Annotation of imach/src/imach.c, revision 1.317
1.317 ! brouard 1: /* $Id: imach.c,v 1.316 2022/05/11 15:11:31 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.317 ! brouard 4: Revision 1.316 2022/05/11 15:11:31 brouard
! 5: Summary: r27
! 6:
1.316 brouard 7: Revision 1.315 2022/05/11 15:06:32 brouard
8: *** empty log message ***
9:
1.315 brouard 10: Revision 1.314 2022/04/13 17:43:09 brouard
11: * imach.c (Module): Adding link to text data files
12:
1.314 brouard 13: Revision 1.313 2022/04/11 15:57:42 brouard
14: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
15:
1.313 brouard 16: Revision 1.312 2022/04/05 21:24:39 brouard
17: *** empty log message ***
18:
1.312 brouard 19: Revision 1.311 2022/04/05 21:03:51 brouard
20: Summary: Fixed quantitative covariates
21:
22: Fixed covariates (dummy or quantitative)
23: with missing values have never been allowed but are ERRORS and
24: program quits. Standard deviations of fixed covariates were
25: wrongly computed. Mean and standard deviations of time varying
26: covariates are still not computed.
27:
1.311 brouard 28: Revision 1.310 2022/03/17 08:45:53 brouard
29: Summary: 99r25
30:
31: Improving detection of errors: result lines should be compatible with
32: the model.
33:
1.310 brouard 34: Revision 1.309 2021/05/20 12:39:14 brouard
35: Summary: Version 0.99r24
36:
1.309 brouard 37: Revision 1.308 2021/03/31 13:11:57 brouard
38: Summary: Version 0.99r23
39:
40:
41: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
42:
1.308 brouard 43: Revision 1.307 2021/03/08 18:11:32 brouard
44: Summary: 0.99r22 fixed bug on result:
45:
1.307 brouard 46: Revision 1.306 2021/02/20 15:44:02 brouard
47: Summary: Version 0.99r21
48:
49: * imach.c (Module): Fix bug on quitting after result lines!
50: (Module): Version 0.99r21
51:
1.306 brouard 52: Revision 1.305 2021/02/20 15:28:30 brouard
53: * imach.c (Module): Fix bug on quitting after result lines!
54:
1.305 brouard 55: Revision 1.304 2021/02/12 11:34:20 brouard
56: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
57:
1.304 brouard 58: Revision 1.303 2021/02/11 19:50:15 brouard
59: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
60:
1.303 brouard 61: Revision 1.302 2020/02/22 21:00:05 brouard
62: * (Module): imach.c Update mle=-3 (for computing Life expectancy
63: and life table from the data without any state)
64:
1.302 brouard 65: Revision 1.301 2019/06/04 13:51:20 brouard
66: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
67:
1.301 brouard 68: Revision 1.300 2019/05/22 19:09:45 brouard
69: Summary: version 0.99r19 of May 2019
70:
1.300 brouard 71: Revision 1.299 2019/05/22 18:37:08 brouard
72: Summary: Cleaned 0.99r19
73:
1.299 brouard 74: Revision 1.298 2019/05/22 18:19:56 brouard
75: *** empty log message ***
76:
1.298 brouard 77: Revision 1.297 2019/05/22 17:56:10 brouard
78: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
79:
1.297 brouard 80: Revision 1.296 2019/05/20 13:03:18 brouard
81: Summary: Projection syntax simplified
82:
83:
84: We can now start projections, forward or backward, from the mean date
85: of inteviews up to or down to a number of years of projection:
86: prevforecast=1 yearsfproj=15.3 mobil_average=0
87: or
88: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
89: or
90: prevbackcast=1 yearsbproj=12.3 mobil_average=1
91: or
92: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
93:
1.296 brouard 94: Revision 1.295 2019/05/18 09:52:50 brouard
95: Summary: doxygen tex bug
96:
1.295 brouard 97: Revision 1.294 2019/05/16 14:54:33 brouard
98: Summary: There was some wrong lines added
99:
1.294 brouard 100: Revision 1.293 2019/05/09 15:17:34 brouard
101: *** empty log message ***
102:
1.293 brouard 103: Revision 1.292 2019/05/09 14:17:20 brouard
104: Summary: Some updates
105:
1.292 brouard 106: Revision 1.291 2019/05/09 13:44:18 brouard
107: Summary: Before ncovmax
108:
1.291 brouard 109: Revision 1.290 2019/05/09 13:39:37 brouard
110: Summary: 0.99r18 unlimited number of individuals
111:
112: 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.
113:
1.290 brouard 114: Revision 1.289 2018/12/13 09:16:26 brouard
115: Summary: Bug for young ages (<-30) will be in r17
116:
1.289 brouard 117: Revision 1.288 2018/05/02 20:58:27 brouard
118: Summary: Some bugs fixed
119:
1.288 brouard 120: Revision 1.287 2018/05/01 17:57:25 brouard
121: Summary: Bug fixed by providing frequencies only for non missing covariates
122:
1.287 brouard 123: Revision 1.286 2018/04/27 14:27:04 brouard
124: Summary: some minor bugs
125:
1.286 brouard 126: Revision 1.285 2018/04/21 21:02:16 brouard
127: Summary: Some bugs fixed, valgrind tested
128:
1.285 brouard 129: Revision 1.284 2018/04/20 05:22:13 brouard
130: Summary: Computing mean and stdeviation of fixed quantitative variables
131:
1.284 brouard 132: Revision 1.283 2018/04/19 14:49:16 brouard
133: Summary: Some minor bugs fixed
134:
1.283 brouard 135: Revision 1.282 2018/02/27 22:50:02 brouard
136: *** empty log message ***
137:
1.282 brouard 138: Revision 1.281 2018/02/27 19:25:23 brouard
139: Summary: Adding second argument for quitting
140:
1.281 brouard 141: Revision 1.280 2018/02/21 07:58:13 brouard
142: Summary: 0.99r15
143:
144: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
145:
1.280 brouard 146: Revision 1.279 2017/07/20 13:35:01 brouard
147: Summary: temporary working
148:
1.279 brouard 149: Revision 1.278 2017/07/19 14:09:02 brouard
150: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
151:
1.278 brouard 152: Revision 1.277 2017/07/17 08:53:49 brouard
153: Summary: BOM files can be read now
154:
1.277 brouard 155: Revision 1.276 2017/06/30 15:48:31 brouard
156: Summary: Graphs improvements
157:
1.276 brouard 158: Revision 1.275 2017/06/30 13:39:33 brouard
159: Summary: Saito's color
160:
1.275 brouard 161: Revision 1.274 2017/06/29 09:47:08 brouard
162: Summary: Version 0.99r14
163:
1.274 brouard 164: Revision 1.273 2017/06/27 11:06:02 brouard
165: Summary: More documentation on projections
166:
1.273 brouard 167: Revision 1.272 2017/06/27 10:22:40 brouard
168: Summary: Color of backprojection changed from 6 to 5(yellow)
169:
1.272 brouard 170: Revision 1.271 2017/06/27 10:17:50 brouard
171: Summary: Some bug with rint
172:
1.271 brouard 173: Revision 1.270 2017/05/24 05:45:29 brouard
174: *** empty log message ***
175:
1.270 brouard 176: Revision 1.269 2017/05/23 08:39:25 brouard
177: Summary: Code into subroutine, cleanings
178:
1.269 brouard 179: Revision 1.268 2017/05/18 20:09:32 brouard
180: Summary: backprojection and confidence intervals of backprevalence
181:
1.268 brouard 182: Revision 1.267 2017/05/13 10:25:05 brouard
183: Summary: temporary save for backprojection
184:
1.267 brouard 185: Revision 1.266 2017/05/13 07:26:12 brouard
186: Summary: Version 0.99r13 (improvements and bugs fixed)
187:
1.266 brouard 188: Revision 1.265 2017/04/26 16:22:11 brouard
189: Summary: imach 0.99r13 Some bugs fixed
190:
1.265 brouard 191: Revision 1.264 2017/04/26 06:01:29 brouard
192: Summary: Labels in graphs
193:
1.264 brouard 194: Revision 1.263 2017/04/24 15:23:15 brouard
195: Summary: to save
196:
1.263 brouard 197: Revision 1.262 2017/04/18 16:48:12 brouard
198: *** empty log message ***
199:
1.262 brouard 200: Revision 1.261 2017/04/05 10:14:09 brouard
201: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
202:
1.261 brouard 203: Revision 1.260 2017/04/04 17:46:59 brouard
204: Summary: Gnuplot indexations fixed (humm)
205:
1.260 brouard 206: Revision 1.259 2017/04/04 13:01:16 brouard
207: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
208:
1.259 brouard 209: Revision 1.258 2017/04/03 10:17:47 brouard
210: Summary: Version 0.99r12
211:
212: Some cleanings, conformed with updated documentation.
