:45:30 brouard
289: Summary: Fixing some arrays, still debugging
290:
1.227 brouard 291: Revision 1.226 2016/07/12 18:42:34 brouard
292: Summary: temp
293:
1.226 brouard 294: Revision 1.225 2016/07/12 08:40:03 brouard
295: Summary: saving but not running
296:
1.225 brouard 297: Revision 1.224 2016/07/01 13:16:01 brouard
298: Summary: Fixes
299:
1.224 brouard 300: Revision 1.223 2016/02/19 09:23:35 brouard
301: Summary: temporary
302:
1.223 brouard 303: Revision 1.222 2016/02/17 08:14:50 brouard
304: Summary: Probably last 0.98 stable version 0.98r6
305:
1.222 brouard 306: Revision 1.221 2016/02/15 23:35:36 brouard
307: Summary: minor bug
308:
1.220 brouard 309: Revision 1.219 2016/02/15 00:48:12 brouard
310: *** empty log message ***
311:
1.219 brouard 312: Revision 1.218 2016/02/12 11:29:23 brouard
313: Summary: 0.99 Back projections
314:
1.218 brouard 315: Revision 1.217 2015/12/23 17:18:31 brouard
316: Summary: Experimental backcast
317:
1.217 brouard 318: Revision 1.216 2015/12/18 17:32:11 brouard
319: Summary: 0.98r4 Warning and status=-2
320:
321: Version 0.98r4 is now:
322: - displaying an error when status is -1, date of interview unknown and date of death known;
323: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
324: Older changes concerning s=-2, dating from 2005 have been supersed.
325:
1.216 brouard 326: Revision 1.215 2015/12/16 08:52:24 brouard
327: Summary: 0.98r4 working
328:
1.215 brouard 329: Revision 1.214 2015/12/16 06:57:54 brouard
330: Summary: temporary not working
331:
1.214 brouard 332: Revision 1.213 2015/12/11 18:22:17 brouard
333: Summary: 0.98r4
334:
1.213 brouard 335: Revision 1.212 2015/11/21 12:47:24 brouard
336: Summary: minor typo
337:
1.212 brouard 338: Revision 1.211 2015/11/21 12:41:11 brouard
339: Summary: 0.98r3 with some graph of projected cross-sectional
340:
341: Author: Nicolas Brouard
342:
1.211 brouard 343: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 344: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 345: Summary: Adding ftolpl parameter
346: Author: N Brouard
347:
348: We had difficulties to get smoothed confidence intervals. It was due
349: to the period prevalence which wasn't computed accurately. The inner
350: parameter ftolpl is now an outer parameter of the .imach parameter
351: file after estepm. If ftolpl is small 1.e-4 and estepm too,
352: computation are long.
353:
1.209 brouard 354: Revision 1.208 2015/11/17 14:31:57 brouard
355: Summary: temporary
356:
1.208 brouard 357: Revision 1.207 2015/10/27 17:36:57 brouard
358: *** empty log message ***
359:
1.207 brouard 360: Revision 1.206 2015/10/24 07:14:11 brouard
361: *** empty log message ***
362:
1.206 brouard 363: Revision 1.205 2015/10/23 15:50:53 brouard
364: Summary: 0.98r3 some clarification for graphs on likelihood contributions
365:
1.205 brouard 366: Revision 1.204 2015/10/01 16:20:26 brouard
367: Summary: Some new graphs of contribution to likelihood
368:
1.204 brouard 369: Revision 1.203 2015/09/30 17:45:14 brouard
370: Summary: looking at better estimation of the hessian
371:
372: Also a better criteria for convergence to the period prevalence And
373: therefore adding the number of years needed to converge. (The
374: prevalence in any alive state shold sum to one
375:
1.203 brouard 376: Revision 1.202 2015/09/22 19:45:16 brouard
377: Summary: Adding some overall graph on contribution to likelihood. Might change
378:
1.202 brouard 379: Revision 1.201 2015/09/15 17:34:58 brouard
380: Summary: 0.98r0
381:
382: - Some new graphs like suvival functions
383: - Some bugs fixed like model=1+age+V2.
384:
1.201 brouard 385: Revision 1.200 2015/09/09 16:53:55 brouard
386: Summary: Big bug thanks to Flavia
387:
388: Even model=1+age+V2. did not work anymore
389:
1.200 brouard 390: Revision 1.199 2015/09/07 14:09:23 brouard
391: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
392:
1.199 brouard 393: Revision 1.198 2015/09/03 07:14:39 brouard
394: Summary: 0.98q5 Flavia
395:
1.198 brouard 396: Revision 1.197 2015/09/01 18:24:39 brouard
397: *** empty log message ***
398:
1.197 brouard 399: Revision 1.196 2015/08/18 23:17:52 brouard
400: Summary: 0.98q5
401:
1.196 brouard 402: Revision 1.195 2015/08/18 16:28:39 brouard
403: Summary: Adding a hack for testing purpose
404:
405: After reading the title, ftol and model lines, if the comment line has
406: a q, starting with #q, the answer at the end of the run is quit. It
407: permits to run test files in batch with ctest. The former workaround was
408: $ echo q | imach foo.imach
409:
1.195 brouard 410: Revision 1.194 2015/08/18 13:32:00 brouard
411: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
412:
1.194 brouard 413: Revision 1.193 2015/08/04 07:17:42 brouard
414: Summary: 0.98q4
415:
1.193 brouard 416: Revision 1.192 2015/07/16 16:49:02 brouard
417: Summary: Fixing some outputs
418:
1.192 brouard 419: Revision 1.191 2015/07/14 10:00:33 brouard
420: Summary: Some fixes
421:
1.191 brouard 422: Revision 1.190 2015/05/05 08:51:13 brouard
423: Summary: Adding digits in output parameters (7 digits instead of 6)
424:
425: Fix 1+age+.
426:
1.190 brouard 427: Revision 1.189 2015/04/30 14:45:16 brouard
428: Summary: 0.98q2
429:
1.189 brouard 430: Revision 1.188 2015/04/30 08:27:53 brouard
431: *** empty log message ***
432:
1.188 brouard 433: Revision 1.187 2015/04/29 09:11:15 brouard
434: *** empty log message ***
435:
1.187 brouard 436: Revision 1.186 2015/04/23 12:01:52 brouard
437: Summary: V1*age is working now, version 0.98q1
438:
439: Some codes had been disabled in order to simplify and Vn*age was
440: working in the optimization phase, ie, giving correct MLE parameters,
441: but, as usual, outputs were not correct and program core dumped.
442:
1.186 brouard 443: Revision 1.185 2015/03/11 13:26:42 brouard
444: Summary: Inclusion of compile and links command line for Intel Compiler
445:
1.185 brouard 446: Revision 1.184 2015/03/11 11:52:39 brouard
447: Summary: Back from Windows 8. Intel Compiler
448:
1.184 brouard 449: Revision 1.183 2015/03/10 20:34:32 brouard
450: Summary: 0.98q0, trying with directest, mnbrak fixed
451:
452: We use directest instead of original Powell test; probably no
453: incidence on the results, but better justifications;
454: We fixed Numerical Recipes mnbrak routine which was wrong and gave
455: wrong results.
456:
1.183 brouard 457: Revision 1.182 2015/02/12 08:19:57 brouard
458: Summary: Trying to keep directest which seems simpler and more general
459: Author: Nicolas Brouard
460:
1.182 brouard 461: Revision 1.181 2015/02/11 23:22:24 brouard
462: Summary: Comments on Powell added
463:
464: Author:
465:
1.181 brouard 466: Revision 1.180 2015/02/11 17:33:45 brouard
467: Summary: Finishing move from main to function (hpijx and prevalence_limit)
468:
1.180 brouard 469: Revision 1.179 2015/01/04 09:57:06 brouard
470: Summary: back to OS/X
471:
1.179 brouard 472: Revision 1.178 2015/01/04 09:35:48 brouard
473: *** empty log message ***
474:
1.178 brouard 475: Revision 1.177 2015/01/03 18:40:56 brouard
476: Summary: Still testing ilc32 on OSX
477:
1.177 brouard 478: Revision 1.176 2015/01/03 16:45:04 brouard
479: *** empty log message ***
480:
1.176 brouard 481: Revision 1.175 2015/01/03 16:33:42 brouard
482: *** empty log message ***
483:
1.175 brouard 484: Revision 1.174 2015/01/03 16:15:49 brouard
485: Summary: Still in cross-compilation
486:
1.174 brouard 487: Revision 1.173 2015/01/03 12:06:26 brouard
488: Summary: trying to detect cross-compilation
489:
1.173 brouard 490: Revision 1.172 2014/12/27 12:07:47 brouard
491: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
492:
1.172 brouard 493: Revision 1.171 2014/12/23 13:26:59 brouard
494: Summary: Back from Visual C
495:
496: Still problem with utsname.h on Windows
497:
1.171 brouard 498: Revision 1.170 2014/12/23 11:17:12 brouard
499: Summary: Cleaning some \%% back to %%
500:
501: The escape was mandatory for a specific compiler (which one?), but too many warnings.
502:
1.170 brouard 503: Revision 1.169 2014/12/22 23:08:31 brouard
504: Summary: 0.98p
505:
506: Outputs some informations on compiler used, OS etc. Testing on different platforms.
507:
1.169 brouard 508: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 509: Summary: update
1.169 brouard 510:
1.168 brouard 511: Revision 1.167 2014/12/22 13:50:56 brouard
512: Summary: Testing uname and compiler version and if compiled 32 or 64
513:
514: Testing on Linux 64
515:
1.167 brouard 516: Revision 1.166 2014/12/22 11:40:47 brouard
517: *** empty log message ***
518:
1.166 brouard 519: Revision 1.165 2014/12/16 11:20:36 brouard
520: Summary: After compiling on Visual C
521:
522: * imach.c (Module): Merging 1.61 to 1.162
523:
1.165 brouard 524: Revision 1.164 2014/12/16 10:52:11 brouard
525: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
526:
527: * imach.c (Module): Merging 1.61 to 1.162
528:
1.164 brouard 529: Revision 1.163 2014/12/16 10:30:11 brouard
530: * imach.c (Module): Merging 1.61 to 1.162
531:
1.163 brouard 532: Revision 1.162 2014/09/25 11:43:39 brouard
533: Summary: temporary backup 0.99!
534:
1.162 brouard 535: Revision 1.1 2014/09/16 11:06:58 brouard
536: Summary: With some code (wrong) for nlopt
537:
538: Author:
539:
540: Revision 1.161 2014/09/15 20:41:41 brouard
541: Summary: Problem with macro SQR on Intel compiler
542:
1.161 brouard 543: Revision 1.160 2014/09/02 09:24:05 brouard
544: *** empty log message ***
545:
1.160 brouard 546: Revision 1.159 2014/09/01 10:34:10 brouard
547: Summary: WIN32
548: Author: Brouard
549:
1.159 brouard 550: Revision 1.158 2014/08/27 17:11:51 brouard
551: *** empty log message ***
552:
1.158 brouard 553: Revision 1.157 2014/08/27 16:26:55 brouard
554: Summary: Preparing windows Visual studio version
555: Author: Brouard
556:
557: In order to compile on Visual studio, time.h is now correct and time_t
558: and tm struct should be used. difftime should be used but sometimes I
559: just make the differences in raw time format (time(&now).
560: Trying to suppress #ifdef LINUX
561: Add xdg-open for __linux in order to open default browser.
562:
1.157 brouard 563: Revision 1.156 2014/08/25 20:10:10 brouard
564: *** empty log message ***
565:
1.156 brouard 566: Revision 1.155 2014/08/25 18:32:34 brouard
567: Summary: New compile, minor changes
568: Author: Brouard
569:
1.155 brouard 570: Revision 1.154 2014/06/20 17:32:08 brouard
571: Summary: Outputs now all graphs of convergence to period prevalence
572:
1.154 brouard 573: Revision 1.153 2014/06/20 16:45:46 brouard
574: Summary: If 3 live state, convergence to period prevalence on same graph
575: Author: Brouard
576:
1.153 brouard 577: Revision 1.152 2014/06/18 17:54:09 brouard
578: Summary: open browser, use gnuplot on same dir than imach if not found in the path
579:
1.152 brouard 580: Revision 1.151 2014/06/18 16:43:30 brouard
581: *** empty log message ***
582:
1.151 brouard 583: Revision 1.150 2014/06/18 16:42:35 brouard
584: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
585: Author: brouard
586:
1.150 brouard 587: Revision 1.149 2014/06/18 15:51:14 brouard
588: Summary: Some fixes in parameter files errors
589: Author: Nicolas Brouard
590:
1.149 brouard 591: Revision 1.148 2014/06/17 17:38:48 brouard
592: Summary: Nothing new
593: Author: Brouard
594:
595: Just a new packaging for OS/X version 0.98nS
596:
1.148 brouard 597: Revision 1.147 2014/06/16 10:33:11 brouard
598: *** empty log message ***
599:
1.147 brouard 600: Revision 1.146 2014/06/16 10:20:28 brouard
601: Summary: Merge
602: Author: Brouard
603:
604: Merge, before building revised version.
605:
1.146 brouard 606: Revision 1.145 2014/06/10 21:23:15 brouard
607: Summary: Debugging with valgrind
608: Author: Nicolas Brouard
609:
610: Lot of changes in order to output the results with some covariates
611: After the Edimburgh REVES conference 2014, it seems mandatory to
612: improve the code.
613: No more memory valgrind error but a lot has to be done in order to
614: continue the work of splitting the code into subroutines.
615: Also, decodemodel has been improved. Tricode is still not
616: optimal. nbcode should be improved. Documentation has been added in
617: the source code.
618:
1.144 brouard 619: Revision 1.143 2014/01/26 09:45:38 brouard
620: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
621:
622: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
623: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
624:
1.143 brouard 625: Revision 1.142 2014/01/26 03:57:36 brouard
626: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
627:
628: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
629:
1.142 brouard 630: Revision 1.141 2014/01/26 02:42:01 brouard
631: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
632:
1.141 brouard 633: Revision 1.140 2011/09/02 10:37:54 brouard
634: Summary: times.h is ok with mingw32 now.
635:
1.140 brouard 636: Revision 1.139 2010/06/14 07:50:17 brouard
637: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
638: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
639:
1.139 brouard 640: Revision 1.138 2010/04/30 18:19:40 brouard
641: *** empty log message ***
642:
1.138 brouard 643: Revision 1.137 2010/04/29 18:11:38 brouard
644: (Module): Checking covariates for more complex models
645: than V1+V2. A lot of change to be done. Unstable.
646:
1.137 brouard 647: Revision 1.136 2010/04/26 20:30:53 brouard
648: (Module): merging some libgsl code. Fixing computation
649: of likelione (using inter/intrapolation if mle = 0) in order to
650: get same likelihood as if mle=1.
651: Some cleaning of code and comments added.
652:
1.136 brouard 653: Revision 1.135 2009/10/29 15:33:14 brouard
654: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
655:
1.135 brouard 656: Revision 1.134 2009/10/29 13:18:53 brouard
657: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
658:
1.134 brouard 659: Revision 1.133 2009/07/06 10:21:25 brouard
660: just nforces
661:
1.133 brouard 662: Revision 1.132 2009/07/06 08:22:05 brouard
663: Many tings
664:
1.132 brouard 665: Revision 1.131 2009/06/20 16:22:47 brouard
666: Some dimensions resccaled
667:
1.131 brouard 668: Revision 1.130 2009/05/26 06:44:34 brouard
669: (Module): Max Covariate is now set to 20 instead of 8. A
670: lot of cleaning with variables initialized to 0. Trying to make
671: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
672:
1.130 brouard 673: Revision 1.129 2007/08/31 13:49:27 lievre
674: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
675:
1.129 lievre 676: Revision 1.128 2006/06/30 13:02:05 brouard
677: (Module): Clarifications on computing e.j
678:
1.128 brouard 679: Revision 1.127 2006/04/28 18:11:50 brouard
680: (Module): Yes the sum of survivors was wrong since
681: imach-114 because nhstepm was no more computed in the age
682: loop. Now we define nhstepma in the age loop.
683: (Module): In order to speed up (in case of numerous covariates) we
684: compute health expectancies (without variances) in a first step
685: and then all the health expectancies with variances or standard
686: deviation (needs data from the Hessian matrices) which slows the
687: computation.
688: In the future we should be able to stop the program is only health
689: expectancies and graph are needed without standard deviations.
690:
1.127 brouard 691: Revision 1.126 2006/04/28 17:23:28 brouard
692: (Module): Yes the sum of survivors was wrong since
693: imach-114 because nhstepm was no more computed in the age
694: loop. Now we define nhstepma in the age loop.
695: Version 0.98h
696:
1.126 brouard 697: Revision 1.125 2006/04/04 15:20:31 lievre
698: Errors in calculation of health expectancies. Age was not initialized.
699: Forecasting file added.
700:
701: Revision 1.124 2006/03/22 17:13:53 lievre
702: Parameters are printed with %lf instead of %f (more numbers after the comma).
703: The log-likelihood is printed in the log file
704:
705: Revision 1.123 2006/03/20 10:52:43 brouard
706: * imach.c (Module): <title> changed, corresponds to .htm file
707: name. <head> headers where missing.
708:
709: * imach.c (Module): Weights can have a decimal point as for
710: English (a comma might work with a correct LC_NUMERIC environment,
711: otherwise the weight is truncated).
712: Modification of warning when the covariates values are not 0 or
713: 1.
714: Version 0.98g
715:
716: Revision 1.122 2006/03/20 09:45:41 brouard
717: (Module): Weights can have a decimal point as for
718: English (a comma might work with a correct LC_NUMERIC environment,
719: otherwise the weight is truncated).
720: Modification of warning when the covariates values are not 0 or
721: 1.
722: Version 0.98g
723:
724: Revision 1.121 2006/03/16 17:45:01 lievre
725: * imach.c (Module): Comments concerning covariates added
726:
727: * imach.c (Module): refinements in the computation of lli if
728: status=-2 in order to have more reliable computation if stepm is
729: not 1 month. Version 0.98f
730:
731: Revision 1.120 2006/03/16 15:10:38 lievre
732: (Module): refinements in the computation of lli if
733: status=-2 in order to have more reliable computation if stepm is
734: not 1 month. Version 0.98f
735:
736: Revision 1.119 2006/03/15 17:42:26 brouard
737: (Module): Bug if status = -2, the loglikelihood was
738: computed as likelihood omitting the logarithm. Version O.98e
739:
740: Revision 1.118 2006/03/14 18:20:07 brouard
741: (Module): varevsij Comments added explaining the second
742: table of variances if popbased=1 .
743: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
744: (Module): Function pstamp added
745: (Module): Version 0.98d
746:
747: Revision 1.117 2006/03/14 17:16:22 brouard
748: (Module): varevsij Comments added explaining the second
749: table of variances if popbased=1 .
750: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
751: (Module): Function pstamp added
752: (Module): Version 0.98d
753:
754: Revision 1.116 2006/03/06 10:29:27 brouard
755: (Module): Variance-covariance wrong links and
756: varian-covariance of ej. is needed (Saito).
757:
758: Revision 1.115 2006/02/27 12:17:45 brouard
759: (Module): One freematrix added in mlikeli! 0.98c
760:
761: Revision 1.114 2006/02/26 12:57:58 brouard
762: (Module): Some improvements in processing parameter
763: filename with strsep.
764:
765: Revision 1.113 2006/02/24 14:20:24 brouard
766: (Module): Memory leaks checks with valgrind and:
767: datafile was not closed, some imatrix were not freed and on matrix
768: allocation too.
769:
770: Revision 1.112 2006/01/30 09:55:26 brouard
771: (Module): Back to gnuplot.exe instead of wgnuplot.exe
772:
773: Revision 1.111 2006/01/25 20:38:18 brouard
774: (Module): Lots of cleaning and bugs added (Gompertz)
775: (Module): Comments can be added in data file. Missing date values
776: can be a simple dot '.'.
777:
778: Revision 1.110 2006/01/25 00:51:50 brouard
779: (Module): Lots of cleaning and bugs added (Gompertz)
780:
781: Revision 1.109 2006/01/24 19:37:15 brouard
782: (Module): Comments (lines starting with a #) are allowed in data.
783:
784: Revision 1.108 2006/01/19 18:05:42 lievre
785: Gnuplot problem appeared...
786: To be fixed
787:
788: Revision 1.107 2006/01/19 16:20:37 brouard
789: Test existence of gnuplot in imach path
790:
791: Revision 1.106 2006/01/19 13:24:36 brouard
792: Some cleaning and links added in html output
793:
794: Revision 1.105 2006/01/05 20:23:19 lievre
795: *** empty log message ***
796:
797: Revision 1.104 2005/09/30 16:11:43 lievre
798: (Module): sump fixed, loop imx fixed, and simplifications.
799: (Module): If the status is missing at the last wave but we know
800: that the person is alive, then we can code his/her status as -2
801: (instead of missing=-1 in earlier versions) and his/her
802: contributions to the likelihood is 1 - Prob of dying from last
803: health status (= 1-p13= p11+p12 in the easiest case of somebody in
804: the healthy state at last known wave). Version is 0.98
805:
806: Revision 1.103 2005/09/30 15:54:49 lievre
807: (Module): sump fixed, loop imx fixed, and simplifications.
808:
809: Revision 1.102 2004/09/15 17:31:30 brouard
810: Add the possibility to read data file including tab characters.
811:
812: Revision 1.101 2004/09/15 10:38:38 brouard
813: Fix on curr_time
814:
815: Revision 1.100 2004/07/12 18:29:06 brouard
816: Add version for Mac OS X. Just define UNIX in Makefile
817:
818: Revision 1.99 2004/06/05 08:57:40 brouard
819: *** empty log message ***
820:
821: Revision 1.98 2004/05/16 15:05:56 brouard
822: New version 0.97 . First attempt to estimate force of mortality
823: directly from the data i.e. without the need of knowing the health
824: state at each age, but using a Gompertz model: log u =a + b*age .
825: This is the basic analysis of mortality and should be done before any
826: other analysis, in order to test if the mortality estimated from the
827: cross-longitudinal survey is different from the mortality estimated
828: from other sources like vital statistic data.
829:
830: The same imach parameter file can be used but the option for mle should be -3.
831:
1.133 brouard 832: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 833: former routines in order to include the new code within the former code.
834:
835: The output is very simple: only an estimate of the intercept and of
836: the slope with 95% confident intervals.
837:
838: Current limitations:
839: A) Even if you enter covariates, i.e. with the
840: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
841: B) There is no computation of Life Expectancy nor Life Table.
842:
843: Revision 1.97 2004/02/20 13:25:42 lievre
844: Version 0.96d. Population forecasting command line is (temporarily)
845: suppressed.
846:
847: Revision 1.96 2003/07/15 15:38:55 brouard
848: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
849: rewritten within the same printf. Workaround: many printfs.
850:
851: Revision 1.95 2003/07/08 07:54:34 brouard
852: * imach.c (Repository):
853: (Repository): Using imachwizard code to output a more meaningful covariance
854: matrix (cov(a12,c31) instead of numbers.
855:
856: Revision 1.94 2003/06/27 13:00:02 brouard
857: Just cleaning
858:
859: Revision 1.93 2003/06/25 16:33:55 brouard
860: (Module): On windows (cygwin) function asctime_r doesn't
861: exist so I changed back to asctime which exists.
862: (Module): Version 0.96b
863:
864: Revision 1.92 2003/06/25 16:30:45 brouard
865: (Module): On windows (cygwin) function asctime_r doesn't
866: exist so I changed back to asctime which exists.
867:
868: Revision 1.91 2003/06/25 15:30:29 brouard
869: * imach.c (Repository): Duplicated warning errors corrected.
870: (Repository): Elapsed time after each iteration is now output. It
871: helps to forecast when convergence will be reached. Elapsed time
872: is stamped in powell. We created a new html file for the graphs
873: concerning matrix of covariance. It has extension -cov.htm.
874:
875: Revision 1.90 2003/06/24 12:34:15 brouard
876: (Module): Some bugs corrected for windows. Also, when
877: mle=-1 a template is output in file "or"mypar.txt with the design
878: of the covariance matrix to be input.
879:
880: Revision 1.89 2003/06/24 12:30:52 brouard
881: (Module): Some bugs corrected for windows. Also, when
882: mle=-1 a template is output in file "or"mypar.txt with the design
883: of the covariance matrix to be input.
884:
885: Revision 1.88 2003/06/23 17:54:56 brouard
886: * 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.
887:
888: Revision 1.87 2003/06/18 12:26:01 brouard
889: Version 0.96
890:
891: Revision 1.86 2003/06/17 20:04:08 brouard
892: (Module): Change position of html and gnuplot routines and added
893: routine fileappend.
894:
895: Revision 1.85 2003/06/17 13:12:43 brouard
896: * imach.c (Repository): Check when date of death was earlier that
897: current date of interview. It may happen when the death was just
898: prior to the death. In this case, dh was negative and likelihood
899: was wrong (infinity). We still send an "Error" but patch by
900: assuming that the date of death was just one stepm after the
901: interview.
902: (Repository): Because some people have very long ID (first column)
903: we changed int to long in num[] and we added a new lvector for
904: memory allocation. But we also truncated to 8 characters (left
905: truncation)
906: (Repository): No more line truncation errors.
907:
908: Revision 1.84 2003/06/13 21:44:43 brouard
909: * imach.c (Repository): Replace "freqsummary" at a correct
910: place. It differs from routine "prevalence" which may be called
911: many times. Probs is memory consuming and must be used with
912: parcimony.
913: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
914:
915: Revision 1.83 2003/06/10 13:39:11 lievre
916: *** empty log message ***
917:
918: Revision 1.82 2003/06/05 15:57:20 brouard
919: Add log in imach.c and fullversion number is now printed.
920:
921: */
922: /*
923: Interpolated Markov Chain
924:
925: Short summary of the programme:
926:
1.227 brouard 927: This program computes Healthy Life Expectancies or State-specific
928: (if states aren't health statuses) Expectancies from
929: cross-longitudinal data. Cross-longitudinal data consist in:
930:
931: -1- a first survey ("cross") where individuals from different ages
932: are interviewed on their health status or degree of disability (in
933: the case of a health survey which is our main interest)
934:
935: -2- at least a second wave of interviews ("longitudinal") which
936: measure each change (if any) in individual health status. Health
937: expectancies are computed from the time spent in each health state
938: according to a model. More health states you consider, more time is
939: necessary to reach the Maximum Likelihood of the parameters involved
940: in the model. The simplest model is the multinomial logistic model
941: where pij is the probability to be observed in state j at the second
942: wave conditional to be observed in state i at the first
943: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
944: etc , where 'age' is age and 'sex' is a covariate. If you want to
945: have a more complex model than "constant and age", you should modify
946: the program where the markup *Covariates have to be included here
947: again* invites you to do it. More covariates you add, slower the
1.126 brouard 948: convergence.
949:
950: The advantage of this computer programme, compared to a simple
951: multinomial logistic model, is clear when the delay between waves is not
952: identical for each individual. Also, if a individual missed an
953: intermediate interview, the information is lost, but taken into
954: account using an interpolation or extrapolation.
955:
956: hPijx is the probability to be observed in state i at age x+h
957: conditional to the observed state i at age x. The delay 'h' can be
958: split into an exact number (nh*stepm) of unobserved intermediate
959: states. This elementary transition (by month, quarter,
960: semester or year) is modelled as a multinomial logistic. The hPx
961: matrix is simply the matrix product of nh*stepm elementary matrices
962: and the contribution of each individual to the likelihood is simply
963: hPijx.
964:
965: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 966: of the life expectancies. It also computes the period (stable) prevalence.
967:
968: Back prevalence and projections:
1.227 brouard 969:
970: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
971: double agemaxpar, double ftolpl, int *ncvyearp, double
972: dateprev1,double dateprev2, int firstpass, int lastpass, int
973: mobilavproj)
974:
975: Computes the back prevalence limit for any combination of
976: covariate values k at any age between ageminpar and agemaxpar and
977: returns it in **bprlim. In the loops,
978:
979: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
980: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
981:
982: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 983: Computes for any combination of covariates k and any age between bage and fage
984: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
985: oldm=oldms;savm=savms;
1.227 brouard 986:
1.267 brouard 987: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 988: Computes the transition matrix starting at age 'age' over
989: 'nhstepm*hstepm*stepm' months (i.e. until
990: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 991: nhstepm*hstepm matrices.
992:
993: Returns p3mat[i][j][h] after calling
994: p3mat[i][j][h]=matprod2(newm,
995: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
996: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
997: oldm);
1.226 brouard 998:
999: Important routines
1000:
1001: - func (or funcone), computes logit (pij) distinguishing
1002: o fixed variables (single or product dummies or quantitative);
1003: o varying variables by:
1004: (1) wave (single, product dummies, quantitative),
1005: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1006: % fixed dummy (treated) or quantitative (not done because time-consuming);
1007: % varying dummy (not done) or quantitative (not done);
1008: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1009: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1010: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1011: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1012: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1013:
1.226 brouard 1014:
1015:
1.133 brouard 1016: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1017: Institut national d'études démographiques, Paris.
1.126 brouard 1018: This software have been partly granted by Euro-REVES, a concerted action
1019: from the European Union.
1020: It is copyrighted identically to a GNU software product, ie programme and
1021: software can be distributed freely for non commercial use. Latest version
1022: can be accessed at http://euroreves.ined.fr/imach .
1023:
1024: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1025: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1026:
1027: **********************************************************************/
1028: /*
1029: main
1030: read parameterfile
1031: read datafile
1032: concatwav
1033: freqsummary
1034: if (mle >= 1)
1035: mlikeli
1036: print results files
1037: if mle==1
1038: computes hessian
1039: read end of parameter file: agemin, agemax, bage, fage, estepm
1040: begin-prev-date,...
1041: open gnuplot file
1042: open html file
1.145 brouard 1043: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1044: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1045: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1046: freexexit2 possible for memory heap.
1047:
1048: h Pij x | pij_nom ficrestpij
1049: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1050: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1051: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1052:
1053: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1054: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1055: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1056: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1057: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1058:
1.126 brouard 1059: forecasting if prevfcast==1 prevforecast call prevalence()
1060: health expectancies
1061: Variance-covariance of DFLE
1062: prevalence()
1063: movingaverage()
1064: varevsij()
1065: if popbased==1 varevsij(,popbased)
1066: total life expectancies
1067: Variance of period (stable) prevalence
1068: end
1069: */
1070:
1.187 brouard 1071: /* #define DEBUG */
1072: /* #define DEBUGBRENT */
1.203 brouard 1073: /* #define DEBUGLINMIN */
1074: /* #define DEBUGHESS */
1075: #define DEBUGHESSIJ
1.224 brouard 1076: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1077: #define POWELL /* Instead of NLOPT */
1.224 brouard 1078: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1079: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1080: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1081:
1082: #include <math.h>
1083: #include <stdio.h>
1084: #include <stdlib.h>
1085: #include <string.h>
1.226 brouard 1086: #include <ctype.h>
1.159 brouard 1087:
1088: #ifdef _WIN32
1089: #include <io.h>
1.172 brouard 1090: #include <windows.h>
1091: #include <tchar.h>
1.159 brouard 1092: #else
1.126 brouard 1093: #include <unistd.h>
1.159 brouard 1094: #endif
1.126 brouard 1095:
1096: #include <limits.h>
1097: #include <sys/types.h>
1.171 brouard 1098:
1099: #if defined(__GNUC__)
1100: #include <sys/utsname.h> /* Doesn't work on Windows */
1101: #endif
1102:
1.126 brouard 1103: #include <sys/stat.h>
1104: #include <errno.h>
1.159 brouard 1105: /* extern int errno; */
1.126 brouard 1106:
1.157 brouard 1107: /* #ifdef LINUX */
1108: /* #include <time.h> */
1109: /* #include "timeval.h" */
1110: /* #else */
1111: /* #include <sys/time.h> */
1112: /* #endif */
1113:
1.126 brouard 1114: #include <time.h>
1115:
1.136 brouard 1116: #ifdef GSL
1117: #include <gsl/gsl_errno.h>
1118: #include <gsl/gsl_multimin.h>
1119: #endif
1120:
1.167 brouard 1121:
1.162 brouard 1122: #ifdef NLOPT
1123: #include <nlopt.h>
1124: typedef struct {
1125: double (* function)(double [] );
1126: } myfunc_data ;
1127: #endif
1128:
1.126 brouard 1129: /* #include <libintl.h> */
1130: /* #define _(String) gettext (String) */
1131:
1.251 brouard 1132: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1133:
1134: #define GNUPLOTPROGRAM "gnuplot"
1135: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1136: #define FILENAMELENGTH 132
1137:
1138: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1139: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1140:
1.144 brouard 1141: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1142: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1143:
1144: #define NINTERVMAX 8
1.144 brouard 1145: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1146: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1147: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1148: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1149: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1150: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1151: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1152: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1153: /* #define AGESUP 130 */
1.288 brouard 1154: /* #define AGESUP 150 */
1155: #define AGESUP 200
1.268 brouard 1156: #define AGEINF 0
1.218 brouard 1157: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1158: #define AGEBASE 40
1.194 brouard 1159: #define AGEOVERFLOW 1.e20
1.164 brouard 1160: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1161: #ifdef _WIN32
1162: #define DIRSEPARATOR '\\'
1163: #define CHARSEPARATOR "\\"
1164: #define ODIRSEPARATOR '/'
1165: #else
1.126 brouard 1166: #define DIRSEPARATOR '/'
1167: #define CHARSEPARATOR "/"
1168: #define ODIRSEPARATOR '\\'
1169: #endif
1170:
1.312 ! brouard 1171: /* $Id: imach.c,v 1.311 2022/04/05 21:03:51 brouard Exp $ */
1.126 brouard 1172: /* $State: Exp $ */
1.196 brouard 1173: #include "version.h"
1174: char version[]=__IMACH_VERSION__;
1.308 brouard 1175: char copyright[]="March 2021,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021, INED 2000-2021";
1.312 ! brouard 1176: char fullversion[]="$Revision: 1.311 $ $Date: 2022/04/05 21:03:51 $";
1.126 brouard 1177: char strstart[80];
1178: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1179: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1180: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1181: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1182: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1183: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1184: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1185: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1186: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1187: int cptcovprodnoage=0; /**< Number of covariate products without age */
1188: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1189: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1190: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1191: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1192: int nsd=0; /**< Total number of single dummy variables (output) */
1193: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1194: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1195: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1196: int ntveff=0; /**< ntveff number of effective time varying variables */
1197: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1198: int cptcov=0; /* Working variable */
1.290 brouard 1199: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1200: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1201: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1202: int nlstate=2; /* Number of live states */
1203: int ndeath=1; /* Number of dead states */
1.130 brouard 1204: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1205: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1206: int popbased=0;
1207:
1208: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1209: int maxwav=0; /* Maxim number of waves */
1210: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1211: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1212: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1213: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1214: int mle=1, weightopt=0;
1.126 brouard 1215: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1216: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1217: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1218: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1219: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1220: int selected(int kvar); /* Is covariate kvar selected for printing results */
1221:
1.130 brouard 1222: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1223: double **matprod2(); /* test */
1.126 brouard 1224: double **oldm, **newm, **savm; /* Working pointers to matrices */
1225: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1226: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1227:
1.136 brouard 1228: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1229: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1230: FILE *ficlog, *ficrespow;
1.130 brouard 1231: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1232: double fretone; /* Only one call to likelihood */
1.130 brouard 1233: long ipmx=0; /* Number of contributions */
1.126 brouard 1234: double sw; /* Sum of weights */
1235: char filerespow[FILENAMELENGTH];
1236: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1237: FILE *ficresilk;
1238: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1239: FILE *ficresprobmorprev;
1240: FILE *fichtm, *fichtmcov; /* Html File */
1241: FILE *ficreseij;
1242: char filerese[FILENAMELENGTH];
1243: FILE *ficresstdeij;
1244: char fileresstde[FILENAMELENGTH];
1245: FILE *ficrescveij;
1246: char filerescve[FILENAMELENGTH];
1247: FILE *ficresvij;
1248: char fileresv[FILENAMELENGTH];
1.269 brouard 1249:
1.126 brouard 1250: char title[MAXLINE];
1.234 brouard 1251: char model[MAXLINE]; /**< The model line */
1.217 brouard 1252: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1253: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1254: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1255: char command[FILENAMELENGTH];
1256: int outcmd=0;
1257:
1.217 brouard 1258: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1259: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1260: char filelog[FILENAMELENGTH]; /* Log file */
1261: char filerest[FILENAMELENGTH];
1262: char fileregp[FILENAMELENGTH];
1263: char popfile[FILENAMELENGTH];
1264:
1265: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1266:
1.157 brouard 1267: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1268: /* struct timezone tzp; */
1269: /* extern int gettimeofday(); */
1270: struct tm tml, *gmtime(), *localtime();
1271:
1272: extern time_t time();
1273:
1274: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1275: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1276: struct tm tm;
1277:
1.126 brouard 1278: char strcurr[80], strfor[80];
1279:
1280: char *endptr;
1281: long lval;
1282: double dval;
1283:
1284: #define NR_END 1
1285: #define FREE_ARG char*
1286: #define FTOL 1.0e-10
1287:
1288: #define NRANSI
1.240 brouard 1289: #define ITMAX 200
1290: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1291:
1292: #define TOL 2.0e-4
1293:
1294: #define CGOLD 0.3819660
1295: #define ZEPS 1.0e-10
1296: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1297:
1298: #define GOLD 1.618034
1299: #define GLIMIT 100.0
1300: #define TINY 1.0e-20
1301:
1302: static double maxarg1,maxarg2;
1303: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1304: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1305:
1306: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1307: #define rint(a) floor(a+0.5)
1.166 brouard 1308: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1309: #define mytinydouble 1.0e-16
1.166 brouard 1310: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1311: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1312: /* static double dsqrarg; */
1313: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1314: static double sqrarg;
1315: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1316: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1317: int agegomp= AGEGOMP;
1318:
1319: int imx;
1320: int stepm=1;
1321: /* Stepm, step in month: minimum step interpolation*/
1322:
1323: int estepm;
1324: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1325:
1326: int m,nb;
1327: long *num;
1.197 brouard 1328: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1329: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1330: covariate for which somebody answered excluding
1331: undefined. Usually 2: 0 and 1. */
1332: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1333: covariate for which somebody answered including
1334: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1335: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1336: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1337: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1338: double *ageexmed,*agecens;
1339: double dateintmean=0;
1.296 brouard 1340: double anprojd, mprojd, jprojd; /* For eventual projections */
1341: double anprojf, mprojf, jprojf;
1.126 brouard 1342:
1.296 brouard 1343: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1344: double anbackf, mbackf, jbackf;
1345: double jintmean,mintmean,aintmean;
1.126 brouard 1346: double *weight;
1347: int **s; /* Status */
1.141 brouard 1348: double *agedc;
1.145 brouard 1349: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1350: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1351: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1352: double **coqvar; /* Fixed quantitative covariate nqv */
1353: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1354: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1355: double idx;
1356: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1357: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1358: /*k 1 2 3 4 5 6 7 8 9 */
1359: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1360: /* Tndvar[k] 1 2 3 4 5 */
1361: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1362: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1363: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1364: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1365: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1366: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1367: /* Tprod[i]=k 4 7 */
1368: /* Tage[i]=k 5 8 */
1369: /* */
1370: /* Type */
1371: /* V 1 2 3 4 5 */
1372: /* F F V V V */
1373: /* D Q D D Q */
1374: /* */
1375: int *TvarsD;
1376: int *TvarsDind;
1377: int *TvarsQ;
1378: int *TvarsQind;
1379:
1.235 brouard 1380: #define MAXRESULTLINES 10
1381: int nresult=0;
1.258 brouard 1382: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1383: int TKresult[MAXRESULTLINES];
1.237 brouard 1384: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1385: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1386: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1387: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1388: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1389: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1390:
1.234 brouard 1391: /* 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 1392: 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 */
1393: 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 */
1394: 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 */
1395: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1396: 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 */
1397: 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 1398: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1399: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1400: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1401: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1402: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1403: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1404: 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 */
1405: 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 */
1406:
1.230 brouard 1407: int *Tvarsel; /**< Selected covariates for output */
1408: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1409: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1410: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1411: 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 1412: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1413: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1414: int *Tage;
1.227 brouard 1415: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1416: 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 1417: 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*/
1418: 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 1419: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1420: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1421: int **Tvard;
1422: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1423: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1424: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1425: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1426: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1427: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1428: double *lsurv, *lpop, *tpop;
1429:
1.231 brouard 1430: #define FD 1; /* Fixed dummy covariate */
1431: #define FQ 2; /* Fixed quantitative covariate */
1432: #define FP 3; /* Fixed product covariate */
1433: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1434: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1435: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1436: #define VD 10; /* Varying dummy covariate */
1437: #define VQ 11; /* Varying quantitative covariate */
1438: #define VP 12; /* Varying product covariate */
1439: #define VPDD 13; /* Varying product dummy*dummy covariate */
1440: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1441: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1442: #define APFD 16; /* Age product * fixed dummy covariate */
1443: #define APFQ 17; /* Age product * fixed quantitative covariate */
1444: #define APVD 18; /* Age product * varying dummy covariate */
1445: #define APVQ 19; /* Age product * varying quantitative covariate */
1446:
1447: #define FTYPE 1; /* Fixed covariate */
1448: #define VTYPE 2; /* Varying covariate (loop in wave) */
1449: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1450:
1451: struct kmodel{
1452: int maintype; /* main type */
1453: int subtype; /* subtype */
1454: };
1455: struct kmodel modell[NCOVMAX];
1456:
1.143 brouard 1457: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1458: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1459:
1460: /**************** split *************************/
1461: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1462: {
1463: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1464: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1465: */
1466: char *ss; /* pointer */
1.186 brouard 1467: int l1=0, l2=0; /* length counters */
1.126 brouard 1468:
1469: l1 = strlen(path ); /* length of path */
1470: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1471: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1472: if ( ss == NULL ) { /* no directory, so determine current directory */
1473: strcpy( name, path ); /* we got the fullname name because no directory */
1474: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1475: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1476: /* get current working directory */
1477: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1478: #ifdef WIN32
1479: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1480: #else
1481: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1482: #endif
1.126 brouard 1483: return( GLOCK_ERROR_GETCWD );
1484: }
1485: /* got dirc from getcwd*/
1486: printf(" DIRC = %s \n",dirc);
1.205 brouard 1487: } else { /* strip directory from path */
1.126 brouard 1488: ss++; /* after this, the filename */
1489: l2 = strlen( ss ); /* length of filename */
1490: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1491: strcpy( name, ss ); /* save file name */
1492: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1493: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1494: printf(" DIRC2 = %s \n",dirc);
1495: }
1496: /* We add a separator at the end of dirc if not exists */
1497: l1 = strlen( dirc ); /* length of directory */
1498: if( dirc[l1-1] != DIRSEPARATOR ){
1499: dirc[l1] = DIRSEPARATOR;
1500: dirc[l1+1] = 0;
1501: printf(" DIRC3 = %s \n",dirc);
1502: }
1503: ss = strrchr( name, '.' ); /* find last / */
1504: if (ss >0){
1505: ss++;
1506: strcpy(ext,ss); /* save extension */
1507: l1= strlen( name);
1508: l2= strlen(ss)+1;
1509: strncpy( finame, name, l1-l2);
1510: finame[l1-l2]= 0;
1511: }
1512:
1513: return( 0 ); /* we're done */
1514: }
1515:
1516:
1517: /******************************************/
1518:
1519: void replace_back_to_slash(char *s, char*t)
1520: {
1521: int i;
1522: int lg=0;
1523: i=0;
1524: lg=strlen(t);
1525: for(i=0; i<= lg; i++) {
1526: (s[i] = t[i]);
1527: if (t[i]== '\\') s[i]='/';
1528: }
1529: }
1530:
1.132 brouard 1531: char *trimbb(char *out, char *in)
1.137 brouard 1532: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1533: char *s;
1534: s=out;
1535: while (*in != '\0'){
1.137 brouard 1536: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1537: in++;
1538: }
1539: *out++ = *in++;
1540: }
1541: *out='\0';
1542: return s;
1543: }
1544:
1.187 brouard 1545: /* char *substrchaine(char *out, char *in, char *chain) */
1546: /* { */
1547: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1548: /* char *s, *t; */
1549: /* t=in;s=out; */
1550: /* while ((*in != *chain) && (*in != '\0')){ */
1551: /* *out++ = *in++; */
1552: /* } */
1553:
1554: /* /\* *in matches *chain *\/ */
1555: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1556: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1557: /* } */
1558: /* in--; chain--; */
1559: /* while ( (*in != '\0')){ */
1560: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1561: /* *out++ = *in++; */
1562: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1563: /* } */
1564: /* *out='\0'; */
1565: /* out=s; */
1566: /* return out; */
1567: /* } */
1568: char *substrchaine(char *out, char *in, char *chain)
1569: {
1570: /* Substract chain 'chain' from 'in', return and output 'out' */
1571: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1572:
1573: char *strloc;
1574:
1575: strcpy (out, in);
1576: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1577: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1578: if(strloc != NULL){
1579: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1580: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1581: /* strcpy (strloc, strloc +strlen(chain));*/
1582: }
1583: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1584: return out;
1585: }
1586:
1587:
1.145 brouard 1588: char *cutl(char *blocc, char *alocc, char *in, char occ)
1589: {
1.187 brouard 1590: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1591: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1592: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1593: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1594: */
1.160 brouard 1595: char *s, *t;
1.145 brouard 1596: t=in;s=in;
1597: while ((*in != occ) && (*in != '\0')){
1598: *alocc++ = *in++;
1599: }
1600: if( *in == occ){
1601: *(alocc)='\0';
1602: s=++in;
1603: }
1604:
1605: if (s == t) {/* occ not found */
1606: *(alocc-(in-s))='\0';
1607: in=s;
1608: }
1609: while ( *in != '\0'){
1610: *blocc++ = *in++;
1611: }
1612:
1613: *blocc='\0';
1614: return t;
1615: }
1.137 brouard 1616: char *cutv(char *blocc, char *alocc, char *in, char occ)
1617: {
1.187 brouard 1618: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1619: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1620: gives blocc="abcdef2ghi" and alocc="j".