213:
1.258 brouard 214: Revision 1.257 2017/03/29 16:53:30 brouard
215: Summary: Temp
216:
1.257 brouard 217: Revision 1.256 2017/03/27 05:50:23 brouard
218: Summary: Temporary
219:
1.256 brouard 220: Revision 1.255 2017/03/08 16:02:28 brouard
221: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
222:
1.255 brouard 223: Revision 1.254 2017/03/08 07:13:00 brouard
224: Summary: Fixing data parameter line
225:
1.254 brouard 226: Revision 1.253 2016/12/15 11:59:41 brouard
227: Summary: 0.99 in progress
228:
1.253 brouard 229: Revision 1.252 2016/09/15 21:15:37 brouard
230: *** empty log message ***
231:
1.252 brouard 232: Revision 1.251 2016/09/15 15:01:13 brouard
233: Summary: not working
234:
1.251 brouard 235: Revision 1.250 2016/09/08 16:07:27 brouard
236: Summary: continue
237:
1.250 brouard 238: Revision 1.249 2016/09/07 17:14:18 brouard
239: Summary: Starting values from frequencies
240:
1.249 brouard 241: Revision 1.248 2016/09/07 14:10:18 brouard
242: *** empty log message ***
243:
1.248 brouard 244: Revision 1.247 2016/09/02 11:11:21 brouard
245: *** empty log message ***
246:
1.247 brouard 247: Revision 1.246 2016/09/02 08:49:22 brouard
248: *** empty log message ***
249:
1.246 brouard 250: Revision 1.245 2016/09/02 07:25:01 brouard
251: *** empty log message ***
252:
1.245 brouard 253: Revision 1.244 2016/09/02 07:17:34 brouard
254: *** empty log message ***
255:
1.244 brouard 256: Revision 1.243 2016/09/02 06:45:35 brouard
257: *** empty log message ***
258:
1.243 brouard 259: Revision 1.242 2016/08/30 15:01:20 brouard
260: Summary: Fixing a lots
261:
1.242 brouard 262: Revision 1.241 2016/08/29 17:17:25 brouard
263: Summary: gnuplot problem in Back projection to fix
264:
1.241 brouard 265: Revision 1.240 2016/08/29 07:53:18 brouard
266: Summary: Better
267:
1.240 brouard 268: Revision 1.239 2016/08/26 15:51:03 brouard
269: Summary: Improvement in Powell output in order to copy and paste
270:
271: Author:
272:
1.239 brouard 273: Revision 1.238 2016/08/26 14:23:35 brouard
274: Summary: Starting tests of 0.99
275:
1.238 brouard 276: Revision 1.237 2016/08/26 09:20:19 brouard
277: Summary: to valgrind
278:
1.237 brouard 279: Revision 1.236 2016/08/25 10:50:18 brouard
280: *** empty log message ***
281:
1.236 brouard 282: Revision 1.235 2016/08/25 06:59:23 brouard
283: *** empty log message ***
284:
1.235 brouard 285: Revision 1.234 2016/08/23 16:51:20 brouard
286: *** empty log message ***
287:
1.234 brouard 288: Revision 1.233 2016/08/23 07:40:50 brouard
289: Summary: not working
290:
1.233 brouard 291: Revision 1.232 2016/08/22 14:20:21 brouard
292: Summary: not working
293:
1.232 brouard 294: Revision 1.231 2016/08/22 07:17:15 brouard
295: Summary: not working
296:
1.231 brouard 297: Revision 1.230 2016/08/22 06:55:53 brouard
298: Summary: Not working
299:
1.230 brouard 300: Revision 1.229 2016/07/23 09:45:53 brouard
301: Summary: Completing for func too
302:
1.229 brouard 303: Revision 1.228 2016/07/22 17:45:30 brouard
304: Summary: Fixing some arrays, still debugging
305:
1.227 brouard 306: Revision 1.226 2016/07/12 18:42:34 brouard
307: Summary: temp
308:
1.226 brouard 309: Revision 1.225 2016/07/12 08:40:03 brouard
310: Summary: saving but not running
311:
1.225 brouard 312: Revision 1.224 2016/07/01 13:16:01 brouard
313: Summary: Fixes
314:
1.224 brouard 315: Revision 1.223 2016/02/19 09:23:35 brouard
316: Summary: temporary
317:
1.223 brouard 318: Revision 1.222 2016/02/17 08:14:50 brouard
319: Summary: Probably last 0.98 stable version 0.98r6
320:
1.222 brouard 321: Revision 1.221 2016/02/15 23:35:36 brouard
322: Summary: minor bug
323:
1.220 brouard 324: Revision 1.219 2016/02/15 00:48:12 brouard
325: *** empty log message ***
326:
1.219 brouard 327: Revision 1.218 2016/02/12 11:29:23 brouard
328: Summary: 0.99 Back projections
329:
1.218 brouard 330: Revision 1.217 2015/12/23 17:18:31 brouard
331: Summary: Experimental backcast
332:
1.217 brouard 333: Revision 1.216 2015/12/18 17:32:11 brouard
334: Summary: 0.98r4 Warning and status=-2
335:
336: Version 0.98r4 is now:
337: - displaying an error when status is -1, date of interview unknown and date of death known;
338: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
339: Older changes concerning s=-2, dating from 2005 have been supersed.