1621: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1622: */
1623: char *s, *t;
1624: t=in;s=in;
1625: while (*in != '\0'){
1626: while( *in == occ){
1627: *blocc++ = *in++;
1628: s=in;
1629: }
1630: *blocc++ = *in++;
1631: }
1632: if (s == t) /* occ not found */
1633: *(blocc-(in-s))='\0';
1634: else
1635: *(blocc-(in-s)-1)='\0';
1636: in=s;
1637: while ( *in != '\0'){
1638: *alocc++ = *in++;
1639: }
1640:
1641: *alocc='\0';
1642: return s;
1643: }
1644:
1.126 brouard 1645: int nbocc(char *s, char occ)
1646: {
1647: int i,j=0;
1648: int lg=20;
1649: i=0;
1650: lg=strlen(s);
1651: for(i=0; i<= lg; i++) {
1.234 brouard 1652: if (s[i] == occ ) j++;
1.126 brouard 1653: }
1654: return j;
1655: }
1656:
1.137 brouard 1657: /* void cutv(char *u,char *v, char*t, char occ) */
1658: /* { */
1659: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1660: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1661: /* gives u="abcdef2ghi" and v="j" *\/ */
1662: /* int i,lg,j,p=0; */
1663: /* i=0; */
1664: /* lg=strlen(t); */
1665: /* for(j=0; j<=lg-1; j++) { */
1666: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1667: /* } */
1.126 brouard 1668:
1.137 brouard 1669: /* for(j=0; j<p; j++) { */
1670: /* (u[j] = t[j]); */
1671: /* } */
1672: /* u[p]='\0'; */
1.126 brouard 1673:
1.137 brouard 1674: /* for(j=0; j<= lg; j++) { */
1675: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1676: /* } */
1677: /* } */
1.126 brouard 1678:
1.160 brouard 1679: #ifdef _WIN32
1680: char * strsep(char **pp, const char *delim)
1681: {
1682: char *p, *q;
1683:
1684: if ((p = *pp) == NULL)
1685: return 0;
1686: if ((q = strpbrk (p, delim)) != NULL)
1687: {
1688: *pp = q + 1;
1689: *q = '\0';
1690: }
1691: else
1692: *pp = 0;
1693: return p;
1694: }
1695: #endif
1696:
1.126 brouard 1697: /********************** nrerror ********************/
1698:
1699: void nrerror(char error_text[])
1700: {
1701: fprintf(stderr,"ERREUR ...\n");
1702: fprintf(stderr,"%s\n",error_text);
1703: exit(EXIT_FAILURE);
1704: }
1705: /*********************** vector *******************/
1706: double *vector(int nl, int nh)
1707: {
1708: double *v;
1709: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1710: if (!v) nrerror("allocation failure in vector");
1711: return v-nl+NR_END;
1712: }
1713:
1714: /************************ free vector ******************/
1715: void free_vector(double*v, int nl, int nh)
1716: {
1717: free((FREE_ARG)(v+nl-NR_END));
1718: }
1719:
1720: /************************ivector *******************************/
1721: int *ivector(long nl,long nh)
1722: {
1723: int *v;
1724: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1725: if (!v) nrerror("allocation failure in ivector");
1726: return v-nl+NR_END;
1727: }
1728:
1729: /******************free ivector **************************/
1730: void free_ivector(int *v, long nl, long nh)
1731: {
1732: free((FREE_ARG)(v+nl-NR_END));
1733: }
1734:
1735: /************************lvector *******************************/
1736: long *lvector(long nl,long nh)
1737: {
1738: long *v;
1739: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1740: if (!v) nrerror("allocation failure in ivector");
1741: return v-nl+NR_END;
1742: }
1743:
1744: /******************free lvector **************************/
1745: void free_lvector(long *v, long nl, long nh)
1746: {
1747: free((FREE_ARG)(v+nl-NR_END));
1748: }
1749:
1750: /******************* imatrix *******************************/
1751: int **imatrix(long nrl, long nrh, long ncl, long nch)
1752: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1753: {
1754: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1755: int **m;
1756:
1757: /* allocate pointers to rows */
1758: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1759: if (!m) nrerror("allocation failure 1 in matrix()");
1760: m += NR_END;
1761: m -= nrl;
1762:
1763:
1764: /* allocate rows and set pointers to them */
1765: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1766: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1767: m[nrl] += NR_END;
1768: m[nrl] -= ncl;
1769:
1770: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1771:
1772: /* return pointer to array of pointers to rows */
1773: return m;
1774: }
1775:
1776: /****************** free_imatrix *************************/
1777: void free_imatrix(m,nrl,nrh,ncl,nch)
1778: int **m;
1779: long nch,ncl,nrh,nrl;
1780: /* free an int matrix allocated by imatrix() */
1781: {
1782: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1783: free((FREE_ARG) (m+nrl-NR_END));
1784: }
1785:
1786: /******************* matrix *******************************/
1787: double **matrix(long nrl, long nrh, long ncl, long nch)
1788: {
1789: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1790: double **m;
1791:
1792: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1793: if (!m) nrerror("allocation failure 1 in matrix()");
1794: m += NR_END;
1795: m -= nrl;
1796:
1797: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1798: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1799: m[nrl] += NR_END;
1800: m[nrl] -= ncl;
1801:
1802: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1803: return m;
1.145 brouard 1804: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1805: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1806: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1807: */
1808: }
1809:
1810: /*************************free matrix ************************/
1811: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1812: {
1813: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1814: free((FREE_ARG)(m+nrl-NR_END));
1815: }
1816:
1817: /******************* ma3x *******************************/
1818: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1819: {
1820: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1821: double ***m;
1822:
1823: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1824: if (!m) nrerror("allocation failure 1 in matrix()");
1825: m += NR_END;
1826: m -= nrl;
1827:
1828: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1829: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1830: m[nrl] += NR_END;
1831: m[nrl] -= ncl;
1832:
1833: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1834:
1835: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1836: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1837: m[nrl][ncl] += NR_END;
1838: m[nrl][ncl] -= nll;
1839: for (j=ncl+1; j<=nch; j++)
1840: m[nrl][j]=m[nrl][j-1]+nlay;
1841:
1842: for (i=nrl+1; i<=nrh; i++) {
1843: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1844: for (j=ncl+1; j<=nch; j++)
1845: m[i][j]=m[i][j-1]+nlay;
1846: }
1847: return m;
1848: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1849: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1850: */
1851: }
1852:
1853: /*************************free ma3x ************************/
1854: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1855: {
1856: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1857: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1858: free((FREE_ARG)(m+nrl-NR_END));
1859: }
1860:
1861: /*************** function subdirf ***********/
1862: char *subdirf(char fileres[])
1863: {
1864: /* Caution optionfilefiname is hidden */
1865: strcpy(tmpout,optionfilefiname);
1866: strcat(tmpout,"/"); /* Add to the right */
1867: strcat(tmpout,fileres);
1868: return tmpout;
1869: }
1870:
1871: /*************** function subdirf2 ***********/
1872: char *subdirf2(char fileres[], char *preop)
1873: {
1874:
1875: /* Caution optionfilefiname is hidden */
1876: strcpy(tmpout,optionfilefiname);
1877: strcat(tmpout,"/");
1878: strcat(tmpout,preop);
1879: strcat(tmpout,fileres);
1880: return tmpout;
1881: }
1882:
1883: /*************** function subdirf3 ***********/
1884: char *subdirf3(char fileres[], char *preop, char *preop2)
1885: {
1886:
1887: /* Caution optionfilefiname is hidden */
1888: strcpy(tmpout,optionfilefiname);
1889: strcat(tmpout,"/");
1890: strcat(tmpout,preop);
1891: strcat(tmpout,preop2);
1892: strcat(tmpout,fileres);
1893: return tmpout;
1894: }
1.213 brouard 1895:
1896: /*************** function subdirfext ***********/
1897: char *subdirfext(char fileres[], char *preop, char *postop)
1898: {
1899:
1900: strcpy(tmpout,preop);
1901: strcat(tmpout,fileres);
1902: strcat(tmpout,postop);
1903: return tmpout;
1904: }
1.126 brouard 1905:
1.213 brouard 1906: /*************** function subdirfext3 ***********/
1907: char *subdirfext3(char fileres[], char *preop, char *postop)
1908: {
1909:
1910: /* Caution optionfilefiname is hidden */
1911: strcpy(tmpout,optionfilefiname);
1912: strcat(tmpout,"/");
1913: strcat(tmpout,preop);
1914: strcat(tmpout,fileres);
1915: strcat(tmpout,postop);
1916: return tmpout;
1917: }
1918:
1.162 brouard 1919: char *asc_diff_time(long time_sec, char ascdiff[])
1920: {
1921: long sec_left, days, hours, minutes;
1922: days = (time_sec) / (60*60*24);
1923: sec_left = (time_sec) % (60*60*24);
1924: hours = (sec_left) / (60*60) ;
1925: sec_left = (sec_left) %(60*60);
1926: minutes = (sec_left) /60;
1927: sec_left = (sec_left) % (60);
1928: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1929: return ascdiff;
1930: }
1931:
1.126 brouard 1932: /***************** f1dim *************************/
1933: extern int ncom;
1934: extern double *pcom,*xicom;
1935: extern double (*nrfunc)(double []);
1936:
1937: double f1dim(double x)
1938: {
1939: int j;
1940: double f;
1941: double *xt;
1942:
1943: xt=vector(1,ncom);
1944: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1945: f=(*nrfunc)(xt);
1946: free_vector(xt,1,ncom);
1947: return f;
1948: }
1949:
1950: /*****************brent *************************/
1951: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1952: {
1953: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1954: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1955: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1956: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1957: * returned function value.
1958: */
1.126 brouard 1959: int iter;
1960: double a,b,d,etemp;
1.159 brouard 1961: double fu=0,fv,fw,fx;
1.164 brouard 1962: double ftemp=0.;
1.126 brouard 1963: double p,q,r,tol1,tol2,u,v,w,x,xm;
1964: double e=0.0;
1965:
1966: a=(ax < cx ? ax : cx);
1967: b=(ax > cx ? ax : cx);
1968: x=w=v=bx;
1969: fw=fv=fx=(*f)(x);
1970: for (iter=1;iter<=ITMAX;iter++) {
1971: xm=0.5*(a+b);
1972: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1973: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1974: printf(".");fflush(stdout);
1975: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1976: #ifdef DEBUGBRENT
1.126 brouard 1977: 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);
1978: 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);
1979: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1980: #endif
1981: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1982: *xmin=x;
1983: return fx;
1984: }
1985: ftemp=fu;
1986: if (fabs(e) > tol1) {
1987: r=(x-w)*(fx-fv);
1988: q=(x-v)*(fx-fw);
1989: p=(x-v)*q-(x-w)*r;
1990: q=2.0*(q-r);
1991: if (q > 0.0) p = -p;
1992: q=fabs(q);
1993: etemp=e;
1994: e=d;
1995: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1996: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1997: else {
1.224 brouard 1998: d=p/q;
1999: u=x+d;
2000: if (u-a < tol2 || b-u < tol2)
2001: d=SIGN(tol1,xm-x);
1.126 brouard 2002: }
2003: } else {
2004: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2005: }
2006: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2007: fu=(*f)(u);
2008: if (fu <= fx) {
2009: if (u >= x) a=x; else b=x;
2010: SHFT(v,w,x,u)
1.183 brouard 2011: SHFT(fv,fw,fx,fu)
2012: } else {
2013: if (u < x) a=u; else b=u;
2014: if (fu <= fw || w == x) {
1.224 brouard 2015: v=w;
2016: w=u;
2017: fv=fw;
2018: fw=fu;
1.183 brouard 2019: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2020: v=u;
2021: fv=fu;
1.183 brouard 2022: }
2023: }
1.126 brouard 2024: }
2025: nrerror("Too many iterations in brent");
2026: *xmin=x;
2027: return fx;
2028: }
2029:
2030: /****************** mnbrak ***********************/
2031:
2032: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2033: double (*func)(double))
1.183 brouard 2034: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2035: the downhill direction (defined by the function as evaluated at the initial points) and returns
2036: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2037: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2038: */
1.126 brouard 2039: double ulim,u,r,q, dum;
2040: double fu;
1.187 brouard 2041:
2042: double scale=10.;
2043: int iterscale=0;
2044:
2045: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2046: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2047:
2048:
2049: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2050: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2051: /* *bx = *ax - (*ax - *bx)/scale; */
2052: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2053: /* } */
2054:
1.126 brouard 2055: if (*fb > *fa) {
2056: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2057: SHFT(dum,*fb,*fa,dum)
2058: }
1.126 brouard 2059: *cx=(*bx)+GOLD*(*bx-*ax);
2060: *fc=(*func)(*cx);
1.183 brouard 2061: #ifdef DEBUG
1.224 brouard 2062: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2063: 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 2064: #endif
1.224 brouard 2065: 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 2066: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2067: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2068: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2069: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2070: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2071: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2072: fu=(*func)(u);
1.163 brouard 2073: #ifdef DEBUG
2074: /* f(x)=A(x-u)**2+f(u) */
2075: double A, fparabu;
2076: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2077: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2078: 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);
2079: 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 2080: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2081: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2082: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2083: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2084: #endif
1.184 brouard 2085: #ifdef MNBRAKORIGINAL
1.183 brouard 2086: #else
1.191 brouard 2087: /* if (fu > *fc) { */
2088: /* #ifdef DEBUG */
2089: /* printf("mnbrak4 fu > fc \n"); */
2090: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2091: /* #endif */
2092: /* /\* 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 *\\/ *\/ */
2093: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2094: /* dum=u; /\* Shifting c and u *\/ */
2095: /* u = *cx; */
2096: /* *cx = dum; */
2097: /* dum = fu; */
2098: /* fu = *fc; */
2099: /* *fc =dum; */
2100: /* } else { /\* end *\/ */
2101: /* #ifdef DEBUG */
2102: /* printf("mnbrak3 fu < fc \n"); */
2103: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2104: /* #endif */
2105: /* dum=u; /\* Shifting c and u *\/ */
2106: /* u = *cx; */
2107: /* *cx = dum; */
2108: /* dum = fu; */
2109: /* fu = *fc; */
2110: /* *fc =dum; */
2111: /* } */
1.224 brouard 2112: #ifdef DEBUGMNBRAK
2113: double A, fparabu;
2114: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2115: fparabu= *fa - A*(*ax-u)*(*ax-u);
2116: 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);
2117: 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 2118: #endif
1.191 brouard 2119: dum=u; /* Shifting c and u */
2120: u = *cx;
2121: *cx = dum;
2122: dum = fu;
2123: fu = *fc;
2124: *fc =dum;
1.183 brouard 2125: #endif
1.162 brouard 2126: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2127: #ifdef DEBUG
1.224 brouard 2128: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2129: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2130: #endif
1.126 brouard 2131: fu=(*func)(u);
2132: if (fu < *fc) {
1.183 brouard 2133: #ifdef DEBUG
1.224 brouard 2134: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2135: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2136: #endif
2137: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2138: SHFT(*fb,*fc,fu,(*func)(u))
2139: #ifdef DEBUG
2140: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2141: #endif
2142: }
1.162 brouard 2143: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2144: #ifdef DEBUG
1.224 brouard 2145: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2146: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2147: #endif
1.126 brouard 2148: u=ulim;
2149: fu=(*func)(u);
1.183 brouard 2150: } else { /* u could be left to b (if r > q parabola has a maximum) */
2151: #ifdef DEBUG
1.224 brouard 2152: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2153: 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 2154: #endif
1.126 brouard 2155: u=(*cx)+GOLD*(*cx-*bx);
2156: fu=(*func)(u);
1.224 brouard 2157: #ifdef DEBUG
2158: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2159: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2160: #endif
1.183 brouard 2161: } /* end tests */
1.126 brouard 2162: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2163: SHFT(*fa,*fb,*fc,fu)
2164: #ifdef DEBUG
1.224 brouard 2165: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2166: 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 2167: #endif
2168: } /* 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 2169: }
2170:
2171: /*************** linmin ************************/
1.162 brouard 2172: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2173: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2174: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2175: the value of func at the returned location p . This is actually all accomplished by calling the
2176: routines mnbrak and brent .*/
1.126 brouard 2177: int ncom;
2178: double *pcom,*xicom;
2179: double (*nrfunc)(double []);
2180:
1.224 brouard 2181: #ifdef LINMINORIGINAL
1.126 brouard 2182: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2183: #else
2184: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2185: #endif
1.126 brouard 2186: {
2187: double brent(double ax, double bx, double cx,
2188: double (*f)(double), double tol, double *xmin);
2189: double f1dim(double x);
2190: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2191: double *fc, double (*func)(double));
2192: int j;
2193: double xx,xmin,bx,ax;
2194: double fx,fb,fa;
1.187 brouard 2195:
1.203 brouard 2196: #ifdef LINMINORIGINAL
2197: #else
2198: double scale=10., axs, xxs; /* Scale added for infinity */
2199: #endif
2200:
1.126 brouard 2201: ncom=n;
2202: pcom=vector(1,n);
2203: xicom=vector(1,n);
2204: nrfunc=func;
2205: for (j=1;j<=n;j++) {
2206: pcom[j]=p[j];
1.202 brouard 2207: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2208: }
1.187 brouard 2209:
1.203 brouard 2210: #ifdef LINMINORIGINAL
2211: xx=1.;
2212: #else
2213: axs=0.0;
2214: xxs=1.;
2215: do{
2216: xx= xxs;
2217: #endif
1.187 brouard 2218: ax=0.;
2219: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2220: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2221: /* 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)) */
2222: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2223: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2224: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2225: /* 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 2226: #ifdef LINMINORIGINAL
2227: #else
2228: if (fx != fx){
1.224 brouard 2229: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2230: printf("|");
2231: fprintf(ficlog,"|");
1.203 brouard 2232: #ifdef DEBUGLINMIN
1.224 brouard 2233: 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 2234: #endif
2235: }
1.224 brouard 2236: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2237: #endif
2238:
1.191 brouard 2239: #ifdef DEBUGLINMIN
2240: 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 2241: 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 2242: #endif
1.224 brouard 2243: #ifdef LINMINORIGINAL
2244: #else
2245: if(fb == fx){ /* Flat function in the direction */
2246: xmin=xx;
2247: *flat=1;
2248: }else{
2249: *flat=0;
2250: #endif
2251: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2252: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2253: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2254: /* fmin = f(p[j] + xmin * xi[j]) */
2255: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2256: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2257: #ifdef DEBUG
1.224 brouard 2258: 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);
2259: 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);
2260: #endif
2261: #ifdef LINMINORIGINAL
2262: #else
2263: }
1.126 brouard 2264: #endif
1.191 brouard 2265: #ifdef DEBUGLINMIN
2266: printf("linmin end ");
1.202 brouard 2267: fprintf(ficlog,"linmin end ");
1.191 brouard 2268: #endif
1.126 brouard 2269: for (j=1;j<=n;j++) {
1.203 brouard 2270: #ifdef LINMINORIGINAL
2271: xi[j] *= xmin;
2272: #else
2273: #ifdef DEBUGLINMIN
2274: if(xxs <1.0)
2275: printf(" before xi[%d]=%12.8f", j,xi[j]);
2276: #endif
2277: 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) */
2278: #ifdef DEBUGLINMIN
2279: if(xxs <1.0)
2280: 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 );
2281: #endif
2282: #endif
1.187 brouard 2283: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2284: }
1.191 brouard 2285: #ifdef DEBUGLINMIN
1.203 brouard 2286: printf("\n");
1.191 brouard 2287: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2288: 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 2289: for (j=1;j<=n;j++) {
1.202 brouard 2290: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2291: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2292: if(j % ncovmodel == 0){
1.191 brouard 2293: printf("\n");
1.202 brouard 2294: fprintf(ficlog,"\n");
2295: }
1.191 brouard 2296: }
1.203 brouard 2297: #else
1.191 brouard 2298: #endif
1.126 brouard 2299: free_vector(xicom,1,n);
2300: free_vector(pcom,1,n);
2301: }
2302:
2303:
2304: /*************** powell ************************/
1.162 brouard 2305: /*
2306: Minimization of a function func of n variables. Input consists of an initial starting point
2307: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2308: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2309: such that failure to decrease by more than this amount on one iteration signals doneness. On
2310: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2311: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2312: */
1.224 brouard 2313: #ifdef LINMINORIGINAL
2314: #else
2315: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2316: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2317: #endif
1.126 brouard 2318: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2319: double (*func)(double []))
2320: {
1.224 brouard 2321: #ifdef LINMINORIGINAL
2322: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2323: double (*func)(double []));
1.224 brouard 2324: #else
1.241 brouard 2325: void linmin(double p[], double xi[], int n, double *fret,
2326: double (*func)(double []),int *flat);
1.224 brouard 2327: #endif
1.239 brouard 2328: int i,ibig,j,jk,k;
1.126 brouard 2329: double del,t,*pt,*ptt,*xit;
1.181 brouard 2330: double directest;
1.126 brouard 2331: double fp,fptt;
2332: double *xits;
2333: int niterf, itmp;
1.224 brouard 2334: #ifdef LINMINORIGINAL
2335: #else
2336:
2337: flatdir=ivector(1,n);
2338: for (j=1;j<=n;j++) flatdir[j]=0;
2339: #endif
1.126 brouard 2340:
2341: pt=vector(1,n);
2342: ptt=vector(1,n);
2343: xit=vector(1,n);
2344: xits=vector(1,n);
2345: *fret=(*func)(p);
2346: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2347: rcurr_time = time(NULL);
1.126 brouard 2348: for (*iter=1;;++(*iter)) {
1.187 brouard 2349: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2350: ibig=0;
2351: del=0.0;
1.157 brouard 2352: rlast_time=rcurr_time;
2353: /* (void) gettimeofday(&curr_time,&tzp); */
2354: rcurr_time = time(NULL);
2355: curr_time = *localtime(&rcurr_time);
2356: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2357: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2358: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2359: for (i=1;i<=n;i++) {
1.126 brouard 2360: fprintf(ficrespow," %.12lf", p[i]);
2361: }
1.239 brouard 2362: fprintf(ficrespow,"\n");fflush(ficrespow);
2363: printf("\n#model= 1 + age ");
2364: fprintf(ficlog,"\n#model= 1 + age ");
2365: if(nagesqr==1){
1.241 brouard 2366: printf(" + age*age ");
2367: fprintf(ficlog," + age*age ");
1.239 brouard 2368: }
2369: for(j=1;j <=ncovmodel-2;j++){
2370: if(Typevar[j]==0) {
2371: printf(" + V%d ",Tvar[j]);
2372: fprintf(ficlog," + V%d ",Tvar[j]);
2373: }else if(Typevar[j]==1) {
2374: printf(" + V%d*age ",Tvar[j]);
2375: fprintf(ficlog," + V%d*age ",Tvar[j]);
2376: }else if(Typevar[j]==2) {
2377: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2378: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2379: }
2380: }
1.126 brouard 2381: printf("\n");
1.239 brouard 2382: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2383: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2384: fprintf(ficlog,"\n");
1.239 brouard 2385: for(i=1,jk=1; i <=nlstate; i++){
2386: for(k=1; k <=(nlstate+ndeath); k++){
2387: if (k != i) {
2388: printf("%d%d ",i,k);
2389: fprintf(ficlog,"%d%d ",i,k);
2390: for(j=1; j <=ncovmodel; j++){
2391: printf("%12.7f ",p[jk]);
2392: fprintf(ficlog,"%12.7f ",p[jk]);
2393: jk++;
2394: }
2395: printf("\n");
2396: fprintf(ficlog,"\n");
2397: }
2398: }
2399: }
1.241 brouard 2400: if(*iter <=3 && *iter >1){
1.157 brouard 2401: tml = *localtime(&rcurr_time);
2402: strcpy(strcurr,asctime(&tml));
2403: rforecast_time=rcurr_time;
1.126 brouard 2404: itmp = strlen(strcurr);
2405: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2406: strcurr[itmp-1]='\0';
1.162 brouard 2407: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2408: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2409: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2410: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2411: forecast_time = *localtime(&rforecast_time);
2412: strcpy(strfor,asctime(&forecast_time));
2413: itmp = strlen(strfor);
2414: if(strfor[itmp-1]=='\n')
2415: strfor[itmp-1]='\0';
2416: 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);
2417: 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 2418: }
2419: }
1.187 brouard 2420: for (i=1;i<=n;i++) { /* For each direction i */
2421: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2422: fptt=(*fret);
2423: #ifdef DEBUG
1.203 brouard 2424: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2425: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2426: #endif
1.203 brouard 2427: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2428: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2429: #ifdef LINMINORIGINAL
1.188 brouard 2430: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2431: #else
2432: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2433: flatdir[i]=flat; /* Function is vanishing in that direction i */
2434: #endif
2435: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2436: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2437: /* because that direction will be replaced unless the gain del is small */
2438: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2439: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2440: /* with the new direction. */
2441: del=fabs(fptt-(*fret));
2442: ibig=i;
1.126 brouard 2443: }
2444: #ifdef DEBUG
2445: printf("%d %.12e",i,(*fret));
2446: fprintf(ficlog,"%d %.12e",i,(*fret));
2447: for (j=1;j<=n;j++) {
1.224 brouard 2448: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2449: printf(" x(%d)=%.12e",j,xit[j]);
2450: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2451: }
2452: for(j=1;j<=n;j++) {
1.225 brouard 2453: printf(" p(%d)=%.12e",j,p[j]);
2454: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2455: }
2456: printf("\n");
2457: fprintf(ficlog,"\n");
2458: #endif
1.187 brouard 2459: } /* end loop on each direction i */
2460: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2461: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2462: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2463: for(j=1;j<=n;j++) {
1.302 brouard 2464: if(flatdir[j] >0){
2465: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2466: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2467: }
2468: /* printf("\n"); */
2469: /* fprintf(ficlog,"\n"); */
2470: }
1.243 brouard 2471: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2472: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2473: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2474: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2475: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2476: /* decreased of more than 3.84 */
2477: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2478: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2479: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2480:
1.188 brouard 2481: /* Starting the program with initial values given by a former maximization will simply change */
2482: /* the scales of the directions and the directions, because the are reset to canonical directions */
2483: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2484: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2485: #ifdef DEBUG
2486: int k[2],l;
2487: k[0]=1;
2488: k[1]=-1;
2489: printf("Max: %.12e",(*func)(p));
2490: fprintf(ficlog,"Max: %.12e",(*func)(p));
2491: for (j=1;j<=n;j++) {
2492: printf(" %.12e",p[j]);
2493: fprintf(ficlog," %.12e",p[j]);
2494: }
2495: printf("\n");
2496: fprintf(ficlog,"\n");
2497: for(l=0;l<=1;l++) {
2498: for (j=1;j<=n;j++) {
2499: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2500: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2501: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2502: }
2503: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2504: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2505: }
2506: #endif
2507:
1.224 brouard 2508: #ifdef LINMINORIGINAL
2509: #else
2510: free_ivector(flatdir,1,n);
2511: #endif
1.126 brouard 2512: free_vector(xit,1,n);
2513: free_vector(xits,1,n);
2514: free_vector(ptt,1,n);
2515: free_vector(pt,1,n);
2516: return;
1.192 brouard 2517: } /* enough precision */
1.240 brouard 2518: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2519: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2520: ptt[j]=2.0*p[j]-pt[j];
2521: xit[j]=p[j]-pt[j];
2522: pt[j]=p[j];
2523: }
1.181 brouard 2524: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2525: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2526: if (*iter <=4) {
1.225 brouard 2527: #else
2528: #endif
1.224 brouard 2529: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2530: #else
1.161 brouard 2531: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2532: #endif
1.162 brouard 2533: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2534: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2535: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2536: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2537: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2538: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2539: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2540: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2541: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2542: /* Even if f3 <f1, directest can be negative and t >0 */
2543: /* mu² and del² are equal when f3=f1 */
2544: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2545: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2546: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2547: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2548: #ifdef NRCORIGINAL
2549: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2550: #else
2551: 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 2552: t= t- del*SQR(fp-fptt);
1.183 brouard 2553: #endif
1.202 brouard 2554: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2555: #ifdef DEBUG
1.181 brouard 2556: 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);
2557: 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 2558: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2559: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2560: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2561: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2562: 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);
2563: 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);
2564: #endif
1.183 brouard 2565: #ifdef POWELLORIGINAL
2566: if (t < 0.0) { /* Then we use it for new direction */
2567: #else
1.182 brouard 2568: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2569: 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 2570: 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 2571: 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 2572: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2573: }
1.181 brouard 2574: if (directest < 0.0) { /* Then we use it for new direction */
2575: #endif
1.191 brouard 2576: #ifdef DEBUGLINMIN
1.234 brouard 2577: printf("Before linmin in direction P%d-P0\n",n);
2578: for (j=1;j<=n;j++) {
2579: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2580: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2581: if(j % ncovmodel == 0){
2582: printf("\n");
2583: fprintf(ficlog,"\n");
2584: }
2585: }
1.224 brouard 2586: #endif
2587: #ifdef LINMINORIGINAL
1.234 brouard 2588: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2589: #else
1.234 brouard 2590: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2591: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2592: #endif
1.234 brouard 2593:
1.191 brouard 2594: #ifdef DEBUGLINMIN
1.234 brouard 2595: for (j=1;j<=n;j++) {
2596: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2597: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2598: if(j % ncovmodel == 0){
2599: printf("\n");
2600: fprintf(ficlog,"\n");
2601: }
2602: }
1.224 brouard 2603: #endif
1.234 brouard 2604: for (j=1;j<=n;j++) {
2605: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2606: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2607: }
1.224 brouard 2608: #ifdef LINMINORIGINAL
2609: #else
1.234 brouard 2610: for (j=1, flatd=0;j<=n;j++) {
2611: if(flatdir[j]>0)
2612: flatd++;
2613: }
2614: if(flatd >0){
1.255 brouard 2615: printf("%d flat directions: ",flatd);
2616: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2617: for (j=1;j<=n;j++) {
2618: if(flatdir[j]>0){
2619: printf("%d ",j);
2620: fprintf(ficlog,"%d ",j);
2621: }
2622: }
2623: printf("\n");
2624: fprintf(ficlog,"\n");
2625: }
1.191 brouard 2626: #endif
1.234 brouard 2627: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2628: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2629:
1.126 brouard 2630: #ifdef DEBUG
1.234 brouard 2631: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2632: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2633: for(j=1;j<=n;j++){
2634: printf(" %lf",xit[j]);
2635: fprintf(ficlog," %lf",xit[j]);
2636: }
2637: printf("\n");
2638: fprintf(ficlog,"\n");
1.126 brouard 2639: #endif
1.192 brouard 2640: } /* end of t or directest negative */
1.224 brouard 2641: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2642: #else
1.234 brouard 2643: } /* end if (fptt < fp) */
1.192 brouard 2644: #endif
1.225 brouard 2645: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2646: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2647: #else
1.224 brouard 2648: #endif
1.234 brouard 2649: } /* loop iteration */
1.126 brouard 2650: }
1.234 brouard 2651:
1.126 brouard 2652: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2653:
1.235 brouard 2654: 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 2655: {
1.279 brouard 2656: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2657: * (and selected quantitative values in nres)
2658: * by left multiplying the unit
2659: * matrix by transitions matrix until convergence is reached with precision ftolpl
2660: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2661: * Wx is row vector: population in state 1, population in state 2, population dead
2662: * or prevalence in state 1, prevalence in state 2, 0
2663: * newm is the matrix after multiplications, its rows are identical at a factor.
2664: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2665: * Output is prlim.
2666: * Initial matrix pimij
2667: */
1.206 brouard 2668: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2669: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2670: /* 0, 0 , 1} */
2671: /*
2672: * and after some iteration: */
2673: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2674: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2675: /* 0, 0 , 1} */
2676: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2677: /* {0.51571254859325999, 0.4842874514067399, */
2678: /* 0.51326036147820708, 0.48673963852179264} */
2679: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2680:
1.126 brouard 2681: int i, ii,j,k;
1.209 brouard 2682: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2683: /* double **matprod2(); */ /* test */
1.218 brouard 2684: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2685: double **newm;
1.209 brouard 2686: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2687: int ncvloop=0;
1.288 brouard 2688: int first=0;
1.169 brouard 2689:
1.209 brouard 2690: min=vector(1,nlstate);
2691: max=vector(1,nlstate);
2692: meandiff=vector(1,nlstate);
2693:
1.218 brouard 2694: /* Starting with matrix unity */
1.126 brouard 2695: for (ii=1;ii<=nlstate+ndeath;ii++)
2696: for (j=1;j<=nlstate+ndeath;j++){
2697: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2698: }
1.169 brouard 2699:
2700: cov[1]=1.;
2701:
2702: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2703: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2704: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2705: ncvloop++;
1.126 brouard 2706: newm=savm;
2707: /* Covariates have to be included here again */
1.138 brouard 2708: cov[2]=agefin;
1.187 brouard 2709: if(nagesqr==1)
2710: cov[3]= agefin*agefin;;
1.234 brouard 2711: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2712: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2713: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2714: /* 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 2715: }
2716: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2717: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2718: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2719: /* 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 2720: }
1.237 brouard 2721: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2722: if(Dummy[Tvar[Tage[k]]]){
2723: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2724: } else{
1.235 brouard 2725: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2726: }
1.235 brouard 2727: /* 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 2728: }
1.237 brouard 2729: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2730: /* 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 2731: if(Dummy[Tvard[k][1]==0]){
2732: if(Dummy[Tvard[k][2]==0]){
2733: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2734: }else{
2735: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2736: }
2737: }else{
2738: if(Dummy[Tvard[k][2]==0]){
2739: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2740: }else{
2741: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2742: }
2743: }
1.234 brouard 2744: }
1.138 brouard 2745: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2746: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2747: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2748: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2749: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2750: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2751: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2752:
1.126 brouard 2753: savm=oldm;
2754: oldm=newm;
1.209 brouard 2755:
2756: for(j=1; j<=nlstate; j++){
2757: max[j]=0.;
2758: min[j]=1.;
2759: }
2760: for(i=1;i<=nlstate;i++){
2761: sumnew=0;
2762: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2763: for(j=1; j<=nlstate; j++){
2764: prlim[i][j]= newm[i][j]/(1-sumnew);
2765: max[j]=FMAX(max[j],prlim[i][j]);
2766: min[j]=FMIN(min[j],prlim[i][j]);
2767: }
2768: }
2769:
1.126 brouard 2770: maxmax=0.;
1.209 brouard 2771: for(j=1; j<=nlstate; j++){
2772: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2773: maxmax=FMAX(maxmax,meandiff[j]);
2774: /* 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 2775: } /* j loop */
1.203 brouard 2776: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2777: /* 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 2778: if(maxmax < ftolpl){
1.209 brouard 2779: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2780: free_vector(min,1,nlstate);
2781: free_vector(max,1,nlstate);
2782: free_vector(meandiff,1,nlstate);
1.126 brouard 2783: return prlim;
2784: }
1.288 brouard 2785: } /* agefin loop */
1.208 brouard 2786: /* After some age loop it doesn't converge */
1.288 brouard 2787: if(!first){
2788: first=1;
2789: 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);
2790: }
2791: 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);
2792:
1.209 brouard 2793: /* 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); */
2794: free_vector(min,1,nlstate);
2795: free_vector(max,1,nlstate);
2796: free_vector(meandiff,1,nlstate);
1.208 brouard 2797:
1.169 brouard 2798: return prlim; /* should not reach here */
1.126 brouard 2799: }
2800:
1.217 brouard 2801:
2802: /**** Back Prevalence limit (stable or period prevalence) ****************/
2803:
1.218 brouard 2804: /* 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) */
2805: /* 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 2806: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2807: {
1.264 brouard 2808: /* 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 2809: matrix by transitions matrix until convergence is reached with precision ftolpl */
2810: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2811: /* Wx is row vector: population in state 1, population in state 2, population dead */
2812: /* or prevalence in state 1, prevalence in state 2, 0 */
2813: /* newm is the matrix after multiplications, its rows are identical at a factor */
2814: /* Initial matrix pimij */
2815: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2816: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2817: /* 0, 0 , 1} */
2818: /*
2819: * and after some iteration: */
2820: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2821: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2822: /* 0, 0 , 1} */
2823: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2824: /* {0.51571254859325999, 0.4842874514067399, */
2825: /* 0.51326036147820708, 0.48673963852179264} */
2826: /* If we start from prlim again, prlim tends to a constant matrix */
2827:
2828: int i, ii,j,k;
1.247 brouard 2829: int first=0;
1.217 brouard 2830: double *min, *max, *meandiff, maxmax,sumnew=0.;
2831: /* double **matprod2(); */ /* test */
2832: double **out, cov[NCOVMAX+1], **bmij();
2833: double **newm;
1.218 brouard 2834: double **dnewm, **doldm, **dsavm; /* for use */
2835: double **oldm, **savm; /* for use */
2836:
1.217 brouard 2837: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2838: int ncvloop=0;
2839:
2840: min=vector(1,nlstate);
2841: max=vector(1,nlstate);
2842: meandiff=vector(1,nlstate);
2843:
1.266 brouard 2844: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2845: oldm=oldms; savm=savms;
2846:
2847: /* Starting with matrix unity */
2848: for (ii=1;ii<=nlstate+ndeath;ii++)
2849: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2850: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2851: }
2852:
2853: cov[1]=1.;
2854:
2855: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2856: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2857: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2858: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2859: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2860: ncvloop++;
1.218 brouard 2861: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2862: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2863: /* Covariates have to be included here again */
2864: cov[2]=agefin;
2865: if(nagesqr==1)
2866: cov[3]= agefin*agefin;;
1.242 brouard 2867: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2868: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2869: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2870: /* 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 2871: }
2872: /* for (k=1; k<=cptcovn;k++) { */
2873: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2874: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2875: /* /\* 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])]); *\/ */
2876: /* } */
2877: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2878: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2879: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2880: /* 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]); */
2881: }
2882: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2883: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2884: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2885: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2886: for (k=1; k<=cptcovage;k++){ /* For product with age */
2887: if(Dummy[Tvar[Tage[k]]]){
2888: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2889: } else{
2890: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2891: }
2892: /* 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]); */
2893: }
2894: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2895: /* 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]); */
2896: if(Dummy[Tvard[k][1]==0]){
2897: if(Dummy[Tvard[k][2]==0]){
2898: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2899: }else{
2900: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2901: }
2902: }else{
2903: if(Dummy[Tvard[k][2]==0]){
2904: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2905: }else{
2906: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2907: }
2908: }
1.217 brouard 2909: }
2910:
2911: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2912: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2913: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2914: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2915: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2916: /* ij should be linked to the correct index of cov */
2917: /* age and covariate values ij are in 'cov', but we need to pass
2918: * ij for the observed prevalence at age and status and covariate
2919: * number: prevacurrent[(int)agefin][ii][ij]
2920: */
2921: /* 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 *\/ */
2922: /* 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 *\/ */
2923: 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 2924: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2925: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2926: /* for(i=1; i<=nlstate+ndeath; i++) { */
2927: /* printf("%d newm= ",i); */
2928: /* for(j=1;j<=nlstate+ndeath;j++) { */
2929: /* printf("%f ",newm[i][j]); */
2930: /* } */
2931: /* printf("oldm * "); */
2932: /* for(j=1;j<=nlstate+ndeath;j++) { */
2933: /* printf("%f ",oldm[i][j]); */
2934: /* } */
1.268 brouard 2935: /* printf(" bmmij "); */
1.266 brouard 2936: /* for(j=1;j<=nlstate+ndeath;j++) { */
2937: /* printf("%f ",pmmij[i][j]); */
2938: /* } */
2939: /* printf("\n"); */
2940: /* } */
2941: /* } */
1.217 brouard 2942: savm=oldm;
2943: oldm=newm;
1.266 brouard 2944:
1.217 brouard 2945: for(j=1; j<=nlstate; j++){
2946: max[j]=0.;
2947: min[j]=1.;
2948: }
2949: for(j=1; j<=nlstate; j++){
2950: for(i=1;i<=nlstate;i++){
1.234 brouard 2951: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2952: bprlim[i][j]= newm[i][j];
2953: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2954: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2955: }
2956: }
1.218 brouard 2957:
1.217 brouard 2958: maxmax=0.;
2959: for(i=1; i<=nlstate; i++){
2960: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2961: maxmax=FMAX(maxmax,meandiff[i]);
2962: /* 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 2963: } /* i loop */
1.217 brouard 2964: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2965: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2966: if(maxmax < ftolpl){
1.220 brouard 2967: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2968: free_vector(min,1,nlstate);
2969: free_vector(max,1,nlstate);
2970: free_vector(meandiff,1,nlstate);
2971: return bprlim;
2972: }
1.288 brouard 2973: } /* agefin loop */
1.217 brouard 2974: /* After some age loop it doesn't converge */
1.288 brouard 2975: if(!first){
1.247 brouard 2976: first=1;
2977: 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\
2978: 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);
2979: }
2980: 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 2981: 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);
2982: /* 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); */
2983: free_vector(min,1,nlstate);
2984: free_vector(max,1,nlstate);
2985: free_vector(meandiff,1,nlstate);
2986:
2987: return bprlim; /* should not reach here */
2988: }
2989:
1.126 brouard 2990: /*************** transition probabilities ***************/
2991:
2992: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2993: {
1.138 brouard 2994: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2995: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2996: model to the ncovmodel covariates (including constant and age).
2997: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2998: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2999: ncth covariate in the global vector x is given by the formula:
3000: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3001: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3002: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3003: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3004: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3005: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3006: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3007: */
3008: double s1, lnpijopii;
1.126 brouard 3009: /*double t34;*/
1.164 brouard 3010: int i,j, nc, ii, jj;
1.126 brouard 3011:
1.223 brouard 3012: for(i=1; i<= nlstate; i++){
3013: for(j=1; j<i;j++){
3014: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3015: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3016: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3017: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3018: }
3019: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3020: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3021: }
3022: for(j=i+1; j<=nlstate+ndeath;j++){
3023: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3024: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3025: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3026: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3027: }
3028: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3029: }
3030: }
1.218 brouard 3031:
1.223 brouard 3032: for(i=1; i<= nlstate; i++){
3033: s1=0;
3034: for(j=1; j<i; j++){
3035: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3036: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3037: }
3038: for(j=i+1; j<=nlstate+ndeath; j++){
3039: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3040: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3041: }
3042: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3043: ps[i][i]=1./(s1+1.);
3044: /* Computing other pijs */
3045: for(j=1; j<i; j++)
3046: ps[i][j]= exp(ps[i][j])*ps[i][i];
3047: for(j=i+1; j<=nlstate+ndeath; j++)
3048: ps[i][j]= exp(ps[i][j])*ps[i][i];
3049: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3050: } /* end i */
1.218 brouard 3051:
1.223 brouard 3052: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3053: for(jj=1; jj<= nlstate+ndeath; jj++){
3054: ps[ii][jj]=0;
3055: ps[ii][ii]=1;
3056: }
3057: }
1.294 brouard 3058:
3059:
1.223 brouard 3060: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3061: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3062: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3063: /* } */
3064: /* printf("\n "); */
3065: /* } */
3066: /* printf("\n ");printf("%lf ",cov[2]);*/
3067: /*
3068: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3069: goto end;*/
1.266 brouard 3070: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3071: }
3072:
1.218 brouard 3073: /*************** backward transition probabilities ***************/
3074:
3075: /* 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 ) */
3076: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3077: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3078: {
1.302 brouard 3079: /* 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 3080: * 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 3081: */
1.218 brouard 3082: int i, ii, j,k;
1.222 brouard 3083:
3084: double **out, **pmij();
3085: double sumnew=0.;
1.218 brouard 3086: double agefin;
1.292 brouard 3087: 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 3088: double **dnewm, **dsavm, **doldm;
3089: double **bbmij;
3090:
1.218 brouard 3091: doldm=ddoldms; /* global pointers */
1.222 brouard 3092: dnewm=ddnewms;
3093: dsavm=ddsavms;
3094:
3095: agefin=cov[2];
1.268 brouard 3096: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3097: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3098: the observed prevalence (with this covariate ij) at beginning of transition */
3099: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3100:
3101: /* P_x */
1.266 brouard 3102: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3103: /* outputs pmmij which is a stochastic matrix in row */
3104:
3105: /* Diag(w_x) */
1.292 brouard 3106: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3107: sumnew=0.;
1.269 brouard 3108: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3109: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3110: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3111: sumnew+=prevacurrent[(int)agefin][ii][ij];
3112: }
3113: if(sumnew >0.01){ /* At least some value in the prevalence */
3114: for (ii=1;ii<=nlstate+ndeath;ii++){
3115: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3116: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3117: }
3118: }else{
3119: for (ii=1;ii<=nlstate+ndeath;ii++){
3120: for (j=1;j<=nlstate+ndeath;j++)
3121: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3122: }
3123: /* if(sumnew <0.9){ */
3124: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3125: /* } */
3126: }
3127: k3=0.0; /* We put the last diagonal to 0 */
3128: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3129: doldm[ii][ii]= k3;
3130: }
3131: /* End doldm, At the end doldm is diag[(w_i)] */
3132:
1.292 brouard 3133: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3134: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3135:
1.292 brouard 3136: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3137: /* 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 3138: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3139: sumnew=0.;
1.222 brouard 3140: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3141: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3142: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3143: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3144: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3145: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3146: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3147: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3148: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3149: /* }else */
1.268 brouard 3150: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3151: } /*End ii */
3152: } /* 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 */
3153:
1.292 brouard 3154: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3155: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3156: /* end bmij */
1.266 brouard 3157: return ps; /*pointer is unchanged */
1.218 brouard 3158: }
1.217 brouard 3159: /*************** transition probabilities ***************/
3160:
1.218 brouard 3161: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3162: {
3163: /* According to parameters values stored in x and the covariate's values stored in cov,
3164: computes the probability to be observed in state j being in state i by appying the
3165: model to the ncovmodel covariates (including constant and age).
3166: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3167: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3168: ncth covariate in the global vector x is given by the formula:
3169: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3170: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3171: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3172: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3173: Outputs ps[i][j] the probability to be observed in j being in j according to
3174: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3175: */
3176: double s1, lnpijopii;
3177: /*double t34;*/
3178: int i,j, nc, ii, jj;
3179:
1.234 brouard 3180: for(i=1; i<= nlstate; i++){
3181: for(j=1; j<i;j++){
3182: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3183: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3184: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3185: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3186: }
3187: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3188: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3189: }
3190: for(j=i+1; j<=nlstate+ndeath;j++){
3191: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3192: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3193: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3194: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3195: }
3196: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3197: }
3198: }
3199:
3200: for(i=1; i<= nlstate; i++){
3201: s1=0;
3202: for(j=1; j<i; j++){
3203: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3204: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3205: }
3206: for(j=i+1; j<=nlstate+ndeath; j++){
3207: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3208: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3209: }
3210: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3211: ps[i][i]=1./(s1+1.);
3212: /* Computing other pijs */
3213: for(j=1; j<i; j++)
3214: ps[i][j]= exp(ps[i][j])*ps[i][i];
3215: for(j=i+1; j<=nlstate+ndeath; j++)
3216: ps[i][j]= exp(ps[i][j])*ps[i][i];
3217: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3218: } /* end i */
3219:
3220: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3221: for(jj=1; jj<= nlstate+ndeath; jj++){
3222: ps[ii][jj]=0;
3223: ps[ii][ii]=1;
3224: }
3225: }
1.296 brouard 3226: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3227: for(jj=1; jj<= nlstate+ndeath; jj++){
3228: s1=0.;
3229: for(ii=1; ii<= nlstate+ndeath; ii++){
3230: s1+=ps[ii][jj];
3231: }
3232: for(ii=1; ii<= nlstate; ii++){
3233: ps[ii][jj]=ps[ii][jj]/s1;
3234: }
3235: }
3236: /* Transposition */
3237: for(jj=1; jj<= nlstate+ndeath; jj++){
3238: for(ii=jj; ii<= nlstate+ndeath; ii++){
3239: s1=ps[ii][jj];
3240: ps[ii][jj]=ps[jj][ii];
3241: ps[jj][ii]=s1;
3242: }
3243: }
3244: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3245: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3246: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3247: /* } */
3248: /* printf("\n "); */
3249: /* } */
3250: /* printf("\n ");printf("%lf ",cov[2]);*/
3251: /*
3252: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3253: goto end;*/
3254: return ps;
1.217 brouard 3255: }
3256:
3257:
1.126 brouard 3258: /**************** Product of 2 matrices ******************/
3259:
1.145 brouard 3260: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3261: {
3262: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3263: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3264: /* in, b, out are matrice of pointers which should have been initialized
3265: before: only the contents of out is modified. The function returns
3266: a pointer to pointers identical to out */
1.145 brouard 3267: int i, j, k;
1.126 brouard 3268: for(i=nrl; i<= nrh; i++)
1.145 brouard 3269: for(k=ncolol; k<=ncoloh; k++){
3270: out[i][k]=0.;
3271: for(j=ncl; j<=nch; j++)
3272: out[i][k] +=in[i][j]*b[j][k];
3273: }
1.126 brouard 3274: return out;
3275: }
3276:
3277:
3278: /************* Higher Matrix Product ***************/
3279:
1.235 brouard 3280: 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 3281: {
1.218 brouard 3282: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3283: 'nhstepm*hstepm*stepm' months (i.e. until
3284: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3285: nhstepm*hstepm matrices.
3286: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3287: (typically every 2 years instead of every month which is too big
3288: for the memory).
3289: Model is determined by parameters x and covariates have to be
3290: included manually here.
3291:
3292: */
3293:
3294: int i, j, d, h, k;
1.131 brouard 3295: double **out, cov[NCOVMAX+1];
1.126 brouard 3296: double **newm;
1.187 brouard 3297: double agexact;
1.214 brouard 3298: double agebegin, ageend;
1.126 brouard 3299:
3300: /* Hstepm could be zero and should return the unit matrix */
3301: for (i=1;i<=nlstate+ndeath;i++)
3302: for (j=1;j<=nlstate+ndeath;j++){
3303: oldm[i][j]=(i==j ? 1.0 : 0.0);
3304: po[i][j][0]=(i==j ? 1.0 : 0.0);
3305: }
3306: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3307: for(h=1; h <=nhstepm; h++){
3308: for(d=1; d <=hstepm; d++){
3309: newm=savm;
3310: /* Covariates have to be included here again */
3311: cov[1]=1.;
1.214 brouard 3312: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3313: cov[2]=agexact;
3314: if(nagesqr==1)
1.227 brouard 3315: cov[3]= agexact*agexact;
1.235 brouard 3316: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3317: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3318: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3319: /* 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)); */
3320: }
3321: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3322: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3323: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3324: /* 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]); */
3325: }
3326: for (k=1; k<=cptcovage;k++){
3327: if(Dummy[Tvar[Tage[k]]]){
3328: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3329: } else{
3330: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3331: }
3332: /* 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]); */
3333: }
3334: for (k=1; k<=cptcovprod;k++){ /* */
3335: /* 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]); */
3336: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3337: }
3338: /* for (k=1; k<=cptcovn;k++) */
3339: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3340: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3341: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3342: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3343: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3344:
3345:
1.126 brouard 3346: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3347: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3348: /* right multiplication of oldm by the current matrix */
1.126 brouard 3349: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3350: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3351: /* if((int)age == 70){ */
3352: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3353: /* for(i=1; i<=nlstate+ndeath; i++) { */
3354: /* printf("%d pmmij ",i); */
3355: /* for(j=1;j<=nlstate+ndeath;j++) { */
3356: /* printf("%f ",pmmij[i][j]); */
3357: /* } */
3358: /* printf(" oldm "); */
3359: /* for(j=1;j<=nlstate+ndeath;j++) { */
3360: /* printf("%f ",oldm[i][j]); */
3361: /* } */
3362: /* printf("\n"); */
3363: /* } */
3364: /* } */
1.126 brouard 3365: savm=oldm;
3366: oldm=newm;
3367: }
3368: for(i=1; i<=nlstate+ndeath; i++)
3369: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3370: po[i][j][h]=newm[i][j];
3371: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3372: }
1.128 brouard 3373: /*printf("h=%d ",h);*/
1.126 brouard 3374: } /* end h */
1.267 brouard 3375: /* printf("\n H=%d \n",h); */
1.126 brouard 3376: return po;
3377: }
3378:
1.217 brouard 3379: /************* Higher Back Matrix Product ***************/
1.218 brouard 3380: /* 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 3381: 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 3382: {
1.266 brouard 3383: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3384: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3385: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3386: nhstepm*hstepm matrices.
3387: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3388: (typically every 2 years instead of every month which is too big
1.217 brouard 3389: for the memory).
1.218 brouard 3390: Model is determined by parameters x and covariates have to be
1.266 brouard 3391: included manually here. Then we use a call to bmij(x and cov)
3392: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3393: */
1.217 brouard 3394:
3395: int i, j, d, h, k;
1.266 brouard 3396: double **out, cov[NCOVMAX+1], **bmij();
3397: double **newm, ***newmm;
1.217 brouard 3398: double agexact;
3399: double agebegin, ageend;
1.222 brouard 3400: double **oldm, **savm;
1.217 brouard 3401:
1.266 brouard 3402: newmm=po; /* To be saved */
3403: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3404: /* Hstepm could be zero and should return the unit matrix */
3405: for (i=1;i<=nlstate+ndeath;i++)
3406: for (j=1;j<=nlstate+ndeath;j++){
3407: oldm[i][j]=(i==j ? 1.0 : 0.0);
3408: po[i][j][0]=(i==j ? 1.0 : 0.0);
3409: }
3410: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3411: for(h=1; h <=nhstepm; h++){
3412: for(d=1; d <=hstepm; d++){
3413: newm=savm;
3414: /* Covariates have to be included here again */
3415: cov[1]=1.;
1.271 brouard 3416: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3417: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3418: cov[2]=agexact;
3419: if(nagesqr==1)
1.222 brouard 3420: cov[3]= agexact*agexact;
1.266 brouard 3421: for (k=1; k<=cptcovn;k++){
3422: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3423: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3424: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3425: /* 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)); */
3426: }
1.267 brouard 3427: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3428: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3429: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3430: /* 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]); */
3431: }
3432: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3433: if(Dummy[Tvar[Tage[k]]]){
3434: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3435: } else{
3436: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3437: }
3438: /* 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]); */
3439: }
3440: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3441: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3442: }
1.217 brouard 3443: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3444: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3445:
1.218 brouard 3446: /* Careful transposed matrix */
1.266 brouard 3447: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3448: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3449: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3450: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3451: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3452: /* if((int)age == 70){ */
3453: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3454: /* for(i=1; i<=nlstate+ndeath; i++) { */
3455: /* printf("%d pmmij ",i); */
3456: /* for(j=1;j<=nlstate+ndeath;j++) { */
3457: /* printf("%f ",pmmij[i][j]); */
3458: /* } */
3459: /* printf(" oldm "); */
3460: /* for(j=1;j<=nlstate+ndeath;j++) { */
3461: /* printf("%f ",oldm[i][j]); */
3462: /* } */
3463: /* printf("\n"); */
3464: /* } */
3465: /* } */
3466: savm=oldm;
3467: oldm=newm;
3468: }
3469: for(i=1; i<=nlstate+ndeath; i++)
3470: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3471: po[i][j][h]=newm[i][j];
1.268 brouard 3472: /* if(h==nhstepm) */
3473: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3474: }
1.268 brouard 3475: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3476: } /* end h */
1.268 brouard 3477: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3478: return po;
3479: }
3480:
3481:
1.162 brouard 3482: #ifdef NLOPT
3483: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3484: double fret;
3485: double *xt;
3486: int j;
3487: myfunc_data *d2 = (myfunc_data *) pd;
3488: /* xt = (p1-1); */
3489: xt=vector(1,n);
3490: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3491:
3492: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3493: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3494: printf("Function = %.12lf ",fret);
3495: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3496: printf("\n");
3497: free_vector(xt,1,n);
3498: return fret;
3499: }
3500: #endif
1.126 brouard 3501:
3502: /*************** log-likelihood *************/
3503: double func( double *x)
3504: {
1.226 brouard 3505: int i, ii, j, k, mi, d, kk;
3506: int ioffset=0;
3507: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3508: double **out;
3509: double lli; /* Individual log likelihood */
3510: int s1, s2;
1.228 brouard 3511: 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 3512: double bbh, survp;
3513: long ipmx;
3514: double agexact;
3515: /*extern weight */
3516: /* We are differentiating ll according to initial status */
3517: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3518: /*for(i=1;i<imx;i++)
3519: printf(" %d\n",s[4][i]);
3520: */
1.162 brouard 3521:
1.226 brouard 3522: ++countcallfunc;
1.162 brouard 3523:
1.226 brouard 3524: cov[1]=1.;
1.126 brouard 3525:
1.226 brouard 3526: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3527: ioffset=0;
1.226 brouard 3528: if(mle==1){
3529: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3530: /* Computes the values of the ncovmodel covariates of the model
3531: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3532: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3533: to be observed in j being in i according to the model.
3534: */
1.243 brouard 3535: ioffset=2+nagesqr ;
1.233 brouard 3536: /* Fixed */
1.234 brouard 3537: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3538: 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)*/
3539: }
1.226 brouard 3540: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3541: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3542: has been calculated etc */
3543: /* For an individual i, wav[i] gives the number of effective waves */
3544: /* We compute the contribution to Likelihood of each effective transition
3545: mw[mi][i] is real wave of the mi th effectve wave */
3546: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3547: s2=s[mw[mi+1][i]][i];
3548: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3549: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3550: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3551: */
3552: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3553: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3554: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3555: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3556: }
3557: for (ii=1;ii<=nlstate+ndeath;ii++)
3558: for (j=1;j<=nlstate+ndeath;j++){
3559: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3560: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3561: }
3562: for(d=0; d<dh[mi][i]; d++){
3563: newm=savm;
3564: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3565: cov[2]=agexact;
3566: if(nagesqr==1)
3567: cov[3]= agexact*agexact; /* Should be changed here */
3568: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3569: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3570: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3571: else
3572: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3573: }
3574: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3575: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3576: savm=oldm;
3577: oldm=newm;
3578: } /* end mult */
3579:
3580: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3581: /* But now since version 0.9 we anticipate for bias at large stepm.
3582: * If stepm is larger than one month (smallest stepm) and if the exact delay
3583: * (in months) between two waves is not a multiple of stepm, we rounded to
3584: * the nearest (and in case of equal distance, to the lowest) interval but now
3585: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3586: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3587: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3588: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3589: * -stepm/2 to stepm/2 .
3590: * For stepm=1 the results are the same as for previous versions of Imach.
3591: * For stepm > 1 the results are less biased than in previous versions.
3592: */
1.234 brouard 3593: s1=s[mw[mi][i]][i];
3594: s2=s[mw[mi+1][i]][i];
3595: bbh=(double)bh[mi][i]/(double)stepm;
3596: /* bias bh is positive if real duration
3597: * is higher than the multiple of stepm and negative otherwise.
3598: */
3599: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3600: if( s2 > nlstate){
3601: /* i.e. if s2 is a death state and if the date of death is known
3602: then the contribution to the likelihood is the probability to
3603: die between last step unit time and current step unit time,
3604: which is also equal to probability to die before dh
3605: minus probability to die before dh-stepm .
3606: In version up to 0.92 likelihood was computed
3607: as if date of death was unknown. Death was treated as any other
3608: health state: the date of the interview describes the actual state
3609: and not the date of a change in health state. The former idea was
3610: to consider that at each interview the state was recorded
3611: (healthy, disable or death) and IMaCh was corrected; but when we
3612: introduced the exact date of death then we should have modified
3613: the contribution of an exact death to the likelihood. This new
3614: contribution is smaller and very dependent of the step unit
3615: stepm. It is no more the probability to die between last interview
3616: and month of death but the probability to survive from last
3617: interview up to one month before death multiplied by the
3618: probability to die within a month. Thanks to Chris
3619: Jackson for correcting this bug. Former versions increased
3620: mortality artificially. The bad side is that we add another loop
3621: which slows down the processing. The difference can be up to 10%
3622: lower mortality.
3623: */
3624: /* If, at the beginning of the maximization mostly, the
3625: cumulative probability or probability to be dead is
3626: constant (ie = 1) over time d, the difference is equal to
3627: 0. out[s1][3] = savm[s1][3]: probability, being at state
3628: s1 at precedent wave, to be dead a month before current
3629: wave is equal to probability, being at state s1 at
3630: precedent wave, to be dead at mont of the current
3631: wave. Then the observed probability (that this person died)
3632: is null according to current estimated parameter. In fact,
3633: it should be very low but not zero otherwise the log go to
3634: infinity.
3635: */
1.183 brouard 3636: /* #ifdef INFINITYORIGINAL */
3637: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3638: /* #else */
3639: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3640: /* lli=log(mytinydouble); */
3641: /* else */
3642: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3643: /* #endif */
1.226 brouard 3644: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3645:
1.226 brouard 3646: } else if ( s2==-1 ) { /* alive */
3647: for (j=1,survp=0. ; j<=nlstate; j++)
3648: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3649: /*survp += out[s1][j]; */
3650: lli= log(survp);
3651: }
3652: else if (s2==-4) {
3653: for (j=3,survp=0. ; j<=nlstate; j++)
3654: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3655: lli= log(survp);
3656: }
3657: else if (s2==-5) {
3658: for (j=1,survp=0. ; j<=2; j++)
3659: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3660: lli= log(survp);
3661: }
3662: else{
3663: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3664: /* 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 */
3665: }
3666: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3667: /*if(lli ==000.0)*/
3668: /*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); */
3669: ipmx +=1;
3670: sw += weight[i];
3671: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3672: /* if (lli < log(mytinydouble)){ */
3673: /* 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); */
3674: /* 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]); */
3675: /* } */
3676: } /* end of wave */
3677: } /* end of individual */
3678: } else if(mle==2){
3679: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3680: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3681: for(mi=1; mi<= wav[i]-1; mi++){
3682: for (ii=1;ii<=nlstate+ndeath;ii++)
3683: for (j=1;j<=nlstate+ndeath;j++){
3684: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3685: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3686: }
3687: for(d=0; d<=dh[mi][i]; d++){
3688: newm=savm;
3689: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3690: cov[2]=agexact;
3691: if(nagesqr==1)
3692: cov[3]= agexact*agexact;
3693: for (kk=1; kk<=cptcovage;kk++) {
3694: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3695: }
3696: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3697: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3698: savm=oldm;
3699: oldm=newm;
3700: } /* end mult */
3701:
3702: s1=s[mw[mi][i]][i];
3703: s2=s[mw[mi+1][i]][i];
3704: bbh=(double)bh[mi][i]/(double)stepm;
3705: 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 */
3706: ipmx +=1;
3707: sw += weight[i];
3708: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3709: } /* end of wave */
3710: } /* end of individual */
3711: } else if(mle==3){ /* exponential inter-extrapolation */
3712: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3713: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3714: for(mi=1; mi<= wav[i]-1; mi++){
3715: for (ii=1;ii<=nlstate+ndeath;ii++)
3716: for (j=1;j<=nlstate+ndeath;j++){
3717: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3718: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3719: }
3720: for(d=0; d<dh[mi][i]; d++){
3721: newm=savm;
3722: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3723: cov[2]=agexact;
3724: if(nagesqr==1)
3725: cov[3]= agexact*agexact;
3726: for (kk=1; kk<=cptcovage;kk++) {
3727: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3728: }
3729: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3730: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3731: savm=oldm;
3732: oldm=newm;
3733: } /* end mult */
3734:
3735: s1=s[mw[mi][i]][i];
3736: s2=s[mw[mi+1][i]][i];
3737: bbh=(double)bh[mi][i]/(double)stepm;
3738: 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 */
3739: ipmx +=1;
3740: sw += weight[i];
3741: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3742: } /* end of wave */
3743: } /* end of individual */
3744: }else if (mle==4){ /* ml=4 no inter-extrapolation */
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: if( s2 > nlstate){
3772: lli=log(out[s1][s2] - savm[s1][s2]);
3773: } else if ( s2==-1 ) { /* alive */
3774: for (j=1,survp=0. ; j<=nlstate; j++)
3775: survp += out[s1][j];
3776: lli= log(survp);
3777: }else{
3778: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3779: }
3780: ipmx +=1;
3781: sw += weight[i];
3782: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3783: /* 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 3784: } /* end of wave */
3785: } /* end of individual */
3786: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3787: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3788: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3789: for(mi=1; mi<= wav[i]-1; mi++){
3790: for (ii=1;ii<=nlstate+ndeath;ii++)
3791: for (j=1;j<=nlstate+ndeath;j++){
3792: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3793: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3794: }
3795: for(d=0; d<dh[mi][i]; d++){
3796: newm=savm;
3797: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3798: cov[2]=agexact;
3799: if(nagesqr==1)
3800: cov[3]= agexact*agexact;
3801: for (kk=1; kk<=cptcovage;kk++) {
3802: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3803: }
1.126 brouard 3804:
1.226 brouard 3805: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3806: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3807: savm=oldm;
3808: oldm=newm;
3809: } /* end mult */
3810:
3811: s1=s[mw[mi][i]][i];
3812: s2=s[mw[mi+1][i]][i];
3813: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3814: ipmx +=1;
3815: sw += weight[i];
3816: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3817: /*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]);*/
3818: } /* end of wave */
3819: } /* end of individual */
3820: } /* End of if */
3821: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3822: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3823: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3824: return -l;
1.126 brouard 3825: }
3826:
3827: /*************** log-likelihood *************/
3828: double funcone( double *x)
3829: {
1.228 brouard 3830: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3831: int i, ii, j, k, mi, d, kk;
1.228 brouard 3832: int ioffset=0;
1.131 brouard 3833: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3834: double **out;
3835: double lli; /* Individual log likelihood */
3836: double llt;
3837: int s1, s2;
1.228 brouard 3838: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3839:
1.126 brouard 3840: double bbh, survp;
1.187 brouard 3841: double agexact;
1.214 brouard 3842: double agebegin, ageend;
1.126 brouard 3843: /*extern weight */
3844: /* We are differentiating ll according to initial status */
3845: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3846: /*for(i=1;i<imx;i++)
3847: printf(" %d\n",s[4][i]);
3848: */
3849: cov[1]=1.;
3850:
3851: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3852: ioffset=0;
3853: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3854: /* ioffset=2+nagesqr+cptcovage; */
3855: ioffset=2+nagesqr;
1.232 brouard 3856: /* Fixed */
1.224 brouard 3857: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3858: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 3859: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.232 brouard 3860: 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)*/
3861: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3862: /* cov[2+6]=covar[Tvar[6]][i]; */
3863: /* cov[2+6]=covar[2][i]; V2 */
3864: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3865: /* cov[2+7]=covar[Tvar[7]][i]; */
3866: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3867: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3868: /* cov[2+9]=covar[Tvar[9]][i]; */
3869: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3870: }
1.232 brouard 3871: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3872: /* 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?)*\/ */
3873: /* } */
1.231 brouard 3874: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3875: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3876: /* } */
1.225 brouard 3877:
1.233 brouard 3878:
3879: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3880: /* Wave varying (but not age varying) */
3881: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3882: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3883: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3884: }
1.232 brouard 3885: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3886: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3887: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3888: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3889: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3890: /* 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 3891: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3892: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3893: /* /\* 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]); *\/ */
3894: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3895: /* } */
1.126 brouard 3896: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3897: for (j=1;j<=nlstate+ndeath;j++){
3898: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3899: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3900: }
1.214 brouard 3901:
3902: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3903: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3904: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3905: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3906: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3907: and mw[mi+1][i]. dh depends on stepm.*/
3908: newm=savm;
1.247 brouard 3909: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3910: cov[2]=agexact;
3911: if(nagesqr==1)
3912: cov[3]= agexact*agexact;
3913: for (kk=1; kk<=cptcovage;kk++) {
3914: if(!FixedV[Tvar[Tage[kk]]])
3915: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3916: else
3917: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3918: }
3919: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3920: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3921: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3922: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3923: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3924: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3925: savm=oldm;
3926: oldm=newm;
1.126 brouard 3927: } /* end mult */
3928:
3929: s1=s[mw[mi][i]][i];
3930: s2=s[mw[mi+1][i]][i];
1.217 brouard 3931: /* if(s2==-1){ */
1.268 brouard 3932: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3933: /* /\* exit(1); *\/ */
3934: /* } */
1.126 brouard 3935: bbh=(double)bh[mi][i]/(double)stepm;
3936: /* bias is positive if real duration
3937: * is higher than the multiple of stepm and negative otherwise.
3938: */
3939: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3940: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3941: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3942: for (j=1,survp=0. ; j<=nlstate; j++)
3943: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3944: lli= log(survp);
1.126 brouard 3945: }else if (mle==1){
1.242 brouard 3946: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3947: } else if(mle==2){
1.242 brouard 3948: 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 3949: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3950: 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 3951: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3952: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3953: } else{ /* mle=0 back to 1 */
1.242 brouard 3954: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3955: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3956: } /* End of if */
3957: ipmx +=1;
3958: sw += weight[i];
3959: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3960: /*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 3961: if(globpr){
1.246 brouard 3962: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3963: %11.6f %11.6f %11.6f ", \
1.242 brouard 3964: 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 3965: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3966: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3967: llt +=ll[k]*gipmx/gsw;
3968: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3969: }
3970: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3971: }
1.232 brouard 3972: } /* end of wave */
3973: } /* end of individual */
3974: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3975: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3976: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3977: if(globpr==0){ /* First time we count the contributions and weights */
3978: gipmx=ipmx;
3979: gsw=sw;
3980: }
3981: return -l;
1.126 brouard 3982: }
3983:
3984:
3985: /*************** function likelione ***********/
1.292 brouard 3986: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3987: {
3988: /* This routine should help understanding what is done with
3989: the selection of individuals/waves and
3990: to check the exact contribution to the likelihood.
3991: Plotting could be done.
3992: */
3993: int k;
3994:
3995: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3996: strcpy(fileresilk,"ILK_");
1.202 brouard 3997: strcat(fileresilk,fileresu);
1.126 brouard 3998: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3999: printf("Problem with resultfile: %s\n", fileresilk);
4000: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4001: }
1.214 brouard 4002: 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");
4003: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4004: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4005: for(k=1; k<=nlstate; k++)
4006: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4007: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4008: }
4009:
1.292 brouard 4010: *fretone=(*func)(p);
1.126 brouard 4011: if(*globpri !=0){
4012: fclose(ficresilk);
1.205 brouard 4013: if (mle ==0)
4014: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4015: else if(mle >=1)
4016: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4017: 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 4018: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4019:
4020: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4021: 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 4022: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4023: }
1.207 brouard 4024: 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 4025: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4026: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4027: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4028: fflush(fichtm);
1.205 brouard 4029: }
1.126 brouard 4030: return;
4031: }
4032:
4033:
4034: /*********** Maximum Likelihood Estimation ***************/
4035:
4036: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4037: {
1.165 brouard 4038: int i,j, iter=0;
1.126 brouard 4039: double **xi;
4040: double fret;
4041: double fretone; /* Only one call to likelihood */
4042: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4043:
4044: #ifdef NLOPT
4045: int creturn;
4046: nlopt_opt opt;
4047: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4048: double *lb;
4049: double minf; /* the minimum objective value, upon return */
4050: double * p1; /* Shifted parameters from 0 instead of 1 */
4051: myfunc_data dinst, *d = &dinst;
4052: #endif
4053:
4054:
1.126 brouard 4055: xi=matrix(1,npar,1,npar);
4056: for (i=1;i<=npar;i++)
4057: for (j=1;j<=npar;j++)
4058: xi[i][j]=(i==j ? 1.0 : 0.0);
4059: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4060: strcpy(filerespow,"POW_");
1.126 brouard 4061: strcat(filerespow,fileres);
4062: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4063: printf("Problem with resultfile: %s\n", filerespow);
4064: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4065: }
4066: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4067: for (i=1;i<=nlstate;i++)
4068: for(j=1;j<=nlstate+ndeath;j++)
4069: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4070: fprintf(ficrespow,"\n");
1.162 brouard 4071: #ifdef POWELL
1.126 brouard 4072: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4073: #endif
1.126 brouard 4074:
1.162 brouard 4075: #ifdef NLOPT
4076: #ifdef NEWUOA
4077: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4078: #else
4079: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4080: #endif
4081: lb=vector(0,npar-1);
4082: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4083: nlopt_set_lower_bounds(opt, lb);
4084: nlopt_set_initial_step1(opt, 0.1);
4085:
4086: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4087: d->function = func;
4088: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4089: nlopt_set_min_objective(opt, myfunc, d);
4090: nlopt_set_xtol_rel(opt, ftol);
4091: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4092: printf("nlopt failed! %d\n",creturn);
4093: }
4094: else {
4095: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4096: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4097: iter=1; /* not equal */
4098: }
4099: nlopt_destroy(opt);
4100: #endif
1.126 brouard 4101: free_matrix(xi,1,npar,1,npar);
4102: fclose(ficrespow);
1.203 brouard 4103: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4104: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4105: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4106:
4107: }
4108:
4109: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4110: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4111: {
4112: double **a,**y,*x,pd;
1.203 brouard 4113: /* double **hess; */
1.164 brouard 4114: int i, j;
1.126 brouard 4115: int *indx;
4116:
4117: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4118: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4119: void lubksb(double **a, int npar, int *indx, double b[]) ;
4120: void ludcmp(double **a, int npar, int *indx, double *d) ;
4121: double gompertz(double p[]);
1.203 brouard 4122: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4123:
4124: printf("\nCalculation of the hessian matrix. Wait...\n");
4125: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4126: for (i=1;i<=npar;i++){
1.203 brouard 4127: printf("%d-",i);fflush(stdout);
4128: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4129:
4130: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4131:
4132: /* printf(" %f ",p[i]);
4133: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4134: }
4135:
4136: for (i=1;i<=npar;i++) {
4137: for (j=1;j<=npar;j++) {
4138: if (j>i) {
1.203 brouard 4139: printf(".%d-%d",i,j);fflush(stdout);
4140: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4141: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4142:
4143: hess[j][i]=hess[i][j];
4144: /*printf(" %lf ",hess[i][j]);*/
4145: }
4146: }
4147: }
4148: printf("\n");
4149: fprintf(ficlog,"\n");
4150:
4151: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4152: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4153:
4154: a=matrix(1,npar,1,npar);
4155: y=matrix(1,npar,1,npar);
4156: x=vector(1,npar);
4157: indx=ivector(1,npar);
4158: for (i=1;i<=npar;i++)
4159: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4160: ludcmp(a,npar,indx,&pd);
4161:
4162: for (j=1;j<=npar;j++) {
4163: for (i=1;i<=npar;i++) x[i]=0;
4164: x[j]=1;
4165: lubksb(a,npar,indx,x);
4166: for (i=1;i<=npar;i++){
4167: matcov[i][j]=x[i];
4168: }
4169: }
4170:
4171: printf("\n#Hessian matrix#\n");
4172: fprintf(ficlog,"\n#Hessian matrix#\n");
4173: for (i=1;i<=npar;i++) {
4174: for (j=1;j<=npar;j++) {
1.203 brouard 4175: printf("%.6e ",hess[i][j]);
4176: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4177: }
4178: printf("\n");
4179: fprintf(ficlog,"\n");
4180: }
4181:
1.203 brouard 4182: /* printf("\n#Covariance matrix#\n"); */
4183: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4184: /* for (i=1;i<=npar;i++) { */
4185: /* for (j=1;j<=npar;j++) { */
4186: /* printf("%.6e ",matcov[i][j]); */
4187: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4188: /* } */
4189: /* printf("\n"); */
4190: /* fprintf(ficlog,"\n"); */
4191: /* } */
4192:
1.126 brouard 4193: /* Recompute Inverse */
1.203 brouard 4194: /* for (i=1;i<=npar;i++) */
4195: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4196: /* ludcmp(a,npar,indx,&pd); */
4197:
4198: /* printf("\n#Hessian matrix recomputed#\n"); */
4199:
4200: /* for (j=1;j<=npar;j++) { */
4201: /* for (i=1;i<=npar;i++) x[i]=0; */
4202: /* x[j]=1; */
4203: /* lubksb(a,npar,indx,x); */
4204: /* for (i=1;i<=npar;i++){ */
4205: /* y[i][j]=x[i]; */
4206: /* printf("%.3e ",y[i][j]); */
4207: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4208: /* } */
4209: /* printf("\n"); */
4210: /* fprintf(ficlog,"\n"); */
4211: /* } */
4212:
4213: /* Verifying the inverse matrix */
4214: #ifdef DEBUGHESS
4215: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4216:
1.203 brouard 4217: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4218: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4219:
4220: for (j=1;j<=npar;j++) {
4221: for (i=1;i<=npar;i++){
1.203 brouard 4222: printf("%.2f ",y[i][j]);
4223: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4224: }
4225: printf("\n");
4226: fprintf(ficlog,"\n");
4227: }
1.203 brouard 4228: #endif
1.126 brouard 4229:
4230: free_matrix(a,1,npar,1,npar);
4231: free_matrix(y,1,npar,1,npar);
4232: free_vector(x,1,npar);
4233: free_ivector(indx,1,npar);
1.203 brouard 4234: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4235:
4236:
4237: }
4238:
4239: /*************** hessian matrix ****************/
4240: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4241: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4242: int i;
4243: int l=1, lmax=20;
1.203 brouard 4244: double k1,k2, res, fx;
1.132 brouard 4245: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4246: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4247: int k=0,kmax=10;
4248: double l1;
4249:
4250: fx=func(x);
4251: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4252: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4253: l1=pow(10,l);
4254: delts=delt;
4255: for(k=1 ; k <kmax; k=k+1){
4256: delt = delta*(l1*k);
4257: p2[theta]=x[theta] +delt;
1.145 brouard 4258: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4259: p2[theta]=x[theta]-delt;
4260: k2=func(p2)-fx;
4261: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4262: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4263:
1.203 brouard 4264: #ifdef DEBUGHESSII
1.126 brouard 4265: 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);
4266: 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);
4267: #endif
4268: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4269: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4270: k=kmax;
4271: }
4272: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4273: k=kmax; l=lmax*10;
1.126 brouard 4274: }
4275: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4276: delts=delt;
4277: }
1.203 brouard 4278: } /* End loop k */
1.126 brouard 4279: }
4280: delti[theta]=delts;
4281: return res;
4282:
4283: }
4284:
1.203 brouard 4285: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4286: {
4287: int i;
1.164 brouard 4288: int l=1, lmax=20;
1.126 brouard 4289: double k1,k2,k3,k4,res,fx;
1.132 brouard 4290: double p2[MAXPARM+1];
1.203 brouard 4291: int k, kmax=1;
4292: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4293:
4294: int firstime=0;
1.203 brouard 4295:
1.126 brouard 4296: fx=func(x);
1.203 brouard 4297: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4298: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4299: p2[thetai]=x[thetai]+delti[thetai]*k;
4300: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4301: k1=func(p2)-fx;
4302:
1.203 brouard 4303: p2[thetai]=x[thetai]+delti[thetai]*k;
4304: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4305: k2=func(p2)-fx;
4306:
1.203 brouard 4307: p2[thetai]=x[thetai]-delti[thetai]*k;
4308: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4309: k3=func(p2)-fx;
4310:
1.203 brouard 4311: p2[thetai]=x[thetai]-delti[thetai]*k;
4312: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4313: k4=func(p2)-fx;
1.203 brouard 4314: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4315: if(k1*k2*k3*k4 <0.){
1.208 brouard 4316: firstime=1;
1.203 brouard 4317: kmax=kmax+10;
1.208 brouard 4318: }
4319: if(kmax >=10 || firstime ==1){
1.246 brouard 4320: 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);
4321: 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 4322: 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);
4323: 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);
4324: }
4325: #ifdef DEBUGHESSIJ
4326: v1=hess[thetai][thetai];
4327: v2=hess[thetaj][thetaj];
4328: cv12=res;
4329: /* Computing eigen value of Hessian matrix */
4330: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4331: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4332: if ((lc2 <0) || (lc1 <0) ){
4333: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4334: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4335: 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);
4336: 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);
4337: }
1.126 brouard 4338: #endif
4339: }
4340: return res;
4341: }
4342:
1.203 brouard 4343: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4344: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4345: /* { */
4346: /* int i; */
4347: /* int l=1, lmax=20; */
4348: /* double k1,k2,k3,k4,res,fx; */
4349: /* double p2[MAXPARM+1]; */
4350: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4351: /* int k=0,kmax=10; */
4352: /* double l1; */
4353:
4354: /* fx=func(x); */
4355: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4356: /* l1=pow(10,l); */
4357: /* delts=delt; */
4358: /* for(k=1 ; k <kmax; k=k+1){ */
4359: /* delt = delti*(l1*k); */
4360: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4361: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4362: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4363: /* k1=func(p2)-fx; */
4364:
4365: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4366: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4367: /* k2=func(p2)-fx; */
4368:
4369: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4370: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4371: /* k3=func(p2)-fx; */
4372:
4373: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4374: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4375: /* k4=func(p2)-fx; */
4376: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4377: /* #ifdef DEBUGHESSIJ */
4378: /* 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); */
4379: /* 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); */
4380: /* #endif */
4381: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4382: /* k=kmax; */
4383: /* } */
4384: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4385: /* k=kmax; l=lmax*10; */
4386: /* } */
4387: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4388: /* delts=delt; */
4389: /* } */
4390: /* } /\* End loop k *\/ */
4391: /* } */
4392: /* delti[theta]=delts; */
4393: /* return res; */
4394: /* } */
4395:
4396:
1.126 brouard 4397: /************** Inverse of matrix **************/
4398: void ludcmp(double **a, int n, int *indx, double *d)
4399: {
4400: int i,imax,j,k;
4401: double big,dum,sum,temp;
4402: double *vv;
4403:
4404: vv=vector(1,n);
4405: *d=1.0;
4406: for (i=1;i<=n;i++) {
4407: big=0.0;
4408: for (j=1;j<=n;j++)
4409: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4410: if (big == 0.0){
4411: printf(" Singular Hessian matrix at row %d:\n",i);
4412: for (j=1;j<=n;j++) {
4413: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4414: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4415: }
4416: fflush(ficlog);
4417: fclose(ficlog);
4418: nrerror("Singular matrix in routine ludcmp");
4419: }
1.126 brouard 4420: vv[i]=1.0/big;
4421: }
4422: for (j=1;j<=n;j++) {
4423: for (i=1;i<j;i++) {
4424: sum=a[i][j];
4425: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4426: a[i][j]=sum;
4427: }
4428: big=0.0;
4429: for (i=j;i<=n;i++) {
4430: sum=a[i][j];
4431: for (k=1;k<j;k++)
4432: sum -= a[i][k]*a[k][j];
4433: a[i][j]=sum;
4434: if ( (dum=vv[i]*fabs(sum)) >= big) {
4435: big=dum;
4436: imax=i;
4437: }
4438: }
4439: if (j != imax) {
4440: for (k=1;k<=n;k++) {
4441: dum=a[imax][k];
4442: a[imax][k]=a[j][k];
4443: a[j][k]=dum;
4444: }
4445: *d = -(*d);
4446: vv[imax]=vv[j];
4447: }
4448: indx[j]=imax;
4449: if (a[j][j] == 0.0) a[j][j]=TINY;
4450: if (j != n) {
4451: dum=1.0/(a[j][j]);
4452: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4453: }
4454: }
4455: free_vector(vv,1,n); /* Doesn't work */
4456: ;
4457: }
4458:
4459: void lubksb(double **a, int n, int *indx, double b[])
4460: {
4461: int i,ii=0,ip,j;
4462: double sum;
4463:
4464: for (i=1;i<=n;i++) {
4465: ip=indx[i];
4466: sum=b[ip];
4467: b[ip]=b[i];
4468: if (ii)
4469: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4470: else if (sum) ii=i;
4471: b[i]=sum;
4472: }
4473: for (i=n;i>=1;i--) {
4474: sum=b[i];
4475: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4476: b[i]=sum/a[i][i];
4477: }
4478: }
4479:
4480: void pstamp(FILE *fichier)
4481: {
1.196 brouard 4482: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4483: }
4484:
1.297 brouard 4485: void date2dmy(double date,double *day, double *month, double *year){
4486: double yp=0., yp1=0., yp2=0.;
4487:
4488: yp1=modf(date,&yp);/* extracts integral of date in yp and
4489: fractional in yp1 */
4490: *year=yp;
4491: yp2=modf((yp1*12),&yp);
4492: *month=yp;
4493: yp1=modf((yp2*30.5),&yp);
4494: *day=yp;
4495: if(*day==0) *day=1;
4496: if(*month==0) *month=1;
4497: }
4498:
1.253 brouard 4499:
4500:
1.126 brouard 4501: /************ Frequencies ********************/
1.251 brouard 4502: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4503: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4504: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4505: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4506:
1.265 brouard 4507: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4508: int iind=0, iage=0;
4509: int mi; /* Effective wave */
4510: int first;
4511: double ***freq; /* Frequencies */
1.268 brouard 4512: 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 */
4513: 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 4514: double *meanq, *stdq, *idq;
1.226 brouard 4515: double **meanqt;
4516: double *pp, **prop, *posprop, *pospropt;
4517: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4518: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4519: double agebegin, ageend;
4520:
4521: pp=vector(1,nlstate);
1.251 brouard 4522: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4523: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4524: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4525: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4526: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4527: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4528: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4529: meanqt=matrix(1,lastpass,1,nqtveff);
4530: strcpy(fileresp,"P_");
4531: strcat(fileresp,fileresu);
4532: /*strcat(fileresphtm,fileresu);*/
4533: if((ficresp=fopen(fileresp,"w"))==NULL) {
4534: printf("Problem with prevalence resultfile: %s\n", fileresp);
4535: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4536: exit(0);
4537: }
1.240 brouard 4538:
1.226 brouard 4539: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4540: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4541: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4542: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4543: fflush(ficlog);
4544: exit(70);
4545: }
4546: else{
4547: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4548: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4549: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4550: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4551: }
1.237 brouard 4552: 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 4553:
1.226 brouard 4554: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4555: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4556: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4557: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4558: fflush(ficlog);
4559: exit(70);
1.240 brouard 4560: } else{
1.226 brouard 4561: 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 4562: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4563: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4564: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4565: }
1.240 brouard 4566: 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);
4567:
1.253 brouard 4568: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4569: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4570: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4571: j1=0;
1.126 brouard 4572:
1.227 brouard 4573: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4574: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4575: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4576:
4577:
1.226 brouard 4578: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4579: reference=low_education V1=0,V2=0
4580: med_educ V1=1 V2=0,
4581: high_educ V1=0 V2=1
4582: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4583: */
1.249 brouard 4584: dateintsum=0;
4585: k2cpt=0;
4586:
1.253 brouard 4587: if(cptcoveff == 0 )
1.265 brouard 4588: nl=1; /* Constant and age model only */
1.253 brouard 4589: else
4590: nl=2;
1.265 brouard 4591:
4592: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4593: /* Loop on nj=1 or 2 if dummy covariates j!=0
4594: * Loop on j1(1 to 2**cptcoveff) covariate combination
4595: * freq[s1][s2][iage] =0.