340:
1.216 brouard 341: Revision 1.215 2015/12/16 08:52:24 brouard
342: Summary: 0.98r4 working
343:
1.215 brouard 344: Revision 1.214 2015/12/16 06:57:54 brouard
345: Summary: temporary not working
346:
1.214 brouard 347: Revision 1.213 2015/12/11 18:22:17 brouard
348: Summary: 0.98r4
349:
1.213 brouard 350: Revision 1.212 2015/11/21 12:47:24 brouard
351: Summary: minor typo
352:
1.212 brouard 353: Revision 1.211 2015/11/21 12:41:11 brouard
354: Summary: 0.98r3 with some graph of projected cross-sectional
355:
356: Author: Nicolas Brouard
357:
1.211 brouard 358: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 359: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 360: Summary: Adding ftolpl parameter
361: Author: N Brouard
362:
363: We had difficulties to get smoothed confidence intervals. It was due
364: to the period prevalence which wasn't computed accurately. The inner
365: parameter ftolpl is now an outer parameter of the .imach parameter
366: file after estepm. If ftolpl is small 1.e-4 and estepm too,
367: computation are long.
368:
1.209 brouard 369: Revision 1.208 2015/11/17 14:31:57 brouard
370: Summary: temporary
371:
1.208 brouard 372: Revision 1.207 2015/10/27 17:36:57 brouard
373: *** empty log message ***
374:
1.207 brouard 375: Revision 1.206 2015/10/24 07:14:11 brouard
376: *** empty log message ***
377:
1.206 brouard 378: Revision 1.205 2015/10/23 15:50:53 brouard
379: Summary: 0.98r3 some clarification for graphs on likelihood contributions
380:
1.205 brouard 381: Revision 1.204 2015/10/01 16:20:26 brouard
382: Summary: Some new graphs of contribution to likelihood
383:
1.204 brouard 384: Revision 1.203 2015/09/30 17:45:14 brouard
385: Summary: looking at better estimation of the hessian
386:
387: Also a better criteria for convergence to the period prevalence And
388: therefore adding the number of years needed to converge. (The
389: prevalence in any alive state shold sum to one
390:
1.203 brouard 391: Revision 1.202 2015/09/22 19:45:16 brouard
392: Summary: Adding some overall graph on contribution to likelihood. Might change
393:
1.202 brouard 394: Revision 1.201 2015/09/15 17:34:58 brouard
395: Summary: 0.98r0
396:
397: - Some new graphs like suvival functions
398: - Some bugs fixed like model=1+age+V2.
399:
1.201 brouard 400: Revision 1.200 2015/09/09 16:53:55 brouard
401: Summary: Big bug thanks to Flavia
402:
403: Even model=1+age+V2. did not work anymore
404:
1.200 brouard 405: Revision 1.199 2015/09/07 14:09:23 brouard
406: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
407:
1.199 brouard 408: Revision 1.198 2015/09/03 07:14:39 brouard
409: Summary: 0.98q5 Flavia
410:
1.198 brouard 411: Revision 1.197 2015/09/01 18:24:39 brouard
412: *** empty log message ***
413:
1.197 brouard 414: Revision 1.196 2015/08/18 23:17:52 brouard
415: Summary: 0.98q5
416:
1.196 brouard 417: Revision 1.195 2015/08/18 16:28:39 brouard
418: Summary: Adding a hack for testing purpose
419:
420: After reading the title, ftol and model lines, if the comment line has
421: a q, starting with #q, the answer at the end of the run is quit. It
422: permits to run test files in batch with ctest. The former workaround was
423: $ echo q | imach foo.imach
424:
1.195 brouard 425: Revision 1.194 2015/08/18 13:32:00 brouard
426: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
427:
1.194 brouard 428: Revision 1.193 2015/08/04 07:17:42 brouard
429: Summary: 0.98q4
430:
1.193 brouard 431: Revision 1.192 2015/07/16 16:49:02 brouard
432: Summary: Fixing some outputs
433:
1.192 brouard 434: Revision 1.191 2015/07/14 10:00:33 brouard
435: Summary: Some fixes
436:
1.191 brouard 437: Revision 1.190 2015/05/05 08:51:13 brouard
438: Summary: Adding digits in output parameters (7 digits instead of 6)
439:
440: Fix 1+age+.
441:
1.190 brouard 442: Revision 1.189 2015/04/30 14:45:16 brouard
443: Summary: 0.98q2
444:
1.189 brouard 445: Revision 1.188 2015/04/30 08:27:53 brouard
446: *** empty log message ***
447:
1.188 brouard 448: Revision 1.187 2015/04/29 09:11:15 brouard
449: *** empty log message ***
450:
1.187 brouard 451: Revision 1.186 2015/04/23 12:01:52 brouard
452: Summary: V1*age is working now, version 0.98q1
453:
454: Some codes had been disabled in order to simplify and Vn*age was
455: working in the optimization phase, ie, giving correct MLE parameters,
456: but, as usual, outputs were not correct and program core dumped.
457:
1.186 brouard 458: Revision 1.185 2015/03/11 13:26:42 brouard
459: Summary: Inclusion of compile and links command line for Intel Compiler
460:
1.185 brouard 461: Revision 1.184 2015/03/11 11:52:39 brouard
462: Summary: Back from Windows 8. Intel Compiler
463:
1.184 brouard 464: Revision 1.183 2015/03/10 20:34:32 brouard
465: Summary: 0.98q0, trying with directest, mnbrak fixed
466:
467: We use directest instead of original Powell test; probably no
468: incidence on the results, but better justifications;
469: We fixed Numerical Recipes mnbrak routine which was wrong and gave
470: wrong results.
471:
1.183 brouard 472: Revision 1.182 2015/02/12 08:19:57 brouard
473: Summary: Trying to keep directest which seems simpler and more general
474: Author: Nicolas Brouard
475:
1.182 brouard 476: Revision 1.181 2015/02/11 23:22:24 brouard
477: Summary: Comments on Powell added
478:
479: Author:
480:
1.181 brouard 481: Revision 1.180 2015/02/11 17:33:45 brouard
482: Summary: Finishing move from main to function (hpijx and prevalence_limit)
483:
1.180 brouard 484: Revision 1.179 2015/01/04 09:57:06 brouard
485: Summary: back to OS/X
486:
1.179 brouard 487: Revision 1.178 2015/01/04 09:35:48 brouard
488: *** empty log message ***
489:
1.178 brouard 490: Revision 1.177 2015/01/03 18:40:56 brouard
491: Summary: Still testing ilc32 on OSX
492:
1.177 brouard 493: Revision 1.176 2015/01/03 16:45:04 brouard
494: *** empty log message ***
495:
1.176 brouard 496: Revision 1.175 2015/01/03 16:33:42 brouard
497: *** empty log message ***
498:
1.175 brouard 499: Revision 1.174 2015/01/03 16:15:49 brouard
500: Summary: Still in cross-compilation
501:
1.174 brouard 502: Revision 1.173 2015/01/03 12:06:26 brouard
503: Summary: trying to detect cross-compilation
504:
1.173 brouard 505: Revision 1.172 2014/12/27 12:07:47 brouard
506: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
507:
1.172 brouard 508: Revision 1.171 2014/12/23 13:26:59 brouard
509: Summary: Back from Visual C
510:
511: Still problem with utsname.h on Windows
512:
1.171 brouard 513: Revision 1.170 2014/12/23 11:17:12 brouard
514: Summary: Cleaning some \%% back to %%
515:
516: The escape was mandatory for a specific compiler (which one?), but too many warnings.