4596: * Loop on iind
4597: * ++freq[s1][s2][iage] weighted
4598: * end iind
4599: * if covariate and j!0
4600: * headers Variable on one line
4601: * endif cov j!=0
4602: * header of frequency table by age
4603: * Loop on age
4604: * pp[s1]+=freq[s1][s2][iage] weighted
4605: * pos+=freq[s1][s2][iage] weighted
4606: * Loop on s1 initial state
4607: * fprintf(ficresp
4608: * end s1
4609: * end age
4610: * if j!=0 computes starting values
4611: * end compute starting values
4612: * end j1
4613: * end nl
4614: */
1.253 brouard 4615: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4616: if(nj==1)
4617: j=0; /* First pass for the constant */
1.265 brouard 4618: else{
1.253 brouard 4619: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4620: }
1.251 brouard 4621: first=1;
1.265 brouard 4622: 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 4623: posproptt=0.;
4624: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4625: scanf("%d", i);*/
4626: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4627: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4628: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4629: freq[i][s2][m]=0;
1.251 brouard 4630:
4631: for (i=1; i<=nlstate; i++) {
1.240 brouard 4632: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4633: prop[i][m]=0;
4634: posprop[i]=0;
4635: pospropt[i]=0;
4636: }
1.283 brouard 4637: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4638: idq[z1]=0.;
4639: meanq[z1]=0.;
4640: stdq[z1]=0.;
1.283 brouard 4641: }
4642: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4643: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4644: /* meanqt[m][z1]=0.; */
4645: /* } */
4646: /* } */
1.251 brouard 4647: /* dateintsum=0; */
4648: /* k2cpt=0; */
4649:
1.265 brouard 4650: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4651: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4652: bool=1;
4653: if(j !=0){
4654: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4655: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4656: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4657: /* if(Tvaraff[z1] ==-20){ */
4658: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4659: /* }else if(Tvaraff[z1] ==-10){ */
4660: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4661: /* }else */
4662: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4663: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4664: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4665: /* 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",
4666: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4667: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4668: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4669: } /* Onlyf fixed */
4670: } /* end z1 */
4671: } /* cptcovn > 0 */
4672: } /* end any */
4673: }/* end j==0 */
1.265 brouard 4674: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4675: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4676: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4677: m=mw[mi][iind];
4678: if(j!=0){
4679: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4680: for (z1=1; z1<=cptcoveff; z1++) {
4681: if( Fixed[Tmodelind[z1]]==1){
4682: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4683: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4684: value is -1, we don't select. It differs from the
4685: constant and age model which counts them. */
4686: bool=0; /* not selected */
4687: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4688: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4689: bool=0;
4690: }
4691: }
4692: }
4693: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4694: } /* end j==0 */
4695: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4696: if(bool==1){ /*Selected */
1.251 brouard 4697: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4698: and mw[mi+1][iind]. dh depends on stepm. */
4699: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4700: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4701: if(m >=firstpass && m <=lastpass){
4702: k2=anint[m][iind]+(mint[m][iind]/12.);
4703: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4704: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4705: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4706: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4707: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4708: if (m<lastpass) {
4709: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4710: /* 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]); */
4711: if(s[m][iind]==-1)
4712: 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.));
4713: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.311 brouard 4714: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4715: if(!isnan(covar[ncovcol+z1][iind])){
4716: idq[z1]=idq[z1]+weight[iind];
4717: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4718: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4719: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4720: }
1.284 brouard 4721: }
1.251 brouard 4722: /* if((int)agev[m][iind] == 55) */
4723: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4724: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4725: 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 4726: }
1.251 brouard 4727: } /* end if between passes */
4728: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4729: dateintsum=dateintsum+k2; /* on all covariates ?*/
4730: k2cpt++;
4731: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4732: }
1.251 brouard 4733: }else{
4734: bool=1;
4735: }/* end bool 2 */
4736: } /* end m */
1.284 brouard 4737: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4738: /* idq[z1]=idq[z1]+weight[iind]; */
4739: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4740: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4741: /* } */
1.251 brouard 4742: } /* end bool */
4743: } /* end iind = 1 to imx */
4744: /* prop[s][age] is feeded for any initial and valid live state as well as
4745: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4746:
4747:
4748: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4749: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4750: pstamp(ficresp);
1.251 brouard 4751: if (cptcoveff>0 && j!=0){
1.265 brouard 4752: pstamp(ficresp);
1.251 brouard 4753: printf( "\n#********** Variable ");
4754: fprintf(ficresp, "\n#********** Variable ");
4755: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4756: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4757: fprintf(ficlog, "\n#********** Variable ");
4758: for (z1=1; z1<=cptcoveff; z1++){
4759: if(!FixedV[Tvaraff[z1]]){
4760: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4761: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4762: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4763: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4764: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4765: }else{
1.251 brouard 4766: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4767: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4768: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4769: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4770: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4771: }
4772: }
4773: printf( "**********\n#");
4774: fprintf(ficresp, "**********\n#");
4775: fprintf(ficresphtm, "**********</h3>\n");
4776: fprintf(ficresphtmfr, "**********</h3>\n");
4777: fprintf(ficlog, "**********\n");
4778: }
1.284 brouard 4779: /*
4780: Printing means of quantitative variables if any
4781: */
4782: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 4783: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 ! brouard 4784: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 4785: if(weightopt==1){
4786: printf(" Weighted mean and standard deviation of");
4787: fprintf(ficlog," Weighted mean and standard deviation of");
4788: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4789: }
1.311 brouard 4790: /* mu = \frac{w x}{\sum w}
4791: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
4792: */
4793: printf(" fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
4794: fprintf(ficlog," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
4795: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
1.284 brouard 4796: }
4797: /* for (z1=1; z1<= nqtveff; z1++) { */
4798: /* for(m=1;m<=lastpass;m++){ */
4799: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4800: /* } */
4801: /* } */
1.283 brouard 4802:
1.251 brouard 4803: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4804: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4805: fprintf(ficresp, " Age");
4806: 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 4807: for(i=1; i<=nlstate;i++) {
1.265 brouard 4808: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4809: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4810: }
1.265 brouard 4811: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4812: fprintf(ficresphtm, "\n");
4813:
4814: /* Header of frequency table by age */
4815: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4816: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4817: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4818: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4819: if(s2!=0 && m!=0)
4820: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4821: }
1.226 brouard 4822: }
1.251 brouard 4823: fprintf(ficresphtmfr, "\n");
4824:
4825: /* For each age */
4826: for(iage=iagemin; iage <= iagemax+3; iage++){
4827: fprintf(ficresphtm,"<tr>");
4828: if(iage==iagemax+1){
4829: fprintf(ficlog,"1");
4830: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4831: }else if(iage==iagemax+2){
4832: fprintf(ficlog,"0");
4833: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4834: }else if(iage==iagemax+3){
4835: fprintf(ficlog,"Total");
4836: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4837: }else{
1.240 brouard 4838: if(first==1){
1.251 brouard 4839: first=0;
4840: printf("See log file for details...\n");
4841: }
4842: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4843: fprintf(ficlog,"Age %d", iage);
4844: }
1.265 brouard 4845: for(s1=1; s1 <=nlstate ; s1++){
4846: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4847: pp[s1] += freq[s1][m][iage];
1.251 brouard 4848: }
1.265 brouard 4849: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4850: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4851: pos += freq[s1][m][iage];
4852: if(pp[s1]>=1.e-10){
1.251 brouard 4853: if(first==1){
1.265 brouard 4854: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4855: }
1.265 brouard 4856: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4857: }else{
4858: if(first==1)
1.265 brouard 4859: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4860: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4861: }
4862: }
4863:
1.265 brouard 4864: for(s1=1; s1 <=nlstate ; s1++){
4865: /* posprop[s1]=0; */
4866: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4867: pp[s1] += freq[s1][m][iage];
4868: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4869:
4870: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4871: pos += pp[s1]; /* pos is the total number of transitions until this age */
4872: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4873: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4874: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4875: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4876: }
4877:
4878: /* Writing ficresp */
4879: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4880: if( iage <= iagemax){
4881: fprintf(ficresp," %d",iage);
4882: }
4883: }else if( nj==2){
4884: if( iage <= iagemax){
4885: fprintf(ficresp," %d",iage);
4886: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4887: }
1.240 brouard 4888: }
1.265 brouard 4889: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4890: if(pos>=1.e-5){
1.251 brouard 4891: if(first==1)
1.265 brouard 4892: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4893: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4894: }else{
4895: if(first==1)
1.265 brouard 4896: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4897: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4898: }
4899: if( iage <= iagemax){
4900: if(pos>=1.e-5){
1.265 brouard 4901: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4902: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4903: }else if( nj==2){
4904: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4905: }
4906: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4907: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4908: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4909: } else{
4910: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4911: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4912: }
1.240 brouard 4913: }
1.265 brouard 4914: pospropt[s1] +=posprop[s1];
4915: } /* end loop s1 */
1.251 brouard 4916: /* pospropt=0.; */
1.265 brouard 4917: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4918: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4919: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4920: if(first==1){
1.265 brouard 4921: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4922: }
1.265 brouard 4923: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4924: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4925: }
1.265 brouard 4926: if(s1!=0 && m!=0)
4927: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4928: }
1.265 brouard 4929: } /* end loop s1 */
1.251 brouard 4930: posproptt=0.;
1.265 brouard 4931: for(s1=1; s1 <=nlstate; s1++){
4932: posproptt += pospropt[s1];
1.251 brouard 4933: }
4934: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4935: fprintf(ficresphtm,"</tr>\n");
4936: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4937: if(iage <= iagemax)
4938: fprintf(ficresp,"\n");
1.240 brouard 4939: }
1.251 brouard 4940: if(first==1)
4941: printf("Others in log...\n");
4942: fprintf(ficlog,"\n");
4943: } /* end loop age iage */
1.265 brouard 4944:
1.251 brouard 4945: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4946: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4947: if(posproptt < 1.e-5){
1.265 brouard 4948: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4949: }else{
1.265 brouard 4950: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4951: }
1.226 brouard 4952: }
1.251 brouard 4953: fprintf(ficresphtm,"</tr>\n");
4954: fprintf(ficresphtm,"</table>\n");
4955: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4956: if(posproptt < 1.e-5){
1.251 brouard 4957: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4958: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4959: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4960: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4961: invalidvarcomb[j1]=1;
1.226 brouard 4962: }else{
1.251 brouard 4963: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4964: invalidvarcomb[j1]=0;
1.226 brouard 4965: }
1.251 brouard 4966: fprintf(ficresphtmfr,"</table>\n");
4967: fprintf(ficlog,"\n");
4968: if(j!=0){
4969: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4970: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4971: for(k=1; k <=(nlstate+ndeath); k++){
4972: if (k != i) {
1.265 brouard 4973: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4974: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4975: if(j1==1){ /* All dummy covariates to zero */
4976: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4977: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4978: printf("%d%d ",i,k);
4979: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4980: 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]));
4981: 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]));
4982: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4983: }
1.253 brouard 4984: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4985: for(iage=iagemin; iage <= iagemax+3; iage++){
4986: x[iage]= (double)iage;
4987: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4988: /* 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 4989: }
1.268 brouard 4990: /* Some are not finite, but linreg will ignore these ages */
4991: no=0;
1.253 brouard 4992: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4993: pstart[s1]=b;
4994: pstart[s1-1]=a;
1.252 brouard 4995: }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 */
4996: 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]);
4997: 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 4998: 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 4999: printf("%d%d ",i,k);
5000: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5001: 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 5002: }else{ /* Other cases, like quantitative fixed or varying covariates */
5003: ;
5004: }
5005: /* printf("%12.7f )", param[i][jj][k]); */
5006: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5007: s1++;
1.251 brouard 5008: } /* end jj */
5009: } /* end k!= i */
5010: } /* end k */
1.265 brouard 5011: } /* end i, s1 */
1.251 brouard 5012: } /* end j !=0 */
5013: } /* end selected combination of covariate j1 */
5014: if(j==0){ /* We can estimate starting values from the occurences in each case */
5015: printf("#Freqsummary: Starting values for the constants:\n");
5016: fprintf(ficlog,"\n");
1.265 brouard 5017: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5018: for(k=1; k <=(nlstate+ndeath); k++){
5019: if (k != i) {
5020: printf("%d%d ",i,k);
5021: fprintf(ficlog,"%d%d ",i,k);
5022: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5023: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5024: if(jj==1){ /* Age has to be done */
1.265 brouard 5025: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5026: 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]));
5027: 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 5028: }
5029: /* printf("%12.7f )", param[i][jj][k]); */
5030: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5031: s1++;
1.250 brouard 5032: }
1.251 brouard 5033: printf("\n");
5034: fprintf(ficlog,"\n");
1.250 brouard 5035: }
5036: }
1.284 brouard 5037: } /* end of state i */
1.251 brouard 5038: printf("#Freqsummary\n");
5039: fprintf(ficlog,"\n");
1.265 brouard 5040: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5041: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5042: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5043: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5044: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5045: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5046: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5047: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5048: /* } */
5049: }
1.265 brouard 5050: } /* end loop s1 */
1.251 brouard 5051:
5052: printf("\n");
5053: fprintf(ficlog,"\n");
5054: } /* end j=0 */
1.249 brouard 5055: } /* end j */
1.252 brouard 5056:
1.253 brouard 5057: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5058: for(i=1, jk=1; i <=nlstate; i++){
5059: for(j=1; j <=nlstate+ndeath; j++){
5060: if(j!=i){
5061: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5062: printf("%1d%1d",i,j);
5063: fprintf(ficparo,"%1d%1d",i,j);
5064: for(k=1; k<=ncovmodel;k++){
5065: /* printf(" %lf",param[i][j][k]); */
5066: /* fprintf(ficparo," %lf",param[i][j][k]); */
5067: p[jk]=pstart[jk];
5068: printf(" %f ",pstart[jk]);
5069: fprintf(ficparo," %f ",pstart[jk]);
5070: jk++;
5071: }
5072: printf("\n");
5073: fprintf(ficparo,"\n");
5074: }
5075: }
5076: }
5077: } /* end mle=-2 */
1.226 brouard 5078: dateintmean=dateintsum/k2cpt;
1.296 brouard 5079: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5080:
1.226 brouard 5081: fclose(ficresp);
5082: fclose(ficresphtm);
5083: fclose(ficresphtmfr);
1.283 brouard 5084: free_vector(idq,1,nqfveff);
1.226 brouard 5085: free_vector(meanq,1,nqfveff);
1.284 brouard 5086: free_vector(stdq,1,nqfveff);
1.226 brouard 5087: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5088: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5089: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5090: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5091: free_vector(pospropt,1,nlstate);
5092: free_vector(posprop,1,nlstate);
1.251 brouard 5093: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5094: free_vector(pp,1,nlstate);
5095: /* End of freqsummary */
5096: }
1.126 brouard 5097:
1.268 brouard 5098: /* Simple linear regression */
5099: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5100:
5101: /* y=a+bx regression */
5102: double sumx = 0.0; /* sum of x */
5103: double sumx2 = 0.0; /* sum of x**2 */
5104: double sumxy = 0.0; /* sum of x * y */
5105: double sumy = 0.0; /* sum of y */
5106: double sumy2 = 0.0; /* sum of y**2 */
5107: double sume2 = 0.0; /* sum of square or residuals */
5108: double yhat;
5109:
5110: double denom=0;
5111: int i;
5112: int ne=*no;
5113:
5114: for ( i=ifi, ne=0;i<=ila;i++) {
5115: if(!isfinite(x[i]) || !isfinite(y[i])){
5116: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5117: continue;
5118: }
5119: ne=ne+1;
5120: sumx += x[i];
5121: sumx2 += x[i]*x[i];
5122: sumxy += x[i] * y[i];
5123: sumy += y[i];
5124: sumy2 += y[i]*y[i];
5125: denom = (ne * sumx2 - sumx*sumx);
5126: /* 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); */
5127: }
5128:
5129: denom = (ne * sumx2 - sumx*sumx);
5130: if (denom == 0) {
5131: // vertical, slope m is infinity
5132: *b = INFINITY;
5133: *a = 0;
5134: if (r) *r = 0;
5135: return 1;
5136: }
5137:
5138: *b = (ne * sumxy - sumx * sumy) / denom;
5139: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5140: if (r!=NULL) {
5141: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5142: sqrt((sumx2 - sumx*sumx/ne) *
5143: (sumy2 - sumy*sumy/ne));
5144: }
5145: *no=ne;
5146: for ( i=ifi, ne=0;i<=ila;i++) {
5147: if(!isfinite(x[i]) || !isfinite(y[i])){
5148: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5149: continue;
5150: }
5151: ne=ne+1;
5152: yhat = y[i] - *a -*b* x[i];
5153: sume2 += yhat * yhat ;
5154:
5155: denom = (ne * sumx2 - sumx*sumx);
5156: /* 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); */
5157: }
5158: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5159: *sa= *sb * sqrt(sumx2/ne);
5160:
5161: return 0;
5162: }
5163:
1.126 brouard 5164: /************ Prevalence ********************/
1.227 brouard 5165: 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)
5166: {
5167: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5168: in each health status at the date of interview (if between dateprev1 and dateprev2).
5169: We still use firstpass and lastpass as another selection.
5170: */
1.126 brouard 5171:
1.227 brouard 5172: int i, m, jk, j1, bool, z1,j, iv;
5173: int mi; /* Effective wave */
5174: int iage;
5175: double agebegin, ageend;
5176:
5177: double **prop;
5178: double posprop;
5179: double y2; /* in fractional years */
5180: int iagemin, iagemax;
5181: int first; /** to stop verbosity which is redirected to log file */
5182:
5183: iagemin= (int) agemin;
5184: iagemax= (int) agemax;
5185: /*pp=vector(1,nlstate);*/
1.251 brouard 5186: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5187: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5188: j1=0;
1.222 brouard 5189:
1.227 brouard 5190: /*j=cptcoveff;*/
5191: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5192:
1.288 brouard 5193: first=0;
1.227 brouard 5194: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5195: for (i=1; i<=nlstate; i++)
1.251 brouard 5196: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5197: prop[i][iage]=0.0;
5198: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5199: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5200: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5201:
5202: for (i=1; i<=imx; i++) { /* Each individual */
5203: bool=1;
5204: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5205: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5206: m=mw[mi][i];
5207: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5208: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5209: for (z1=1; z1<=cptcoveff; z1++){
5210: if( Fixed[Tmodelind[z1]]==1){
5211: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5212: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5213: bool=0;
5214: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5215: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5216: bool=0;
5217: }
5218: }
5219: if(bool==1){ /* Otherwise we skip that wave/person */
5220: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5221: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5222: if(m >=firstpass && m <=lastpass){
5223: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5224: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5225: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5226: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5227: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5228: 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);
5229: exit(1);
5230: }
5231: if (s[m][i]>0 && s[m][i]<=nlstate) {
5232: /*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]]);*/
5233: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5234: prop[s[m][i]][iagemax+3] += weight[i];
5235: } /* end valid statuses */
5236: } /* end selection of dates */
5237: } /* end selection of waves */
5238: } /* end bool */
5239: } /* end wave */
5240: } /* end individual */
5241: for(i=iagemin; i <= iagemax+3; i++){
5242: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5243: posprop += prop[jk][i];
5244: }
5245:
5246: for(jk=1; jk <=nlstate ; jk++){
5247: if( i <= iagemax){
5248: if(posprop>=1.e-5){
5249: probs[i][jk][j1]= prop[jk][i]/posprop;
5250: } else{
1.288 brouard 5251: if(!first){
5252: first=1;
1.266 brouard 5253: 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]);
5254: }else{
1.288 brouard 5255: 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 5256: }
5257: }
5258: }
5259: }/* end jk */
5260: }/* end i */
1.222 brouard 5261: /*} *//* end i1 */
1.227 brouard 5262: } /* end j1 */
1.222 brouard 5263:
1.227 brouard 5264: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5265: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5266: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5267: } /* End of prevalence */
1.126 brouard 5268:
5269: /************* Waves Concatenation ***************/
5270:
5271: 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)
5272: {
1.298 brouard 5273: /* 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 5274: Death is a valid wave (if date is known).
5275: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5276: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5277: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5278: */
1.126 brouard 5279:
1.224 brouard 5280: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5281: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5282: double sum=0., jmean=0.;*/
1.224 brouard 5283: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5284: int j, k=0,jk, ju, jl;
5285: double sum=0.;
5286: first=0;
1.214 brouard 5287: firstwo=0;
1.217 brouard 5288: firsthree=0;
1.218 brouard 5289: firstfour=0;
1.164 brouard 5290: jmin=100000;
1.126 brouard 5291: jmax=-1;
5292: jmean=0.;
1.224 brouard 5293:
5294: /* Treating live states */
1.214 brouard 5295: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5296: mi=0; /* First valid wave */
1.227 brouard 5297: mli=0; /* Last valid wave */
1.309 brouard 5298: m=firstpass; /* Loop on waves */
5299: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5300: 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 */
5301: mli=m-1;/* mw[++mi][i]=m-1; */
5302: }else if(s[m][i]>=1 || s[m][i]==-4 || s[m][i]==-5){ /* Since 0.98r4 if status=-2 vital status is really unknown, wave should be skipped */
1.309 brouard 5303: mw[++mi][i]=m; /* Valid wave: incrementing mi and updating mi; mw[mi] is the wave number of mi_th valid transition */
1.227 brouard 5304: mli=m;
1.224 brouard 5305: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5306: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5307: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5308: }
1.309 brouard 5309: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5310: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5311: break;
1.224 brouard 5312: #else
1.309 brouard 5313: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){ /* case -2 (vital status unknown is warned later */
1.227 brouard 5314: if(firsthree == 0){
1.302 brouard 5315: 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 5316: firsthree=1;
5317: }
1.302 brouard 5318: 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);
1.309 brouard 5319: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5320: mli=m;
5321: }
5322: if(s[m][i]==-2){ /* Vital status is really unknown */
5323: nbwarn++;
1.309 brouard 5324: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5325: 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);
5326: 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);
5327: }
5328: break;
5329: }
5330: break;
1.224 brouard 5331: #endif
1.227 brouard 5332: }/* End m >= lastpass */
1.126 brouard 5333: }/* end while */
1.224 brouard 5334:
1.227 brouard 5335: /* 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 5336: /* After last pass */
1.224 brouard 5337: /* Treating death states */
1.214 brouard 5338: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5339: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5340: /* } */
1.126 brouard 5341: mi++; /* Death is another wave */
5342: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5343: /* Only death is a correct wave */
1.126 brouard 5344: mw[mi][i]=m;
1.257 brouard 5345: } /* else not in a death state */
1.224 brouard 5346: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5347: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5348: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5349: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* month of death occured before last wave month and status should have been death instead of -1 */
1.227 brouard 5350: nbwarn++;
5351: if(firstfiv==0){
1.309 brouard 5352: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 5353: firstfiv=1;
5354: }else{
1.309 brouard 5355: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 5356: }
1.309 brouard 5357: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5358: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5359: nberr++;
5360: if(firstwo==0){
1.309 brouard 5361: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 5362: firstwo=1;
5363: }
1.309 brouard 5364: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 5365: }
1.257 brouard 5366: }else{ /* if date of interview is unknown */
1.227 brouard 5367: /* death is known but not confirmed by death status at any wave */
5368: if(firstfour==0){
1.309 brouard 5369: printf("Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 5370: firstfour=1;
5371: }
1.309 brouard 5372: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.214 brouard 5373: }
1.224 brouard 5374: } /* end if date of death is known */
5375: #endif
1.309 brouard 5376: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5377: /* wav[i]=mw[mi][i]; */
1.126 brouard 5378: if(mi==0){
5379: nbwarn++;
5380: if(first==0){
1.227 brouard 5381: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5382: first=1;
1.126 brouard 5383: }
5384: if(first==1){
1.227 brouard 5385: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5386: }
5387: } /* end mi==0 */
5388: } /* End individuals */
1.214 brouard 5389: /* wav and mw are no more changed */
1.223 brouard 5390:
1.214 brouard 5391:
1.126 brouard 5392: for(i=1; i<=imx; i++){
5393: for(mi=1; mi<wav[i];mi++){
5394: if (stepm <=0)
1.227 brouard 5395: dh[mi][i]=1;
1.126 brouard 5396: else{
1.260 brouard 5397: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5398: if (agedc[i] < 2*AGESUP) {
5399: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5400: if(j==0) j=1; /* Survives at least one month after exam */
5401: else if(j<0){
5402: nberr++;
5403: 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]);
5404: j=1; /* Temporary Dangerous patch */
5405: 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);
5406: 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]);
5407: 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);
5408: }
5409: k=k+1;
5410: if (j >= jmax){
5411: jmax=j;
5412: ijmax=i;
5413: }
5414: if (j <= jmin){
5415: jmin=j;
5416: ijmin=i;
5417: }
5418: sum=sum+j;
5419: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5420: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5421: }
5422: }
5423: else{
5424: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5425: /* 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 5426:
1.227 brouard 5427: k=k+1;
5428: if (j >= jmax) {
5429: jmax=j;
5430: ijmax=i;
5431: }
5432: else if (j <= jmin){
5433: jmin=j;
5434: ijmin=i;
5435: }
5436: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5437: /*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]);*/
5438: if(j<0){
5439: nberr++;
5440: 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]);
5441: 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]);
5442: }
5443: sum=sum+j;
5444: }
5445: jk= j/stepm;
5446: jl= j -jk*stepm;
5447: ju= j -(jk+1)*stepm;
5448: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5449: if(jl==0){
5450: dh[mi][i]=jk;
5451: bh[mi][i]=0;
5452: }else{ /* We want a negative bias in order to only have interpolation ie
5453: * to avoid the price of an extra matrix product in likelihood */
5454: dh[mi][i]=jk+1;
5455: bh[mi][i]=ju;
5456: }
5457: }else{
5458: if(jl <= -ju){
5459: dh[mi][i]=jk;
5460: bh[mi][i]=jl; /* bias is positive if real duration
5461: * is higher than the multiple of stepm and negative otherwise.
5462: */
5463: }
5464: else{
5465: dh[mi][i]=jk+1;
5466: bh[mi][i]=ju;
5467: }
5468: if(dh[mi][i]==0){
5469: dh[mi][i]=1; /* At least one step */
5470: bh[mi][i]=ju; /* At least one step */
5471: /* 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);*/
5472: }
5473: } /* end if mle */
1.126 brouard 5474: }
5475: } /* end wave */
5476: }
5477: jmean=sum/k;
5478: 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 5479: 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 5480: }
1.126 brouard 5481:
5482: /*********** Tricode ****************************/
1.220 brouard 5483: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5484: {
5485: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5486: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5487: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5488: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5489: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5490: */
1.130 brouard 5491:
1.242 brouard 5492: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5493: int modmaxcovj=0; /* Modality max of covariates j */
5494: int cptcode=0; /* Modality max of covariates j */
5495: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5496:
5497:
1.242 brouard 5498: /* cptcoveff=0; */
5499: /* *cptcov=0; */
1.126 brouard 5500:
1.242 brouard 5501: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5502: for (k=1; k <= maxncov; k++)
5503: for(j=1; j<=2; j++)
5504: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5505:
1.242 brouard 5506: /* Loop on covariates without age and products and no quantitative variable */
5507: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5508: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5509: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5510: switch(Fixed[k]) {
5511: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5512: modmaxcovj=0;
5513: modmincovj=0;
1.242 brouard 5514: 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*/
5515: ij=(int)(covar[Tvar[k]][i]);
5516: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5517: * If product of Vn*Vm, still boolean *:
5518: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5519: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5520: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5521: modality of the nth covariate of individual i. */
5522: if (ij > modmaxcovj)
5523: modmaxcovj=ij;
5524: else if (ij < modmincovj)
5525: modmincovj=ij;
1.287 brouard 5526: if (ij <0 || ij >1 ){
1.311 brouard 5527: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5528: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5529: fflush(ficlog);
5530: exit(1);
1.287 brouard 5531: }
5532: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5533: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5534: exit(1);
5535: }else
5536: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5537: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5538: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5539: /* getting the maximum value of the modality of the covariate
5540: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5541: female ies 1, then modmaxcovj=1.
5542: */
5543: } /* end for loop on individuals i */
5544: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5545: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5546: cptcode=modmaxcovj;
5547: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5548: /*for (i=0; i<=cptcode; i++) {*/
5549: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5550: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5551: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5552: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5553: if( j != -1){
5554: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5555: covariate for which somebody answered excluding
5556: undefined. Usually 2: 0 and 1. */
5557: }
5558: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5559: covariate for which somebody answered including
5560: undefined. Usually 3: -1, 0 and 1. */
5561: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5562: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5563: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5564:
1.242 brouard 5565: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5566: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5567: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5568: /* modmincovj=3; modmaxcovj = 7; */
5569: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5570: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5571: /* defining two dummy variables: variables V1_1 and V1_2.*/
5572: /* nbcode[Tvar[j]][ij]=k; */
5573: /* nbcode[Tvar[j]][1]=0; */
5574: /* nbcode[Tvar[j]][2]=1; */
5575: /* nbcode[Tvar[j]][3]=2; */
5576: /* To be continued (not working yet). */
5577: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5578:
5579: /* 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*/
5580: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5581: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5582: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5583: /*, could be restored in the future */
5584: 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 5585: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5586: break;
5587: }
5588: ij++;
1.287 brouard 5589: 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 5590: cptcode = ij; /* New max modality for covar j */
5591: } /* end of loop on modality i=-1 to 1 or more */
5592: break;
5593: case 1: /* Testing on varying covariate, could be simple and
5594: * should look at waves or product of fixed *
5595: * varying. No time to test -1, assuming 0 and 1 only */
5596: ij=0;
5597: for(i=0; i<=1;i++){
5598: nbcode[Tvar[k]][++ij]=i;
5599: }
5600: break;
5601: default:
5602: break;
5603: } /* end switch */
5604: } /* end dummy test */
1.311 brouard 5605: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5606: 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*/
5607: if(isnan(covar[Tvar[k]][i])){
5608: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5609: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5610: fflush(ficlog);
5611: exit(1);
5612: }
5613: }
5614: }
1.287 brouard 5615: } /* 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 5616:
5617: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5618: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5619: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5620: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5621: 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 */
5622: 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 */
5623: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5624: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5625:
5626: ij=0;
5627: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5628: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5629: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5630: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5631: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5632: /* If product not in single variable we don't print results */
5633: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5634: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5635: 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*/
5636: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5637: 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 */
5638: if(Fixed[k]!=0)
5639: anyvaryingduminmodel=1;
5640: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5641: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5642: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5643: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5644: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5645: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5646: }
5647: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5648: /* ij--; */
5649: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5650: *cptcov=ij; /*Number of total real effective covariates: effective
5651: * because they can be excluded from the model and real
5652: * if in the model but excluded because missing values, but how to get k from ij?*/
5653: for(j=ij+1; j<= cptcovt; j++){
5654: Tvaraff[j]=0;
5655: Tmodelind[j]=0;
5656: }
5657: for(j=ntveff+1; j<= cptcovt; j++){
5658: TmodelInvind[j]=0;
5659: }
5660: /* To be sorted */
5661: ;
5662: }
1.126 brouard 5663:
1.145 brouard 5664:
1.126 brouard 5665: /*********** Health Expectancies ****************/
5666:
1.235 brouard 5667: 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 5668:
5669: {
5670: /* Health expectancies, no variances */
1.164 brouard 5671: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5672: int nhstepma, nstepma; /* Decreasing with age */
5673: double age, agelim, hf;
5674: double ***p3mat;
5675: double eip;
5676:
1.238 brouard 5677: /* pstamp(ficreseij); */
1.126 brouard 5678: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5679: fprintf(ficreseij,"# Age");
5680: for(i=1; i<=nlstate;i++){
5681: for(j=1; j<=nlstate;j++){
5682: fprintf(ficreseij," e%1d%1d ",i,j);
5683: }
5684: fprintf(ficreseij," e%1d. ",i);
5685: }
5686: fprintf(ficreseij,"\n");
5687:
5688:
5689: if(estepm < stepm){
5690: printf ("Problem %d lower than %d\n",estepm, stepm);
5691: }
5692: else hstepm=estepm;
5693: /* We compute the life expectancy from trapezoids spaced every estepm months
5694: * This is mainly to measure the difference between two models: for example
5695: * if stepm=24 months pijx are given only every 2 years and by summing them
5696: * we are calculating an estimate of the Life Expectancy assuming a linear
5697: * progression in between and thus overestimating or underestimating according
5698: * to the curvature of the survival function. If, for the same date, we
5699: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5700: * to compare the new estimate of Life expectancy with the same linear
5701: * hypothesis. A more precise result, taking into account a more precise
5702: * curvature will be obtained if estepm is as small as stepm. */
5703:
5704: /* For example we decided to compute the life expectancy with the smallest unit */
5705: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5706: nhstepm is the number of hstepm from age to agelim
5707: nstepm is the number of stepm from age to agelin.
1.270 brouard 5708: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5709: and note for a fixed period like estepm months */
5710: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5711: survival function given by stepm (the optimization length). Unfortunately it
5712: means that if the survival funtion is printed only each two years of age and if
5713: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5714: results. So we changed our mind and took the option of the best precision.
5715: */
5716: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5717:
5718: agelim=AGESUP;
5719: /* If stepm=6 months */
5720: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5721: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5722:
5723: /* nhstepm age range expressed in number of stepm */
5724: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5725: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5726: /* if (stepm >= YEARM) hstepm=1;*/
5727: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5728: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5729:
5730: for (age=bage; age<=fage; age ++){
5731: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5732: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5733: /* if (stepm >= YEARM) hstepm=1;*/
5734: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5735:
5736: /* If stepm=6 months */
5737: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5738: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5739:
1.235 brouard 5740: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5741:
5742: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5743:
5744: printf("%d|",(int)age);fflush(stdout);
5745: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5746:
5747: /* Computing expectancies */
5748: for(i=1; i<=nlstate;i++)
5749: for(j=1; j<=nlstate;j++)
5750: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5751: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5752:
5753: /* 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]);*/
5754:
5755: }
5756:
5757: fprintf(ficreseij,"%3.0f",age );
5758: for(i=1; i<=nlstate;i++){
5759: eip=0;
5760: for(j=1; j<=nlstate;j++){
5761: eip +=eij[i][j][(int)age];
5762: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5763: }
5764: fprintf(ficreseij,"%9.4f", eip );
5765: }
5766: fprintf(ficreseij,"\n");
5767:
5768: }
5769: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5770: printf("\n");
5771: fprintf(ficlog,"\n");
5772:
5773: }
5774:
1.235 brouard 5775: 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 5776:
5777: {
5778: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5779: to initial status i, ei. .
1.126 brouard 5780: */
5781: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5782: int nhstepma, nstepma; /* Decreasing with age */
5783: double age, agelim, hf;
5784: double ***p3matp, ***p3matm, ***varhe;
5785: double **dnewm,**doldm;
5786: double *xp, *xm;
5787: double **gp, **gm;
5788: double ***gradg, ***trgradg;
5789: int theta;
5790:
5791: double eip, vip;
5792:
5793: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5794: xp=vector(1,npar);
5795: xm=vector(1,npar);
5796: dnewm=matrix(1,nlstate*nlstate,1,npar);
5797: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5798:
5799: pstamp(ficresstdeij);
5800: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5801: fprintf(ficresstdeij,"# Age");
5802: for(i=1; i<=nlstate;i++){
5803: for(j=1; j<=nlstate;j++)
5804: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5805: fprintf(ficresstdeij," e%1d. ",i);
5806: }
5807: fprintf(ficresstdeij,"\n");
5808:
5809: pstamp(ficrescveij);
5810: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5811: fprintf(ficrescveij,"# Age");
5812: for(i=1; i<=nlstate;i++)
5813: for(j=1; j<=nlstate;j++){
5814: cptj= (j-1)*nlstate+i;
5815: for(i2=1; i2<=nlstate;i2++)
5816: for(j2=1; j2<=nlstate;j2++){
5817: cptj2= (j2-1)*nlstate+i2;
5818: if(cptj2 <= cptj)
5819: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5820: }
5821: }
5822: fprintf(ficrescveij,"\n");
5823:
5824: if(estepm < stepm){
5825: printf ("Problem %d lower than %d\n",estepm, stepm);
5826: }
5827: else hstepm=estepm;
5828: /* We compute the life expectancy from trapezoids spaced every estepm months
5829: * This is mainly to measure the difference between two models: for example
5830: * if stepm=24 months pijx are given only every 2 years and by summing them
5831: * we are calculating an estimate of the Life Expectancy assuming a linear
5832: * progression in between and thus overestimating or underestimating according
5833: * to the curvature of the survival function. If, for the same date, we
5834: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5835: * to compare the new estimate of Life expectancy with the same linear
5836: * hypothesis. A more precise result, taking into account a more precise
5837: * curvature will be obtained if estepm is as small as stepm. */
5838:
5839: /* For example we decided to compute the life expectancy with the smallest unit */
5840: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5841: nhstepm is the number of hstepm from age to agelim
5842: nstepm is the number of stepm from age to agelin.
5843: Look at hpijx to understand the reason of that which relies in memory size
5844: and note for a fixed period like estepm months */
5845: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5846: survival function given by stepm (the optimization length). Unfortunately it
5847: means that if the survival funtion is printed only each two years of age and if
5848: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5849: results. So we changed our mind and took the option of the best precision.
5850: */
5851: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5852:
5853: /* If stepm=6 months */
5854: /* nhstepm age range expressed in number of stepm */
5855: agelim=AGESUP;
5856: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5857: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5858: /* if (stepm >= YEARM) hstepm=1;*/
5859: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5860:
5861: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5862: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5863: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5864: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5865: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5866: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5867:
5868: for (age=bage; age<=fage; age ++){
5869: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5870: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5871: /* if (stepm >= YEARM) hstepm=1;*/
5872: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5873:
1.126 brouard 5874: /* If stepm=6 months */
5875: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5876: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5877:
5878: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5879:
1.126 brouard 5880: /* Computing Variances of health expectancies */
5881: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5882: decrease memory allocation */
5883: for(theta=1; theta <=npar; theta++){
5884: for(i=1; i<=npar; i++){
1.222 brouard 5885: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5886: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5887: }
1.235 brouard 5888: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5889: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5890:
1.126 brouard 5891: for(j=1; j<= nlstate; j++){
1.222 brouard 5892: for(i=1; i<=nlstate; i++){
5893: for(h=0; h<=nhstepm-1; h++){
5894: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5895: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5896: }
5897: }
1.126 brouard 5898: }
1.218 brouard 5899:
1.126 brouard 5900: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5901: for(h=0; h<=nhstepm-1; h++){
5902: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5903: }
1.126 brouard 5904: }/* End theta */
5905:
5906:
5907: for(h=0; h<=nhstepm-1; h++)
5908: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5909: for(theta=1; theta <=npar; theta++)
5910: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5911:
1.218 brouard 5912:
1.222 brouard 5913: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5914: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5915: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5916:
1.222 brouard 5917: printf("%d|",(int)age);fflush(stdout);
5918: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5919: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5920: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5921: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5922: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5923: for(ij=1;ij<=nlstate*nlstate;ij++)
5924: for(ji=1;ji<=nlstate*nlstate;ji++)
5925: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5926: }
5927: }
1.218 brouard 5928:
1.126 brouard 5929: /* Computing expectancies */
1.235 brouard 5930: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5931: for(i=1; i<=nlstate;i++)
5932: for(j=1; j<=nlstate;j++)
1.222 brouard 5933: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5934: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5935:
1.222 brouard 5936: /* 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 5937:
1.222 brouard 5938: }
1.269 brouard 5939:
5940: /* Standard deviation of expectancies ij */
1.126 brouard 5941: fprintf(ficresstdeij,"%3.0f",age );
5942: for(i=1; i<=nlstate;i++){
5943: eip=0.;
5944: vip=0.;
5945: for(j=1; j<=nlstate;j++){
1.222 brouard 5946: eip += eij[i][j][(int)age];
5947: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5948: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5949: 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 5950: }
5951: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5952: }
5953: fprintf(ficresstdeij,"\n");
1.218 brouard 5954:
1.269 brouard 5955: /* Variance of expectancies ij */
1.126 brouard 5956: fprintf(ficrescveij,"%3.0f",age );
5957: for(i=1; i<=nlstate;i++)
5958: for(j=1; j<=nlstate;j++){
1.222 brouard 5959: cptj= (j-1)*nlstate+i;
5960: for(i2=1; i2<=nlstate;i2++)
5961: for(j2=1; j2<=nlstate;j2++){
5962: cptj2= (j2-1)*nlstate+i2;
5963: if(cptj2 <= cptj)
5964: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5965: }
1.126 brouard 5966: }
5967: fprintf(ficrescveij,"\n");
1.218 brouard 5968:
1.126 brouard 5969: }
5970: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5971: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5972: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5973: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5974: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5975: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5976: printf("\n");
5977: fprintf(ficlog,"\n");
1.218 brouard 5978:
1.126 brouard 5979: free_vector(xm,1,npar);
5980: free_vector(xp,1,npar);
5981: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5982: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5983: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5984: }
1.218 brouard 5985:
1.126 brouard 5986: /************ Variance ******************/
1.235 brouard 5987: 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 5988: {
1.279 brouard 5989: /** Variance of health expectancies
5990: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5991: * double **newm;
5992: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
5993: */
1.218 brouard 5994:
5995: /* int movingaverage(); */
5996: double **dnewm,**doldm;
5997: double **dnewmp,**doldmp;
5998: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 5999: int first=0;
1.218 brouard 6000: int k;
6001: double *xp;
1.279 brouard 6002: double **gp, **gm; /**< for var eij */
6003: double ***gradg, ***trgradg; /**< for var eij */
6004: double **gradgp, **trgradgp; /**< for var p point j */
6005: double *gpp, *gmp; /**< for var p point j */
6006: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6007: double ***p3mat;
6008: double age,agelim, hf;
6009: /* double ***mobaverage; */
6010: int theta;
6011: char digit[4];
6012: char digitp[25];
6013:
6014: char fileresprobmorprev[FILENAMELENGTH];
6015:
6016: if(popbased==1){
6017: if(mobilav!=0)
6018: strcpy(digitp,"-POPULBASED-MOBILAV_");
6019: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6020: }
6021: else
6022: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6023:
1.218 brouard 6024: /* if (mobilav!=0) { */
6025: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6026: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6027: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6028: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6029: /* } */
6030: /* } */
6031:
6032: strcpy(fileresprobmorprev,"PRMORPREV-");
6033: sprintf(digit,"%-d",ij);
6034: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6035: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6036: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6037: strcat(fileresprobmorprev,fileresu);
6038: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6039: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6040: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6041: }
6042: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6043: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6044: pstamp(ficresprobmorprev);
6045: 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 6046: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6047: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6048: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6049: }
6050: for(j=1;j<=cptcoveff;j++)
6051: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6052: fprintf(ficresprobmorprev,"\n");
6053:
1.218 brouard 6054: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6055: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6056: fprintf(ficresprobmorprev," p.%-d SE",j);
6057: for(i=1; i<=nlstate;i++)
6058: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6059: }
6060: fprintf(ficresprobmorprev,"\n");
6061:
6062: fprintf(ficgp,"\n# Routine varevsij");
6063: fprintf(ficgp,"\nunset title \n");
6064: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6065: 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");
6066: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6067:
1.218 brouard 6068: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6069: pstamp(ficresvij);
6070: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6071: if(popbased==1)
6072: 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);
6073: else
6074: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6075: fprintf(ficresvij,"# Age");
6076: for(i=1; i<=nlstate;i++)
6077: for(j=1; j<=nlstate;j++)
6078: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6079: fprintf(ficresvij,"\n");
6080:
6081: xp=vector(1,npar);
6082: dnewm=matrix(1,nlstate,1,npar);
6083: doldm=matrix(1,nlstate,1,nlstate);
6084: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6085: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6086:
6087: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6088: gpp=vector(nlstate+1,nlstate+ndeath);
6089: gmp=vector(nlstate+1,nlstate+ndeath);
6090: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6091:
1.218 brouard 6092: if(estepm < stepm){
6093: printf ("Problem %d lower than %d\n",estepm, stepm);
6094: }
6095: else hstepm=estepm;
6096: /* For example we decided to compute the life expectancy with the smallest unit */
6097: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6098: nhstepm is the number of hstepm from age to agelim
6099: nstepm is the number of stepm from age to agelim.
6100: Look at function hpijx to understand why because of memory size limitations,
6101: we decided (b) to get a life expectancy respecting the most precise curvature of the
6102: survival function given by stepm (the optimization length). Unfortunately it
6103: means that if the survival funtion is printed every two years of age and if
6104: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6105: results. So we changed our mind and took the option of the best precision.
6106: */
6107: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6108: agelim = AGESUP;
6109: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6110: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6111: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6112: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6113: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6114: gp=matrix(0,nhstepm,1,nlstate);
6115: gm=matrix(0,nhstepm,1,nlstate);
6116:
6117:
6118: for(theta=1; theta <=npar; theta++){
6119: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6120: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6121: }
1.279 brouard 6122: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6123: * returns into prlim .
1.288 brouard 6124: */
1.242 brouard 6125: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6126:
6127: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6128: if (popbased==1) {
6129: if(mobilav ==0){
6130: for(i=1; i<=nlstate;i++)
6131: prlim[i][i]=probs[(int)age][i][ij];
6132: }else{ /* mobilav */
6133: for(i=1; i<=nlstate;i++)
6134: prlim[i][i]=mobaverage[(int)age][i][ij];
6135: }
6136: }
1.295 brouard 6137: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6138: */
6139: 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 6140: /**< 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 6141: * at horizon h in state j including mortality.
6142: */
1.218 brouard 6143: for(j=1; j<= nlstate; j++){
6144: for(h=0; h<=nhstepm; h++){
6145: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6146: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6147: }
6148: }
1.279 brouard 6149: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6150: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6151: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6152: */
6153: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6154: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6155: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6156: }
6157:
6158: /* Again with minus shift */
1.218 brouard 6159:
6160: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6161: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6162:
1.242 brouard 6163: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6164:
6165: if (popbased==1) {
6166: if(mobilav ==0){
6167: for(i=1; i<=nlstate;i++)
6168: prlim[i][i]=probs[(int)age][i][ij];
6169: }else{ /* mobilav */
6170: for(i=1; i<=nlstate;i++)
6171: prlim[i][i]=mobaverage[(int)age][i][ij];
6172: }
6173: }
6174:
1.235 brouard 6175: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6176:
6177: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6178: for(h=0; h<=nhstepm; h++){
6179: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6180: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6181: }
6182: }
6183: /* This for computing probability of death (h=1 means
6184: computed over hstepm matrices product = hstepm*stepm months)
6185: as a weighted average of prlim.
6186: */
6187: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6188: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6189: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6190: }
1.279 brouard 6191: /* end shifting computations */
6192:
6193: /**< Computing gradient matrix at horizon h
6194: */
1.218 brouard 6195: for(j=1; j<= nlstate; j++) /* vareij */
6196: for(h=0; h<=nhstepm; h++){
6197: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6198: }
1.279 brouard 6199: /**< Gradient of overall mortality p.3 (or p.j)
6200: */
6201: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6202: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6203: }
6204:
6205: } /* End theta */
1.279 brouard 6206:
6207: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6208: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6209:
6210: for(h=0; h<=nhstepm; h++) /* veij */
6211: for(j=1; j<=nlstate;j++)
6212: for(theta=1; theta <=npar; theta++)
6213: trgradg[h][j][theta]=gradg[h][theta][j];
6214:
6215: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6216: for(theta=1; theta <=npar; theta++)
6217: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6218: /**< as well as its transposed matrix
6219: */
1.218 brouard 6220:
6221: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6222: for(i=1;i<=nlstate;i++)
6223: for(j=1;j<=nlstate;j++)
6224: vareij[i][j][(int)age] =0.;
1.279 brouard 6225:
6226: /* Computing trgradg by matcov by gradg at age and summing over h
6227: * and k (nhstepm) formula 15 of article
6228: * Lievre-Brouard-Heathcote
6229: */
6230:
1.218 brouard 6231: for(h=0;h<=nhstepm;h++){
6232: for(k=0;k<=nhstepm;k++){
6233: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6234: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6235: for(i=1;i<=nlstate;i++)
6236: for(j=1;j<=nlstate;j++)
6237: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6238: }
6239: }
6240:
1.279 brouard 6241: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6242: * p.j overall mortality formula 49 but computed directly because
6243: * we compute the grad (wix pijx) instead of grad (pijx),even if
6244: * wix is independent of theta.
6245: */
1.218 brouard 6246: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6247: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6248: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6249: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6250: varppt[j][i]=doldmp[j][i];
6251: /* end ppptj */
6252: /* x centered again */
6253:
1.242 brouard 6254: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6255:
6256: if (popbased==1) {
6257: if(mobilav ==0){
6258: for(i=1; i<=nlstate;i++)
6259: prlim[i][i]=probs[(int)age][i][ij];
6260: }else{ /* mobilav */
6261: for(i=1; i<=nlstate;i++)
6262: prlim[i][i]=mobaverage[(int)age][i][ij];
6263: }
6264: }
6265:
6266: /* This for computing probability of death (h=1 means
6267: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6268: as a weighted average of prlim.
6269: */
1.235 brouard 6270: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6271: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6272: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6273: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6274: }
6275: /* end probability of death */
6276:
6277: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6278: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6279: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6280: for(i=1; i<=nlstate;i++){
6281: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6282: }
6283: }
6284: fprintf(ficresprobmorprev,"\n");
6285:
6286: fprintf(ficresvij,"%.0f ",age );
6287: for(i=1; i<=nlstate;i++)
6288: for(j=1; j<=nlstate;j++){
6289: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6290: }
6291: fprintf(ficresvij,"\n");
6292: free_matrix(gp,0,nhstepm,1,nlstate);
6293: free_matrix(gm,0,nhstepm,1,nlstate);
6294: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6295: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6296: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6297: } /* End age */
6298: free_vector(gpp,nlstate+1,nlstate+ndeath);
6299: free_vector(gmp,nlstate+1,nlstate+ndeath);
6300: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6301: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6302: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6303: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6304: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6305: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6306: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6307: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6308: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6309: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6310: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6311: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6312: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6313: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6314: 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);
6315: /* 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 6316: */
1.218 brouard 6317: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6318: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6319:
1.218 brouard 6320: free_vector(xp,1,npar);
6321: free_matrix(doldm,1,nlstate,1,nlstate);
6322: free_matrix(dnewm,1,nlstate,1,npar);
6323: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6324: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6325: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6326: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6327: fclose(ficresprobmorprev);
6328: fflush(ficgp);
6329: fflush(fichtm);
6330: } /* end varevsij */
1.126 brouard 6331:
6332: /************ Variance of prevlim ******************/
1.269 brouard 6333: 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 6334: {
1.205 brouard 6335: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6336: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6337:
1.268 brouard 6338: double **dnewmpar,**doldm;
1.126 brouard 6339: int i, j, nhstepm, hstepm;
6340: double *xp;
6341: double *gp, *gm;
6342: double **gradg, **trgradg;
1.208 brouard 6343: double **mgm, **mgp;
1.126 brouard 6344: double age,agelim;
6345: int theta;
6346:
6347: pstamp(ficresvpl);
1.288 brouard 6348: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6349: fprintf(ficresvpl,"# Age ");
6350: if(nresult >=1)
6351: fprintf(ficresvpl," Result# ");
1.126 brouard 6352: for(i=1; i<=nlstate;i++)
6353: fprintf(ficresvpl," %1d-%1d",i,i);
6354: fprintf(ficresvpl,"\n");
6355:
6356: xp=vector(1,npar);
1.268 brouard 6357: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6358: doldm=matrix(1,nlstate,1,nlstate);
6359:
6360: hstepm=1*YEARM; /* Every year of age */
6361: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6362: agelim = AGESUP;
6363: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6364: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6365: if (stepm >= YEARM) hstepm=1;
6366: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6367: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6368: mgp=matrix(1,npar,1,nlstate);
6369: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6370: gp=vector(1,nlstate);
6371: gm=vector(1,nlstate);
6372:
6373: for(theta=1; theta <=npar; theta++){
6374: for(i=1; i<=npar; i++){ /* Computes gradient */
6375: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6376: }
1.288 brouard 6377: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6378: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6379: /* else */
6380: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6381: for(i=1;i<=nlstate;i++){
1.126 brouard 6382: gp[i] = prlim[i][i];
1.208 brouard 6383: mgp[theta][i] = prlim[i][i];
6384: }
1.126 brouard 6385: for(i=1; i<=npar; i++) /* Computes gradient */
6386: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6387: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6388: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6389: /* else */
6390: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6391: for(i=1;i<=nlstate;i++){
1.126 brouard 6392: gm[i] = prlim[i][i];
1.208 brouard 6393: mgm[theta][i] = prlim[i][i];
6394: }
1.126 brouard 6395: for(i=1;i<=nlstate;i++)
6396: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6397: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6398: } /* End theta */
6399:
6400: trgradg =matrix(1,nlstate,1,npar);
6401:
6402: for(j=1; j<=nlstate;j++)
6403: for(theta=1; theta <=npar; theta++)
6404: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6405: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6406: /* printf("\nmgm mgp %d ",(int)age); */
6407: /* for(j=1; j<=nlstate;j++){ */
6408: /* printf(" %d ",j); */
6409: /* for(theta=1; theta <=npar; theta++) */
6410: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6411: /* printf("\n "); */
6412: /* } */
6413: /* } */
6414: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6415: /* printf("\n gradg %d ",(int)age); */
6416: /* for(j=1; j<=nlstate;j++){ */
6417: /* printf("%d ",j); */
6418: /* for(theta=1; theta <=npar; theta++) */
6419: /* printf("%d %lf ",theta,gradg[theta][j]); */
6420: /* printf("\n "); */
6421: /* } */
6422: /* } */
1.126 brouard 6423:
6424: for(i=1;i<=nlstate;i++)
6425: varpl[i][(int)age] =0.;
1.209 brouard 6426: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6427: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6428: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6429: }else{
1.268 brouard 6430: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6431: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6432: }
1.126 brouard 6433: for(i=1;i<=nlstate;i++)
6434: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6435:
6436: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6437: if(nresult >=1)
6438: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6439: for(i=1; i<=nlstate;i++){
1.126 brouard 6440: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6441: /* for(j=1;j<=nlstate;j++) */
6442: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6443: }
1.126 brouard 6444: fprintf(ficresvpl,"\n");
6445: free_vector(gp,1,nlstate);
6446: free_vector(gm,1,nlstate);
1.208 brouard 6447: free_matrix(mgm,1,npar,1,nlstate);
6448: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6449: free_matrix(gradg,1,npar,1,nlstate);
6450: free_matrix(trgradg,1,nlstate,1,npar);
6451: } /* End age */
6452:
6453: free_vector(xp,1,npar);
6454: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6455: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6456:
6457: }
6458:
6459:
6460: /************ Variance of backprevalence limit ******************/
1.269 brouard 6461: 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 6462: {
6463: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6464: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6465:
6466: double **dnewmpar,**doldm;
6467: int i, j, nhstepm, hstepm;
6468: double *xp;
6469: double *gp, *gm;
6470: double **gradg, **trgradg;
6471: double **mgm, **mgp;
6472: double age,agelim;
6473: int theta;
6474:
6475: pstamp(ficresvbl);
6476: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6477: fprintf(ficresvbl,"# Age ");
6478: if(nresult >=1)
6479: fprintf(ficresvbl," Result# ");
6480: for(i=1; i<=nlstate;i++)
6481: fprintf(ficresvbl," %1d-%1d",i,i);
6482: fprintf(ficresvbl,"\n");
6483:
6484: xp=vector(1,npar);
6485: dnewmpar=matrix(1,nlstate,1,npar);
6486: doldm=matrix(1,nlstate,1,nlstate);
6487:
6488: hstepm=1*YEARM; /* Every year of age */
6489: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6490: agelim = AGEINF;
6491: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6492: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6493: if (stepm >= YEARM) hstepm=1;
6494: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6495: gradg=matrix(1,npar,1,nlstate);
6496: mgp=matrix(1,npar,1,nlstate);
6497: mgm=matrix(1,npar,1,nlstate);
6498: gp=vector(1,nlstate);
6499: gm=vector(1,nlstate);
6500:
6501: for(theta=1; theta <=npar; theta++){
6502: for(i=1; i<=npar; i++){ /* Computes gradient */
6503: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6504: }
6505: if(mobilavproj > 0 )
6506: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6507: else
6508: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6509: for(i=1;i<=nlstate;i++){
6510: gp[i] = bprlim[i][i];
6511: mgp[theta][i] = bprlim[i][i];
6512: }
6513: for(i=1; i<=npar; i++) /* Computes gradient */
6514: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6515: if(mobilavproj > 0 )
6516: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6517: else
6518: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6519: for(i=1;i<=nlstate;i++){
6520: gm[i] = bprlim[i][i];
6521: mgm[theta][i] = bprlim[i][i];
6522: }
6523: for(i=1;i<=nlstate;i++)
6524: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6525: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6526: } /* End theta */
6527:
6528: trgradg =matrix(1,nlstate,1,npar);
6529:
6530: for(j=1; j<=nlstate;j++)
6531: for(theta=1; theta <=npar; theta++)
6532: trgradg[j][theta]=gradg[theta][j];
6533: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6534: /* printf("\nmgm mgp %d ",(int)age); */
6535: /* for(j=1; j<=nlstate;j++){ */
6536: /* printf(" %d ",j); */
6537: /* for(theta=1; theta <=npar; theta++) */
6538: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6539: /* printf("\n "); */
6540: /* } */
6541: /* } */
6542: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6543: /* printf("\n gradg %d ",(int)age); */
6544: /* for(j=1; j<=nlstate;j++){ */
6545: /* printf("%d ",j); */
6546: /* for(theta=1; theta <=npar; theta++) */
6547: /* printf("%d %lf ",theta,gradg[theta][j]); */
6548: /* printf("\n "); */
6549: /* } */
6550: /* } */
6551:
6552: for(i=1;i<=nlstate;i++)
6553: varbpl[i][(int)age] =0.;
6554: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6555: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6556: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6557: }else{
6558: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6559: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6560: }
6561: for(i=1;i<=nlstate;i++)
6562: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6563:
6564: fprintf(ficresvbl,"%.0f ",age );
6565: if(nresult >=1)
6566: fprintf(ficresvbl,"%d ",nres );
6567: for(i=1; i<=nlstate;i++)
6568: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6569: fprintf(ficresvbl,"\n");
6570: free_vector(gp,1,nlstate);
6571: free_vector(gm,1,nlstate);
6572: free_matrix(mgm,1,npar,1,nlstate);
6573: free_matrix(mgp,1,npar,1,nlstate);
6574: free_matrix(gradg,1,npar,1,nlstate);
6575: free_matrix(trgradg,1,nlstate,1,npar);
6576: } /* End age */
6577:
6578: free_vector(xp,1,npar);
6579: free_matrix(doldm,1,nlstate,1,npar);
6580: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6581:
6582: }
6583:
6584: /************ Variance of one-step probabilities ******************/
6585: 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 6586: {
6587: int i, j=0, k1, l1, tj;
6588: int k2, l2, j1, z1;
6589: int k=0, l;
6590: int first=1, first1, first2;
6591: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6592: double **dnewm,**doldm;
6593: double *xp;
6594: double *gp, *gm;
6595: double **gradg, **trgradg;
6596: double **mu;
6597: double age, cov[NCOVMAX+1];
6598: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6599: int theta;
6600: char fileresprob[FILENAMELENGTH];
6601: char fileresprobcov[FILENAMELENGTH];
6602: char fileresprobcor[FILENAMELENGTH];
6603: double ***varpij;
6604:
6605: strcpy(fileresprob,"PROB_");
6606: strcat(fileresprob,fileres);
6607: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6608: printf("Problem with resultfile: %s\n", fileresprob);
6609: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6610: }
6611: strcpy(fileresprobcov,"PROBCOV_");
6612: strcat(fileresprobcov,fileresu);
6613: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6614: printf("Problem with resultfile: %s\n", fileresprobcov);
6615: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6616: }
6617: strcpy(fileresprobcor,"PROBCOR_");
6618: strcat(fileresprobcor,fileresu);
6619: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6620: printf("Problem with resultfile: %s\n", fileresprobcor);
6621: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6622: }
6623: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6624: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6625: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6626: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6627: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6628: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6629: pstamp(ficresprob);
6630: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6631: fprintf(ficresprob,"# Age");
6632: pstamp(ficresprobcov);
6633: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6634: fprintf(ficresprobcov,"# Age");
6635: pstamp(ficresprobcor);
6636: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6637: fprintf(ficresprobcor,"# Age");
1.126 brouard 6638:
6639:
1.222 brouard 6640: for(i=1; i<=nlstate;i++)
6641: for(j=1; j<=(nlstate+ndeath);j++){
6642: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6643: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6644: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6645: }
6646: /* fprintf(ficresprob,"\n");
6647: fprintf(ficresprobcov,"\n");
6648: fprintf(ficresprobcor,"\n");
6649: */
6650: xp=vector(1,npar);
6651: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6652: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6653: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6654: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6655: first=1;
6656: fprintf(ficgp,"\n# Routine varprob");
6657: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6658: fprintf(fichtm,"\n");
6659:
1.288 brouard 6660: 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 6661: 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);
6662: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6663: and drawn. It helps understanding how is the covariance between two incidences.\
6664: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6665: 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 6666: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6667: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6668: standard deviations wide on each axis. <br>\
6669: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6670: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6671: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6672:
1.222 brouard 6673: cov[1]=1;
6674: /* tj=cptcoveff; */
1.225 brouard 6675: tj = (int) pow(2,cptcoveff);
1.222 brouard 6676: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6677: j1=0;
1.224 brouard 6678: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6679: if (cptcovn>0) {
6680: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6681: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6682: fprintf(ficresprob, "**********\n#\n");
6683: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6684: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6685: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6686:
1.222 brouard 6687: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6688: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6689: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6690:
6691:
1.222 brouard 6692: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6693: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6694: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6695:
1.222 brouard 6696: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6697: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6698: fprintf(ficresprobcor, "**********\n#");
6699: if(invalidvarcomb[j1]){
6700: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6701: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6702: continue;
6703: }
6704: }
6705: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6706: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6707: gp=vector(1,(nlstate)*(nlstate+ndeath));
6708: gm=vector(1,(nlstate)*(nlstate+ndeath));
6709: for (age=bage; age<=fage; age ++){
6710: cov[2]=age;
6711: if(nagesqr==1)
6712: cov[3]= age*age;
6713: for (k=1; k<=cptcovn;k++) {
6714: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6715: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6716: * 1 1 1 1 1
6717: * 2 2 1 1 1
6718: * 3 1 2 1 1
6719: */
6720: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6721: }
6722: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6723: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6724: for (k=1; k<=cptcovprod;k++)
6725: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6726:
6727:
1.222 brouard 6728: for(theta=1; theta <=npar; theta++){
6729: for(i=1; i<=npar; i++)
6730: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6731:
1.222 brouard 6732: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6733:
1.222 brouard 6734: k=0;
6735: for(i=1; i<= (nlstate); i++){
6736: for(j=1; j<=(nlstate+ndeath);j++){
6737: k=k+1;
6738: gp[k]=pmmij[i][j];
6739: }
6740: }
1.220 brouard 6741:
1.222 brouard 6742: for(i=1; i<=npar; i++)
6743: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6744:
1.222 brouard 6745: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6746: k=0;
6747: for(i=1; i<=(nlstate); i++){
6748: for(j=1; j<=(nlstate+ndeath);j++){
6749: k=k+1;
6750: gm[k]=pmmij[i][j];
6751: }
6752: }
1.220 brouard 6753:
1.222 brouard 6754: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6755: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6756: }
1.126 brouard 6757:
1.222 brouard 6758: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6759: for(theta=1; theta <=npar; theta++)
6760: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6761:
1.222 brouard 6762: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6763: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6764:
1.222 brouard 6765: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6766:
1.222 brouard 6767: k=0;
6768: for(i=1; i<=(nlstate); i++){
6769: for(j=1; j<=(nlstate+ndeath);j++){
6770: k=k+1;
6771: mu[k][(int) age]=pmmij[i][j];
6772: }
6773: }
6774: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6775: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6776: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6777:
1.222 brouard 6778: /*printf("\n%d ",(int)age);
6779: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6780: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6781: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6782: }*/
1.220 brouard 6783:
1.222 brouard 6784: fprintf(ficresprob,"\n%d ",(int)age);
6785: fprintf(ficresprobcov,"\n%d ",(int)age);
6786: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6787:
1.222 brouard 6788: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6789: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6790: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6791: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6792: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6793: }
6794: i=0;
6795: for (k=1; k<=(nlstate);k++){
6796: for (l=1; l<=(nlstate+ndeath);l++){
6797: i++;
6798: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6799: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6800: for (j=1; j<=i;j++){
6801: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6802: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6803: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6804: }
6805: }
6806: }/* end of loop for state */
6807: } /* end of loop for age */
6808: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6809: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6810: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6811: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6812:
6813: /* Confidence intervalle of pij */
6814: /*
6815: fprintf(ficgp,"\nunset parametric;unset label");
6816: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6817: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6818: 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);
6819: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6820: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6821: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6822: */
6823:
6824: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6825: first1=1;first2=2;
6826: for (k2=1; k2<=(nlstate);k2++){
6827: for (l2=1; l2<=(nlstate+ndeath);l2++){
6828: if(l2==k2) continue;
6829: j=(k2-1)*(nlstate+ndeath)+l2;
6830: for (k1=1; k1<=(nlstate);k1++){
6831: for (l1=1; l1<=(nlstate+ndeath);l1++){
6832: if(l1==k1) continue;
6833: i=(k1-1)*(nlstate+ndeath)+l1;
6834: if(i<=j) continue;
6835: for (age=bage; age<=fage; age ++){
6836: if ((int)age %5==0){
6837: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6838: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6839: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6840: mu1=mu[i][(int) age]/stepm*YEARM ;
6841: mu2=mu[j][(int) age]/stepm*YEARM;
6842: c12=cv12/sqrt(v1*v2);
6843: /* Computing eigen value of matrix of covariance */
6844: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6845: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6846: if ((lc2 <0) || (lc1 <0) ){
6847: if(first2==1){
6848: first1=0;
6849: 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);
6850: }
6851: 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);
6852: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6853: /* lc2=fabs(lc2); */
6854: }
1.220 brouard 6855:
1.222 brouard 6856: /* Eigen vectors */
1.280 brouard 6857: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6858: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6859: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6860: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6861: }else
6862: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6863: /*v21=sqrt(1.-v11*v11); *//* error */
6864: v21=(lc1-v1)/cv12*v11;
6865: v12=-v21;
6866: v22=v11;
6867: tnalp=v21/v11;
6868: if(first1==1){
6869: first1=0;
6870: 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);
6871: }
6872: 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);
6873: /*printf(fignu*/
6874: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6875: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6876: if(first==1){
6877: first=0;
6878: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6879: fprintf(ficgp,"\nset parametric;unset label");
6880: 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);
6881: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6882: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6883: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6884: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6885: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6886: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6887: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6888: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6889: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6890: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6891: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6892: 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 6893: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6894: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6895: }else{
6896: first=0;
6897: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6898: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6899: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6900: 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 6901: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6902: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6903: }/* if first */
6904: } /* age mod 5 */
6905: } /* end loop age */
6906: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6907: first=1;
6908: } /*l12 */
6909: } /* k12 */
6910: } /*l1 */
6911: }/* k1 */
6912: } /* loop on combination of covariates j1 */
6913: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6914: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6915: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6916: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6917: free_vector(xp,1,npar);
6918: fclose(ficresprob);
6919: fclose(ficresprobcov);
6920: fclose(ficresprobcor);
6921: fflush(ficgp);
6922: fflush(fichtmcov);
6923: }
1.126 brouard 6924:
6925:
6926: /******************* Printing html file ***********/
1.201 brouard 6927: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6928: int lastpass, int stepm, int weightopt, char model[],\
6929: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6930: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6931: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6932: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6933: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6934:
6935: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6936: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6937: </ul>");
1.237 brouard 6938: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6939: </ul>", model);
1.214 brouard 6940: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6941: 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",
6942: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6943: 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 6944: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6945: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6946: fprintf(fichtm,"\
6947: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6948: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6949: fprintf(fichtm,"\
1.217 brouard 6950: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6951: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6952: fprintf(fichtm,"\
1.288 brouard 6953: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6954: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6955: fprintf(fichtm,"\
1.288 brouard 6956: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6957: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6958: fprintf(fichtm,"\
1.211 brouard 6959: - (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 6960: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6961: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6962: if(prevfcast==1){
6963: fprintf(fichtm,"\
6964: - Prevalence projections by age and states: \
1.201 brouard 6965: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6966: }
1.126 brouard 6967:
6968:
1.225 brouard 6969: m=pow(2,cptcoveff);
1.222 brouard 6970: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6971:
1.264 brouard 6972: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6973:
6974: jj1=0;
6975:
6976: fprintf(fichtm," \n<ul>");
6977: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6978: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6979: if(m != 1 && TKresult[nres]!= k1)
6980: continue;
6981: jj1++;
6982: if (cptcovn > 0) {
6983: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6984: for (cpt=1; cpt<=cptcoveff;cpt++){
6985: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6986: }
6987: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6988: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6989: }
6990: fprintf(fichtm,"\">");
6991:
6992: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6993: fprintf(fichtm,"************ Results for covariates");
6994: for (cpt=1; cpt<=cptcoveff;cpt++){
6995: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6996: }
6997: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6998: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6999: }
7000: if(invalidvarcomb[k1]){
7001: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7002: continue;
7003: }
7004: fprintf(fichtm,"</a></li>");
7005: } /* cptcovn >0 */
7006: }
7007: fprintf(fichtm," \n</ul>");
7008:
1.222 brouard 7009: jj1=0;
1.237 brouard 7010:
7011: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7012: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7013: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7014: continue;
1.220 brouard 7015:
1.222 brouard 7016: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7017: jj1++;
7018: if (cptcovn > 0) {
1.264 brouard 7019: fprintf(fichtm,"\n<p><a name=\"rescov");
7020: for (cpt=1; cpt<=cptcoveff;cpt++){
7021: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7022: }
7023: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7024: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7025: }
7026: fprintf(fichtm,"\"</a>");
7027:
1.222 brouard 7028: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7029: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7030: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7031: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7032: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7033: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7034: }
1.237 brouard 7035: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7036: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7037: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7038: }
7039:
1.230 brouard 7040: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 7041: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
7042: if(invalidvarcomb[k1]){
7043: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7044: printf("\nCombination (%d) ignored because no cases \n",k1);
7045: continue;
7046: }
7047: }
7048: /* aij, bij */
1.259 brouard 7049: 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 7050: <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 7051: /* Pij */
1.241 brouard 7052: 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> \
7053: <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 7054: /* Quasi-incidences */
7055: 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 7056: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7057: 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 7058: 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> \
7059: <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 7060: /* Survival functions (period) in state j */
7061: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7062: 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 7063: <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 7064: }
7065: /* State specific survival functions (period) */
7066: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7067: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7068: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7069: <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 7070: }
1.288 brouard 7071: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7072: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7073: 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> \
7074: <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 7075: }
1.296 brouard 7076: if(prevbcast==1){
1.288 brouard 7077: /* Backward prevalence in each health state */
1.222 brouard 7078: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7079: 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 7080: <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 7081: }
1.217 brouard 7082: }
1.222 brouard 7083: if(prevfcast==1){
1.288 brouard 7084: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7085: for(cpt=1; cpt<=nlstate;cpt++){
1.288 brouard 7086: 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 7087: <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 7088: }
7089: }
1.296 brouard 7090: if(prevbcast==1){
1.268 brouard 7091: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7092: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7093: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7094: 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 \
7095: 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) \
7096: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
7097: <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 7098: }
7099: }
1.220 brouard 7100:
1.222 brouard 7101: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 7102: 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> \
7103: <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 7104: }
7105: /* } /\* end i1 *\/ */
7106: }/* End k1 */
7107: fprintf(fichtm,"</ul>");
1.126 brouard 7108:
1.222 brouard 7109: fprintf(fichtm,"\
1.126 brouard 7110: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7111: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7112: - 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 7113: But because parameters are usually highly correlated (a higher incidence of disability \
7114: and a higher incidence of recovery can give very close observed transition) it might \
7115: be very useful to look not only at linear confidence intervals estimated from the \
7116: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7117: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7118: covariance matrix of the one-step probabilities. \
7119: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7120:
1.222 brouard 7121: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7122: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7123: fprintf(fichtm,"\
1.126 brouard 7124: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7125: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7126:
1.222 brouard 7127: fprintf(fichtm,"\
1.126 brouard 7128: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7129: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7130: fprintf(fichtm,"\
1.126 brouard 7131: - 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): \
7132: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7133: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7134: fprintf(fichtm,"\
1.126 brouard 7135: - (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): \
7136: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7137: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7138: fprintf(fichtm,"\
1.288 brouard 7139: - 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 7140: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7141: fprintf(fichtm,"\
1.128 brouard 7142: - 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 7143: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7144: fprintf(fichtm,"\
1.288 brouard 7145: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7146: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7147:
7148: /* if(popforecast==1) fprintf(fichtm,"\n */
7149: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7150: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7151: /* <br>",fileres,fileres,fileres,fileres); */
7152: /* else */
7153: /* 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 7154: fflush(fichtm);
7155: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7156:
1.225 brouard 7157: m=pow(2,cptcoveff);
1.222 brouard 7158: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7159:
1.222 brouard 7160: jj1=0;
1.237 brouard 7161:
1.241 brouard 7162: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7163: for(k1=1; k1<=m;k1++){
1.253 brouard 7164: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7165: continue;
1.222 brouard 7166: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7167: jj1++;
1.126 brouard 7168: if (cptcovn > 0) {
7169: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7170: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7171: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7172: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7173: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7174: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7175: }
7176:
1.126 brouard 7177: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7178:
1.222 brouard 7179: if(invalidvarcomb[k1]){
7180: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7181: continue;
7182: }
1.126 brouard 7183: }
7184: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7185: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 7186: 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 7187: <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 7188: }
7189: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 7190: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
7191: true period expectancies (those weighted with period prevalences are also\
7192: drawn in addition to the population based expectancies computed using\
1.241 brouard 7193: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
7194: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7195: /* } /\* end i1 *\/ */
7196: }/* End k1 */
1.241 brouard 7197: }/* End nres */
1.222 brouard 7198: fprintf(fichtm,"</ul>");
7199: fflush(fichtm);
1.126 brouard 7200: }
7201:
7202: /******************* Gnuplot file **************/
1.296 brouard 7203: 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 7204:
7205: char dirfileres[132],optfileres[132];
1.264 brouard 7206: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7207: 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 7208: int lv=0, vlv=0, kl=0;
1.130 brouard 7209: int ng=0;
1.201 brouard 7210: int vpopbased;
1.223 brouard 7211: int ioffset; /* variable offset for columns */
1.270 brouard 7212: int iyearc=1; /* variable column for year of projection */
7213: int iagec=1; /* variable column for age of projection */
1.235 brouard 7214: int nres=0; /* Index of resultline */
1.266 brouard 7215: int istart=1; /* For starting graphs in projections */
1.219 brouard 7216:
1.126 brouard 7217: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7218: /* printf("Problem with file %s",optionfilegnuplot); */
7219: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7220: /* } */
7221:
7222: /*#ifdef windows */
7223: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7224: /*#endif */
1.225 brouard 7225: m=pow(2,cptcoveff);
1.126 brouard 7226:
1.274 brouard 7227: /* diagram of the model */
7228: fprintf(ficgp,"\n#Diagram of the model \n");
7229: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7230: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7231: 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);
7232:
7233: 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);
7234: fprintf(ficgp,"\n#show arrow\nunset label\n");
7235: 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);
7236: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7237: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7238: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7239: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7240:
1.202 brouard 7241: /* Contribution to likelihood */
7242: /* Plot the probability implied in the likelihood */
1.223 brouard 7243: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7244: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7245: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7246: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7247: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7248: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7249: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7250: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7251: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7252: 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));
7253: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7254: 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));
7255: for (i=1; i<= nlstate ; i ++) {
7256: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7257: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7258: 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);
7259: for (j=2; j<= nlstate+ndeath ; j ++) {
7260: 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);
7261: }
7262: fprintf(ficgp,";\nset out; unset ylabel;\n");
7263: }
7264: /* 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 */
7265: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7266: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7267: fprintf(ficgp,"\nset out;unset log\n");
7268: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7269:
1.126 brouard 7270: strcpy(dirfileres,optionfilefiname);
7271: strcpy(optfileres,"vpl");
1.223 brouard 7272: /* 1eme*/
1.238 brouard 7273: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7274: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7275: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7276: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7277: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7278: continue;
7279: /* We are interested in selected combination by the resultline */
1.246 brouard 7280: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7281: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7282: strcpy(gplotlabel,"(");
1.238 brouard 7283: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7284: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7285: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7286: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7287: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7288: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7289: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7290: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7291: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7292: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7293: }
7294: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7295: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7296: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7297: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7298: }
7299: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7300: /* printf("\n#\n"); */
1.238 brouard 7301: fprintf(ficgp,"\n#\n");
7302: if(invalidvarcomb[k1]){
1.260 brouard 7303: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7304: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7305: continue;
7306: }
1.235 brouard 7307:
1.241 brouard 7308: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7309: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7310: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7311: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7312: 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);
7313: /* 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); */
7314: /* k1-1 error should be nres-1*/
1.238 brouard 7315: for (i=1; i<= nlstate ; i ++) {
7316: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7317: else fprintf(ficgp," %%*lf (%%*lf)");
7318: }
1.288 brouard 7319: 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 7320: for (i=1; i<= nlstate ; i ++) {
7321: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7322: else fprintf(ficgp," %%*lf (%%*lf)");
7323: }
1.260 brouard 7324: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7325: for (i=1; i<= nlstate ; i ++) {
7326: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7327: else fprintf(ficgp," %%*lf (%%*lf)");
7328: }
1.265 brouard 7329: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7330:
7331: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7332: if(cptcoveff ==0){
1.271 brouard 7333: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7334: }else{
7335: kl=0;
7336: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7337: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7338: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7339: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7340: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7341: vlv= nbcode[Tvaraff[k]][lv];
7342: kl++;
7343: /* 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 *\/ */
7344: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7345: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7346: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7347: if(k==cptcoveff){
7348: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7349: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7350: }else{
7351: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7352: kl++;
7353: }
7354: } /* end covariate */
7355: } /* end if no covariate */
7356:
1.296 brouard 7357: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7358: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7359: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7360: if(cptcoveff ==0){
1.245 brouard 7361: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7362: }else{
7363: kl=0;
7364: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7365: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7366: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7367: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7368: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7369: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7370: kl++;
1.238 brouard 7371: /* 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 *\/ */
7372: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7373: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7374: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7375: if(k==cptcoveff){
1.245 brouard 7376: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7377: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7378: }else{
7379: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7380: kl++;
7381: }
7382: } /* end covariate */
7383: } /* end if no covariate */
1.296 brouard 7384: if(prevbcast == 1){
1.268 brouard 7385: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7386: /* k1-1 error should be nres-1*/
7387: for (i=1; i<= nlstate ; i ++) {
7388: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7389: else fprintf(ficgp," %%*lf (%%*lf)");
7390: }
1.271 brouard 7391: 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 7392: for (i=1; i<= nlstate ; i ++) {
7393: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7394: else fprintf(ficgp," %%*lf (%%*lf)");
7395: }
1.276 brouard 7396: 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 7397: for (i=1; i<= nlstate ; i ++) {
7398: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7399: else fprintf(ficgp," %%*lf (%%*lf)");
7400: }
1.274 brouard 7401: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7402: } /* end if backprojcast */
1.296 brouard 7403: } /* end if prevbcast */
1.276 brouard 7404: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7405: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7406: } /* nres */
1.201 brouard 7407: } /* k1 */
7408: } /* cpt */
1.235 brouard 7409:
7410:
1.126 brouard 7411: /*2 eme*/
1.238 brouard 7412: for (k1=1; k1<= m ; k1 ++){
7413: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7414: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7415: continue;
7416: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7417: strcpy(gplotlabel,"(");
1.238 brouard 7418: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7419: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7420: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7421: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7422: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7423: vlv= nbcode[Tvaraff[k]][lv];
7424: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7425: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7426: }
1.237 brouard 7427: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7428: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7429: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7430: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7431: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7432: }
1.264 brouard 7433: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7434: fprintf(ficgp,"\n#\n");
1.223 brouard 7435: if(invalidvarcomb[k1]){
7436: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7437: continue;
7438: }
1.219 brouard 7439:
1.241 brouard 7440: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7441: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7442: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7443: if(vpopbased==0){
1.238 brouard 7444: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7445: }else
1.238 brouard 7446: fprintf(ficgp,"\nreplot ");
7447: for (i=1; i<= nlstate+1 ; i ++) {
7448: k=2*i;
1.261 brouard 7449: 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 7450: for (j=1; j<= nlstate+1 ; j ++) {
7451: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7452: else fprintf(ficgp," %%*lf (%%*lf)");
7453: }
7454: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7455: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7456: 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 7457: for (j=1; j<= nlstate+1 ; j ++) {
7458: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7459: else fprintf(ficgp," %%*lf (%%*lf)");
7460: }
7461: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7462: 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 7463: for (j=1; j<= nlstate+1 ; j ++) {
7464: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7465: else fprintf(ficgp," %%*lf (%%*lf)");
7466: }
7467: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7468: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7469: } /* state */
7470: } /* vpopbased */
1.264 brouard 7471: 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 7472: } /* end nres */
7473: } /* k1 end 2 eme*/
7474:
7475:
7476: /*3eme*/
7477: for (k1=1; k1<= m ; k1 ++){
7478: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7479: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7480: continue;
7481:
7482: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7483: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7484: strcpy(gplotlabel,"(");
1.238 brouard 7485: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7486: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7487: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7488: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7489: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7490: vlv= nbcode[Tvaraff[k]][lv];
7491: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7492: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7493: }
7494: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7495: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7496: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7497: }
1.264 brouard 7498: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7499: fprintf(ficgp,"\n#\n");
7500: if(invalidvarcomb[k1]){
7501: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7502: continue;
7503: }
7504:
7505: /* k=2+nlstate*(2*cpt-2); */
7506: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7507: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7508: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7509: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7510: 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 7511: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7512: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7513: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7514: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7515: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7516: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7517:
1.238 brouard 7518: */
7519: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7520: 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 7521: /* 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 7522:
1.238 brouard 7523: }
1.261 brouard 7524: 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 7525: }
1.264 brouard 7526: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7527: } /* end nres */
7528: } /* end kl 3eme */
1.126 brouard 7529:
1.223 brouard 7530: /* 4eme */
1.201 brouard 7531: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7532: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7533: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7534: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7535: continue;
1.238 brouard 7536: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7537: strcpy(gplotlabel,"(");
1.238 brouard 7538: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7539: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7540: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7541: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7542: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7543: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7544: vlv= nbcode[Tvaraff[k]][lv];
7545: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7546: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7547: }
7548: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7549: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7550: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7551: }
1.264 brouard 7552: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7553: fprintf(ficgp,"\n#\n");
7554: if(invalidvarcomb[k1]){
7555: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7556: continue;
1.223 brouard 7557: }
1.238 brouard 7558:
1.241 brouard 7559: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7560: 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 7561: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7562: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7563: k=3;
7564: for (i=1; i<= nlstate ; i ++){
7565: if(i==1){
7566: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7567: }else{
7568: fprintf(ficgp,", '' ");
7569: }
7570: l=(nlstate+ndeath)*(i-1)+1;
7571: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7572: for (j=2; j<= nlstate+ndeath ; j ++)
7573: fprintf(ficgp,"+$%d",k+l+j-1);
7574: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7575: } /* nlstate */
1.264 brouard 7576: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7577: } /* end cpt state*/
7578: } /* end nres */
7579: } /* end covariate k1 */
7580:
1.220 brouard 7581: /* 5eme */
1.201 brouard 7582: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7583: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7584: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7585: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7586: continue;
1.238 brouard 7587: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7588: strcpy(gplotlabel,"(");
1.238 brouard 7589: 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);
7590: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7591: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7592: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7593: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7594: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7595: vlv= nbcode[Tvaraff[k]][lv];
7596: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7597: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7598: }
7599: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7600: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7601: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7602: }
1.264 brouard 7603: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7604: fprintf(ficgp,"\n#\n");
7605: if(invalidvarcomb[k1]){
7606: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7607: continue;
7608: }
1.227 brouard 7609:
1.241 brouard 7610: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7611: 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 7612: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7613: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7614: k=3;
7615: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7616: if(j==1)
7617: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7618: else
7619: fprintf(ficgp,", '' ");
7620: l=(nlstate+ndeath)*(cpt-1) +j;
7621: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7622: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7623: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7624: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7625: } /* nlstate */
7626: fprintf(ficgp,", '' ");
7627: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7628: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7629: l=(nlstate+ndeath)*(cpt-1) +j;
7630: if(j < nlstate)
7631: fprintf(ficgp,"$%d +",k+l);
7632: else
7633: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7634: }
1.264 brouard 7635: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7636: } /* end cpt state*/
7637: } /* end covariate */
7638: } /* end nres */
1.227 brouard 7639:
1.220 brouard 7640: /* 6eme */
1.202 brouard 7641: /* CV preval stable (period) for each covariate */
1.237 brouard 7642: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7643: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7644: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7645: continue;
1.255 brouard 7646: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7647: strcpy(gplotlabel,"(");
1.288 brouard 7648: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7649: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7650: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7651: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7652: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7653: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7654: vlv= nbcode[Tvaraff[k]][lv];
7655: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7656: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7657: }
1.237 brouard 7658: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7659: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7660: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7661: }
1.264 brouard 7662: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7663: fprintf(ficgp,"\n#\n");
1.223 brouard 7664: if(invalidvarcomb[k1]){
1.227 brouard 7665: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7666: continue;
1.223 brouard 7667: }
1.227 brouard 7668:
1.241 brouard 7669: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7670: 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 7671: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7672: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7673: k=3; /* Offset */
1.255 brouard 7674: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7675: if(i==1)
7676: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7677: else
7678: fprintf(ficgp,", '' ");
1.255 brouard 7679: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7680: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7681: for (j=2; j<= nlstate ; j ++)
7682: fprintf(ficgp,"+$%d",k+l+j-1);
7683: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7684: } /* nlstate */
1.264 brouard 7685: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7686: } /* end cpt state*/
7687: } /* end covariate */
1.227 brouard 7688:
7689:
1.220 brouard 7690: /* 7eme */
1.296 brouard 7691: if(prevbcast == 1){
1.288 brouard 7692: /* CV backward prevalence for each covariate */
1.237 brouard 7693: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7694: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7695: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7696: continue;
1.268 brouard 7697: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7698: strcpy(gplotlabel,"(");
1.288 brouard 7699: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7700: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7701: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7702: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7703: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7704: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7705: vlv= nbcode[Tvaraff[k]][lv];
7706: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7707: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7708: }
1.237 brouard 7709: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7710: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7711: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7712: }
1.264 brouard 7713: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7714: fprintf(ficgp,"\n#\n");
7715: if(invalidvarcomb[k1]){
7716: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7717: continue;
7718: }
7719:
1.241 brouard 7720: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7721: 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 7722: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7723: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7724: k=3; /* Offset */
1.268 brouard 7725: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7726: if(i==1)
7727: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7728: else
7729: fprintf(ficgp,", '' ");
7730: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7731: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7732: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7733: /* 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 7734: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7735: /* for (j=2; j<= nlstate ; j ++) */
7736: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7737: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7738: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7739: } /* nlstate */
1.264 brouard 7740: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7741: } /* end cpt state*/
7742: } /* end covariate */
1.296 brouard 7743: } /* End if prevbcast */
1.218 brouard 7744:
1.223 brouard 7745: /* 8eme */
1.218 brouard 7746: if(prevfcast==1){
1.288 brouard 7747: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7748:
1.237 brouard 7749: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7750: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7751: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7752: continue;
1.211 brouard 7753: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7754: strcpy(gplotlabel,"(");
1.288 brouard 7755: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7756: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7757: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7758: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7759: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7760: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7761: vlv= nbcode[Tvaraff[k]][lv];
7762: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7763: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7764: }
1.237 brouard 7765: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7766: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7767: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7768: }
1.264 brouard 7769: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7770: fprintf(ficgp,"\n#\n");
7771: if(invalidvarcomb[k1]){
7772: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7773: continue;
7774: }
7775:
7776: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7777: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7778: 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 7779: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7780: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7781:
7782: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7783: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7784: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7785: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7786: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7787: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7788: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7789: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7790: if(i==istart){
1.227 brouard 7791: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7792: }else{
7793: fprintf(ficgp,",\\\n '' ");
7794: }
7795: if(cptcoveff ==0){ /* No covariate */
7796: ioffset=2; /* Age is in 2 */
7797: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7798: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7799: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7800: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7801: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7802: if(i==nlstate+1){
1.270 brouard 7803: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7804: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7805: fprintf(ficgp,",\\\n '' ");
7806: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7807: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7808: offyear, \
1.268 brouard 7809: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7810: }else
1.227 brouard 7811: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7812: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7813: }else{ /* more than 2 covariates */
1.270 brouard 7814: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7815: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7816: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7817: iyearc=ioffset-1;
7818: iagec=ioffset;
1.227 brouard 7819: fprintf(ficgp," u %d:(",ioffset);
7820: kl=0;
7821: strcpy(gplotcondition,"(");
7822: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7823: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7824: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7825: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7826: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7827: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7828: kl++;
7829: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7830: kl++;
7831: if(k <cptcoveff && cptcoveff>1)
7832: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7833: }
7834: strcpy(gplotcondition+strlen(gplotcondition),")");
7835: /* 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 *\/ */
7836: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7837: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7838: /* '' 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*/
7839: if(i==nlstate+1){
1.270 brouard 7840: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7841: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7842: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7843: fprintf(ficgp," u %d:(",iagec);
7844: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7845: iyearc, iagec, offyear, \
7846: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7847: /* '' 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 7848: }else{
7849: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7850: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7851: }
7852: } /* end if covariate */
7853: } /* nlstate */
1.264 brouard 7854: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7855: } /* end cpt state*/
7856: } /* end covariate */
7857: } /* End if prevfcast */
1.227 brouard 7858:
1.296 brouard 7859: if(prevbcast==1){
1.268 brouard 7860: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7861:
7862: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7863: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7864: if(m != 1 && TKresult[nres]!= k1)
7865: continue;
7866: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7867: strcpy(gplotlabel,"(");
7868: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7869: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7870: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7871: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7872: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7873: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7874: vlv= nbcode[Tvaraff[k]][lv];
7875: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7876: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7877: }
7878: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7879: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7880: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7881: }
7882: strcpy(gplotlabel+strlen(gplotlabel),")");
7883: fprintf(ficgp,"\n#\n");
7884: if(invalidvarcomb[k1]){
7885: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7886: continue;
7887: }
7888:
7889: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7890: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7891: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7892: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7893: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7894:
7895: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7896: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7897: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7898: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7899: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7900: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7901: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7902: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7903: if(i==istart){
7904: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7905: }else{
7906: fprintf(ficgp,",\\\n '' ");
7907: }
7908: if(cptcoveff ==0){ /* No covariate */
7909: ioffset=2; /* Age is in 2 */
7910: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7911: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7912: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7913: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7914: fprintf(ficgp," u %d:(", ioffset);
7915: if(i==nlstate+1){
1.270 brouard 7916: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7917: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7918: fprintf(ficgp,",\\\n '' ");
7919: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7920: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7921: offbyear, \
7922: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7923: }else
7924: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7925: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7926: }else{ /* more than 2 covariates */
1.270 brouard 7927: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7928: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7929: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7930: iyearc=ioffset-1;
7931: iagec=ioffset;
1.268 brouard 7932: fprintf(ficgp," u %d:(",ioffset);
7933: kl=0;
7934: strcpy(gplotcondition,"(");
7935: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7936: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7937: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7938: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7939: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7940: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7941: kl++;
7942: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7943: kl++;
7944: if(k <cptcoveff && cptcoveff>1)
7945: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7946: }
7947: strcpy(gplotcondition+strlen(gplotcondition),")");
7948: /* 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 *\/ */
7949: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7950: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7951: /* '' 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*/
7952: if(i==nlstate+1){
1.270 brouard 7953: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7954: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7955: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7956: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7957: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7958: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7959: iyearc,iagec,offbyear, \
7960: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7961: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7962: }else{
7963: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7964: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7965: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7966: }
7967: } /* end if covariate */
7968: } /* nlstate */
7969: fprintf(ficgp,"\nset out; unset label;\n");
7970: } /* end cpt state*/
7971: } /* end covariate */
1.296 brouard 7972: } /* End if prevbcast */
1.268 brouard 7973:
1.227 brouard 7974:
1.238 brouard 7975: /* 9eme writing MLE parameters */
7976: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7977: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7978: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7979: for(k=1; k <=(nlstate+ndeath); k++){
7980: if (k != i) {
1.227 brouard 7981: fprintf(ficgp,"# current state %d\n",k);
7982: for(j=1; j <=ncovmodel; j++){
7983: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7984: jk++;
7985: }
7986: fprintf(ficgp,"\n");
1.126 brouard 7987: }
7988: }
1.223 brouard 7989: }
1.187 brouard 7990: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7991:
1.145 brouard 7992: /*goto avoid;*/
1.238 brouard 7993: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7994: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7995: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7996: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7997: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7998: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7999: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8000: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8001: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8002: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8003: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8004: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8005: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8006: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8007: fprintf(ficgp,"#\n");
1.223 brouard 8008: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8009: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8010: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8011: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8012: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8013: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8014: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8015: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8016: continue;
1.264 brouard 8017: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8018: strcpy(gplotlabel,"(");
1.276 brouard 8019: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8020: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8021: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8022: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8023: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8024: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8025: vlv= nbcode[Tvaraff[k]][lv];
8026: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8027: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8028: }
1.237 brouard 8029: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8030: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8031: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8032: }
1.264 brouard 8033: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8034: fprintf(ficgp,"\n#\n");
1.264 brouard 8035: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8036: fprintf(ficgp,"\nset key outside ");
8037: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8038: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8039: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8040: if (ng==1){
8041: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8042: fprintf(ficgp,"\nunset log y");
8043: }else if (ng==2){
8044: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8045: fprintf(ficgp,"\nset log y");
8046: }else if (ng==3){
8047: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8048: fprintf(ficgp,"\nset log y");
8049: }else
8050: fprintf(ficgp,"\nunset title ");
8051: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8052: i=1;
8053: for(k2=1; k2<=nlstate; k2++) {
8054: k3=i;
8055: for(k=1; k<=(nlstate+ndeath); k++) {
8056: if (k != k2){
8057: switch( ng) {
8058: case 1:
8059: if(nagesqr==0)
8060: fprintf(ficgp," p%d+p%d*x",i,i+1);
8061: else /* nagesqr =1 */
8062: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8063: break;
8064: case 2: /* ng=2 */
8065: if(nagesqr==0)
8066: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8067: else /* nagesqr =1 */
8068: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8069: break;
8070: case 3:
8071: if(nagesqr==0)
8072: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8073: else /* nagesqr =1 */
8074: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8075: break;
8076: }
8077: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8078: ijp=1; /* product no age */
8079: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8080: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8081: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8082: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8083: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8084: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8085: if(DummyV[j]==0){
8086: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8087: }else{ /* quantitative */
8088: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8089: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8090: }
8091: ij++;
1.237 brouard 8092: }
1.268 brouard 8093: }
8094: }else if(cptcovprod >0){
8095: if(j==Tprod[ijp]) { /* */
8096: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8097: if(ijp <=cptcovprod) { /* Product */
8098: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8099: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8100: /* 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)]); */
8101: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8102: }else{ /* Vn is dummy and Vm is quanti */
8103: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8104: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8105: }
8106: }else{ /* Vn*Vm Vn is quanti */
8107: if(DummyV[Tvard[ijp][2]]==0){
8108: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8109: }else{ /* Both quanti */
8110: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8111: }
1.237 brouard 8112: }
1.268 brouard 8113: ijp++;
1.237 brouard 8114: }
1.268 brouard 8115: } /* end Tprod */
1.237 brouard 8116: } else{ /* simple covariate */
1.264 brouard 8117: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8118: if(Dummy[j]==0){
8119: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8120: }else{ /* quantitative */
8121: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8122: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8123: }
1.237 brouard 8124: } /* end simple */
8125: } /* end j */
1.223 brouard 8126: }else{
8127: i=i-ncovmodel;
8128: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8129: fprintf(ficgp," (1.");
8130: }
1.227 brouard 8131:
1.223 brouard 8132: if(ng != 1){
8133: fprintf(ficgp,")/(1");
1.227 brouard 8134:
1.264 brouard 8135: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8136: if(nagesqr==0)
1.264 brouard 8137: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8138: else /* nagesqr =1 */
1.264 brouard 8139: 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 8140:
1.223 brouard 8141: ij=1;
8142: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8143: if(cptcovage >0){
8144: if((j-2)==Tage[ij]) { /* Bug valgrind */
8145: if(ij <=cptcovage) { /* Bug valgrind */
8146: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8147: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8148: ij++;
8149: }
8150: }
8151: }else
8152: 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 8153: }
8154: fprintf(ficgp,")");
8155: }
8156: fprintf(ficgp,")");
8157: if(ng ==2)
1.276 brouard 8158: 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 8159: else /* ng= 3 */
1.276 brouard 8160: 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 8161: }else{ /* end ng <> 1 */
8162: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8163: 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 8164: }
8165: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8166: fprintf(ficgp,",");
8167: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8168: fprintf(ficgp,",");
8169: i=i+ncovmodel;
8170: } /* end k */
8171: } /* end k2 */
1.276 brouard 8172: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8173: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8174: } /* end k1 */
1.223 brouard 8175: } /* end ng */
8176: /* avoid: */
8177: fflush(ficgp);
1.126 brouard 8178: } /* end gnuplot */
8179:
8180:
8181: /*************** Moving average **************/
1.219 brouard 8182: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8183: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8184:
1.222 brouard 8185: int i, cpt, cptcod;
8186: int modcovmax =1;
8187: int mobilavrange, mob;
8188: int iage=0;
1.288 brouard 8189: int firstA1=0, firstA2=0;
1.222 brouard 8190:
1.266 brouard 8191: double sum=0., sumr=0.;
1.222 brouard 8192: double age;
1.266 brouard 8193: double *sumnewp, *sumnewm, *sumnewmr;
8194: double *agemingood, *agemaxgood;
8195: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8196:
8197:
1.278 brouard 8198: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8199: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8200:
8201: sumnewp = vector(1,ncovcombmax);
8202: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8203: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8204: agemingood = vector(1,ncovcombmax);
1.266 brouard 8205: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8206: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8207: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8208:
8209: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8210: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8211: sumnewp[cptcod]=0.;
1.266 brouard 8212: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8213: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8214: }
8215: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8216:
1.266 brouard 8217: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8218: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8219: else mobilavrange=mobilav;
8220: for (age=bage; age<=fage; age++)
8221: for (i=1; i<=nlstate;i++)
8222: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8223: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8224: /* We keep the original values on the extreme ages bage, fage and for
8225: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8226: we use a 5 terms etc. until the borders are no more concerned.
8227: */
8228: for (mob=3;mob <=mobilavrange;mob=mob+2){
8229: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8230: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8231: sumnewm[cptcod]=0.;
8232: for (i=1; i<=nlstate;i++){
1.222 brouard 8233: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8234: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8235: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8236: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8237: }
8238: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8239: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8240: } /* end i */
8241: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8242: } /* end cptcod */
1.222 brouard 8243: }/* end age */
8244: }/* end mob */
1.266 brouard 8245: }else{
8246: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8247: return -1;
1.266 brouard 8248: }
8249:
8250: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8251: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8252: if(invalidvarcomb[cptcod]){
8253: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8254: continue;
8255: }
1.219 brouard 8256:
1.266 brouard 8257: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8258: sumnewm[cptcod]=0.;
8259: sumnewmr[cptcod]=0.;
8260: for (i=1; i<=nlstate;i++){
8261: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8262: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8263: }
8264: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8265: agemingoodr[cptcod]=age;
8266: }
8267: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8268: agemingood[cptcod]=age;
8269: }
8270: } /* age */
8271: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8272: sumnewm[cptcod]=0.;
1.266 brouard 8273: sumnewmr[cptcod]=0.;
1.222 brouard 8274: for (i=1; i<=nlstate;i++){
8275: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8276: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8277: }
8278: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8279: agemaxgoodr[cptcod]=age;
1.222 brouard 8280: }
8281: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8282: agemaxgood[cptcod]=age;
8283: }
8284: } /* age */
8285: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8286: /* but they will change */
1.288 brouard 8287: firstA1=0;firstA2=0;
1.266 brouard 8288: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8289: sumnewm[cptcod]=0.;
8290: sumnewmr[cptcod]=0.;
8291: for (i=1; i<=nlstate;i++){
8292: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8293: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8294: }
8295: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8296: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8297: agemaxgoodr[cptcod]=age; /* age min */
8298: for (i=1; i<=nlstate;i++)
8299: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8300: }else{ /* bad we change the value with the values of good ages */
8301: for (i=1; i<=nlstate;i++){
8302: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8303: } /* i */
8304: } /* end bad */
8305: }else{
8306: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8307: agemaxgood[cptcod]=age;
8308: }else{ /* bad we change the value with the values of good ages */
8309: for (i=1; i<=nlstate;i++){
8310: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8311: } /* i */
8312: } /* end bad */
8313: }/* end else */
8314: sum=0.;sumr=0.;
8315: for (i=1; i<=nlstate;i++){
8316: sum+=mobaverage[(int)age][i][cptcod];
8317: sumr+=probs[(int)age][i][cptcod];
8318: }
8319: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8320: if(!firstA1){
8321: firstA1=1;
8322: 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);
8323: }
8324: 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 8325: } /* end bad */
8326: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8327: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8328: if(!firstA2){
8329: firstA2=1;
8330: 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);
8331: }
8332: 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 8333: } /* end bad */
8334: }/* age */
1.266 brouard 8335:
8336: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8337: sumnewm[cptcod]=0.;
1.266 brouard 8338: sumnewmr[cptcod]=0.;
1.222 brouard 8339: for (i=1; i<=nlstate;i++){
8340: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8341: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8342: }
8343: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8344: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8345: agemingoodr[cptcod]=age;
8346: for (i=1; i<=nlstate;i++)
8347: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8348: }else{ /* bad we change the value with the values of good ages */
8349: for (i=1; i<=nlstate;i++){
8350: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8351: } /* i */
8352: } /* end bad */
8353: }else{
8354: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8355: agemingood[cptcod]=age;
8356: }else{ /* bad */
8357: for (i=1; i<=nlstate;i++){
8358: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8359: } /* i */
8360: } /* end bad */
8361: }/* end else */
8362: sum=0.;sumr=0.;
8363: for (i=1; i<=nlstate;i++){
8364: sum+=mobaverage[(int)age][i][cptcod];
8365: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8366: }
1.266 brouard 8367: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8368: 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 8369: } /* end bad */
8370: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8371: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8372: 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 8373: } /* end bad */
8374: }/* age */
1.266 brouard 8375:
1.222 brouard 8376:
8377: for (age=bage; age<=fage; age++){
1.235 brouard 8378: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8379: sumnewp[cptcod]=0.;
8380: sumnewm[cptcod]=0.;
8381: for (i=1; i<=nlstate;i++){
8382: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8383: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8384: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8385: }
8386: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8387: }
8388: /* printf("\n"); */
8389: /* } */
1.266 brouard 8390:
1.222 brouard 8391: /* brutal averaging */
1.266 brouard 8392: /* for (i=1; i<=nlstate;i++){ */
8393: /* for (age=1; age<=bage; age++){ */
8394: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8395: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8396: /* } */
8397: /* for (age=fage; age<=AGESUP; age++){ */
8398: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8399: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8400: /* } */
8401: /* } /\* end i status *\/ */
8402: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8403: /* for (age=1; age<=AGESUP; age++){ */
8404: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8405: /* mobaverage[(int)age][i][cptcod]=0.; */
8406: /* } */
8407: /* } */
1.222 brouard 8408: }/* end cptcod */
1.266 brouard 8409: free_vector(agemaxgoodr,1, ncovcombmax);
8410: free_vector(agemaxgood,1, ncovcombmax);
8411: free_vector(agemingood,1, ncovcombmax);
8412: free_vector(agemingoodr,1, ncovcombmax);
8413: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8414: free_vector(sumnewm,1, ncovcombmax);
8415: free_vector(sumnewp,1, ncovcombmax);
8416: return 0;
8417: }/* End movingaverage */
1.218 brouard 8418:
1.126 brouard 8419:
1.296 brouard 8420:
1.126 brouard 8421: /************** Forecasting ******************/
1.296 brouard 8422: /* 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)*/
8423: 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){
8424: /* dateintemean, mean date of interviews
8425: dateprojd, year, month, day of starting projection
8426: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8427: agemin, agemax range of age
8428: dateprev1 dateprev2 range of dates during which prevalence is computed
8429: */
1.296 brouard 8430: /* double anprojd, mprojd, jprojd; */
8431: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8432: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8433: double agec; /* generic age */
1.296 brouard 8434: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8435: double *popeffectif,*popcount;
8436: double ***p3mat;
1.218 brouard 8437: /* double ***mobaverage; */
1.126 brouard 8438: char fileresf[FILENAMELENGTH];
8439:
8440: agelim=AGESUP;
1.211 brouard 8441: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8442: in each health status at the date of interview (if between dateprev1 and dateprev2).