517:
1.170 brouard 518: Revision 1.169 2014/12/22 23:08:31 brouard
519: Summary: 0.98p
520:
521: Outputs some informations on compiler used, OS etc. Testing on different platforms.
522:
1.169 brouard 523: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 524: Summary: update
1.169 brouard 525:
1.168 brouard 526: Revision 1.167 2014/12/22 13:50:56 brouard
527: Summary: Testing uname and compiler version and if compiled 32 or 64
528:
529: Testing on Linux 64
530:
1.167 brouard 531: Revision 1.166 2014/12/22 11:40:47 brouard
532: *** empty log message ***
533:
1.166 brouard 534: Revision 1.165 2014/12/16 11:20:36 brouard
535: Summary: After compiling on Visual C
536:
537: * imach.c (Module): Merging 1.61 to 1.162
538:
1.165 brouard 539: Revision 1.164 2014/12/16 10:52:11 brouard
540: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
541:
542: * imach.c (Module): Merging 1.61 to 1.162
543:
1.164 brouard 544: Revision 1.163 2014/12/16 10:30:11 brouard
545: * imach.c (Module): Merging 1.61 to 1.162
546:
1.163 brouard 547: Revision 1.162 2014/09/25 11:43:39 brouard
548: Summary: temporary backup 0.99!
549:
1.162 brouard 550: Revision 1.1 2014/09/16 11:06:58 brouard
551: Summary: With some code (wrong) for nlopt
552:
553: Author:
554:
555: Revision 1.161 2014/09/15 20:41:41 brouard
556: Summary: Problem with macro SQR on Intel compiler
557:
1.161 brouard 558: Revision 1.160 2014/09/02 09:24:05 brouard
559: *** empty log message ***
560:
1.160 brouard 561: Revision 1.159 2014/09/01 10:34:10 brouard
562: Summary: WIN32
563: Author: Brouard
564:
1.159 brouard 565: Revision 1.158 2014/08/27 17:11:51 brouard
566: *** empty log message ***
567:
1.158 brouard 568: Revision 1.157 2014/08/27 16:26:55 brouard
569: Summary: Preparing windows Visual studio version
570: Author: Brouard
571:
572: In order to compile on Visual studio, time.h is now correct and time_t
573: and tm struct should be used. difftime should be used but sometimes I
574: just make the differences in raw time format (time(&now).
575: Trying to suppress #ifdef LINUX
576: Add xdg-open for __linux in order to open default browser.
577:
1.157 brouard 578: Revision 1.156 2014/08/25 20:10:10 brouard
579: *** empty log message ***
580:
1.156 brouard 581: Revision 1.155 2014/08/25 18:32:34 brouard
582: Summary: New compile, minor changes
583: Author: Brouard
584:
1.155 brouard 585: Revision 1.154 2014/06/20 17:32:08 brouard
586: Summary: Outputs now all graphs of convergence to period prevalence
587:
1.154 brouard 588: Revision 1.153 2014/06/20 16:45:46 brouard
589: Summary: If 3 live state, convergence to period prevalence on same graph
590: Author: Brouard
591:
1.153 brouard 592: Revision 1.152 2014/06/18 17:54:09 brouard
593: Summary: open browser, use gnuplot on same dir than imach if not found in the path
594:
1.152 brouard 595: Revision 1.151 2014/06/18 16:43:30 brouard
596: *** empty log message ***
597:
1.151 brouard 598: Revision 1.150 2014/06/18 16:42:35 brouard
599: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
600: Author: brouard
601:
1.150 brouard 602: Revision 1.149 2014/06/18 15:51:14 brouard
603: Summary: Some fixes in parameter files errors
604: Author: Nicolas Brouard
605:
1.149 brouard 606: Revision 1.148 2014/06/17 17:38:48 brouard
607: Summary: Nothing new
608: Author: Brouard
609:
610: Just a new packaging for OS/X version 0.98nS
611:
1.148 brouard 612: Revision 1.147 2014/06/16 10:33:11 brouard
613: *** empty log message ***
614:
1.147 brouard 615: Revision 1.146 2014/06/16 10:20:28 brouard
616: Summary: Merge
617: Author: Brouard
618:
619: Merge, before building revised version.
620:
1.146 brouard 621: Revision 1.145 2014/06/10 21:23:15 brouard
622: Summary: Debugging with valgrind
623: Author: Nicolas Brouard
624:
625: Lot of changes in order to output the results with some covariates
626: After the Edimburgh REVES conference 2014, it seems mandatory to
627: improve the code.
628: No more memory valgrind error but a lot has to be done in order to
629: continue the work of splitting the code into subroutines.
630: Also, decodemodel has been improved. Tricode is still not
631: optimal. nbcode should be improved. Documentation has been added in
632: the source code.
633:
1.144 brouard 634: Revision 1.143 2014/01/26 09:45:38 brouard
635: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
636:
637: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
638: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
639:
1.143 brouard 640: Revision 1.142 2014/01/26 03:57:36 brouard
641: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
642:
643: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
644:
1.142 brouard 645: Revision 1.141 2014/01/26 02:42:01 brouard
646: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
647:
1.141 brouard 648: Revision 1.140 2011/09/02 10:37:54 brouard
649: Summary: times.h is ok with mingw32 now.
650:
1.140 brouard 651: Revision 1.139 2010/06/14 07:50:17 brouard
652: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
653: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
654:
1.139 brouard 655: Revision 1.138 2010/04/30 18:19:40 brouard
656: *** empty log message ***
657:
1.138 brouard 658: Revision 1.137 2010/04/29 18:11:38 brouard
659: (Module): Checking covariates for more complex models
660: than V1+V2. A lot of change to be done. Unstable.
661:
1.137 brouard 662: Revision 1.136 2010/04/26 20:30:53 brouard
663: (Module): merging some libgsl code. Fixing computation
664: of likelione (using inter/intrapolation if mle = 0) in order to
665: get same likelihood as if mle=1.
666: Some cleaning of code and comments added.
667:
1.136 brouard 668: Revision 1.135 2009/10/29 15:33:14 brouard
669: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
670:
1.135 brouard 671: Revision 1.134 2009/10/29 13:18:53 brouard
672: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
673:
1.134 brouard 674: Revision 1.133 2009/07/06 10:21:25 brouard
675: just nforces
676:
1.133 brouard 677: Revision 1.132 2009/07/06 08:22:05 brouard
678: Many tings
679:
1.132 brouard 680: Revision 1.131 2009/06/20 16:22:47 brouard
681: Some dimensions resccaled
682:
1.131 brouard 683: Revision 1.130 2009/05/26 06:44:34 brouard
684: (Module): Max Covariate is now set to 20 instead of 8. A
685: lot of cleaning with variables initialized to 0. Trying to make
686: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
687:
1.130 brouard 688: Revision 1.129 2007/08/31 13:49:27 lievre
689: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
690:
1.129 lievre 691: Revision 1.128 2006/06/30 13:02:05 brouard
692: (Module): Clarifications on computing e.j
693:
1.128 brouard 694: Revision 1.127 2006/04/28 18:11:50 brouard
695: (Module): Yes the sum of survivors was wrong since
696: imach-114 because nhstepm was no more computed in the age
697: loop. Now we define nhstepma in the age loop.
698: (Module): In order to speed up (in case of numerous covariates) we
699: compute health expectancies (without variances) in a first step
700: and then all the health expectancies with variances or standard
701: deviation (needs data from the Hessian matrices) which slows the
702: computation.
703: In the future we should be able to stop the program is only health
704: expectancies and graph are needed without standard deviations.
705:
1.127 brouard 706: Revision 1.126 2006/04/28 17:23:28 brouard
707: (Module): Yes the sum of survivors was wrong since
708: imach-114 because nhstepm was no more computed in the age
709: loop. Now we define nhstepma in the age loop.
710: Version 0.98h
711:
1.126 brouard 712: Revision 1.125 2006/04/04 15:20:31 lievre
713: Errors in calculation of health expectancies. Age was not initialized.
714: Forecasting file added.
715:
716: Revision 1.124 2006/03/22 17:13:53 lievre
717: Parameters are printed with %lf instead of %f (more numbers after the comma).
718: The log-likelihood is printed in the log file
719:
720: Revision 1.123 2006/03/20 10:52:43 brouard
721: * imach.c (Module): <title> changed, corresponds to .htm file
722: name. <head> headers where missing.
723:
724: * imach.c (Module): Weights can have a decimal point as for
725: English (a comma might work with a correct LC_NUMERIC environment,
726: otherwise the weight is truncated).
727: Modification of warning when the covariates values are not 0 or
728: 1.
729: Version 0.98g
730:
731: Revision 1.122 2006/03/20 09:45:41 brouard
732: (Module): Weights can have a decimal point as for
733: English (a comma might work with a correct LC_NUMERIC environment,
734: otherwise the weight is truncated).
735: Modification of warning when the covariates values are not 0 or
736: 1.
737: Version 0.98g
738:
739: Revision 1.121 2006/03/16 17:45:01 lievre
740: * imach.c (Module): Comments concerning covariates added
741:
742: * imach.c (Module): refinements in the computation of lli if
743: status=-2 in order to have more reliable computation if stepm is
744: not 1 month. Version 0.98f
745:
746: Revision 1.120 2006/03/16 15:10:38 lievre
747: (Module): refinements in the computation of lli if
748: status=-2 in order to have more reliable computation if stepm is
749: not 1 month. Version 0.98f
750:
751: Revision 1.119 2006/03/15 17:42:26 brouard
752: (Module): Bug if status = -2, the loglikelihood was
753: computed as likelihood omitting the logarithm. Version O.98e
754:
755: Revision 1.118 2006/03/14 18:20:07 brouard
756: (Module): varevsij Comments added explaining the second
757: table of variances if popbased=1 .
758: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
759: (Module): Function pstamp added
760: (Module): Version 0.98d
761:
762: Revision 1.117 2006/03/14 17:16:22 brouard
763: (Module): varevsij Comments added explaining the second
764: table of variances if popbased=1 .
765: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
766: (Module): Function pstamp added
767: (Module): Version 0.98d
768:
769: Revision 1.116 2006/03/06 10:29:27 brouard
770: (Module): Variance-covariance wrong links and
771: varian-covariance of ej. is needed (Saito).
772:
773: Revision 1.115 2006/02/27 12:17:45 brouard
774: (Module): One freematrix added in mlikeli! 0.98c
775:
776: Revision 1.114 2006/02/26 12:57:58 brouard
777: (Module): Some improvements in processing parameter
778: filename with strsep.
779:
780: Revision 1.113 2006/02/24 14:20:24 brouard
781: (Module): Memory leaks checks with valgrind and:
782: datafile was not closed, some imatrix were not freed and on matrix
783: allocation too.
784:
785: Revision 1.112 2006/01/30 09:55:26 brouard
786: (Module): Back to gnuplot.exe instead of wgnuplot.exe
787:
788: Revision 1.111 2006/01/25 20:38:18 brouard
789: (Module): Lots of cleaning and bugs added (Gompertz)
790: (Module): Comments can be added in data file. Missing date values
791: can be a simple dot '.'.
792:
793: Revision 1.110 2006/01/25 00:51:50 brouard
794: (Module): Lots of cleaning and bugs added (Gompertz)
795:
796: Revision 1.109 2006/01/24 19:37:15 brouard
797: (Module): Comments (lines starting with a #) are allowed in data.
798:
799: Revision 1.108 2006/01/19 18:05:42 lievre
800: Gnuplot problem appeared...
801: To be fixed
802:
803: Revision 1.107 2006/01/19 16:20:37 brouard
804: Test existence of gnuplot in imach path
805:
806: Revision 1.106 2006/01/19 13:24:36 brouard
807: Some cleaning and links added in html output
808:
809: Revision 1.105 2006/01/05 20:23:19 lievre
810: *** empty log message ***
811:
812: Revision 1.104 2005/09/30 16:11:43 lievre
813: (Module): sump fixed, loop imx fixed, and simplifications.
814: (Module): If the status is missing at the last wave but we know
815: that the person is alive, then we can code his/her status as -2
816: (instead of missing=-1 in earlier versions) and his/her
817: contributions to the likelihood is 1 - Prob of dying from last
818: health status (= 1-p13= p11+p12 in the easiest case of somebody in
819: the healthy state at last known wave). Version is 0.98
820:
821: Revision 1.103 2005/09/30 15:54:49 lievre
822: (Module): sump fixed, loop imx fixed, and simplifications.
823:
824: Revision 1.102 2004/09/15 17:31:30 brouard
825: Add the possibility to read data file including tab characters.