8443: We still use firstpass and lastpass as another selection.
8444: */
1.214 brouard 8445: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8446: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8447:
1.201 brouard 8448: strcpy(fileresf,"F_");
8449: strcat(fileresf,fileresu);
1.126 brouard 8450: if((ficresf=fopen(fileresf,"w"))==NULL) {
8451: printf("Problem with forecast resultfile: %s\n", fileresf);
8452: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8453: }
1.235 brouard 8454: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8455: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8456:
1.225 brouard 8457: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8458:
8459:
8460: stepsize=(int) (stepm+YEARM-1)/YEARM;
8461: if (stepm<=12) stepsize=1;
8462: if(estepm < stepm){
8463: printf ("Problem %d lower than %d\n",estepm, stepm);
8464: }
1.270 brouard 8465: else{
8466: hstepm=estepm;
8467: }
8468: if(estepm > stepm){ /* Yes every two year */
8469: stepsize=2;
8470: }
1.296 brouard 8471: hstepm=hstepm/stepm;
1.126 brouard 8472:
1.296 brouard 8473:
8474: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8475: /* fractional in yp1 *\/ */
8476: /* aintmean=yp; */
8477: /* yp2=modf((yp1*12),&yp); */
8478: /* mintmean=yp; */
8479: /* yp1=modf((yp2*30.5),&yp); */
8480: /* jintmean=yp; */
8481: /* if(jintmean==0) jintmean=1; */
8482: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8483:
1.296 brouard 8484:
8485: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8486: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8487: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8488: i1=pow(2,cptcoveff);
1.126 brouard 8489: if (cptcovn < 1){i1=1;}
8490:
1.296 brouard 8491: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8492:
8493: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8494:
1.126 brouard 8495: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8496: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8497: for(k=1; k<=i1;k++){
1.253 brouard 8498: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8499: continue;
1.227 brouard 8500: if(invalidvarcomb[k]){
8501: printf("\nCombination (%d) projection ignored because no cases \n",k);
8502: continue;
8503: }
8504: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8505: for(j=1;j<=cptcoveff;j++) {
8506: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8507: }
1.235 brouard 8508: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8509: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8510: }
1.227 brouard 8511: fprintf(ficresf," yearproj age");
8512: for(j=1; j<=nlstate+ndeath;j++){
8513: for(i=1; i<=nlstate;i++)
8514: fprintf(ficresf," p%d%d",i,j);
8515: fprintf(ficresf," wp.%d",j);
8516: }
1.296 brouard 8517: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8518: fprintf(ficresf,"\n");
1.296 brouard 8519: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8520: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8521: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8522: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8523: nhstepm = nhstepm/hstepm;
8524: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8525: oldm=oldms;savm=savms;
1.268 brouard 8526: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8527: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8528: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8529: for (h=0; h<=nhstepm; h++){
8530: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8531: break;
8532: }
8533: }
8534: fprintf(ficresf,"\n");
8535: for(j=1;j<=cptcoveff;j++)
8536: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8537: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8538:
8539: for(j=1; j<=nlstate+ndeath;j++) {
8540: ppij=0.;
8541: for(i=1; i<=nlstate;i++) {
1.278 brouard 8542: if (mobilav>=1)
8543: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8544: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8545: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8546: }
1.268 brouard 8547: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8548: } /* end i */
8549: fprintf(ficresf," %.3f", ppij);
8550: }/* end j */
1.227 brouard 8551: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8552: } /* end agec */
1.266 brouard 8553: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8554: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8555: } /* end yearp */
8556: } /* end k */
1.219 brouard 8557:
1.126 brouard 8558: fclose(ficresf);
1.215 brouard 8559: printf("End of Computing forecasting \n");
8560: fprintf(ficlog,"End of Computing forecasting\n");
8561:
1.126 brouard 8562: }
8563:
1.269 brouard 8564: /************** Back Forecasting ******************/
1.296 brouard 8565: /* 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){ */
8566: 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){
8567: /* back1, year, month, day of starting backprojection
1.267 brouard 8568: agemin, agemax range of age
8569: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8570: anback2 year of end of backprojection (same day and month as back1).
8571: prevacurrent and prev are prevalences.
1.267 brouard 8572: */
8573: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8574: double agec; /* generic age */
1.302 brouard 8575: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8576: double *popeffectif,*popcount;
8577: double ***p3mat;
8578: /* double ***mobaverage; */
8579: char fileresfb[FILENAMELENGTH];
8580:
1.268 brouard 8581: agelim=AGEINF;
1.267 brouard 8582: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8583: in each health status at the date of interview (if between dateprev1 and dateprev2).
8584: We still use firstpass and lastpass as another selection.
8585: */
8586: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8587: /* firstpass, lastpass, stepm, weightopt, model); */
8588:
8589: /*Do we need to compute prevalence again?*/
8590:
8591: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8592:
8593: strcpy(fileresfb,"FB_");
8594: strcat(fileresfb,fileresu);
8595: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8596: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8597: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8598: }
8599: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8600: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8601:
8602: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8603:
8604:
8605: stepsize=(int) (stepm+YEARM-1)/YEARM;
8606: if (stepm<=12) stepsize=1;
8607: if(estepm < stepm){
8608: printf ("Problem %d lower than %d\n",estepm, stepm);
8609: }
1.270 brouard 8610: else{
8611: hstepm=estepm;
8612: }
8613: if(estepm >= stepm){ /* Yes every two year */
8614: stepsize=2;
8615: }
1.267 brouard 8616:
8617: hstepm=hstepm/stepm;
1.296 brouard 8618: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8619: /* fractional in yp1 *\/ */
8620: /* aintmean=yp; */
8621: /* yp2=modf((yp1*12),&yp); */
8622: /* mintmean=yp; */
8623: /* yp1=modf((yp2*30.5),&yp); */
8624: /* jintmean=yp; */
8625: /* if(jintmean==0) jintmean=1; */
8626: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8627:
8628: i1=pow(2,cptcoveff);
8629: if (cptcovn < 1){i1=1;}
8630:
1.296 brouard 8631: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8632: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8633:
8634: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8635:
8636: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8637: for(k=1; k<=i1;k++){
8638: if(i1 != 1 && TKresult[nres]!= k)
8639: continue;
8640: if(invalidvarcomb[k]){
8641: printf("\nCombination (%d) projection ignored because no cases \n",k);
8642: continue;
8643: }
1.268 brouard 8644: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8645: for(j=1;j<=cptcoveff;j++) {
8646: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8647: }
8648: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8649: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8650: }
8651: fprintf(ficresfb," yearbproj age");
8652: for(j=1; j<=nlstate+ndeath;j++){
8653: for(i=1; i<=nlstate;i++)
1.268 brouard 8654: fprintf(ficresfb," b%d%d",i,j);
8655: fprintf(ficresfb," b.%d",j);
1.267 brouard 8656: }
1.296 brouard 8657: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8658: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8659: fprintf(ficresfb,"\n");
1.296 brouard 8660: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8661: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8662: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8663: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8664: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8665: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8666: nhstepm = nhstepm/hstepm;
8667: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8668: oldm=oldms;savm=savms;
1.268 brouard 8669: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8670: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8671: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8672: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8673: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8674: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8675: for (h=0; h<=nhstepm; h++){
1.268 brouard 8676: if (h*hstepm/YEARM*stepm ==-yearp) {
8677: break;
8678: }
8679: }
8680: fprintf(ficresfb,"\n");
8681: for(j=1;j<=cptcoveff;j++)
8682: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8683: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8684: for(i=1; i<=nlstate+ndeath;i++) {
8685: ppij=0.;ppi=0.;
8686: for(j=1; j<=nlstate;j++) {
8687: /* if (mobilav==1) */
1.269 brouard 8688: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8689: ppi=ppi+prevacurrent[(int)agec][j][k];
8690: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8691: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8692: /* else { */
8693: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8694: /* } */
1.268 brouard 8695: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8696: } /* end j */
8697: if(ppi <0.99){
8698: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8699: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8700: }
8701: fprintf(ficresfb," %.3f", ppij);
8702: }/* end j */
1.267 brouard 8703: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8704: } /* end agec */
8705: } /* end yearp */
8706: } /* end k */
1.217 brouard 8707:
1.267 brouard 8708: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8709:
1.267 brouard 8710: fclose(ficresfb);
8711: printf("End of Computing Back forecasting \n");
8712: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8713:
1.267 brouard 8714: }
1.217 brouard 8715:
1.269 brouard 8716: /* Variance of prevalence limit: varprlim */
8717: 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 8718: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8719:
8720: char fileresvpl[FILENAMELENGTH];
8721: FILE *ficresvpl;
8722: double **oldm, **savm;
8723: double **varpl; /* Variances of prevalence limits by age */
8724: int i1, k, nres, j ;
8725:
8726: strcpy(fileresvpl,"VPL_");
8727: strcat(fileresvpl,fileresu);
8728: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8729: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8730: exit(0);
8731: }
1.288 brouard 8732: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8733: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8734:
8735: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8736: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8737:
8738: i1=pow(2,cptcoveff);
8739: if (cptcovn < 1){i1=1;}
8740:
8741: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8742: for(k=1; k<=i1;k++){
8743: if(i1 != 1 && TKresult[nres]!= k)
8744: continue;
8745: fprintf(ficresvpl,"\n#****** ");
8746: printf("\n#****** ");
8747: fprintf(ficlog,"\n#****** ");
8748: for(j=1;j<=cptcoveff;j++) {
8749: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8750: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8751: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8752: }
8753: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8754: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8755: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8756: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8757: }
8758: fprintf(ficresvpl,"******\n");
8759: printf("******\n");
8760: fprintf(ficlog,"******\n");
8761:
8762: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8763: oldm=oldms;savm=savms;
8764: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8765: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8766: /*}*/
8767: }
8768:
8769: fclose(ficresvpl);
1.288 brouard 8770: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8771: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8772:
8773: }
8774: /* Variance of back prevalence: varbprlim */
8775: 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){
8776: /*------- Variance of back (stable) prevalence------*/
8777:
8778: char fileresvbl[FILENAMELENGTH];
8779: FILE *ficresvbl;
8780:
8781: double **oldm, **savm;
8782: double **varbpl; /* Variances of back prevalence limits by age */
8783: int i1, k, nres, j ;
8784:
8785: strcpy(fileresvbl,"VBL_");
8786: strcat(fileresvbl,fileresu);
8787: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8788: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8789: exit(0);
8790: }
8791: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8792: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8793:
8794:
8795: i1=pow(2,cptcoveff);
8796: if (cptcovn < 1){i1=1;}
8797:
8798: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8799: for(k=1; k<=i1;k++){
8800: if(i1 != 1 && TKresult[nres]!= k)
8801: continue;
8802: fprintf(ficresvbl,"\n#****** ");
8803: printf("\n#****** ");
8804: fprintf(ficlog,"\n#****** ");
8805: for(j=1;j<=cptcoveff;j++) {
8806: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8807: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8808: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8809: }
8810: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8811: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8812: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8813: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8814: }
8815: fprintf(ficresvbl,"******\n");
8816: printf("******\n");
8817: fprintf(ficlog,"******\n");
8818:
8819: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8820: oldm=oldms;savm=savms;
8821:
8822: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8823: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8824: /*}*/
8825: }
8826:
8827: fclose(ficresvbl);
8828: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8829: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8830:
8831: } /* End of varbprlim */
8832:
1.126 brouard 8833: /************** Forecasting *****not tested NB*************/
1.227 brouard 8834: /* 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 8835:
1.227 brouard 8836: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8837: /* int *popage; */
8838: /* double calagedatem, agelim, kk1, kk2; */
8839: /* double *popeffectif,*popcount; */
8840: /* double ***p3mat,***tabpop,***tabpopprev; */
8841: /* /\* double ***mobaverage; *\/ */
8842: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8843:
1.227 brouard 8844: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8845: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8846: /* agelim=AGESUP; */
8847: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8848:
1.227 brouard 8849: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8850:
8851:
1.227 brouard 8852: /* strcpy(filerespop,"POP_"); */
8853: /* strcat(filerespop,fileresu); */
8854: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8855: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8856: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8857: /* } */
8858: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8859: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8860:
1.227 brouard 8861: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8862:
1.227 brouard 8863: /* /\* if (mobilav!=0) { *\/ */
8864: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8865: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8866: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8867: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8868: /* /\* } *\/ */
8869: /* /\* } *\/ */
1.126 brouard 8870:
1.227 brouard 8871: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8872: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8873:
1.227 brouard 8874: /* agelim=AGESUP; */
1.126 brouard 8875:
1.227 brouard 8876: /* hstepm=1; */
8877: /* hstepm=hstepm/stepm; */
1.218 brouard 8878:
1.227 brouard 8879: /* if (popforecast==1) { */
8880: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8881: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8882: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8883: /* } */
8884: /* popage=ivector(0,AGESUP); */
8885: /* popeffectif=vector(0,AGESUP); */
8886: /* popcount=vector(0,AGESUP); */
1.126 brouard 8887:
1.227 brouard 8888: /* i=1; */
8889: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8890:
1.227 brouard 8891: /* imx=i; */
8892: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8893: /* } */
1.218 brouard 8894:
1.227 brouard 8895: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8896: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8897: /* k=k+1; */
8898: /* fprintf(ficrespop,"\n#******"); */
8899: /* for(j=1;j<=cptcoveff;j++) { */
8900: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8901: /* } */
8902: /* fprintf(ficrespop,"******\n"); */
8903: /* fprintf(ficrespop,"# Age"); */
8904: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8905: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8906:
1.227 brouard 8907: /* for (cpt=0; cpt<=0;cpt++) { */
8908: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8909:
1.227 brouard 8910: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8911: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8912: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8913:
1.227 brouard 8914: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8915: /* oldm=oldms;savm=savms; */
8916: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8917:
1.227 brouard 8918: /* for (h=0; h<=nhstepm; h++){ */
8919: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8920: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8921: /* } */
8922: /* for(j=1; j<=nlstate+ndeath;j++) { */
8923: /* kk1=0.;kk2=0; */
8924: /* for(i=1; i<=nlstate;i++) { */
8925: /* if (mobilav==1) */
8926: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8927: /* else { */
8928: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8929: /* } */
8930: /* } */
8931: /* if (h==(int)(calagedatem+12*cpt)){ */
8932: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8933: /* /\*fprintf(ficrespop," %.3f", kk1); */
8934: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8935: /* } */
8936: /* } */
8937: /* for(i=1; i<=nlstate;i++){ */
8938: /* kk1=0.; */
8939: /* for(j=1; j<=nlstate;j++){ */
8940: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8941: /* } */
8942: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8943: /* } */
1.218 brouard 8944:
1.227 brouard 8945: /* if (h==(int)(calagedatem+12*cpt)) */
8946: /* for(j=1; j<=nlstate;j++) */
8947: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8948: /* } */
8949: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8950: /* } */
8951: /* } */
1.218 brouard 8952:
1.227 brouard 8953: /* /\******\/ */
1.218 brouard 8954:
1.227 brouard 8955: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8956: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8957: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8958: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8959: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8960:
1.227 brouard 8961: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8962: /* oldm=oldms;savm=savms; */
8963: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8964: /* for (h=0; h<=nhstepm; h++){ */
8965: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8966: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8967: /* } */
8968: /* for(j=1; j<=nlstate+ndeath;j++) { */
8969: /* kk1=0.;kk2=0; */
8970: /* for(i=1; i<=nlstate;i++) { */
8971: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8972: /* } */
8973: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8974: /* } */
8975: /* } */
8976: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8977: /* } */
8978: /* } */
8979: /* } */
8980: /* } */
1.218 brouard 8981:
1.227 brouard 8982: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8983:
1.227 brouard 8984: /* if (popforecast==1) { */
8985: /* free_ivector(popage,0,AGESUP); */
8986: /* free_vector(popeffectif,0,AGESUP); */
8987: /* free_vector(popcount,0,AGESUP); */
8988: /* } */
8989: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8990: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8991: /* fclose(ficrespop); */
8992: /* } /\* End of popforecast *\/ */
1.218 brouard 8993:
1.126 brouard 8994: int fileappend(FILE *fichier, char *optionfich)
8995: {
8996: if((fichier=fopen(optionfich,"a"))==NULL) {
8997: printf("Problem with file: %s\n", optionfich);
8998: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8999: return (0);
9000: }
9001: fflush(fichier);
9002: return (1);
9003: }
9004:
9005:
9006: /**************** function prwizard **********************/
9007: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9008: {
9009:
9010: /* Wizard to print covariance matrix template */
9011:
1.164 brouard 9012: char ca[32], cb[32];
9013: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9014: int numlinepar;
9015:
9016: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9017: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9018: for(i=1; i <=nlstate; i++){
9019: jj=0;
9020: for(j=1; j <=nlstate+ndeath; j++){
9021: if(j==i) continue;
9022: jj++;
9023: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9024: printf("%1d%1d",i,j);
9025: fprintf(ficparo,"%1d%1d",i,j);
9026: for(k=1; k<=ncovmodel;k++){
9027: /* printf(" %lf",param[i][j][k]); */
9028: /* fprintf(ficparo," %lf",param[i][j][k]); */
9029: printf(" 0.");
9030: fprintf(ficparo," 0.");
9031: }
9032: printf("\n");
9033: fprintf(ficparo,"\n");
9034: }
9035: }
9036: printf("# Scales (for hessian or gradient estimation)\n");
9037: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9038: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9039: for(i=1; i <=nlstate; i++){
9040: jj=0;
9041: for(j=1; j <=nlstate+ndeath; j++){
9042: if(j==i) continue;
9043: jj++;
9044: fprintf(ficparo,"%1d%1d",i,j);
9045: printf("%1d%1d",i,j);
9046: fflush(stdout);
9047: for(k=1; k<=ncovmodel;k++){
9048: /* printf(" %le",delti3[i][j][k]); */
9049: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9050: printf(" 0.");
9051: fprintf(ficparo," 0.");
9052: }
9053: numlinepar++;
9054: printf("\n");
9055: fprintf(ficparo,"\n");
9056: }
9057: }
9058: printf("# Covariance matrix\n");
9059: /* # 121 Var(a12)\n\ */
9060: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9061: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9062: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9063: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9064: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9065: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9066: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9067: fflush(stdout);
9068: fprintf(ficparo,"# Covariance matrix\n");
9069: /* # 121 Var(a12)\n\ */
9070: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9071: /* # ...\n\ */
9072: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9073:
9074: for(itimes=1;itimes<=2;itimes++){
9075: jj=0;
9076: for(i=1; i <=nlstate; i++){
9077: for(j=1; j <=nlstate+ndeath; j++){
9078: if(j==i) continue;
9079: for(k=1; k<=ncovmodel;k++){
9080: jj++;
9081: ca[0]= k+'a'-1;ca[1]='\0';
9082: if(itimes==1){
9083: printf("#%1d%1d%d",i,j,k);
9084: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9085: }else{
9086: printf("%1d%1d%d",i,j,k);
9087: fprintf(ficparo,"%1d%1d%d",i,j,k);
9088: /* printf(" %.5le",matcov[i][j]); */
9089: }
9090: ll=0;
9091: for(li=1;li <=nlstate; li++){
9092: for(lj=1;lj <=nlstate+ndeath; lj++){
9093: if(lj==li) continue;
9094: for(lk=1;lk<=ncovmodel;lk++){
9095: ll++;
9096: if(ll<=jj){
9097: cb[0]= lk +'a'-1;cb[1]='\0';
9098: if(ll<jj){
9099: if(itimes==1){
9100: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9101: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9102: }else{
9103: printf(" 0.");
9104: fprintf(ficparo," 0.");
9105: }
9106: }else{
9107: if(itimes==1){
9108: printf(" Var(%s%1d%1d)",ca,i,j);
9109: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9110: }else{
9111: printf(" 0.");
9112: fprintf(ficparo," 0.");
9113: }
9114: }
9115: }
9116: } /* end lk */
9117: } /* end lj */
9118: } /* end li */
9119: printf("\n");
9120: fprintf(ficparo,"\n");
9121: numlinepar++;
9122: } /* end k*/
9123: } /*end j */
9124: } /* end i */
9125: } /* end itimes */
9126:
9127: } /* end of prwizard */
9128: /******************* Gompertz Likelihood ******************************/
9129: double gompertz(double x[])
9130: {
1.302 brouard 9131: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9132: int i,n=0; /* n is the size of the sample */
9133:
1.220 brouard 9134: for (i=1;i<=imx ; i++) {
1.126 brouard 9135: sump=sump+weight[i];
9136: /* sump=sump+1;*/
9137: num=num+1;
9138: }
1.302 brouard 9139: L=0.0;
9140: /* agegomp=AGEGOMP; */
1.126 brouard 9141: /* for (i=0; i<=imx; i++)
9142: 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]);*/
9143:
1.302 brouard 9144: for (i=1;i<=imx ; i++) {
9145: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9146: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9147: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9148: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9149: * +
9150: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9151: */
9152: if (wav[i] > 1 || agedc[i] < AGESUP) {
9153: if (cens[i] == 1){
9154: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9155: } else if (cens[i] == 0){
1.126 brouard 9156: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9157: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9158: } else
9159: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9160: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9161: L=L+A*weight[i];
1.126 brouard 9162: /* 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 9163: }
9164: }
1.126 brouard 9165:
1.302 brouard 9166: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9167:
9168: return -2*L*num/sump;
9169: }
9170:
1.136 brouard 9171: #ifdef GSL
9172: /******************* Gompertz_f Likelihood ******************************/
9173: double gompertz_f(const gsl_vector *v, void *params)
9174: {
1.302 brouard 9175: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9176: double *x= (double *) v->data;
9177: int i,n=0; /* n is the size of the sample */
9178:
9179: for (i=0;i<=imx-1 ; i++) {
9180: sump=sump+weight[i];
9181: /* sump=sump+1;*/
9182: num=num+1;
9183: }
9184:
9185:
9186: /* for (i=0; i<=imx; i++)
9187: 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]);*/
9188: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9189: for (i=1;i<=imx ; i++)
9190: {
9191: if (cens[i] == 1 && wav[i]>1)
9192: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9193:
9194: if (cens[i] == 0 && wav[i]>1)
9195: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9196: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9197:
9198: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9199: if (wav[i] > 1 ) { /* ??? */
9200: LL=LL+A*weight[i];
9201: /* 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]);*/
9202: }
9203: }
9204:
9205: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9206: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9207:
9208: return -2*LL*num/sump;
9209: }
9210: #endif
9211:
1.126 brouard 9212: /******************* Printing html file ***********/
1.201 brouard 9213: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9214: int lastpass, int stepm, int weightopt, char model[],\
9215: int imx, double p[],double **matcov,double agemortsup){
9216: int i,k;
9217:
9218: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9219: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9220: for (i=1;i<=2;i++)
9221: 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 9222: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9223: fprintf(fichtm,"</ul>");
9224:
9225: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9226:
9227: 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>");
9228:
9229: for (k=agegomp;k<(agemortsup-2);k++)
9230: 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]);
9231:
9232:
9233: fflush(fichtm);
9234: }
9235:
9236: /******************* Gnuplot file **************/
1.201 brouard 9237: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9238:
9239: char dirfileres[132],optfileres[132];
1.164 brouard 9240:
1.126 brouard 9241: int ng;
9242:
9243:
9244: /*#ifdef windows */
9245: fprintf(ficgp,"cd \"%s\" \n",pathc);
9246: /*#endif */
9247:
9248:
9249: strcpy(dirfileres,optionfilefiname);
9250: strcpy(optfileres,"vpl");
1.199 brouard 9251: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9252: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9253: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9254: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9255: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9256:
9257: }
9258:
1.136 brouard 9259: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9260: {
1.126 brouard 9261:
1.136 brouard 9262: /*-------- data file ----------*/
9263: FILE *fic;
9264: char dummy[]=" ";
1.240 brouard 9265: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9266: int lstra;
1.136 brouard 9267: int linei, month, year,iout;
1.302 brouard 9268: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9269: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9270: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9271: char *stratrunc;
1.223 brouard 9272:
1.240 brouard 9273: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9274: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9275:
1.240 brouard 9276: for(v=1; v <=ncovcol;v++){
9277: DummyV[v]=0;
9278: FixedV[v]=0;
9279: }
9280: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9281: DummyV[v]=1;
9282: FixedV[v]=0;
9283: }
9284: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9285: DummyV[v]=0;
9286: FixedV[v]=1;
9287: }
9288: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9289: DummyV[v]=1;
9290: FixedV[v]=1;
9291: }
9292: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9293: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9294: 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]);
9295: }
1.126 brouard 9296:
1.136 brouard 9297: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9298: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9299: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9300: }
1.126 brouard 9301:
1.302 brouard 9302: /* Is it a BOM UTF-8 Windows file? */
9303: /* First data line */
9304: linei=0;
9305: while(fgets(line, MAXLINE, fic)) {
9306: noffset=0;
9307: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9308: {
9309: noffset=noffset+3;
9310: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9311: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9312: fflush(ficlog); return 1;
9313: }
9314: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9315: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9316: {
9317: noffset=noffset+2;
1.304 brouard 9318: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
9319: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 9320: fflush(ficlog); return 1;
9321: }
9322: else if( line[0] == 0 && line[1] == 0)
9323: {
9324: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9325: noffset=noffset+4;
1.304 brouard 9326: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
9327: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 9328: fflush(ficlog); return 1;
9329: }
9330: } else{
9331: ;/*printf(" Not a BOM file\n");*/
9332: }
9333: /* If line starts with a # it is a comment */
9334: if (line[noffset] == '#') {
9335: linei=linei+1;
9336: break;
9337: }else{
9338: break;
9339: }
9340: }
9341: fclose(fic);
9342: if((fic=fopen(datafile,"r"))==NULL) {
9343: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9344: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9345: }
9346: /* Not a Bom file */
9347:
1.136 brouard 9348: i=1;
9349: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9350: linei=linei+1;
9351: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9352: if(line[j] == '\t')
9353: line[j] = ' ';
9354: }
9355: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9356: ;
9357: };
9358: line[j+1]=0; /* Trims blanks at end of line */
9359: if(line[0]=='#'){
9360: fprintf(ficlog,"Comment line\n%s\n",line);
9361: printf("Comment line\n%s\n",line);
9362: continue;
9363: }
9364: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9365: strcpy(line, linetmp);
1.223 brouard 9366:
9367: /* Loops on waves */
9368: for (j=maxwav;j>=1;j--){
9369: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9370: cutv(stra, strb, line, ' ');
9371: if(strb[0]=='.') { /* Missing value */
9372: lval=-1;
9373: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9374: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9375: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9376: 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);
9377: 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);
9378: return 1;
9379: }
9380: }else{
9381: errno=0;
9382: /* what_kind_of_number(strb); */
9383: dval=strtod(strb,&endptr);
9384: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9385: /* if(strb != endptr && *endptr == '\0') */
9386: /* dval=dlval; */
9387: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9388: if( strb[0]=='\0' || (*endptr != '\0')){
9389: 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);
9390: 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);
9391: return 1;
9392: }
9393: cotqvar[j][iv][i]=dval;
9394: cotvar[j][ntv+iv][i]=dval;
9395: }
9396: strcpy(line,stra);
1.223 brouard 9397: }/* end loop ntqv */
1.225 brouard 9398:
1.223 brouard 9399: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9400: cutv(stra, strb, line, ' ');
9401: if(strb[0]=='.') { /* Missing value */
9402: lval=-1;
9403: }else{
9404: errno=0;
9405: lval=strtol(strb,&endptr,10);
9406: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9407: if( strb[0]=='\0' || (*endptr != '\0')){
9408: 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);
9409: 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);
9410: return 1;
9411: }
9412: }
9413: if(lval <-1 || lval >1){
9414: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9415: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9416: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9417: For example, for multinomial values like 1, 2 and 3,\n \
9418: build V1=0 V2=0 for the reference value (1),\n \
9419: V1=1 V2=0 for (2) \n \
1.223 brouard 9420: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9421: output of IMaCh is often meaningless.\n \
1.223 brouard 9422: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9423: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9424: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9425: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9426: For example, for multinomial values like 1, 2 and 3,\n \
9427: build V1=0 V2=0 for the reference value (1),\n \
9428: V1=1 V2=0 for (2) \n \
1.223 brouard 9429: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9430: output of IMaCh is often meaningless.\n \
1.223 brouard 9431: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9432: return 1;
9433: }
9434: cotvar[j][iv][i]=(double)(lval);
9435: strcpy(line,stra);
1.223 brouard 9436: }/* end loop ntv */
1.225 brouard 9437:
1.223 brouard 9438: /* Statuses at wave */
1.137 brouard 9439: cutv(stra, strb, line, ' ');
1.223 brouard 9440: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9441: lval=-1;
1.136 brouard 9442: }else{
1.238 brouard 9443: errno=0;
9444: lval=strtol(strb,&endptr,10);
9445: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9446: if( strb[0]=='\0' || (*endptr != '\0')){
9447: 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);
9448: 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);
9449: return 1;
9450: }
1.136 brouard 9451: }
1.225 brouard 9452:
1.136 brouard 9453: s[j][i]=lval;
1.225 brouard 9454:
1.223 brouard 9455: /* Date of Interview */
1.136 brouard 9456: strcpy(line,stra);
9457: cutv(stra, strb,line,' ');
1.169 brouard 9458: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9459: }
1.169 brouard 9460: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9461: month=99;
9462: year=9999;
1.136 brouard 9463: }else{
1.225 brouard 9464: 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);
9465: 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);
9466: return 1;
1.136 brouard 9467: }
9468: anint[j][i]= (double) year;
1.302 brouard 9469: mint[j][i]= (double)month;
9470: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9471: /* 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]); */
9472: /* 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]); */
9473: /* } */
1.136 brouard 9474: strcpy(line,stra);
1.223 brouard 9475: } /* End loop on waves */
1.225 brouard 9476:
1.223 brouard 9477: /* Date of death */
1.136 brouard 9478: cutv(stra, strb,line,' ');
1.169 brouard 9479: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9480: }
1.169 brouard 9481: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9482: month=99;
9483: year=9999;
9484: }else{
1.141 brouard 9485: 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 9486: 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);
9487: return 1;
1.136 brouard 9488: }
9489: andc[i]=(double) year;
9490: moisdc[i]=(double) month;
9491: strcpy(line,stra);
9492:
1.223 brouard 9493: /* Date of birth */
1.136 brouard 9494: cutv(stra, strb,line,' ');
1.169 brouard 9495: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9496: }
1.169 brouard 9497: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9498: month=99;
9499: year=9999;
9500: }else{
1.141 brouard 9501: 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);
9502: 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 9503: return 1;
1.136 brouard 9504: }
9505: if (year==9999) {
1.141 brouard 9506: 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);
9507: 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 9508: return 1;
9509:
1.136 brouard 9510: }
9511: annais[i]=(double)(year);
1.302 brouard 9512: moisnais[i]=(double)(month);
9513: for (j=1;j<=maxwav;j++){
9514: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9515: 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]);
9516: 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]);
9517: }
9518: }
9519:
1.136 brouard 9520: strcpy(line,stra);
1.225 brouard 9521:
1.223 brouard 9522: /* Sample weight */
1.136 brouard 9523: cutv(stra, strb,line,' ');
9524: errno=0;
9525: dval=strtod(strb,&endptr);
9526: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9527: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9528: 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 9529: fflush(ficlog);
9530: return 1;
9531: }
9532: weight[i]=dval;
9533: strcpy(line,stra);
1.225 brouard 9534:
1.223 brouard 9535: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9536: cutv(stra, strb, line, ' ');
9537: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9538: lval=-1;
1.311 brouard 9539: coqvar[iv][i]=NAN;
9540: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9541: }else{
1.225 brouard 9542: errno=0;
9543: /* what_kind_of_number(strb); */
9544: dval=strtod(strb,&endptr);
9545: /* if(strb != endptr && *endptr == '\0') */
9546: /* dval=dlval; */
9547: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9548: if( strb[0]=='\0' || (*endptr != '\0')){
9549: 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);
9550: 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);
9551: return 1;
9552: }
9553: coqvar[iv][i]=dval;
1.226 brouard 9554: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9555: }
9556: strcpy(line,stra);
9557: }/* end loop nqv */
1.136 brouard 9558:
1.223 brouard 9559: /* Covariate values */
1.136 brouard 9560: for (j=ncovcol;j>=1;j--){
9561: cutv(stra, strb,line,' ');
1.223 brouard 9562: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9563: lval=-1;
1.136 brouard 9564: }else{
1.225 brouard 9565: errno=0;
9566: lval=strtol(strb,&endptr,10);
9567: if( strb[0]=='\0' || (*endptr != '\0')){
9568: 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);
9569: 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);
9570: return 1;
9571: }
1.136 brouard 9572: }
9573: if(lval <-1 || lval >1){
1.225 brouard 9574: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9575: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9576: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9577: For example, for multinomial values like 1, 2 and 3,\n \
9578: build V1=0 V2=0 for the reference value (1),\n \
9579: V1=1 V2=0 for (2) \n \
1.136 brouard 9580: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9581: output of IMaCh is often meaningless.\n \
1.136 brouard 9582: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9583: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9584: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9585: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9586: For example, for multinomial values like 1, 2 and 3,\n \
9587: build V1=0 V2=0 for the reference value (1),\n \
9588: V1=1 V2=0 for (2) \n \
1.136 brouard 9589: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9590: output of IMaCh is often meaningless.\n \
1.136 brouard 9591: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9592: return 1;
1.136 brouard 9593: }
9594: covar[j][i]=(double)(lval);
9595: strcpy(line,stra);
9596: }
9597: lstra=strlen(stra);
1.225 brouard 9598:
1.136 brouard 9599: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9600: stratrunc = &(stra[lstra-9]);
9601: num[i]=atol(stratrunc);
9602: }
9603: else
9604: num[i]=atol(stra);
9605: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9606: 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;}*/
9607:
9608: i=i+1;
9609: } /* End loop reading data */
1.225 brouard 9610:
1.136 brouard 9611: *imax=i-1; /* Number of individuals */
9612: fclose(fic);
1.225 brouard 9613:
1.136 brouard 9614: return (0);
1.164 brouard 9615: /* endread: */
1.225 brouard 9616: printf("Exiting readdata: ");
9617: fclose(fic);
9618: return (1);
1.223 brouard 9619: }
1.126 brouard 9620:
1.234 brouard 9621: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9622: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9623: while (*p2 == ' ')
1.234 brouard 9624: p2++;
9625: /* while ((*p1++ = *p2++) !=0) */
9626: /* ; */
9627: /* do */
9628: /* while (*p2 == ' ') */
9629: /* p2++; */
9630: /* while (*p1++ == *p2++); */
9631: *stri=p2;
1.145 brouard 9632: }
9633:
1.235 brouard 9634: int decoderesult ( char resultline[], int nres)
1.230 brouard 9635: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9636: {
1.235 brouard 9637: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9638: char resultsav[MAXLINE];
1.234 brouard 9639: int resultmodel[MAXLINE];
9640: int modelresult[MAXLINE];
1.230 brouard 9641: char stra[80], strb[80], strc[80], strd[80],stre[80];
9642:
1.234 brouard 9643: removefirstspace(&resultline);
1.230 brouard 9644:
9645: if (strstr(resultline,"v") !=0){
9646: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9647: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9648: return 1;
9649: }
9650: trimbb(resultsav, resultline);
9651: if (strlen(resultsav) >1){
9652: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9653: }
1.253 brouard 9654: if(j == 0){ /* Resultline but no = */
9655: TKresult[nres]=0; /* Combination for the nresult and the model */
9656: return (0);
9657: }
1.234 brouard 9658: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.310 brouard 9659: printf("ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
9660: fprintf(ficlog,"ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.234 brouard 9661: }
9662: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9663: if(nbocc(resultsav,'=') >1){
9664: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
1.310 brouard 9665: resultsav= V4=1 V5=25.1 V3=0 stra= V5=25.1 V3=0 strb= V4=1 */
1.234 brouard 9666: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9667: }else
9668: cutl(strc,strd,resultsav,'=');
1.230 brouard 9669: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9670:
1.230 brouard 9671: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9672: Tvarsel[k]=atoi(strc);
9673: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9674: /* cptcovsel++; */
9675: if (nbocc(stra,'=') >0)
9676: strcpy(resultsav,stra); /* and analyzes it */
9677: }
1.235 brouard 9678: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9679: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9680: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9681: match=0;
1.236 brouard 9682: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9683: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9684: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9685: match=1;
9686: break;
9687: }
9688: }
9689: if(match == 0){
1.310 brouard 9690: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9691: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9692: return 1;
1.234 brouard 9693: }
9694: }
9695: }
1.235 brouard 9696: /* Checking for missing or useless values in comparison of current model needs */
9697: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9698: match=0;
1.235 brouard 9699: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9700: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9701: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9702: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9703: ++match;
9704: }
9705: }
9706: }
9707: if(match == 0){
9708: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 9709: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9710: return 1;
1.234 brouard 9711: }else if(match > 1){
9712: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 9713: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9714: return 1;
1.234 brouard 9715: }
9716: }
1.235 brouard 9717:
1.234 brouard 9718: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9719: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9720: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9721: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9722: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9723: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9724: /* 1 0 0 0 */
9725: /* 2 1 0 0 */
9726: /* 3 0 1 0 */
9727: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9728: /* 5 0 0 1 */
9729: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9730: /* 7 0 1 1 */
9731: /* 8 1 1 1 */
1.237 brouard 9732: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9733: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9734: /* V5*age V5 known which value for nres? */
9735: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9736: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9737: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9738: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9739: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9740: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9741: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9742: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9743: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9744: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9745: k4++;;
9746: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9747: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9748: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9749: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9750: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9751: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9752: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9753: k4q++;;
9754: }
9755: }
1.234 brouard 9756:
1.235 brouard 9757: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9758: return (0);
9759: }
1.235 brouard 9760:
1.230 brouard 9761: int decodemodel( char model[], int lastobs)
9762: /**< This routine decodes the model and returns:
1.224 brouard 9763: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9764: * - nagesqr = 1 if age*age in the model, otherwise 0.
9765: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9766: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9767: * - cptcovage number of covariates with age*products =2
9768: * - cptcovs number of simple covariates
9769: * - 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
9770: * which is a new column after the 9 (ncovcol) variables.
9771: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9772: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9773: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9774: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9775: */
1.136 brouard 9776: {
1.238 brouard 9777: int i, j, k, ks, v;
1.227 brouard 9778: int j1, k1, k2, k3, k4;
1.136 brouard 9779: char modelsav[80];
1.145 brouard 9780: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9781: char *strpt;
1.136 brouard 9782:
1.145 brouard 9783: /*removespace(model);*/
1.136 brouard 9784: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9785: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9786: if (strstr(model,"AGE") !=0){
1.192 brouard 9787: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9788: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9789: return 1;
9790: }
1.141 brouard 9791: if (strstr(model,"v") !=0){
9792: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9793: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9794: return 1;
9795: }
1.187 brouard 9796: strcpy(modelsav,model);
9797: if ((strpt=strstr(model,"age*age")) !=0){
9798: printf(" strpt=%s, model=%s\n",strpt, model);
9799: if(strpt != model){
1.234 brouard 9800: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9801: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9802: corresponding column of parameters.\n",model);
1.234 brouard 9803: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9804: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9805: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9806: return 1;
1.225 brouard 9807: }
1.187 brouard 9808: nagesqr=1;
9809: if (strstr(model,"+age*age") !=0)
1.234 brouard 9810: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9811: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9812: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9813: else
1.234 brouard 9814: substrchaine(modelsav, model, "age*age");
1.187 brouard 9815: }else
9816: nagesqr=0;
9817: if (strlen(modelsav) >1){
9818: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9819: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9820: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9821: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9822: * cst, age and age*age
9823: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9824: /* including age products which are counted in cptcovage.
9825: * but the covariates which are products must be treated
9826: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9827: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9828: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9829:
9830:
1.187 brouard 9831: /* Design
9832: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9833: * < ncovcol=8 >
9834: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9835: * k= 1 2 3 4 5 6 7 8
9836: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9837: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9838: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9839: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9840: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9841: * Tage[++cptcovage]=k
9842: * if products, new covar are created after ncovcol with k1
9843: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9844: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9845: * 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
9846: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9847: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9848: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9849: * < ncovcol=8 >
9850: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9851: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9852: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9853: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9854: * p Tprod[1]@2={ 6, 5}
9855: *p Tvard[1][1]@4= {7, 8, 5, 6}
9856: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9857: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9858: *How to reorganize?