826:
827: Revision 1.101 2004/09/15 10:38:38 brouard
828: Fix on curr_time
829:
830: Revision 1.100 2004/07/12 18:29:06 brouard
831: Add version for Mac OS X. Just define UNIX in Makefile
832:
833: Revision 1.99 2004/06/05 08:57:40 brouard
834: *** empty log message ***
835:
836: Revision 1.98 2004/05/16 15:05:56 brouard
837: New version 0.97 . First attempt to estimate force of mortality
838: directly from the data i.e. without the need of knowing the health
839: state at each age, but using a Gompertz model: log u =a + b*age .
840: This is the basic analysis of mortality and should be done before any
841: other analysis, in order to test if the mortality estimated from the
842: cross-longitudinal survey is different from the mortality estimated
843: from other sources like vital statistic data.
844:
845: The same imach parameter file can be used but the option for mle should be -3.
846:
1.133 brouard 847: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 848: former routines in order to include the new code within the former code.
849:
850: The output is very simple: only an estimate of the intercept and of
851: the slope with 95% confident intervals.
852:
853: Current limitations:
854: A) Even if you enter covariates, i.e. with the
855: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
856: B) There is no computation of Life Expectancy nor Life Table.
857:
858: Revision 1.97 2004/02/20 13:25:42 lievre
859: Version 0.96d. Population forecasting command line is (temporarily)
860: suppressed.
861:
862: Revision 1.96 2003/07/15 15:38:55 brouard
863: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
864: rewritten within the same printf. Workaround: many printfs.
865:
866: Revision 1.95 2003/07/08 07:54:34 brouard
867: * imach.c (Repository):
868: (Repository): Using imachwizard code to output a more meaningful covariance
869: matrix (cov(a12,c31) instead of numbers.
870:
871: Revision 1.94 2003/06/27 13:00:02 brouard
872: Just cleaning
873:
874: Revision 1.93 2003/06/25 16:33:55 brouard
875: (Module): On windows (cygwin) function asctime_r doesn't
876: exist so I changed back to asctime which exists.
877: (Module): Version 0.96b
878:
879: Revision 1.92 2003/06/25 16:30:45 brouard
880: (Module): On windows (cygwin) function asctime_r doesn't
881: exist so I changed back to asctime which exists.
882:
883: Revision 1.91 2003/06/25 15:30:29 brouard
884: * imach.c (Repository): Duplicated warning errors corrected.
885: (Repository): Elapsed time after each iteration is now output. It
886: helps to forecast when convergence will be reached. Elapsed time
887: is stamped in powell. We created a new html file for the graphs
888: concerning matrix of covariance. It has extension -cov.htm.
889:
890: Revision 1.90 2003/06/24 12:34:15 brouard
891: (Module): Some bugs corrected for windows. Also, when
892: mle=-1 a template is output in file "or"mypar.txt with the design
893: of the covariance matrix to be input.
894:
895: Revision 1.89 2003/06/24 12:30:52 brouard
896: (Module): Some bugs corrected for windows. Also, when
897: mle=-1 a template is output in file "or"mypar.txt with the design
898: of the covariance matrix to be input.
899:
900: Revision 1.88 2003/06/23 17:54:56 brouard
901: * 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.
902:
903: Revision 1.87 2003/06/18 12:26:01 brouard
904: Version 0.96
905:
906: Revision 1.86 2003/06/17 20:04:08 brouard
907: (Module): Change position of html and gnuplot routines and added
908: routine fileappend.
909:
910: Revision 1.85 2003/06/17 13:12:43 brouard
911: * imach.c (Repository): Check when date of death was earlier that
912: current date of interview. It may happen when the death was just
913: prior to the death. In this case, dh was negative and likelihood
914: was wrong (infinity). We still send an "Error" but patch by
915: assuming that the date of death was just one stepm after the
916: interview.
917: (Repository): Because some people have very long ID (first column)
918: we changed int to long in num[] and we added a new lvector for
919: memory allocation. But we also truncated to 8 characters (left
920: truncation)
921: (Repository): No more line truncation errors.
922:
923: Revision 1.84 2003/06/13 21:44:43 brouard
924: * imach.c (Repository): Replace "freqsummary" at a correct
925: place. It differs from routine "prevalence" which may be called
926: many times. Probs is memory consuming and must be used with
927: parcimony.
928: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
929:
930: Revision 1.83 2003/06/10 13:39:11 lievre
931: *** empty log message ***
932:
933: Revision 1.82 2003/06/05 15:57:20 brouard
934: Add log in imach.c and fullversion number is now printed.
935:
936: */
937: /*
938: Interpolated Markov Chain
939:
940: Short summary of the programme:
941:
1.227 brouard 942: This program computes Healthy Life Expectancies or State-specific
943: (if states aren't health statuses) Expectancies from
944: cross-longitudinal data. Cross-longitudinal data consist in:
945:
946: -1- a first survey ("cross") where individuals from different ages
947: are interviewed on their health status or degree of disability (in
948: the case of a health survey which is our main interest)
949:
950: -2- at least a second wave of interviews ("longitudinal") which
951: measure each change (if any) in individual health status. Health
952: expectancies are computed from the time spent in each health state
953: according to a model. More health states you consider, more time is
954: necessary to reach the Maximum Likelihood of the parameters involved
955: in the model. The simplest model is the multinomial logistic model
956: where pij is the probability to be observed in state j at the second
957: wave conditional to be observed in state i at the first
958: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
959: etc , where 'age' is age and 'sex' is a covariate. If you want to
960: have a more complex model than "constant and age", you should modify
961: the program where the markup *Covariates have to be included here
962: again* invites you to do it. More covariates you add, slower the
1.126 brouard 963: convergence.
964:
965: The advantage of this computer programme, compared to a simple
966: multinomial logistic model, is clear when the delay between waves is not
967: identical for each individual. Also, if a individual missed an
968: intermediate interview, the information is lost, but taken into
969: account using an interpolation or extrapolation.
970:
971: hPijx is the probability to be observed in state i at age x+h
972: conditional to the observed state i at age x. The delay 'h' can be
973: split into an exact number (nh*stepm) of unobserved intermediate
974: states. This elementary transition (by month, quarter,
975: semester or year) is modelled as a multinomial logistic. The hPx
976: matrix is simply the matrix product of nh*stepm elementary matrices
977: and the contribution of each individual to the likelihood is simply
978: hPijx.
979:
980: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 981: of the life expectancies. It also computes the period (stable) prevalence.
982:
983: Back prevalence and projections:
1.227 brouard 984:
985: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
986: double agemaxpar, double ftolpl, int *ncvyearp, double
987: dateprev1,double dateprev2, int firstpass, int lastpass, int
988: mobilavproj)
989:
990: Computes the back prevalence limit for any combination of
991: covariate values k at any age between ageminpar and agemaxpar and
992: returns it in **bprlim. In the loops,
993:
994: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
995: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
996:
997: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 998: Computes for any combination of covariates k and any age between bage and fage
999: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1000: oldm=oldms;savm=savms;
1.227 brouard 1001:
1.267 brouard 1002: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1003: Computes the transition matrix starting at age 'age' over
1004: 'nhstepm*hstepm*stepm' months (i.e. until
1005: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1006: nhstepm*hstepm matrices.