9859: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9860: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9861: * {2, 1, 4, 8, 5, 6, 3, 7}
9862: * Struct []
9863: */
1.225 brouard 9864:
1.187 brouard 9865: /* This loop fills the array Tvar from the string 'model'.*/
9866: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9867: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9868: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9869: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9870: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9871: /* k=1 Tvar[1]=2 (from V2) */
9872: /* k=5 Tvar[5] */
9873: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9874: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9875: /* } */
1.198 brouard 9876: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9877: /*
9878: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9879: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9880: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9881: }
1.187 brouard 9882: cptcovage=0;
9883: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9884: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9885: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9886: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9887: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9888: /*scanf("%d",i);*/
9889: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9890: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9891: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9892: /* covar is not filled and then is empty */
9893: cptcovprod--;
9894: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9895: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9896: Typevar[k]=1; /* 1 for age product */
9897: cptcovage++; /* Sums the number of covariates which include age as a product */
9898: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9899: /*printf("stre=%s ", stre);*/
9900: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9901: cptcovprod--;
9902: cutl(stre,strb,strc,'V');
9903: Tvar[k]=atoi(stre);
9904: Typevar[k]=1; /* 1 for age product */
9905: cptcovage++;
9906: Tage[cptcovage]=k;
9907: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9908: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9909: cptcovn++;
9910: cptcovprodnoage++;k1++;
9911: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9912: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9913: because this model-covariate is a construction we invent a new column
9914: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9915: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9916: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9917: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9918: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9919: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9920: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9921: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9922: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9923: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9924: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9925: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9926: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9927: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9928: for (i=1; i<=lastobs;i++){
9929: /* Computes the new covariate which is a product of
9930: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9931: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9932: }
9933: } /* End age is not in the model */
9934: } /* End if model includes a product */
9935: else { /* no more sum */
9936: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9937: /* scanf("%d",i);*/
9938: cutl(strd,strc,strb,'V');
9939: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9940: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9941: Tvar[k]=atoi(strd);
9942: Typevar[k]=0; /* 0 for simple covariates */
9943: }
9944: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9945: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9946: scanf("%d",i);*/
1.187 brouard 9947: } /* end of loop + on total covariates */
9948: } /* end if strlen(modelsave == 0) age*age might exist */
9949: } /* end if strlen(model == 0) */
1.136 brouard 9950:
9951: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9952: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9953:
1.136 brouard 9954: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9955: printf("cptcovprod=%d ", cptcovprod);
9956: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9957: scanf("%d ",i);*/
9958:
9959:
1.230 brouard 9960: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9961: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9962: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9963: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9964: k = 1 2 3 4 5 6 7 8 9
9965: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9966: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9967: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9968: Dummy[k] 1 0 0 0 3 1 1 2 3
9969: Tmodelind[combination of covar]=k;
1.225 brouard 9970: */
9971: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9972: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9973: /* 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 9974: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9975: printf("Model=%s\n\
9976: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9977: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9978: 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);
9979: fprintf(ficlog,"Model=%s\n\
9980: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9981: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9982: 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 9983: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9984: 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 */
9985: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9986: Fixed[k]= 0;
9987: Dummy[k]= 0;
1.225 brouard 9988: ncoveff++;
1.232 brouard 9989: ncovf++;
1.234 brouard 9990: nsd++;
9991: modell[k].maintype= FTYPE;
9992: TvarsD[nsd]=Tvar[k];
9993: TvarsDind[nsd]=k;
9994: TvarF[ncovf]=Tvar[k];
9995: TvarFind[ncovf]=k;
9996: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9997: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9998: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9999: Fixed[k]= 0;
10000: Dummy[k]= 0;
10001: ncoveff++;
10002: ncovf++;
10003: modell[k].maintype= FTYPE;
10004: TvarF[ncovf]=Tvar[k];
10005: TvarFind[ncovf]=k;
1.230 brouard 10006: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10007: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10008: }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 10009: Fixed[k]= 0;
10010: Dummy[k]= 1;
1.230 brouard 10011: nqfveff++;
1.234 brouard 10012: modell[k].maintype= FTYPE;
10013: modell[k].subtype= FQ;
10014: nsq++;
10015: TvarsQ[nsq]=Tvar[k];
10016: TvarsQind[nsq]=k;
1.232 brouard 10017: ncovf++;
1.234 brouard 10018: TvarF[ncovf]=Tvar[k];
10019: TvarFind[ncovf]=k;
1.231 brouard 10020: 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 10021: 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 10022: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10023: Fixed[k]= 1;
10024: Dummy[k]= 0;
1.225 brouard 10025: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10026: modell[k].maintype= VTYPE;
10027: modell[k].subtype= VD;
10028: nsd++;
10029: TvarsD[nsd]=Tvar[k];
10030: TvarsDind[nsd]=k;
10031: ncovv++; /* Only simple time varying variables */
10032: TvarV[ncovv]=Tvar[k];
1.242 brouard 10033: 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 10034: 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 */
10035: 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 10036: 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);
10037: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10038: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10039: Fixed[k]= 1;
10040: Dummy[k]= 1;
10041: nqtveff++;
10042: modell[k].maintype= VTYPE;
10043: modell[k].subtype= VQ;
10044: ncovv++; /* Only simple time varying variables */
10045: nsq++;
10046: TvarsQ[nsq]=Tvar[k];
10047: TvarsQind[nsq]=k;
10048: TvarV[ncovv]=Tvar[k];
1.242 brouard 10049: 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 10050: 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 */
10051: 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 10052: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10053: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10054: 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 10055: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10056: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10057: ncova++;
10058: TvarA[ncova]=Tvar[k];
10059: TvarAind[ncova]=k;
1.231 brouard 10060: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10061: Fixed[k]= 2;
10062: Dummy[k]= 2;
10063: modell[k].maintype= ATYPE;
10064: modell[k].subtype= APFD;
10065: /* ncoveff++; */
1.227 brouard 10066: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10067: Fixed[k]= 2;
10068: Dummy[k]= 3;
10069: modell[k].maintype= ATYPE;
10070: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10071: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10072: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10073: Fixed[k]= 3;
10074: Dummy[k]= 2;
10075: modell[k].maintype= ATYPE;
10076: modell[k].subtype= APVD; /* Product age * varying dummy */
10077: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10078: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10079: Fixed[k]= 3;
10080: Dummy[k]= 3;
10081: modell[k].maintype= ATYPE;
10082: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10083: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10084: }
10085: }else if (Typevar[k] == 2) { /* product without age */
10086: k1=Tposprod[k];
10087: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10088: if(Tvard[k1][2] <=ncovcol){
10089: Fixed[k]= 1;
10090: Dummy[k]= 0;
10091: modell[k].maintype= FTYPE;
10092: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10093: ncovf++; /* Fixed variables without age */
10094: TvarF[ncovf]=Tvar[k];
10095: TvarFind[ncovf]=k;
10096: }else if(Tvard[k1][2] <=ncovcol+nqv){
10097: Fixed[k]= 0; /* or 2 ?*/
10098: Dummy[k]= 1;
10099: modell[k].maintype= FTYPE;
10100: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10101: ncovf++; /* Varying variables without age */
10102: TvarF[ncovf]=Tvar[k];
10103: TvarFind[ncovf]=k;
10104: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10105: Fixed[k]= 1;
10106: Dummy[k]= 0;
10107: modell[k].maintype= VTYPE;
10108: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10109: ncovv++; /* Varying variables without age */
10110: TvarV[ncovv]=Tvar[k];
10111: TvarVind[ncovv]=k;
10112: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10113: Fixed[k]= 1;
10114: Dummy[k]= 1;
10115: modell[k].maintype= VTYPE;
10116: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10117: ncovv++; /* Varying variables without age */
10118: TvarV[ncovv]=Tvar[k];
10119: TvarVind[ncovv]=k;
10120: }
1.227 brouard 10121: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10122: if(Tvard[k1][2] <=ncovcol){
10123: Fixed[k]= 0; /* or 2 ?*/
10124: Dummy[k]= 1;
10125: modell[k].maintype= FTYPE;
10126: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10127: ncovf++; /* Fixed variables without age */
10128: TvarF[ncovf]=Tvar[k];
10129: TvarFind[ncovf]=k;
10130: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10131: Fixed[k]= 1;
10132: Dummy[k]= 1;
10133: modell[k].maintype= VTYPE;
10134: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10135: ncovv++; /* Varying variables without age */
10136: TvarV[ncovv]=Tvar[k];
10137: TvarVind[ncovv]=k;
10138: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10139: Fixed[k]= 1;
10140: Dummy[k]= 1;
10141: modell[k].maintype= VTYPE;
10142: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10143: ncovv++; /* Varying variables without age */
10144: TvarV[ncovv]=Tvar[k];
10145: TvarVind[ncovv]=k;
10146: ncovv++; /* Varying variables without age */
10147: TvarV[ncovv]=Tvar[k];
10148: TvarVind[ncovv]=k;
10149: }
1.227 brouard 10150: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10151: if(Tvard[k1][2] <=ncovcol){
10152: Fixed[k]= 1;
10153: Dummy[k]= 1;
10154: modell[k].maintype= VTYPE;
10155: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10156: ncovv++; /* Varying variables without age */
10157: TvarV[ncovv]=Tvar[k];
10158: TvarVind[ncovv]=k;
10159: }else if(Tvard[k1][2] <=ncovcol+nqv){
10160: Fixed[k]= 1;
10161: Dummy[k]= 1;
10162: modell[k].maintype= VTYPE;
10163: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10164: ncovv++; /* Varying variables without age */
10165: TvarV[ncovv]=Tvar[k];
10166: TvarVind[ncovv]=k;
10167: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10168: Fixed[k]= 1;
10169: Dummy[k]= 0;
10170: modell[k].maintype= VTYPE;
10171: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10172: ncovv++; /* Varying variables without age */
10173: TvarV[ncovv]=Tvar[k];
10174: TvarVind[ncovv]=k;
10175: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10176: Fixed[k]= 1;
10177: Dummy[k]= 1;
10178: modell[k].maintype= VTYPE;
10179: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10180: ncovv++; /* Varying variables without age */
10181: TvarV[ncovv]=Tvar[k];
10182: TvarVind[ncovv]=k;
10183: }
1.227 brouard 10184: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10185: if(Tvard[k1][2] <=ncovcol){
10186: Fixed[k]= 1;
10187: Dummy[k]= 1;
10188: modell[k].maintype= VTYPE;
10189: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10190: ncovv++; /* Varying variables without age */
10191: TvarV[ncovv]=Tvar[k];
10192: TvarVind[ncovv]=k;
10193: }else if(Tvard[k1][2] <=ncovcol+nqv){
10194: Fixed[k]= 1;
10195: Dummy[k]= 1;
10196: modell[k].maintype= VTYPE;
10197: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10198: ncovv++; /* Varying variables without age */
10199: TvarV[ncovv]=Tvar[k];
10200: TvarVind[ncovv]=k;
10201: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10202: Fixed[k]= 1;
10203: Dummy[k]= 1;
10204: modell[k].maintype= VTYPE;
10205: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10206: ncovv++; /* Varying variables without age */
10207: TvarV[ncovv]=Tvar[k];
10208: TvarVind[ncovv]=k;
10209: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10210: Fixed[k]= 1;
10211: Dummy[k]= 1;
10212: modell[k].maintype= VTYPE;
10213: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10214: ncovv++; /* Varying variables without age */
10215: TvarV[ncovv]=Tvar[k];
10216: TvarVind[ncovv]=k;
10217: }
1.227 brouard 10218: }else{
1.240 brouard 10219: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10220: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10221: } /*end k1*/
1.225 brouard 10222: }else{
1.226 brouard 10223: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10224: 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 10225: }
1.227 brouard 10226: 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 10227: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10228: 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]);
10229: }
10230: /* Searching for doublons in the model */
10231: for(k1=1; k1<= cptcovt;k1++){
10232: for(k2=1; k2 <k1;k2++){
1.285 brouard 10233: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10234: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10235: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10236: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10237: 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]);
10238: 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 10239: return(1);
10240: }
10241: }else if (Typevar[k1] ==2){
10242: k3=Tposprod[k1];
10243: k4=Tposprod[k2];
10244: 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])) ){
10245: 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]]);
10246: 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);
10247: return(1);
10248: }
10249: }
1.227 brouard 10250: }
10251: }
1.225 brouard 10252: }
10253: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10254: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10255: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10256: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10257: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10258: /*endread:*/
1.225 brouard 10259: printf("Exiting decodemodel: ");
10260: return (1);
1.136 brouard 10261: }
10262:
1.169 brouard 10263: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10264: {/* Check ages at death */
1.136 brouard 10265: int i, m;
1.218 brouard 10266: int firstone=0;
10267:
1.136 brouard 10268: for (i=1; i<=imx; i++) {
10269: for(m=2; (m<= maxwav); m++) {
10270: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10271: anint[m][i]=9999;
1.216 brouard 10272: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10273: s[m][i]=-1;
1.136 brouard 10274: }
10275: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10276: *nberr = *nberr + 1;
1.218 brouard 10277: if(firstone == 0){
10278: firstone=1;
1.260 brouard 10279: 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 10280: }
1.262 brouard 10281: 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 10282: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10283: }
10284: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10285: (*nberr)++;
1.259 brouard 10286: 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 10287: 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 10288: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10289: }
10290: }
10291: }
10292:
10293: for (i=1; i<=imx; i++) {
10294: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10295: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10296: 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 10297: if (s[m][i] >= nlstate+1) {
1.169 brouard 10298: if(agedc[i]>0){
10299: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10300: agev[m][i]=agedc[i];
1.214 brouard 10301: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10302: }else {
1.136 brouard 10303: if ((int)andc[i]!=9999){
10304: nbwarn++;
10305: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10306: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10307: agev[m][i]=-1;
10308: }
10309: }
1.169 brouard 10310: } /* agedc > 0 */
1.214 brouard 10311: } /* end if */
1.136 brouard 10312: else if(s[m][i] !=9){ /* Standard case, age in fractional
10313: years but with the precision of a month */
10314: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10315: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10316: agev[m][i]=1;
10317: else if(agev[m][i] < *agemin){
10318: *agemin=agev[m][i];
10319: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10320: }
10321: else if(agev[m][i] >*agemax){
10322: *agemax=agev[m][i];
1.156 brouard 10323: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10324: }
10325: /*agev[m][i]=anint[m][i]-annais[i];*/
10326: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10327: } /* en if 9*/
1.136 brouard 10328: else { /* =9 */
1.214 brouard 10329: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10330: agev[m][i]=1;
10331: s[m][i]=-1;
10332: }
10333: }
1.214 brouard 10334: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10335: agev[m][i]=1;
1.214 brouard 10336: else{
10337: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10338: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10339: agev[m][i]=0;
10340: }
10341: } /* End for lastpass */
10342: }
1.136 brouard 10343:
10344: for (i=1; i<=imx; i++) {
10345: for(m=firstpass; (m<=lastpass); m++){
10346: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10347: (*nberr)++;
1.136 brouard 10348: 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);
10349: 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);
10350: return 1;
10351: }
10352: }
10353: }
10354:
10355: /*for (i=1; i<=imx; i++){
10356: for (m=firstpass; (m<lastpass); m++){
10357: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10358: }
10359:
10360: }*/
10361:
10362:
1.139 brouard 10363: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10364: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10365:
10366: return (0);
1.164 brouard 10367: /* endread:*/
1.136 brouard 10368: printf("Exiting calandcheckages: ");
10369: return (1);
10370: }
10371:
1.172 brouard 10372: #if defined(_MSC_VER)
10373: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10374: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10375: //#include "stdafx.h"
10376: //#include <stdio.h>
10377: //#include <tchar.h>
10378: //#include <windows.h>
10379: //#include <iostream>
10380: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10381:
10382: LPFN_ISWOW64PROCESS fnIsWow64Process;
10383:
10384: BOOL IsWow64()
10385: {
10386: BOOL bIsWow64 = FALSE;
10387:
10388: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10389: // (HANDLE, PBOOL);
10390:
10391: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10392:
10393: HMODULE module = GetModuleHandle(_T("kernel32"));
10394: const char funcName[] = "IsWow64Process";
10395: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10396: GetProcAddress(module, funcName);
10397:
10398: if (NULL != fnIsWow64Process)
10399: {
10400: if (!fnIsWow64Process(GetCurrentProcess(),
10401: &bIsWow64))
10402: //throw std::exception("Unknown error");
10403: printf("Unknown error\n");
10404: }
10405: return bIsWow64 != FALSE;
10406: }
10407: #endif
1.177 brouard 10408:
1.191 brouard 10409: void syscompilerinfo(int logged)
1.292 brouard 10410: {
10411: #include <stdint.h>
10412:
10413: /* #include "syscompilerinfo.h"*/
1.185 brouard 10414: /* command line Intel compiler 32bit windows, XP compatible:*/
10415: /* /GS /W3 /Gy
10416: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10417: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10418: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10419: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10420: */
10421: /* 64 bits */
1.185 brouard 10422: /*
10423: /GS /W3 /Gy
10424: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10425: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10426: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10427: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10428: /* Optimization are useless and O3 is slower than O2 */
10429: /*
10430: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10431: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10432: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10433: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10434: */
1.186 brouard 10435: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10436: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10437: /PDB:"visual studio
10438: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10439: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10440: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10441: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10442: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10443: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10444: uiAccess='false'"
10445: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10446: /NOLOGO /TLBID:1
10447: */
1.292 brouard 10448:
10449:
1.177 brouard 10450: #if defined __INTEL_COMPILER
1.178 brouard 10451: #if defined(__GNUC__)
10452: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10453: #endif
1.177 brouard 10454: #elif defined(__GNUC__)
1.179 brouard 10455: #ifndef __APPLE__
1.174 brouard 10456: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10457: #endif
1.177 brouard 10458: struct utsname sysInfo;
1.178 brouard 10459: int cross = CROSS;
10460: if (cross){
10461: printf("Cross-");
1.191 brouard 10462: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10463: }
1.174 brouard 10464: #endif
10465:
1.191 brouard 10466: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10467: #if defined(__clang__)
1.191 brouard 10468: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10469: #endif
10470: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10471: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10472: #endif
10473: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10474: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10475: #endif
10476: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10477: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10478: #endif
10479: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10480: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10481: #endif
10482: #if defined(_MSC_VER)
1.191 brouard 10483: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10484: #endif
10485: #if defined(__PGI)
1.191 brouard 10486: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10487: #endif
10488: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10489: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10490: #endif
1.191 brouard 10491: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10492:
1.167 brouard 10493: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10494: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10495: // Windows (x64 and x86)
1.191 brouard 10496: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10497: #elif __unix__ // all unices, not all compilers
10498: // Unix
1.191 brouard 10499: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10500: #elif __linux__
10501: // linux
1.191 brouard 10502: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10503: #elif __APPLE__
1.174 brouard 10504: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10505: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10506: #endif
10507:
10508: /* __MINGW32__ */
10509: /* __CYGWIN__ */
10510: /* __MINGW64__ */
10511: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10512: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10513: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10514: /* _WIN64 // Defined for applications for Win64. */
10515: /* _M_X64 // Defined for compilations that target x64 processors. */
10516: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10517:
1.167 brouard 10518: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10519: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10520: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10521: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10522: #else
1.191 brouard 10523: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10524: #endif
10525:
1.169 brouard 10526: #if defined(__GNUC__)
10527: # if defined(__GNUC_PATCHLEVEL__)
10528: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10529: + __GNUC_MINOR__ * 100 \
10530: + __GNUC_PATCHLEVEL__)
10531: # else
10532: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10533: + __GNUC_MINOR__ * 100)
10534: # endif
1.174 brouard 10535: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10536: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10537:
10538: if (uname(&sysInfo) != -1) {
10539: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10540: 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 10541: }
10542: else
10543: perror("uname() error");
1.179 brouard 10544: //#ifndef __INTEL_COMPILER
10545: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10546: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10547: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10548: #endif
1.169 brouard 10549: #endif
1.172 brouard 10550:
1.286 brouard 10551: // void main ()
1.172 brouard 10552: // {
1.169 brouard 10553: #if defined(_MSC_VER)
1.174 brouard 10554: if (IsWow64()){
1.191 brouard 10555: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10556: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10557: }
10558: else{
1.191 brouard 10559: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10560: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10561: }
1.172 brouard 10562: // printf("\nPress Enter to continue...");
10563: // getchar();
10564: // }
10565:
1.169 brouard 10566: #endif
10567:
1.167 brouard 10568:
1.219 brouard 10569: }
1.136 brouard 10570:
1.219 brouard 10571: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10572: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10573: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10574: /* double ftolpl = 1.e-10; */
1.180 brouard 10575: double age, agebase, agelim;
1.203 brouard 10576: double tot;
1.180 brouard 10577:
1.202 brouard 10578: strcpy(filerespl,"PL_");
10579: strcat(filerespl,fileresu);
10580: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10581: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10582: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10583: }
1.288 brouard 10584: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10585: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10586: pstamp(ficrespl);
1.288 brouard 10587: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10588: fprintf(ficrespl,"#Age ");
10589: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10590: fprintf(ficrespl,"\n");
1.180 brouard 10591:
1.219 brouard 10592: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10593:
1.219 brouard 10594: agebase=ageminpar;
10595: agelim=agemaxpar;
1.180 brouard 10596:
1.227 brouard 10597: /* i1=pow(2,ncoveff); */
1.234 brouard 10598: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10599: if (cptcovn < 1){i1=1;}
1.180 brouard 10600:
1.238 brouard 10601: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10602: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10603: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10604: continue;
1.235 brouard 10605:
1.238 brouard 10606: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10607: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10608: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10609: /* k=k+1; */
10610: /* to clean */
10611: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10612: fprintf(ficrespl,"#******");
10613: printf("#******");
10614: fprintf(ficlog,"#******");
10615: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10616: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10617: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10618: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10619: }
10620: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10621: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10622: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10623: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10624: }
10625: fprintf(ficrespl,"******\n");
10626: printf("******\n");
10627: fprintf(ficlog,"******\n");
10628: if(invalidvarcomb[k]){
10629: printf("\nCombination (%d) ignored because no case \n",k);
10630: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10631: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10632: continue;
10633: }
1.219 brouard 10634:
1.238 brouard 10635: fprintf(ficrespl,"#Age ");
10636: for(j=1;j<=cptcoveff;j++) {
10637: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10638: }
10639: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10640: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10641:
1.238 brouard 10642: for (age=agebase; age<=agelim; age++){
10643: /* for (age=agebase; age<=agebase; age++){ */
10644: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10645: fprintf(ficrespl,"%.0f ",age );
10646: for(j=1;j<=cptcoveff;j++)
10647: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10648: tot=0.;
10649: for(i=1; i<=nlstate;i++){
10650: tot += prlim[i][i];
10651: fprintf(ficrespl," %.5f", prlim[i][i]);
10652: }
10653: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10654: } /* Age */
10655: /* was end of cptcod */
10656: } /* cptcov */
10657: } /* nres */
1.219 brouard 10658: return 0;
1.180 brouard 10659: }
10660:
1.218 brouard 10661: 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 10662: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10663:
10664: /* Computes the back prevalence limit for any combination of covariate values
10665: * at any age between ageminpar and agemaxpar
10666: */
1.235 brouard 10667: int i, j, k, i1, nres=0 ;
1.217 brouard 10668: /* double ftolpl = 1.e-10; */
10669: double age, agebase, agelim;
10670: double tot;
1.218 brouard 10671: /* double ***mobaverage; */
10672: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10673:
10674: strcpy(fileresplb,"PLB_");
10675: strcat(fileresplb,fileresu);
10676: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10677: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10678: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10679: }
1.288 brouard 10680: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10681: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10682: pstamp(ficresplb);
1.288 brouard 10683: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10684: fprintf(ficresplb,"#Age ");
10685: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10686: fprintf(ficresplb,"\n");
10687:
1.218 brouard 10688:
10689: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10690:
10691: agebase=ageminpar;
10692: agelim=agemaxpar;
10693:
10694:
1.227 brouard 10695: i1=pow(2,cptcoveff);
1.218 brouard 10696: if (cptcovn < 1){i1=1;}
1.227 brouard 10697:
1.238 brouard 10698: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10699: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10700: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10701: continue;
10702: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10703: fprintf(ficresplb,"#******");
10704: printf("#******");
10705: fprintf(ficlog,"#******");
10706: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10707: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10708: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10709: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10710: }
10711: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10712: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10713: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10714: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10715: }
10716: fprintf(ficresplb,"******\n");
10717: printf("******\n");
10718: fprintf(ficlog,"******\n");
10719: if(invalidvarcomb[k]){
10720: printf("\nCombination (%d) ignored because no cases \n",k);
10721: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10722: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10723: continue;
10724: }
1.218 brouard 10725:
1.238 brouard 10726: fprintf(ficresplb,"#Age ");
10727: for(j=1;j<=cptcoveff;j++) {
10728: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10729: }
10730: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10731: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10732:
10733:
1.238 brouard 10734: for (age=agebase; age<=agelim; age++){
10735: /* for (age=agebase; age<=agebase; age++){ */
10736: if(mobilavproj > 0){
10737: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10738: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10739: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10740: }else if (mobilavproj == 0){
10741: 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);
10742: 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);
10743: exit(1);
10744: }else{
10745: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10746: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10747: /* printf("TOTOT\n"); */
10748: /* exit(1); */
1.238 brouard 10749: }
10750: fprintf(ficresplb,"%.0f ",age );
10751: for(j=1;j<=cptcoveff;j++)
10752: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10753: tot=0.;
10754: for(i=1; i<=nlstate;i++){
10755: tot += bprlim[i][i];
10756: fprintf(ficresplb," %.5f", bprlim[i][i]);
10757: }
10758: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10759: } /* Age */
10760: /* was end of cptcod */
1.255 brouard 10761: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10762: } /* end of any combination */
10763: } /* end of nres */
1.218 brouard 10764: /* hBijx(p, bage, fage); */
10765: /* fclose(ficrespijb); */
10766:
10767: return 0;
1.217 brouard 10768: }
1.218 brouard 10769:
1.180 brouard 10770: int hPijx(double *p, int bage, int fage){
10771: /*------------- h Pij x at various ages ------------*/
10772:
10773: int stepsize;
10774: int agelim;
10775: int hstepm;
10776: int nhstepm;
1.235 brouard 10777: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10778:
10779: double agedeb;
10780: double ***p3mat;
10781:
1.201 brouard 10782: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10783: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10784: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10785: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10786: }
10787: printf("Computing pij: result on file '%s' \n", filerespij);
10788: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10789:
10790: stepsize=(int) (stepm+YEARM-1)/YEARM;
10791: /*if (stepm<=24) stepsize=2;*/
10792:
10793: agelim=AGESUP;
10794: hstepm=stepsize*YEARM; /* Every year of age */
10795: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10796:
1.180 brouard 10797: /* hstepm=1; aff par mois*/
10798: pstamp(ficrespij);
10799: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10800: i1= pow(2,cptcoveff);
1.218 brouard 10801: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10802: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10803: /* k=k+1; */
1.235 brouard 10804: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10805: for(k=1; k<=i1;k++){
1.253 brouard 10806: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10807: continue;
1.183 brouard 10808: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10809: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10810: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10811: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10812: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10813: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10814: }
1.183 brouard 10815: fprintf(ficrespij,"******\n");
10816:
10817: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10818: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10819: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10820:
10821: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10822:
1.183 brouard 10823: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10824: oldm=oldms;savm=savms;
1.235 brouard 10825: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10826: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10827: for(i=1; i<=nlstate;i++)
10828: for(j=1; j<=nlstate+ndeath;j++)
10829: fprintf(ficrespij," %1d-%1d",i,j);
10830: fprintf(ficrespij,"\n");
10831: for (h=0; h<=nhstepm; h++){
10832: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10833: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10834: for(i=1; i<=nlstate;i++)
10835: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10836: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10837: fprintf(ficrespij,"\n");
10838: }
1.183 brouard 10839: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10840: fprintf(ficrespij,"\n");
10841: }
1.180 brouard 10842: /*}*/
10843: }
1.218 brouard 10844: return 0;
1.180 brouard 10845: }
1.218 brouard 10846:
10847: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10848: /*------------- h Bij x at various ages ------------*/
10849:
10850: int stepsize;
1.218 brouard 10851: /* int agelim; */
10852: int ageminl;
1.217 brouard 10853: int hstepm;
10854: int nhstepm;
1.238 brouard 10855: int h, i, i1, j, k, nres;
1.218 brouard 10856:
1.217 brouard 10857: double agedeb;
10858: double ***p3mat;
1.218 brouard 10859:
10860: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10861: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10862: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10863: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10864: }
10865: printf("Computing pij back: result on file '%s' \n", filerespijb);
10866: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10867:
10868: stepsize=(int) (stepm+YEARM-1)/YEARM;
10869: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10870:
1.218 brouard 10871: /* agelim=AGESUP; */
1.289 brouard 10872: ageminl=AGEINF; /* was 30 */
1.218 brouard 10873: hstepm=stepsize*YEARM; /* Every year of age */
10874: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10875:
10876: /* hstepm=1; aff par mois*/
10877: pstamp(ficrespijb);
1.255 brouard 10878: 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 10879: i1= pow(2,cptcoveff);
1.218 brouard 10880: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10881: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10882: /* k=k+1; */
1.238 brouard 10883: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10884: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10885: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10886: continue;
10887: fprintf(ficrespijb,"\n#****** ");
10888: for(j=1;j<=cptcoveff;j++)
10889: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10890: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10891: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10892: }
10893: fprintf(ficrespijb,"******\n");
1.264 brouard 10894: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10895: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10896: continue;
10897: }
10898:
10899: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10900: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10901: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10902: 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 */
10903: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10904:
10905: /* nhstepm=nhstepm*YEARM; aff par mois*/
10906:
1.266 brouard 10907: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10908: /* and memory limitations if stepm is small */
10909:
1.238 brouard 10910: /* oldm=oldms;savm=savms; */
10911: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10912: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10913: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10914: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10915: for(i=1; i<=nlstate;i++)
10916: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10917: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10918: fprintf(ficrespijb,"\n");
1.238 brouard 10919: for (h=0; h<=nhstepm; h++){
10920: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10921: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10922: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10923: for(i=1; i<=nlstate;i++)
10924: for(j=1; j<=nlstate+ndeath;j++)
10925: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10926: fprintf(ficrespijb,"\n");
10927: }
10928: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10929: fprintf(ficrespijb,"\n");
10930: } /* end age deb */
10931: } /* end combination */
10932: } /* end nres */
1.218 brouard 10933: return 0;
10934: } /* hBijx */
1.217 brouard 10935:
1.180 brouard 10936:
1.136 brouard 10937: /***********************************************/
10938: /**************** Main Program *****************/
10939: /***********************************************/
10940:
10941: int main(int argc, char *argv[])
10942: {
10943: #ifdef GSL
10944: const gsl_multimin_fminimizer_type *T;
10945: size_t iteri = 0, it;
10946: int rval = GSL_CONTINUE;
10947: int status = GSL_SUCCESS;
10948: double ssval;
10949: #endif
10950: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10951: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10952: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10953: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10954: int jj, ll, li, lj, lk;
1.136 brouard 10955: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10956: int num_filled;
1.136 brouard 10957: int itimes;
10958: int NDIM=2;
10959: int vpopbased=0;
1.235 brouard 10960: int nres=0;
1.258 brouard 10961: int endishere=0;
1.277 brouard 10962: int noffset=0;
1.274 brouard 10963: int ncurrv=0; /* Temporary variable */
10964:
1.164 brouard 10965: char ca[32], cb[32];
1.136 brouard 10966: /* FILE *fichtm; *//* Html File */
10967: /* FILE *ficgp;*/ /*Gnuplot File */
10968: struct stat info;
1.191 brouard 10969: double agedeb=0.;
1.194 brouard 10970:
10971: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10972: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10973:
1.165 brouard 10974: double fret;
1.191 brouard 10975: double dum=0.; /* Dummy variable */
1.136 brouard 10976: double ***p3mat;
1.218 brouard 10977: /* double ***mobaverage; */
1.164 brouard 10978:
10979: char line[MAXLINE];
1.197 brouard 10980: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10981:
1.234 brouard 10982: char modeltemp[MAXLINE];
1.230 brouard 10983: char resultline[MAXLINE];
10984:
1.136 brouard 10985: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10986: char *tok, *val; /* pathtot */
1.290 brouard 10987: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 10988: int c, h , cpt, c2;
1.191 brouard 10989: int jl=0;
10990: int i1, j1, jk, stepsize=0;
1.194 brouard 10991: int count=0;
10992:
1.164 brouard 10993: int *tab;
1.136 brouard 10994: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 10995: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
10996: /* double anprojf, mprojf, jprojf; */
10997: /* double jintmean,mintmean,aintmean; */
10998: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
10999: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11000: double yrfproj= 10.0; /* Number of years of forward projections */
11001: double yrbproj= 10.0; /* Number of years of backward projections */
11002: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11003: int mobilav=0,popforecast=0;
1.191 brouard 11004: int hstepm=0, nhstepm=0;
1.136 brouard 11005: int agemortsup;
11006: float sumlpop=0.;
11007: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11008: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11009:
1.191 brouard 11010: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11011: double ftolpl=FTOL;
11012: double **prlim;
1.217 brouard 11013: double **bprlim;
1.136 brouard 11014: double ***param; /* Matrix of parameters */
1.251 brouard 11015: double ***paramstart; /* Matrix of starting parameter values */
11016: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11017: double **matcov; /* Matrix of covariance */
1.203 brouard 11018: double **hess; /* Hessian matrix */
1.136 brouard 11019: double ***delti3; /* Scale */
11020: double *delti; /* Scale */
11021: double ***eij, ***vareij;
11022: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11023:
1.136 brouard 11024: double *epj, vepp;
1.164 brouard 11025:
1.273 brouard 11026: double dateprev1, dateprev2;
1.296 brouard 11027: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11028: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11029:
1.217 brouard 11030:
1.136 brouard 11031: double **ximort;
1.145 brouard 11032: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11033: int *dcwave;
11034:
1.164 brouard 11035: char z[1]="c";
1.136 brouard 11036:
11037: /*char *strt;*/
11038: char strtend[80];
1.126 brouard 11039:
1.164 brouard 11040:
1.126 brouard 11041: /* setlocale (LC_ALL, ""); */
11042: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11043: /* textdomain (PACKAGE); */
11044: /* setlocale (LC_CTYPE, ""); */
11045: /* setlocale (LC_MESSAGES, ""); */
11046:
11047: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11048: rstart_time = time(NULL);
11049: /* (void) gettimeofday(&start_time,&tzp);*/
11050: start_time = *localtime(&rstart_time);
1.126 brouard 11051: curr_time=start_time;
1.157 brouard 11052: /*tml = *localtime(&start_time.tm_sec);*/
11053: /* strcpy(strstart,asctime(&tml)); */
11054: strcpy(strstart,asctime(&start_time));
1.126 brouard 11055:
11056: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11057: /* tp.tm_sec = tp.tm_sec +86400; */
11058: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11059: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11060: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11061: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11062: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11063: /* strt=asctime(&tmg); */
11064: /* printf("Time(after) =%s",strstart); */
11065: /* (void) time (&time_value);
11066: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11067: * tm = *localtime(&time_value);
11068: * strstart=asctime(&tm);
11069: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11070: */
11071:
11072: nberr=0; /* Number of errors and warnings */
11073: nbwarn=0;
1.184 brouard 11074: #ifdef WIN32
11075: _getcwd(pathcd, size);
11076: #else
1.126 brouard 11077: getcwd(pathcd, size);
1.184 brouard 11078: #endif
1.191 brouard 11079: syscompilerinfo(0);
1.196 brouard 11080: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11081: if(argc <=1){
11082: printf("\nEnter the parameter file name: ");
1.205 brouard 11083: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11084: printf("ERROR Empty parameter file name\n");
11085: goto end;
11086: }
1.126 brouard 11087: i=strlen(pathr);
11088: if(pathr[i-1]=='\n')
11089: pathr[i-1]='\0';
1.156 brouard 11090: i=strlen(pathr);
1.205 brouard 11091: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11092: pathr[i-1]='\0';
1.205 brouard 11093: }
11094: i=strlen(pathr);
11095: if( i==0 ){
11096: printf("ERROR Empty parameter file name\n");
11097: goto end;
11098: }
11099: for (tok = pathr; tok != NULL; ){
1.126 brouard 11100: printf("Pathr |%s|\n",pathr);
11101: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11102: printf("val= |%s| pathr=%s\n",val,pathr);
11103: strcpy (pathtot, val);
11104: if(pathr[0] == '\0') break; /* Dirty */
11105: }
11106: }
1.281 brouard 11107: else if (argc<=2){
11108: strcpy(pathtot,argv[1]);
11109: }
1.126 brouard 11110: else{
11111: strcpy(pathtot,argv[1]);
1.281 brouard 11112: strcpy(z,argv[2]);
11113: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11114: }
11115: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11116: /*cygwin_split_path(pathtot,path,optionfile);
11117: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11118: /* cutv(path,optionfile,pathtot,'\\');*/
11119:
11120: /* Split argv[0], imach program to get pathimach */
11121: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11122: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11123: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11124: /* strcpy(pathimach,argv[0]); */
11125: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11126: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11127: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11128: #ifdef WIN32
11129: _chdir(path); /* Can be a relative path */
11130: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11131: #else
1.126 brouard 11132: chdir(path); /* Can be a relative path */
1.184 brouard 11133: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11134: #endif
11135: printf("Current directory %s!\n",pathcd);
1.126 brouard 11136: strcpy(command,"mkdir ");
11137: strcat(command,optionfilefiname);
11138: if((outcmd=system(command)) != 0){
1.169 brouard 11139: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11140: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11141: /* fclose(ficlog); */
11142: /* exit(1); */
11143: }
11144: /* if((imk=mkdir(optionfilefiname))<0){ */
11145: /* perror("mkdir"); */
11146: /* } */
11147:
11148: /*-------- arguments in the command line --------*/
11149:
1.186 brouard 11150: /* Main Log file */
1.126 brouard 11151: strcat(filelog, optionfilefiname);
11152: strcat(filelog,".log"); /* */
11153: if((ficlog=fopen(filelog,"w"))==NULL) {
11154: printf("Problem with logfile %s\n",filelog);
11155: goto end;
11156: }
11157: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11158: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11159: fprintf(ficlog,"\nEnter the parameter file name: \n");
11160: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11161: path=%s \n\
11162: optionfile=%s\n\
11163: optionfilext=%s\n\
1.156 brouard 11164: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11165:
1.197 brouard 11166: syscompilerinfo(1);
1.167 brouard 11167:
1.126 brouard 11168: printf("Local time (at start):%s",strstart);
11169: fprintf(ficlog,"Local time (at start): %s",strstart);
11170: fflush(ficlog);
11171: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11172: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11173:
11174: /* */
11175: strcpy(fileres,"r");
11176: strcat(fileres, optionfilefiname);
1.201 brouard 11177: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11178: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11179: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11180:
1.186 brouard 11181: /* Main ---------arguments file --------*/
1.126 brouard 11182:
11183: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11184: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11185: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11186: fflush(ficlog);
1.149 brouard 11187: /* goto end; */
11188: exit(70);
1.126 brouard 11189: }
11190:
11191: strcpy(filereso,"o");
1.201 brouard 11192: strcat(filereso,fileresu);
1.126 brouard 11193: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11194: printf("Problem with Output resultfile: %s\n", filereso);
11195: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11196: fflush(ficlog);
11197: goto end;
11198: }
1.278 brouard 11199: /*-------- Rewriting parameter file ----------*/
11200: strcpy(rfileres,"r"); /* "Rparameterfile */
11201: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11202: strcat(rfileres,"."); /* */
11203: strcat(rfileres,optionfilext); /* Other files have txt extension */
11204: if((ficres =fopen(rfileres,"w"))==NULL) {
11205: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11206: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11207: fflush(ficlog);
11208: goto end;
11209: }
11210: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11211:
1.278 brouard 11212:
1.126 brouard 11213: /* Reads comments: lines beginning with '#' */
11214: numlinepar=0;
1.277 brouard 11215: /* Is it a BOM UTF-8 Windows file? */
11216: /* First parameter line */
1.197 brouard 11217: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11218: noffset=0;
11219: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11220: {
11221: noffset=noffset+3;
11222: printf("# File is an UTF8 Bom.\n"); // 0xBF
11223: }
1.302 brouard 11224: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11225: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11226: {
11227: noffset=noffset+2;
11228: printf("# File is an UTF16BE BOM file\n");
11229: }
11230: else if( line[0] == 0 && line[1] == 0)
11231: {
11232: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11233: noffset=noffset+4;
11234: printf("# File is an UTF16BE BOM file\n");
11235: }
11236: } else{
11237: ;/*printf(" Not a BOM file\n");*/
11238: }
11239:
1.197 brouard 11240: /* If line starts with a # it is a comment */
1.277 brouard 11241: if (line[noffset] == '#') {
1.197 brouard 11242: numlinepar++;
11243: fputs(line,stdout);
11244: fputs(line,ficparo);
1.278 brouard 11245: fputs(line,ficres);
1.197 brouard 11246: fputs(line,ficlog);
11247: continue;
11248: }else
11249: break;
11250: }
11251: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11252: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11253: if (num_filled != 5) {
11254: printf("Should be 5 parameters\n");
1.283 brouard 11255: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11256: }
1.126 brouard 11257: numlinepar++;
1.197 brouard 11258: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11259: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11260: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11261: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11262: }
11263: /* Second parameter line */
11264: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11265: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11266: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11267: if (line[0] == '#') {
11268: numlinepar++;
1.283 brouard 11269: printf("%s",line);
11270: fprintf(ficres,"%s",line);
11271: fprintf(ficparo,"%s",line);
11272: fprintf(ficlog,"%s",line);
1.197 brouard 11273: continue;
11274: }else
11275: break;
11276: }
1.223 brouard 11277: 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", \
11278: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11279: if (num_filled != 11) {
11280: 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 11281: printf("but line=%s\n",line);
1.283 brouard 11282: 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");
11283: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11284: }
1.286 brouard 11285: if( lastpass > maxwav){
11286: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11287: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11288: fflush(ficlog);
11289: goto end;
11290: }
11291: 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 11292: 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 11293: 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 11294: 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 11295: }
1.203 brouard 11296: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11297: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11298: /* Third parameter line */
11299: while(fgets(line, MAXLINE, ficpar)) {
11300: /* If line starts with a # it is a comment */
11301: if (line[0] == '#') {
11302: numlinepar++;
1.283 brouard 11303: printf("%s",line);
11304: fprintf(ficres,"%s",line);
11305: fprintf(ficparo,"%s",line);
11306: fprintf(ficlog,"%s",line);
1.197 brouard 11307: continue;
11308: }else
11309: break;
11310: }
1.201 brouard 11311: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11312: if (num_filled != 1){
1.302 brouard 11313: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11314: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11315: model[0]='\0';
11316: goto end;
11317: }
11318: else{
11319: if (model[0]=='+'){
11320: for(i=1; i<=strlen(model);i++)
11321: modeltemp[i-1]=model[i];
1.201 brouard 11322: strcpy(model,modeltemp);
1.197 brouard 11323: }
11324: }
1.199 brouard 11325: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11326: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11327: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11328: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11329: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11330: }
11331: /* 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); */
11332: /* numlinepar=numlinepar+3; /\* In general *\/ */
11333: /* 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 11334: /* 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); */
11335: /* 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 11336: fflush(ficlog);
1.190 brouard 11337: /* if(model[0]=='#'|| model[0]== '\0'){ */
11338: if(model[0]=='#'){
1.279 brouard 11339: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11340: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11341: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11342: if(mle != -1){
1.279 brouard 11343: 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 11344: exit(1);
11345: }
11346: }
1.126 brouard 11347: while((c=getc(ficpar))=='#' && c!= EOF){
11348: ungetc(c,ficpar);
11349: fgets(line, MAXLINE, ficpar);
11350: numlinepar++;
1.195 brouard 11351: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11352: z[0]=line[1];
11353: }
11354: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11355: fputs(line, stdout);
11356: //puts(line);
1.126 brouard 11357: fputs(line,ficparo);
11358: fputs(line,ficlog);
11359: }
11360: ungetc(c,ficpar);
11361:
11362:
1.290 brouard 11363: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11364: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11365: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11366: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11367: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11368: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11369: v1+v2*age+v2*v3 makes cptcovn = 3
11370: */
11371: if (strlen(model)>1)
1.187 brouard 11372: 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 11373: else
1.187 brouard 11374: ncovmodel=2; /* Constant and age */
1.133 brouard 11375: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11376: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11377: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11378: 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);
11379: 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);
11380: fflush(stdout);
11381: fclose (ficlog);
11382: goto end;
11383: }
1.126 brouard 11384: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11385: delti=delti3[1][1];
11386: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11387: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11388: /* We could also provide initial parameters values giving by simple logistic regression
11389: * only one way, that is without matrix product. We will have nlstate maximizations */
11390: /* for(i=1;i<nlstate;i++){ */
11391: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11392: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11393: /* } */
1.126 brouard 11394: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11395: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11396: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11397: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11398: fclose (ficparo);
11399: fclose (ficlog);
11400: goto end;
11401: exit(0);
1.220 brouard 11402: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11403: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11404: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11405: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11406: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11407: matcov=matrix(1,npar,1,npar);
1.203 brouard 11408: hess=matrix(1,npar,1,npar);
1.220 brouard 11409: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11410: /* Read guessed parameters */
1.126 brouard 11411: /* Reads comments: lines beginning with '#' */
11412: while((c=getc(ficpar))=='#' && c!= EOF){
11413: ungetc(c,ficpar);
11414: fgets(line, MAXLINE, ficpar);
11415: numlinepar++;
1.141 brouard 11416: fputs(line,stdout);
1.126 brouard 11417: fputs(line,ficparo);
11418: fputs(line,ficlog);
11419: }
11420: ungetc(c,ficpar);
11421:
11422: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11423: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11424: for(i=1; i <=nlstate; i++){
1.234 brouard 11425: j=0;
1.126 brouard 11426: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11427: if(jj==i) continue;
11428: j++;
1.292 brouard 11429: while((c=getc(ficpar))=='#' && c!= EOF){
11430: ungetc(c,ficpar);
11431: fgets(line, MAXLINE, ficpar);
11432: numlinepar++;
11433: fputs(line,stdout);
11434: fputs(line,ficparo);
11435: fputs(line,ficlog);
11436: }
11437: ungetc(c,ficpar);
1.234 brouard 11438: fscanf(ficpar,"%1d%1d",&i1,&j1);
11439: if ((i1 != i) || (j1 != jj)){
11440: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11441: It might be a problem of design; if ncovcol and the model are correct\n \
11442: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11443: exit(1);
11444: }
11445: fprintf(ficparo,"%1d%1d",i1,j1);
11446: if(mle==1)
11447: printf("%1d%1d",i,jj);
11448: fprintf(ficlog,"%1d%1d",i,jj);
11449: for(k=1; k<=ncovmodel;k++){
11450: fscanf(ficpar," %lf",¶m[i][j][k]);
11451: if(mle==1){
11452: printf(" %lf",param[i][j][k]);
11453: fprintf(ficlog," %lf",param[i][j][k]);
11454: }
11455: else
11456: fprintf(ficlog," %lf",param[i][j][k]);
11457: fprintf(ficparo," %lf",param[i][j][k]);
11458: }
11459: fscanf(ficpar,"\n");
11460: numlinepar++;
11461: if(mle==1)
11462: printf("\n");
11463: fprintf(ficlog,"\n");
11464: fprintf(ficparo,"\n");
1.126 brouard 11465: }
11466: }
11467: fflush(ficlog);
1.234 brouard 11468:
1.251 brouard 11469: /* Reads parameters values */
1.126 brouard 11470: p=param[1][1];
1.251 brouard 11471: pstart=paramstart[1][1];
1.126 brouard 11472:
11473: /* Reads comments: lines beginning with '#' */
11474: while((c=getc(ficpar))=='#' && c!= EOF){
11475: ungetc(c,ficpar);
11476: fgets(line, MAXLINE, ficpar);
11477: numlinepar++;
1.141 brouard 11478: fputs(line,stdout);
1.126 brouard 11479: fputs(line,ficparo);
11480: fputs(line,ficlog);
11481: }
11482: ungetc(c,ficpar);
11483:
11484: for(i=1; i <=nlstate; i++){
11485: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11486: fscanf(ficpar,"%1d%1d",&i1,&j1);
11487: if ( (i1-i) * (j1-j) != 0){
11488: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11489: exit(1);
11490: }
11491: printf("%1d%1d",i,j);
11492: fprintf(ficparo,"%1d%1d",i1,j1);
11493: fprintf(ficlog,"%1d%1d",i1,j1);
11494: for(k=1; k<=ncovmodel;k++){
11495: fscanf(ficpar,"%le",&delti3[i][j][k]);
11496: printf(" %le",delti3[i][j][k]);
11497: fprintf(ficparo," %le",delti3[i][j][k]);
11498: fprintf(ficlog," %le",delti3[i][j][k]);
11499: }
11500: fscanf(ficpar,"\n");
11501: numlinepar++;
11502: printf("\n");
11503: fprintf(ficparo,"\n");
11504: fprintf(ficlog,"\n");
1.126 brouard 11505: }
11506: }
11507: fflush(ficlog);
1.234 brouard 11508:
1.145 brouard 11509: /* Reads covariance matrix */
1.126 brouard 11510: delti=delti3[1][1];
1.220 brouard 11511:
11512:
1.126 brouard 11513: /* 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 11514:
1.126 brouard 11515: /* Reads comments: lines beginning with '#' */
11516: while((c=getc(ficpar))=='#' && c!= EOF){
11517: ungetc(c,ficpar);
11518: fgets(line, MAXLINE, ficpar);
11519: numlinepar++;
1.141 brouard 11520: fputs(line,stdout);
1.126 brouard 11521: fputs(line,ficparo);
11522: fputs(line,ficlog);
11523: }
11524: ungetc(c,ficpar);
1.220 brouard 11525:
1.126 brouard 11526: matcov=matrix(1,npar,1,npar);
1.203 brouard 11527: hess=matrix(1,npar,1,npar);
1.131 brouard 11528: for(i=1; i <=npar; i++)
11529: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11530:
1.194 brouard 11531: /* Scans npar lines */
1.126 brouard 11532: for(i=1; i <=npar; i++){
1.226 brouard 11533: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11534: if(count != 3){
1.226 brouard 11535: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11536: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11537: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11538: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11539: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11540: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11541: exit(1);
1.220 brouard 11542: }else{
1.226 brouard 11543: if(mle==1)
11544: printf("%1d%1d%d",i1,j1,jk);
11545: }
11546: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11547: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11548: for(j=1; j <=i; j++){
1.226 brouard 11549: fscanf(ficpar," %le",&matcov[i][j]);
11550: if(mle==1){
11551: printf(" %.5le",matcov[i][j]);
11552: }
11553: fprintf(ficlog," %.5le",matcov[i][j]);
11554: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11555: }
11556: fscanf(ficpar,"\n");
11557: numlinepar++;
11558: if(mle==1)
1.220 brouard 11559: printf("\n");
1.126 brouard 11560: fprintf(ficlog,"\n");
11561: fprintf(ficparo,"\n");
11562: }
1.194 brouard 11563: /* End of read covariance matrix npar lines */
1.126 brouard 11564: for(i=1; i <=npar; i++)
11565: for(j=i+1;j<=npar;j++)
1.226 brouard 11566: matcov[i][j]=matcov[j][i];
1.126 brouard 11567:
11568: if(mle==1)
11569: printf("\n");
11570: fprintf(ficlog,"\n");
11571:
11572: fflush(ficlog);
11573:
11574: } /* End of mle != -3 */
1.218 brouard 11575:
1.186 brouard 11576: /* Main data
11577: */
1.290 brouard 11578: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11579: /* num=lvector(1,n); */
11580: /* moisnais=vector(1,n); */
11581: /* annais=vector(1,n); */
11582: /* moisdc=vector(1,n); */
11583: /* andc=vector(1,n); */
11584: /* weight=vector(1,n); */
11585: /* agedc=vector(1,n); */
11586: /* cod=ivector(1,n); */
11587: /* for(i=1;i<=n;i++){ */
11588: num=lvector(firstobs,lastobs);
11589: moisnais=vector(firstobs,lastobs);
11590: annais=vector(firstobs,lastobs);
11591: moisdc=vector(firstobs,lastobs);
11592: andc=vector(firstobs,lastobs);
11593: weight=vector(firstobs,lastobs);
11594: agedc=vector(firstobs,lastobs);
11595: cod=ivector(firstobs,lastobs);
11596: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11597: num[i]=0;
11598: moisnais[i]=0;
11599: annais[i]=0;
11600: moisdc[i]=0;
11601: andc[i]=0;
11602: agedc[i]=0;
11603: cod[i]=0;
11604: weight[i]=1.0; /* Equal weights, 1 by default */
11605: }
1.290 brouard 11606: mint=matrix(1,maxwav,firstobs,lastobs);
11607: anint=matrix(1,maxwav,firstobs,lastobs);
11608: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11609: tab=ivector(1,NCOVMAX);
1.144 brouard 11610: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11611: 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 11612:
1.136 brouard 11613: /* Reads data from file datafile */
11614: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11615: goto end;
11616:
11617: /* Calculation of the number of parameters from char model */
1.234 brouard 11618: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11619: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11620: k=3 V4 Tvar[k=3]= 4 (from V4)
11621: k=2 V1 Tvar[k=2]= 1 (from V1)
11622: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11623: */
11624:
11625: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11626: TvarsDind=ivector(1,NCOVMAX); /* */
11627: TvarsD=ivector(1,NCOVMAX); /* */
11628: TvarsQind=ivector(1,NCOVMAX); /* */
11629: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11630: TvarF=ivector(1,NCOVMAX); /* */
11631: TvarFind=ivector(1,NCOVMAX); /* */
11632: TvarV=ivector(1,NCOVMAX); /* */
11633: TvarVind=ivector(1,NCOVMAX); /* */
11634: TvarA=ivector(1,NCOVMAX); /* */
11635: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11636: TvarFD=ivector(1,NCOVMAX); /* */
11637: TvarFDind=ivector(1,NCOVMAX); /* */
11638: TvarFQ=ivector(1,NCOVMAX); /* */
11639: TvarFQind=ivector(1,NCOVMAX); /* */
11640: TvarVD=ivector(1,NCOVMAX); /* */
11641: TvarVDind=ivector(1,NCOVMAX); /* */
11642: TvarVQ=ivector(1,NCOVMAX); /* */
11643: TvarVQind=ivector(1,NCOVMAX); /* */
11644:
1.230 brouard 11645: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11646: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11647: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11648: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11649: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11650: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11651: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11652: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11653: */
11654: /* For model-covariate k tells which data-covariate to use but
11655: because this model-covariate is a construction we invent a new column
11656: ncovcol + k1
11657: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11658: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11659: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11660: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11661: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11662: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11663: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11664: */
1.145 brouard 11665: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11666: 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 11667: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11668: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11669: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11670: 4 covariates (3 plus signs)
11671: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11672: */
1.230 brouard 11673: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11674: * individual dummy, fixed or varying:
11675: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11676: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11677: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11678: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11679: * Tmodelind[1]@9={9,0,3,2,}*/
11680: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11681: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11682: * individual quantitative, fixed or varying:
11683: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11684: * 3, 1, 0, 0, 0, 0, 0, 0},
11685: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11686: /* Main decodemodel */
11687:
1.187 brouard 11688:
1.223 brouard 11689: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11690: goto end;
11691:
1.137 brouard 11692: if((double)(lastobs-imx)/(double)imx > 1.10){
11693: nbwarn++;
11694: 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);
11695: 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);
11696: }
1.136 brouard 11697: /* if(mle==1){*/
1.137 brouard 11698: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11699: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11700: }
11701:
11702: /*-calculation of age at interview from date of interview and age at death -*/
11703: agev=matrix(1,maxwav,1,imx);
11704:
11705: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11706: goto end;
11707:
1.126 brouard 11708:
1.136 brouard 11709: agegomp=(int)agemin;
1.290 brouard 11710: free_vector(moisnais,firstobs,lastobs);
11711: free_vector(annais,firstobs,lastobs);
1.126 brouard 11712: /* free_matrix(mint,1,maxwav,1,n);
11713: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11714: /* free_vector(moisdc,1,n); */
11715: /* free_vector(andc,1,n); */
1.145 brouard 11716: /* */
11717:
1.126 brouard 11718: wav=ivector(1,imx);
1.214 brouard 11719: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11720: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11721: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11722: 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.*/
11723: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11724: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11725:
11726: /* Concatenates waves */
1.214 brouard 11727: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11728: Death is a valid wave (if date is known).