1007:
1008: Returns p3mat[i][j][h] after calling
1009: p3mat[i][j][h]=matprod2(newm,
1010: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1011: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1012: oldm);
1.226 brouard 1013:
1014: Important routines
1015:
1016: - func (or funcone), computes logit (pij) distinguishing
1017: o fixed variables (single or product dummies or quantitative);
1018: o varying variables by:
1019: (1) wave (single, product dummies, quantitative),
1020: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1021: % fixed dummy (treated) or quantitative (not done because time-consuming);
1022: % varying dummy (not done) or quantitative (not done);
1023: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1024: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1025: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1026: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1027: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1028:
1.226 brouard 1029:
1030:
1.133 brouard 1031: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1032: Institut national d'études démographiques, Paris.
1.126 brouard 1033: This software have been partly granted by Euro-REVES, a concerted action
1034: from the European Union.
1035: It is copyrighted identically to a GNU software product, ie programme and
1036: software can be distributed freely for non commercial use. Latest version
1037: can be accessed at http://euroreves.ined.fr/imach .
1038:
1039: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1040: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1041:
1042: **********************************************************************/
1043: /*
1044: main
1045: read parameterfile
1046: read datafile
1047: concatwav
1048: freqsummary
1049: if (mle >= 1)
1050: mlikeli
1051: print results files
1052: if mle==1
1053: computes hessian
1054: read end of parameter file: agemin, agemax, bage, fage, estepm
1055: begin-prev-date,...
1056: open gnuplot file
1057: open html file
1.145 brouard 1058: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1059: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1060: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1061: freexexit2 possible for memory heap.
1062:
1063: h Pij x | pij_nom ficrestpij
1064: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1065: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1066: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1067:
1068: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1069: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1070: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1071: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1072: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1073:
1.126 brouard 1074: forecasting if prevfcast==1 prevforecast call prevalence()
1075: health expectancies
1076: Variance-covariance of DFLE
1077: prevalence()
1078: movingaverage()
1079: varevsij()
1080: if popbased==1 varevsij(,popbased)
1081: total life expectancies
1082: Variance of period (stable) prevalence
1083: end
1084: */
1085:
1.187 brouard 1086: /* #define DEBUG */
1087: /* #define DEBUGBRENT */
1.203 brouard 1088: /* #define DEBUGLINMIN */
1089: /* #define DEBUGHESS */
1090: #define DEBUGHESSIJ
1.224 brouard 1091: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1092: #define POWELL /* Instead of NLOPT */
1.224 brouard 1093: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1094: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1095: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1096:
1097: #include <math.h>
1098: #include <stdio.h>
1099: #include <stdlib.h>
1100: #include <string.h>
1.226 brouard 1101: #include <ctype.h>
1.159 brouard 1102:
1103: #ifdef _WIN32
1104: #include <io.h>
1.172 brouard 1105: #include <windows.h>
1106: #include <tchar.h>
1.159 brouard 1107: #else
1.126 brouard 1108: #include <unistd.h>
1.159 brouard 1109: #endif
1.126 brouard 1110:
1111: #include <limits.h>
1112: #include <sys/types.h>
1.171 brouard 1113:
1114: #if defined(__GNUC__)
1115: #include <sys/utsname.h> /* Doesn't work on Windows */
1116: #endif
1117:
1.126 brouard 1118: #include <sys/stat.h>
1119: #include <errno.h>
1.159 brouard 1120: /* extern int errno; */
1.126 brouard 1121:
1.157 brouard 1122: /* #ifdef LINUX */
1123: /* #include <time.h> */
1124: /* #include "timeval.h" */
1125: /* #else */
1126: /* #include <sys/time.h> */
1127: /* #endif */
1128:
1.126 brouard 1129: #include <time.h>
1130:
1.136 brouard 1131: #ifdef GSL
1132: #include <gsl/gsl_errno.h>
1133: #include <gsl/gsl_multimin.h>
1134: #endif
1135:
1.167 brouard 1136:
1.162 brouard 1137: #ifdef NLOPT
1138: #include <nlopt.h>
1139: typedef struct {
1140: double (* function)(double [] );
1141: } myfunc_data ;
1142: #endif
1143:
1.126 brouard 1144: /* #include <libintl.h> */
1145: /* #define _(String) gettext (String) */
1146:
1.251 brouard 1147: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1148:
1149: #define GNUPLOTPROGRAM "gnuplot"
1150: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1151: #define FILENAMELENGTH 132
1152:
1153: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1154: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1155:
1.144 brouard 1156: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1157: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1158:
1159: #define NINTERVMAX 8
1.144 brouard 1160: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1161: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1162: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1163: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1164: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1165: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1166: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1167: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1168: /* #define AGESUP 130 */
1.288 brouard 1169: /* #define AGESUP 150 */
1170: #define AGESUP 200
1.268 brouard 1171: #define AGEINF 0
1.218 brouard 1172: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1173: #define AGEBASE 40
1.194 brouard 1174: #define AGEOVERFLOW 1.e20
1.164 brouard 1175: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1176: #ifdef _WIN32
1177: #define DIRSEPARATOR '\\'
1178: #define CHARSEPARATOR "\\"
1179: #define ODIRSEPARATOR '/'
1180: #else
1.126 brouard 1181: #define DIRSEPARATOR '/'
1182: #define CHARSEPARATOR "/"
1183: #define ODIRSEPARATOR '\\'
1184: #endif
1185:
1.317 ! brouard 1186: /* $Id: imach.c,v 1.316 2022/05/11 15:11:31 brouard Exp $ */
1.126 brouard 1187: /* $State: Exp $ */
1.196 brouard 1188: #include "version.h"
1189: char version[]=__IMACH_VERSION__;
1.316 brouard 1190: char copyright[]="May 2022,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021-202, INED 2000-2022";
1.317 ! brouard 1191: char fullversion[]="$Revision: 1.316 $ $Date: 2022/05/11 15:11:31 $";
1.126 brouard 1192: char strstart[80];
1193: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1194: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1195: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1196: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1197: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1198: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1199: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1200: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1201: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1202: int cptcovprodnoage=0; /**< Number of covariate products without age */
1203: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1204: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1205: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1206: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1207: int nsd=0; /**< Total number of single dummy variables (output) */
1208: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1209: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1210: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1211: int ntveff=0; /**< ntveff number of effective time varying variables */
1212: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1213: int cptcov=0; /* Working variable */
1.290 brouard 1214: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1215: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1216: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1217: int nlstate=2; /* Number of live states */
1218: int ndeath=1; /* Number of dead states */
1.130 brouard 1219: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1220: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1221: int popbased=0;
1222:
1223: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1224: int maxwav=0; /* Maxim number of waves */
1225: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1226: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1227: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1228: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1229: int mle=1, weightopt=0;
1.126 brouard 1230: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1231: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1232: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1233: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1234: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1235: int selected(int kvar); /* Is covariate kvar selected for printing results */
1236:
1.130 brouard 1237: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1238: double **matprod2(); /* test */
1.126 brouard 1239: double **oldm, **newm, **savm; /* Working pointers to matrices */
1240: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1241: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1242:
1.136 brouard 1243: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1244: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1245: FILE *ficlog, *ficrespow;