11729: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11730: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11731: and mw[mi+1][i]. dh depends on stepm.
11732: */
11733:
1.126 brouard 11734: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11735: /* Concatenates waves */
1.145 brouard 11736:
1.290 brouard 11737: free_vector(moisdc,firstobs,lastobs);
11738: free_vector(andc,firstobs,lastobs);
1.215 brouard 11739:
1.126 brouard 11740: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11741: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11742: ncodemax[1]=1;
1.145 brouard 11743: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11744: cptcoveff=0;
1.220 brouard 11745: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11746: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11747: }
11748:
11749: ncovcombmax=pow(2,cptcoveff);
11750: invalidvarcomb=ivector(1, ncovcombmax);
11751: for(i=1;i<ncovcombmax;i++)
11752: invalidvarcomb[i]=0;
11753:
1.211 brouard 11754: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11755: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11756: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11757:
1.200 brouard 11758: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11759: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11760: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11761: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11762: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11763: * (currently 0 or 1) in the data.
11764: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11765: * corresponding modality (h,j).
11766: */
11767:
1.145 brouard 11768: h=0;
11769: /*if (cptcovn > 0) */
1.126 brouard 11770: m=pow(2,cptcoveff);
11771:
1.144 brouard 11772: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11773: * For k=4 covariates, h goes from 1 to m=2**k
11774: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11775: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11776: * h\k 1 2 3 4
1.143 brouard 11777: *______________________________
11778: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11779: * 2 2 1 1 1
11780: * 3 i=2 1 2 1 1
11781: * 4 2 2 1 1
11782: * 5 i=3 1 i=2 1 2 1
11783: * 6 2 1 2 1
11784: * 7 i=4 1 2 2 1
11785: * 8 2 2 2 1
1.197 brouard 11786: * 9 i=5 1 i=3 1 i=2 1 2
11787: * 10 2 1 1 2
11788: * 11 i=6 1 2 1 2
11789: * 12 2 2 1 2
11790: * 13 i=7 1 i=4 1 2 2
11791: * 14 2 1 2 2
11792: * 15 i=8 1 2 2 2
11793: * 16 2 2 2 2
1.143 brouard 11794: */
1.212 brouard 11795: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11796: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11797: * and the value of each covariate?
11798: * V1=1, V2=1, V3=2, V4=1 ?
11799: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11800: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11801: * In order to get the real value in the data, we use nbcode
11802: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11803: * We are keeping this crazy system in order to be able (in the future?)
11804: * to have more than 2 values (0 or 1) for a covariate.
11805: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11806: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11807: * bbbbbbbb
11808: * 76543210
11809: * h-1 00000101 (6-1=5)
1.219 brouard 11810: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11811: * &
11812: * 1 00000001 (1)
1.219 brouard 11813: * 00000000 = 1 & ((h-1) >> (k-1))
11814: * +1= 00000001 =1
1.211 brouard 11815: *
11816: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11817: * h' 1101 =2^3+2^2+0x2^1+2^0
11818: * >>k' 11
11819: * & 00000001
11820: * = 00000001
11821: * +1 = 00000010=2 = codtabm(14,3)
11822: * Reverse h=6 and m=16?
11823: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11824: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11825: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11826: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11827: * V3=decodtabm(14,3,2**4)=2
11828: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11829: *(h-1) >> (j-1) 0011 =13 >> 2
11830: * &1 000000001
11831: * = 000000001
11832: * +1= 000000010 =2
11833: * 2211
11834: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11835: * V3=2
1.220 brouard 11836: * codtabm and decodtabm are identical
1.211 brouard 11837: */
11838:
1.145 brouard 11839:
11840: free_ivector(Ndum,-1,NCOVMAX);
11841:
11842:
1.126 brouard 11843:
1.186 brouard 11844: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11845: strcpy(optionfilegnuplot,optionfilefiname);
11846: if(mle==-3)
1.201 brouard 11847: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11848: strcat(optionfilegnuplot,".gp");
11849:
11850: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11851: printf("Problem with file %s",optionfilegnuplot);
11852: }
11853: else{
1.204 brouard 11854: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11855: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11856: //fprintf(ficgp,"set missing 'NaNq'\n");
11857: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11858: }
11859: /* fclose(ficgp);*/
1.186 brouard 11860:
11861:
11862: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11863:
11864: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11865: if(mle==-3)
1.201 brouard 11866: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11867: strcat(optionfilehtm,".htm");
11868: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11869: printf("Problem with %s \n",optionfilehtm);
11870: exit(0);
1.126 brouard 11871: }
11872:
11873: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11874: strcat(optionfilehtmcov,"-cov.htm");
11875: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11876: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11877: }
11878: else{
11879: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11880: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11881: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11882: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11883: }
11884:
1.213 brouard 11885: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本å¦è¡“振興会 (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \
1.204 brouard 11886: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11887: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11888: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11889: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11890: \n\
11891: <hr size=\"2\" color=\"#EC5E5E\">\
11892: <ul><li><h4>Parameter files</h4>\n\
11893: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11894: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11895: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11896: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11897: - Date and time at start: %s</ul>\n",\
11898: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11899: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11900: fileres,fileres,\
11901: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11902: fflush(fichtm);
11903:
11904: strcpy(pathr,path);
11905: strcat(pathr,optionfilefiname);
1.184 brouard 11906: #ifdef WIN32
11907: _chdir(optionfilefiname); /* Move to directory named optionfile */
11908: #else
1.126 brouard 11909: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11910: #endif
11911:
1.126 brouard 11912:
1.220 brouard 11913: /* Calculates basic frequencies. Computes observed prevalence at single age
11914: and for any valid combination of covariates
1.126 brouard 11915: and prints on file fileres'p'. */
1.251 brouard 11916: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11917: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11918:
11919: fprintf(fichtm,"\n");
1.286 brouard 11920: 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 11921: ftol, stepm);
11922: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11923: ncurrv=1;
11924: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11925: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11926: ncurrv=i;
11927: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11928: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11929: ncurrv=i;
11930: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11931: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11932: ncurrv=i;
11933: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11934: 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", \
11935: nlstate, ndeath, maxwav, mle, weightopt);
11936:
11937: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11938: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11939:
11940:
11941: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11942: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11943: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11944: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11945: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11946: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11947: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11948: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11949: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11950:
1.126 brouard 11951: /* For Powell, parameters are in a vector p[] starting at p[1]
11952: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11953: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11954:
11955: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11956: /* For mortality only */
1.126 brouard 11957: if (mle==-3){
1.136 brouard 11958: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11959: for(i=1;i<=NDIM;i++)
11960: for(j=1;j<=NDIM;j++)
11961: ximort[i][j]=0.;
1.186 brouard 11962: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11963: cens=ivector(firstobs,lastobs);
11964: ageexmed=vector(firstobs,lastobs);
11965: agecens=vector(firstobs,lastobs);
11966: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11967:
1.126 brouard 11968: for (i=1; i<=imx; i++){
11969: dcwave[i]=-1;
11970: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11971: if (s[m][i]>nlstate) {
11972: dcwave[i]=m;
11973: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11974: break;
11975: }
1.126 brouard 11976: }
1.226 brouard 11977:
1.126 brouard 11978: for (i=1; i<=imx; i++) {
11979: if (wav[i]>0){
1.226 brouard 11980: ageexmed[i]=agev[mw[1][i]][i];
11981: j=wav[i];
11982: agecens[i]=1.;
11983:
11984: if (ageexmed[i]> 1 && wav[i] > 0){
11985: agecens[i]=agev[mw[j][i]][i];
11986: cens[i]= 1;
11987: }else if (ageexmed[i]< 1)
11988: cens[i]= -1;
11989: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11990: cens[i]=0 ;
1.126 brouard 11991: }
11992: else cens[i]=-1;
11993: }
11994:
11995: for (i=1;i<=NDIM;i++) {
11996: for (j=1;j<=NDIM;j++)
1.226 brouard 11997: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11998: }
11999:
1.302 brouard 12000: p[1]=0.0268; p[NDIM]=0.083;
12001: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12002:
12003:
1.136 brouard 12004: #ifdef GSL
12005: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12006: #else
1.126 brouard 12007: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12008: #endif
1.201 brouard 12009: strcpy(filerespow,"POW-MORT_");
12010: strcat(filerespow,fileresu);
1.126 brouard 12011: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12012: printf("Problem with resultfile: %s\n", filerespow);
12013: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12014: }
1.136 brouard 12015: #ifdef GSL
12016: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12017: #else
1.126 brouard 12018: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12019: #endif
1.126 brouard 12020: /* for (i=1;i<=nlstate;i++)
12021: for(j=1;j<=nlstate+ndeath;j++)
12022: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12023: */
12024: fprintf(ficrespow,"\n");
1.136 brouard 12025: #ifdef GSL
12026: /* gsl starts here */
12027: T = gsl_multimin_fminimizer_nmsimplex;
12028: gsl_multimin_fminimizer *sfm = NULL;
12029: gsl_vector *ss, *x;
12030: gsl_multimin_function minex_func;
12031:
12032: /* Initial vertex size vector */
12033: ss = gsl_vector_alloc (NDIM);
12034:
12035: if (ss == NULL){
12036: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12037: }
12038: /* Set all step sizes to 1 */
12039: gsl_vector_set_all (ss, 0.001);
12040:
12041: /* Starting point */
1.126 brouard 12042:
1.136 brouard 12043: x = gsl_vector_alloc (NDIM);
12044:
12045: if (x == NULL){
12046: gsl_vector_free(ss);
12047: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12048: }
12049:
12050: /* Initialize method and iterate */
12051: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12052: /* gsl_vector_set(x, 0, 0.0268); */
12053: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12054: gsl_vector_set(x, 0, p[1]);
12055: gsl_vector_set(x, 1, p[2]);
12056:
12057: minex_func.f = &gompertz_f;
12058: minex_func.n = NDIM;
12059: minex_func.params = (void *)&p; /* ??? */
12060:
12061: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12062: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12063:
12064: printf("Iterations beginning .....\n\n");
12065: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12066:
12067: iteri=0;
12068: while (rval == GSL_CONTINUE){
12069: iteri++;
12070: status = gsl_multimin_fminimizer_iterate(sfm);
12071:
12072: if (status) printf("error: %s\n", gsl_strerror (status));
12073: fflush(0);
12074:
12075: if (status)
12076: break;
12077:
12078: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12079: ssval = gsl_multimin_fminimizer_size (sfm);
12080:
12081: if (rval == GSL_SUCCESS)
12082: printf ("converged to a local maximum at\n");
12083:
12084: printf("%5d ", iteri);
12085: for (it = 0; it < NDIM; it++){
12086: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12087: }
12088: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12089: }
12090:
12091: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12092:
12093: gsl_vector_free(x); /* initial values */
12094: gsl_vector_free(ss); /* inital step size */
12095: for (it=0; it<NDIM; it++){
12096: p[it+1]=gsl_vector_get(sfm->x,it);
12097: fprintf(ficrespow," %.12lf", p[it]);
12098: }
12099: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12100: #endif
12101: #ifdef POWELL
12102: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12103: #endif
1.126 brouard 12104: fclose(ficrespow);
12105:
1.203 brouard 12106: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12107:
12108: for(i=1; i <=NDIM; i++)
12109: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12110: matcov[i][j]=matcov[j][i];
1.126 brouard 12111:
12112: printf("\nCovariance matrix\n ");
1.203 brouard 12113: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12114: for(i=1; i <=NDIM; i++) {
12115: for(j=1;j<=NDIM;j++){
1.220 brouard 12116: printf("%f ",matcov[i][j]);
12117: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12118: }
1.203 brouard 12119: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12120: }
12121:
12122: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12123: for (i=1;i<=NDIM;i++) {
1.126 brouard 12124: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12125: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12126: }
1.302 brouard 12127: lsurv=vector(agegomp,AGESUP);
12128: lpop=vector(agegomp,AGESUP);
12129: tpop=vector(agegomp,AGESUP);
1.126 brouard 12130: lsurv[agegomp]=100000;
12131:
12132: for (k=agegomp;k<=AGESUP;k++) {
12133: agemortsup=k;
12134: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12135: }
12136:
12137: for (k=agegomp;k<agemortsup;k++)
12138: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12139:
12140: for (k=agegomp;k<agemortsup;k++){
12141: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12142: sumlpop=sumlpop+lpop[k];
12143: }
12144:
12145: tpop[agegomp]=sumlpop;
12146: for (k=agegomp;k<(agemortsup-3);k++){
12147: /* tpop[k+1]=2;*/
12148: tpop[k+1]=tpop[k]-lpop[k];
12149: }
12150:
12151:
12152: printf("\nAge lx qx dx Lx Tx e(x)\n");
12153: for (k=agegomp;k<(agemortsup-2);k++)
12154: 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]);
12155:
12156:
12157: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12158: ageminpar=50;
12159: agemaxpar=100;
1.194 brouard 12160: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12161: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12162: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12163: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12164: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12165: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12166: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12167: }else{
12168: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12169: 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 12170: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12171: }
1.201 brouard 12172: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12173: stepm, weightopt,\
12174: model,imx,p,matcov,agemortsup);
12175:
1.302 brouard 12176: free_vector(lsurv,agegomp,AGESUP);
12177: free_vector(lpop,agegomp,AGESUP);
12178: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12179: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12180: free_ivector(dcwave,firstobs,lastobs);
12181: free_vector(agecens,firstobs,lastobs);
12182: free_vector(ageexmed,firstobs,lastobs);
12183: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12184: #ifdef GSL
1.136 brouard 12185: #endif
1.186 brouard 12186: } /* Endof if mle==-3 mortality only */
1.205 brouard 12187: /* Standard */
12188: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12189: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12190: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12191: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12192: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12193: for (k=1; k<=npar;k++)
12194: printf(" %d %8.5f",k,p[k]);
12195: printf("\n");
1.205 brouard 12196: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12197: /* mlikeli uses func not funcone */
1.247 brouard 12198: /* for(i=1;i<nlstate;i++){ */
12199: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12200: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12201: /* } */
1.205 brouard 12202: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12203: }
12204: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12205: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12206: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12207: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12208: }
12209: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12210: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12211: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12212: for (k=1; k<=npar;k++)
12213: printf(" %d %8.5f",k,p[k]);
12214: printf("\n");
12215:
12216: /*--------- results files --------------*/
1.283 brouard 12217: /* 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 12218:
12219:
12220: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12221: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12222: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12223: for(i=1,jk=1; i <=nlstate; i++){
12224: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12225: if (k != i) {
12226: printf("%d%d ",i,k);
12227: fprintf(ficlog,"%d%d ",i,k);
12228: fprintf(ficres,"%1d%1d ",i,k);
12229: for(j=1; j <=ncovmodel; j++){
12230: printf("%12.7f ",p[jk]);
12231: fprintf(ficlog,"%12.7f ",p[jk]);
12232: fprintf(ficres,"%12.7f ",p[jk]);
12233: jk++;
12234: }
12235: printf("\n");
12236: fprintf(ficlog,"\n");
12237: fprintf(ficres,"\n");
12238: }
1.126 brouard 12239: }
12240: }
1.203 brouard 12241: if(mle != 0){
12242: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12243: ftolhess=ftol; /* Usually correct */
1.203 brouard 12244: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12245: 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");
12246: 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");
12247: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12248: for(k=1; k <=(nlstate+ndeath); k++){
12249: if (k != i) {
12250: printf("%d%d ",i,k);
12251: fprintf(ficlog,"%d%d ",i,k);
12252: for(j=1; j <=ncovmodel; j++){
12253: 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]));
12254: 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]));
12255: jk++;
12256: }
12257: printf("\n");
12258: fprintf(ficlog,"\n");
12259: }
12260: }
1.193 brouard 12261: }
1.203 brouard 12262: } /* end of hesscov and Wald tests */
1.225 brouard 12263:
1.203 brouard 12264: /* */
1.126 brouard 12265: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12266: printf("# Scales (for hessian or gradient estimation)\n");
12267: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12268: for(i=1,jk=1; i <=nlstate; i++){
12269: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12270: if (j!=i) {
12271: fprintf(ficres,"%1d%1d",i,j);
12272: printf("%1d%1d",i,j);
12273: fprintf(ficlog,"%1d%1d",i,j);
12274: for(k=1; k<=ncovmodel;k++){
12275: printf(" %.5e",delti[jk]);
12276: fprintf(ficlog," %.5e",delti[jk]);
12277: fprintf(ficres," %.5e",delti[jk]);
12278: jk++;
12279: }
12280: printf("\n");
12281: fprintf(ficlog,"\n");
12282: fprintf(ficres,"\n");
12283: }
1.126 brouard 12284: }
12285: }
12286:
12287: 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 12288: if(mle >= 1) /* To big for the screen */
1.126 brouard 12289: 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");
12290: 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");
12291: /* # 121 Var(a12)\n\ */
12292: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12293: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12294: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12295: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12296: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12297: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12298: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12299:
12300:
12301: /* Just to have a covariance matrix which will be more understandable
12302: even is we still don't want to manage dictionary of variables
12303: */
12304: for(itimes=1;itimes<=2;itimes++){
12305: jj=0;
12306: for(i=1; i <=nlstate; i++){
1.225 brouard 12307: for(j=1; j <=nlstate+ndeath; j++){
12308: if(j==i) continue;
12309: for(k=1; k<=ncovmodel;k++){
12310: jj++;
12311: ca[0]= k+'a'-1;ca[1]='\0';
12312: if(itimes==1){
12313: if(mle>=1)
12314: printf("#%1d%1d%d",i,j,k);
12315: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12316: fprintf(ficres,"#%1d%1d%d",i,j,k);
12317: }else{
12318: if(mle>=1)
12319: printf("%1d%1d%d",i,j,k);
12320: fprintf(ficlog,"%1d%1d%d",i,j,k);
12321: fprintf(ficres,"%1d%1d%d",i,j,k);
12322: }
12323: ll=0;
12324: for(li=1;li <=nlstate; li++){
12325: for(lj=1;lj <=nlstate+ndeath; lj++){
12326: if(lj==li) continue;
12327: for(lk=1;lk<=ncovmodel;lk++){
12328: ll++;
12329: if(ll<=jj){
12330: cb[0]= lk +'a'-1;cb[1]='\0';
12331: if(ll<jj){
12332: if(itimes==1){
12333: if(mle>=1)
12334: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12335: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12336: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12337: }else{
12338: if(mle>=1)
12339: printf(" %.5e",matcov[jj][ll]);
12340: fprintf(ficlog," %.5e",matcov[jj][ll]);
12341: fprintf(ficres," %.5e",matcov[jj][ll]);
12342: }
12343: }else{
12344: if(itimes==1){
12345: if(mle>=1)
12346: printf(" Var(%s%1d%1d)",ca,i,j);
12347: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12348: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12349: }else{
12350: if(mle>=1)
12351: printf(" %.7e",matcov[jj][ll]);
12352: fprintf(ficlog," %.7e",matcov[jj][ll]);
12353: fprintf(ficres," %.7e",matcov[jj][ll]);
12354: }
12355: }
12356: }
12357: } /* end lk */
12358: } /* end lj */
12359: } /* end li */
12360: if(mle>=1)
12361: printf("\n");
12362: fprintf(ficlog,"\n");
12363: fprintf(ficres,"\n");
12364: numlinepar++;
12365: } /* end k*/
12366: } /*end j */
1.126 brouard 12367: } /* end i */
12368: } /* end itimes */
12369:
12370: fflush(ficlog);
12371: fflush(ficres);
1.225 brouard 12372: while(fgets(line, MAXLINE, ficpar)) {
12373: /* If line starts with a # it is a comment */
12374: if (line[0] == '#') {
12375: numlinepar++;
12376: fputs(line,stdout);
12377: fputs(line,ficparo);
12378: fputs(line,ficlog);
1.299 brouard 12379: fputs(line,ficres);
1.225 brouard 12380: continue;
12381: }else
12382: break;
12383: }
12384:
1.209 brouard 12385: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12386: /* ungetc(c,ficpar); */
12387: /* fgets(line, MAXLINE, ficpar); */
12388: /* fputs(line,stdout); */
12389: /* fputs(line,ficparo); */
12390: /* } */
12391: /* ungetc(c,ficpar); */
1.126 brouard 12392:
12393: estepm=0;
1.209 brouard 12394: 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 12395:
12396: if (num_filled != 6) {
12397: 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);
12398: 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);
12399: goto end;
12400: }
12401: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12402: }
12403: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12404: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12405:
1.209 brouard 12406: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12407: if (estepm==0 || estepm < stepm) estepm=stepm;
12408: if (fage <= 2) {
12409: bage = ageminpar;
12410: fage = agemaxpar;
12411: }
12412:
12413: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12414: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12415: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12416:
1.186 brouard 12417: /* Other stuffs, more or less useful */
1.254 brouard 12418: while(fgets(line, MAXLINE, ficpar)) {
12419: /* If line starts with a # it is a comment */
12420: if (line[0] == '#') {
12421: numlinepar++;
12422: fputs(line,stdout);
12423: fputs(line,ficparo);
12424: fputs(line,ficlog);
1.299 brouard 12425: fputs(line,ficres);
1.254 brouard 12426: continue;
12427: }else
12428: break;
12429: }
12430:
12431: 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){
12432:
12433: if (num_filled != 7) {
12434: 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);
12435: 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);
12436: goto end;
12437: }
12438: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12439: 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);
12440: 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);
12441: 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 12442: }
1.254 brouard 12443:
12444: while(fgets(line, MAXLINE, ficpar)) {
12445: /* If line starts with a # it is a comment */
12446: if (line[0] == '#') {
12447: numlinepar++;
12448: fputs(line,stdout);
12449: fputs(line,ficparo);
12450: fputs(line,ficlog);
1.299 brouard 12451: fputs(line,ficres);
1.254 brouard 12452: continue;
12453: }else
12454: break;
1.126 brouard 12455: }
12456:
12457:
12458: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12459: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12460:
1.254 brouard 12461: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12462: if (num_filled != 1) {
12463: 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);
12464: 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);
12465: goto end;
12466: }
12467: printf("pop_based=%d\n",popbased);
12468: fprintf(ficlog,"pop_based=%d\n",popbased);
12469: fprintf(ficparo,"pop_based=%d\n",popbased);
12470: fprintf(ficres,"pop_based=%d\n",popbased);
12471: }
12472:
1.258 brouard 12473: /* Results */
1.307 brouard 12474: endishere=0;
1.258 brouard 12475: nresult=0;
1.308 brouard 12476: parameterline=0;
1.258 brouard 12477: do{
12478: if(!fgets(line, MAXLINE, ficpar)){
12479: endishere=1;
1.308 brouard 12480: parameterline=15;
1.258 brouard 12481: }else if (line[0] == '#') {
12482: /* If line starts with a # it is a comment */
1.254 brouard 12483: numlinepar++;
12484: fputs(line,stdout);
12485: fputs(line,ficparo);
12486: fputs(line,ficlog);
1.299 brouard 12487: fputs(line,ficres);
1.254 brouard 12488: continue;
1.258 brouard 12489: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12490: parameterline=11;
1.296 brouard 12491: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12492: parameterline=12;
1.307 brouard 12493: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12494: parameterline=13;
1.307 brouard 12495: }
1.258 brouard 12496: else{
12497: parameterline=14;
1.254 brouard 12498: }
1.308 brouard 12499: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12500: case 11:
1.296 brouard 12501: 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)){
12502: 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 12503: 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);
12504: 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);
12505: 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);
12506: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12507: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12508: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12509: prvforecast = 1;
12510: }
12511: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.302 brouard 12512: printf("prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12513: fprintf(ficlog,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12514: fprintf(ficres,"prevforecast=%d yearsfproj=%lf.2 mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12515: prvforecast = 2;
12516: }
12517: else {
12518: 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);
12519: 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);
12520: goto end;
1.258 brouard 12521: }
1.254 brouard 12522: break;
1.258 brouard 12523: case 12:
1.296 brouard 12524: 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)){
12525: 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);
12526: 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);
12527: 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);
12528: 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);
12529: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12530: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12531: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12532: prvbackcast = 1;
12533: }
12534: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.302 brouard 12535: printf("prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12536: fprintf(ficlog,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12537: fprintf(ficres,"prevbackcast=%d yearsbproj=%lf.2 mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12538: prvbackcast = 2;
12539: }
12540: else {
12541: 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);
12542: 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);
12543: goto end;
1.258 brouard 12544: }
1.230 brouard 12545: break;
1.258 brouard 12546: case 13:
1.307 brouard 12547: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12548: nresult++; /* Sum of resultlines */
12549: printf("Result %d: result:%s\n",nresult, resultline);
12550: if(nresult > MAXRESULTLINES){
12551: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINES,nresult,rfileres);
12552: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINES,nresult,rfileres);
12553: goto end;
12554: }
1.310 brouard 12555: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.307 brouard 12556: fprintf(ficparo,"result: %s\n",resultline);
12557: fprintf(ficres,"result: %s\n",resultline);
12558: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12559: } else
12560: goto end;
1.307 brouard 12561: break;
12562: case 14:
12563: printf("Error: Unknown command '%s'\n",line);
12564: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
12565: if(ncovmodel >=2 && nresult==0 ){
12566: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12567: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12568: }
1.307 brouard 12569: /* goto end; */
12570: break;
1.308 brouard 12571: case 15:
12572: printf("End of resultlines.\n");
12573: fprintf(ficlog,"End of resultlines.\n");
12574: break;
12575: default: /* parameterline =0 */
1.307 brouard 12576: nresult=1;
12577: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12578: } /* End switch parameterline */
12579: }while(endishere==0); /* End do */
1.126 brouard 12580:
1.230 brouard 12581: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12582: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12583:
12584: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12585: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12586: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12587: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12588: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12589: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12590: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12591: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12592: }else{
1.270 brouard 12593: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12594: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12595: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12596: if(prvforecast==1){
12597: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12598: jprojd=jproj1;
12599: mprojd=mproj1;
12600: anprojd=anproj1;
12601: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12602: jprojf=jproj2;
12603: mprojf=mproj2;
12604: anprojf=anproj2;
12605: } else if(prvforecast == 2){
12606: dateprojd=dateintmean;
12607: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12608: dateprojf=dateintmean+yrfproj;
12609: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12610: }
12611: if(prvbackcast==1){
12612: datebackd=(jback1+12*mback1+365*anback1)/365;
12613: jbackd=jback1;
12614: mbackd=mback1;
12615: anbackd=anback1;
12616: datebackf=(jback2+12*mback2+365*anback2)/365;
12617: jbackf=jback2;
12618: mbackf=mback2;
12619: anbackf=anback2;
12620: } else if(prvbackcast == 2){
12621: datebackd=dateintmean;
12622: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12623: datebackf=dateintmean-yrbproj;
12624: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12625: }
12626:
12627: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12628: }
12629: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12630: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12631: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12632:
1.225 brouard 12633: /*------------ free_vector -------------*/
12634: /* chdir(path); */
1.220 brouard 12635:
1.215 brouard 12636: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12637: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12638: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12639: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12640: free_lvector(num,firstobs,lastobs);
12641: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12642: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12643: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12644: fclose(ficparo);
12645: fclose(ficres);
1.220 brouard 12646:
12647:
1.186 brouard 12648: /* Other results (useful)*/
1.220 brouard 12649:
12650:
1.126 brouard 12651: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12652: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12653: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12654: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12655: fclose(ficrespl);
12656:
12657: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12658: /*#include "hpijx.h"*/
12659: hPijx(p, bage, fage);
1.145 brouard 12660: fclose(ficrespij);
1.227 brouard 12661:
1.220 brouard 12662: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12663: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12664: k=1;
1.126 brouard 12665: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12666:
1.269 brouard 12667: /* Prevalence for each covariate combination in probs[age][status][cov] */
12668: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12669: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12670: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12671: for(k=1;k<=ncovcombmax;k++)
12672: probs[i][j][k]=0.;
1.269 brouard 12673: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12674: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12675: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12676: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12677: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12678: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12679: for(k=1;k<=ncovcombmax;k++)
12680: mobaverages[i][j][k]=0.;
1.219 brouard 12681: mobaverage=mobaverages;
12682: if (mobilav!=0) {
1.235 brouard 12683: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12684: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12685: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12686: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12687: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12688: }
1.269 brouard 12689: } else if (mobilavproj !=0) {
1.235 brouard 12690: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12691: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12692: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12693: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12694: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12695: }
1.269 brouard 12696: }else{
12697: printf("Internal error moving average\n");
12698: fflush(stdout);
12699: exit(1);
1.219 brouard 12700: }
12701: }/* end if moving average */
1.227 brouard 12702:
1.126 brouard 12703: /*---------- Forecasting ------------------*/
1.296 brouard 12704: if(prevfcast==1){
12705: /* /\* if(stepm ==1){*\/ */
12706: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12707: /*This done previously after freqsummary.*/
12708: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12709: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12710:
12711: /* } else if (prvforecast==2){ */
12712: /* /\* if(stepm ==1){*\/ */
12713: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12714: /* } */
12715: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12716: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12717: }
1.269 brouard 12718:
1.296 brouard 12719: /* Prevbcasting */
12720: if(prevbcast==1){
1.219 brouard 12721: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12722: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12723: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12724:
12725: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12726:
12727: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12728:
1.219 brouard 12729: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12730: fclose(ficresplb);
12731:
1.222 brouard 12732: hBijx(p, bage, fage, mobaverage);
12733: fclose(ficrespijb);
1.219 brouard 12734:
1.296 brouard 12735: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12736: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12737: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12738: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12739: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12740: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12741:
12742:
1.269 brouard 12743: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12744:
12745:
1.269 brouard 12746: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12747: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12748: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12749: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12750: } /* end Prevbcasting */
1.268 brouard 12751:
1.186 brouard 12752:
12753: /* ------ Other prevalence ratios------------ */
1.126 brouard 12754:
1.215 brouard 12755: free_ivector(wav,1,imx);
12756: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12757: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12758: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12759:
12760:
1.127 brouard 12761: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12762:
1.201 brouard 12763: strcpy(filerese,"E_");
12764: strcat(filerese,fileresu);
1.126 brouard 12765: if((ficreseij=fopen(filerese,"w"))==NULL) {
12766: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12767: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12768: }
1.208 brouard 12769: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12770: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12771:
12772: pstamp(ficreseij);
1.219 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;
1.219 brouard 12781: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12782: printf("\n#****** ");
1.225 brouard 12783: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12784: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12785: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12786: }
12787: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12788: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12789: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12790: }
12791: fprintf(ficreseij,"******\n");
1.235 brouard 12792: printf("******\n");
1.219 brouard 12793:
12794: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12795: oldm=oldms;savm=savms;
1.235 brouard 12796: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12797:
1.219 brouard 12798: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12799: }
12800: fclose(ficreseij);
1.208 brouard 12801: printf("done evsij\n");fflush(stdout);
12802: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12803:
1.218 brouard 12804:
1.227 brouard 12805: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12806:
1.201 brouard 12807: strcpy(filerest,"T_");
12808: strcat(filerest,fileresu);
1.127 brouard 12809: if((ficrest=fopen(filerest,"w"))==NULL) {
12810: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12811: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12812: }
1.208 brouard 12813: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12814: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12815: strcpy(fileresstde,"STDE_");
12816: strcat(fileresstde,fileresu);
1.126 brouard 12817: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12818: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12819: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12820: }
1.227 brouard 12821: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12822: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12823:
1.201 brouard 12824: strcpy(filerescve,"CVE_");
12825: strcat(filerescve,fileresu);
1.126 brouard 12826: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12827: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12828: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12829: }
1.227 brouard 12830: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12831: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12832:
1.201 brouard 12833: strcpy(fileresv,"V_");
12834: strcat(fileresv,fileresu);
1.126 brouard 12835: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12836: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12837: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12838: }
1.227 brouard 12839: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12840: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12841:
1.235 brouard 12842: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12843: if (cptcovn < 1){i1=1;}
12844:
12845: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12846: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12847: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12848: continue;
1.242 brouard 12849: printf("\n#****** Result for:");
12850: fprintf(ficrest,"\n#****** Result for:");
12851: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12852: for(j=1;j<=cptcoveff;j++){
12853: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12854: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12855: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12856: }
1.235 brouard 12857: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12858: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12859: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12860: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12861: }
1.208 brouard 12862: fprintf(ficrest,"******\n");
1.227 brouard 12863: fprintf(ficlog,"******\n");
12864: printf("******\n");
1.208 brouard 12865:
12866: fprintf(ficresstdeij,"\n#****** ");
12867: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12868: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12869: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12870: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12871: }
1.235 brouard 12872: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12873: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12874: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12875: }
1.208 brouard 12876: fprintf(ficresstdeij,"******\n");
12877: fprintf(ficrescveij,"******\n");
12878:
12879: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12880: /* pstamp(ficresvij); */
1.225 brouard 12881: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12882: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12883: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12884: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12885: }
1.208 brouard 12886: fprintf(ficresvij,"******\n");
12887:
12888: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12889: oldm=oldms;savm=savms;
1.235 brouard 12890: printf(" cvevsij ");
12891: fprintf(ficlog, " cvevsij ");
12892: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12893: printf(" end cvevsij \n ");
12894: fprintf(ficlog, " end cvevsij \n ");
12895:
12896: /*
12897: */
12898: /* goto endfree; */
12899:
12900: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12901: pstamp(ficrest);
12902:
1.269 brouard 12903: epj=vector(1,nlstate+1);
1.208 brouard 12904: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12905: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12906: cptcod= 0; /* To be deleted */
12907: printf("varevsij vpopbased=%d \n",vpopbased);
12908: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12909: 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 12910: 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 ");
12911: if(vpopbased==1)
12912: 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);
12913: else
1.288 brouard 12914: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12915: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12916: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12917: fprintf(ficrest,"\n");
12918: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12919: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12920: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12921: for(age=bage; age <=fage ;age++){
1.235 brouard 12922: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12923: if (vpopbased==1) {
12924: if(mobilav ==0){
12925: for(i=1; i<=nlstate;i++)
12926: prlim[i][i]=probs[(int)age][i][k];
12927: }else{ /* mobilav */
12928: for(i=1; i<=nlstate;i++)
12929: prlim[i][i]=mobaverage[(int)age][i][k];
12930: }
12931: }
1.219 brouard 12932:
1.227 brouard 12933: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12934: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12935: /* printf(" age %4.0f ",age); */
12936: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12937: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12938: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12939: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12940: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12941: }
12942: epj[nlstate+1] +=epj[j];
12943: }
12944: /* printf(" age %4.0f \n",age); */
1.219 brouard 12945:
1.227 brouard 12946: for(i=1, vepp=0.;i <=nlstate;i++)
12947: for(j=1;j <=nlstate;j++)
12948: vepp += vareij[i][j][(int)age];
12949: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12950: for(j=1;j <=nlstate;j++){
12951: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12952: }
12953: fprintf(ficrest,"\n");
12954: }
1.208 brouard 12955: } /* End vpopbased */
1.269 brouard 12956: free_vector(epj,1,nlstate+1);
1.208 brouard 12957: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12958: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12959: printf("done selection\n");fflush(stdout);
12960: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12961:
1.235 brouard 12962: } /* End k selection */
1.227 brouard 12963:
12964: printf("done State-specific expectancies\n");fflush(stdout);
12965: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12966:
1.288 brouard 12967: /* variance-covariance of forward period prevalence*/
1.269 brouard 12968: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12969:
1.227 brouard 12970:
1.290 brouard 12971: free_vector(weight,firstobs,lastobs);
1.227 brouard 12972: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12973: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12974: free_matrix(anint,1,maxwav,firstobs,lastobs);
12975: free_matrix(mint,1,maxwav,firstobs,lastobs);
12976: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12977: free_ivector(tab,1,NCOVMAX);
12978: fclose(ficresstdeij);
12979: fclose(ficrescveij);
12980: fclose(ficresvij);
12981: fclose(ficrest);
12982: fclose(ficpar);
12983:
12984:
1.126 brouard 12985: /*---------- End : free ----------------*/
1.219 brouard 12986: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 12987: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
12988: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12989: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12990: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12991: } /* mle==-3 arrives here for freeing */
1.227 brouard 12992: /* endfree:*/
12993: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12994: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12995: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 12996: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
12997: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
12998: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
12999: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13000: free_matrix(matcov,1,npar,1,npar);
13001: free_matrix(hess,1,npar,1,npar);
13002: /*free_vector(delti,1,npar);*/
13003: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13004: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13005: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13006: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13007:
13008: free_ivector(ncodemax,1,NCOVMAX);
13009: free_ivector(ncodemaxwundef,1,NCOVMAX);
13010: free_ivector(Dummy,-1,NCOVMAX);
13011: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13012: free_ivector(DummyV,1,NCOVMAX);
13013: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13014: free_ivector(Typevar,-1,NCOVMAX);
13015: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13016: free_ivector(TvarsQ,1,NCOVMAX);
13017: free_ivector(TvarsQind,1,NCOVMAX);
13018: free_ivector(TvarsD,1,NCOVMAX);
13019: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13020: free_ivector(TvarFD,1,NCOVMAX);
13021: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13022: free_ivector(TvarF,1,NCOVMAX);
13023: free_ivector(TvarFind,1,NCOVMAX);
13024: free_ivector(TvarV,1,NCOVMAX);
13025: free_ivector(TvarVind,1,NCOVMAX);
13026: free_ivector(TvarA,1,NCOVMAX);
13027: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13028: free_ivector(TvarFQ,1,NCOVMAX);
13029: free_ivector(TvarFQind,1,NCOVMAX);
13030: free_ivector(TvarVD,1,NCOVMAX);
13031: free_ivector(TvarVDind,1,NCOVMAX);
13032: free_ivector(TvarVQ,1,NCOVMAX);
13033: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13034: free_ivector(Tvarsel,1,NCOVMAX);
13035: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13036: free_ivector(Tposprod,1,NCOVMAX);
13037: free_ivector(Tprod,1,NCOVMAX);
13038: free_ivector(Tvaraff,1,NCOVMAX);
13039: free_ivector(invalidvarcomb,1,ncovcombmax);
13040: free_ivector(Tage,1,NCOVMAX);
13041: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13042: free_ivector(TmodelInvind,1,NCOVMAX);
13043: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13044:
13045: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13046: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13047: fflush(fichtm);
13048: fflush(ficgp);
13049:
1.227 brouard 13050:
1.126 brouard 13051: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13052: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13053: 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 13054: }else{
13055: printf("End of Imach\n");
13056: fprintf(ficlog,"End of Imach\n");
13057: }
13058: printf("See log file on %s\n",filelog);
13059: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13060: /*(void) gettimeofday(&end_time,&tzp);*/
13061: rend_time = time(NULL);
13062: end_time = *localtime(&rend_time);
13063: /* tml = *localtime(&end_time.tm_sec); */
13064: strcpy(strtend,asctime(&end_time));
1.126 brouard 13065: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13066: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13067: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13068:
1.157 brouard 13069: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13070: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13071: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13072: /* printf("Total time was %d uSec.\n", total_usecs);*/
13073: /* if(fileappend(fichtm,optionfilehtm)){ */
13074: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13075: fclose(fichtm);
13076: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13077: fclose(fichtmcov);
13078: fclose(ficgp);
13079: fclose(ficlog);
13080: /*------ End -----------*/
1.227 brouard 13081:
1.281 brouard 13082:
13083: /* Executes gnuplot */
1.227 brouard 13084:
13085: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13086: #ifdef WIN32
1.227 brouard 13087: if (_chdir(pathcd) != 0)
13088: printf("Can't move to directory %s!\n",path);
13089: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13090: #else
1.227 brouard 13091: if(chdir(pathcd) != 0)
13092: printf("Can't move to directory %s!\n", path);
13093: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13094: #endif
1.126 brouard 13095: printf("Current directory %s!\n",pathcd);
13096: /*strcat(plotcmd,CHARSEPARATOR);*/
13097: sprintf(plotcmd,"gnuplot");
1.157 brouard 13098: #ifdef _WIN32
1.126 brouard 13099: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13100: #endif
13101: if(!stat(plotcmd,&info)){
1.158 brouard 13102: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13103: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13104: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13105: }else
13106: strcpy(pplotcmd,plotcmd);
1.157 brouard 13107: #ifdef __unix
1.126 brouard 13108: strcpy(plotcmd,GNUPLOTPROGRAM);
13109: if(!stat(plotcmd,&info)){
1.158 brouard 13110: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13111: }else
13112: strcpy(pplotcmd,plotcmd);
13113: #endif
13114: }else
13115: strcpy(pplotcmd,plotcmd);
13116:
13117: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13118: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13119: strcpy(pplotcmd,plotcmd);
1.227 brouard 13120:
1.126 brouard 13121: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13122: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13123: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13124: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13125: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13126: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13127: strcpy(plotcmd,pplotcmd);
13128: }
1.126 brouard 13129: }
1.158 brouard 13130: printf(" Successful, please wait...");
1.126 brouard 13131: while (z[0] != 'q') {
13132: /* chdir(path); */
1.154 brouard 13133: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13134: scanf("%s",z);
13135: /* if (z[0] == 'c') system("./imach"); */
13136: if (z[0] == 'e') {
1.158 brouard 13137: #ifdef __APPLE__
1.152 brouard 13138: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13139: #elif __linux
13140: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13141: #else
1.152 brouard 13142: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13143: #endif
13144: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13145: system(pplotcmd);
1.126 brouard 13146: }
13147: else if (z[0] == 'g') system(plotcmd);
13148: else if (z[0] == 'q') exit(0);
13149: }
1.227 brouard 13150: end:
1.126 brouard 13151: while (z[0] != 'q') {
1.195 brouard 13152: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13153: scanf("%s",z);
13154: }
1.283 brouard 13155: printf("End\n");
1.282 brouard 13156: exit(0);
1.126 brouard 13157: }
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