1.130 brouard 1246: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1247: double fretone; /* Only one call to likelihood */
1.130 brouard 1248: long ipmx=0; /* Number of contributions */
1.126 brouard 1249: double sw; /* Sum of weights */
1250: char filerespow[FILENAMELENGTH];
1251: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1252: FILE *ficresilk;
1253: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1254: FILE *ficresprobmorprev;
1255: FILE *fichtm, *fichtmcov; /* Html File */
1256: FILE *ficreseij;
1257: char filerese[FILENAMELENGTH];
1258: FILE *ficresstdeij;
1259: char fileresstde[FILENAMELENGTH];
1260: FILE *ficrescveij;
1261: char filerescve[FILENAMELENGTH];
1262: FILE *ficresvij;
1263: char fileresv[FILENAMELENGTH];
1.269 brouard 1264:
1.126 brouard 1265: char title[MAXLINE];
1.234 brouard 1266: char model[MAXLINE]; /**< The model line */
1.217 brouard 1267: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1268: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1269: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1270: char command[FILENAMELENGTH];
1271: int outcmd=0;
1272:
1.217 brouard 1273: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1274: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1275: char filelog[FILENAMELENGTH]; /* Log file */
1276: char filerest[FILENAMELENGTH];
1277: char fileregp[FILENAMELENGTH];
1278: char popfile[FILENAMELENGTH];
1279:
1280: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1281:
1.157 brouard 1282: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1283: /* struct timezone tzp; */
1284: /* extern int gettimeofday(); */
1285: struct tm tml, *gmtime(), *localtime();
1286:
1287: extern time_t time();
1288:
1289: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1290: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1291: struct tm tm;
1292:
1.126 brouard 1293: char strcurr[80], strfor[80];
1294:
1295: char *endptr;
1296: long lval;
1297: double dval;
1298:
1299: #define NR_END 1
1300: #define FREE_ARG char*
1301: #define FTOL 1.0e-10
1302:
1303: #define NRANSI
1.240 brouard 1304: #define ITMAX 200
1305: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1306:
1307: #define TOL 2.0e-4
1308:
1309: #define CGOLD 0.3819660
1310: #define ZEPS 1.0e-10
1311: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1312:
1313: #define GOLD 1.618034
1314: #define GLIMIT 100.0
1315: #define TINY 1.0e-20
1316:
1317: static double maxarg1,maxarg2;
1318: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1319: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1320:
1321: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1322: #define rint(a) floor(a+0.5)
1.166 brouard 1323: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1324: #define mytinydouble 1.0e-16
1.166 brouard 1325: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1326: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1327: /* static double dsqrarg; */
1328: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1329: static double sqrarg;
1330: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1331: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1332: int agegomp= AGEGOMP;
1333:
1334: int imx;
1335: int stepm=1;
1336: /* Stepm, step in month: minimum step interpolation*/
1337:
1338: int estepm;
1339: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1340:
1341: int m,nb;
1342: long *num;
1.197 brouard 1343: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1344: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1345: covariate for which somebody answered excluding
1346: undefined. Usually 2: 0 and 1. */
1347: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1348: covariate for which somebody answered including
1349: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1350: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1351: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1352: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1353: double *ageexmed,*agecens;
1354: double dateintmean=0;
1.296 brouard 1355: double anprojd, mprojd, jprojd; /* For eventual projections */
1356: double anprojf, mprojf, jprojf;
1.126 brouard 1357:
1.296 brouard 1358: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1359: double anbackf, mbackf, jbackf;
1360: double jintmean,mintmean,aintmean;
1.126 brouard 1361: double *weight;
1362: int **s; /* Status */
1.141 brouard 1363: double *agedc;
1.145 brouard 1364: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1365: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1366: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1367: double **coqvar; /* Fixed quantitative covariate nqv */
1368: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1369: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1370: double idx;
1371: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1372: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1373: /*k 1 2 3 4 5 6 7 8 9 */
1374: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1375: /* Tndvar[k] 1 2 3 4 5 */
1376: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1377: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1378: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1379: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1380: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1381: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1382: /* Tprod[i]=k 4 7 */
1383: /* Tage[i]=k 5 8 */
1384: /* */
1385: /* Type */
1386: /* V 1 2 3 4 5 */
1387: /* F F V V V */
1388: /* D Q D D Q */
1389: /* */
1390: int *TvarsD;
1391: int *TvarsDind;
1392: int *TvarsQ;
1393: int *TvarsQind;
1394:
1.235 brouard 1395: #define MAXRESULTLINES 10
1396: int nresult=0;
1.258 brouard 1397: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1398: int TKresult[MAXRESULTLINES];
1.237 brouard 1399: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1400: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1401: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1402: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1403: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1404: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1405:
1.234 brouard 1406: /* 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 1407: 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 */
1408: 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 */
1409: 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 */
1410: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1411: 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 */
1412: 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 1413: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1414: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1415: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1416: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1417: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1418: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1419: 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 */
1420: 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 */
1421:
1.230 brouard 1422: int *Tvarsel; /**< Selected covariates for output */
1423: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1424: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1425: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1426: 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 1427: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1428: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1429: int *Tage;
1.227 brouard 1430: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1431: 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 1432: 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*/
1433: 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 1434: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1435: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1436: int **Tvard;
1437: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1438: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1439: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1440: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1441: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1442: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1443: double *lsurv, *lpop, *tpop;
1444:
1.231 brouard 1445: #define FD 1; /* Fixed dummy covariate */
1446: #define FQ 2; /* Fixed quantitative covariate */
1447: #define FP 3; /* Fixed product covariate */
1448: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1449: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1450: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1451: #define VD 10; /* Varying dummy covariate */
1452: #define VQ 11; /* Varying quantitative covariate */
1453: #define VP 12; /* Varying product covariate */
1454: #define VPDD 13; /* Varying product dummy*dummy covariate */
1455: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1456: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1457: #define APFD 16; /* Age product * fixed dummy covariate */
1458: #define APFQ 17; /* Age product * fixed quantitative covariate */
1459: #define APVD 18; /* Age product * varying dummy covariate */
1460: #define APVQ 19; /* Age product * varying quantitative covariate */
1461:
1462: #define FTYPE 1; /* Fixed covariate */
1463: #define VTYPE 2; /* Varying covariate (loop in wave) */
1464: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1465:
1466: struct kmodel{
1467: int maintype; /* main type */
1468: int subtype; /* subtype */
1469: };
1470: struct kmodel modell[NCOVMAX];
1471:
1.143 brouard 1472: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1473: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1474:
1475: /**************** split *************************/
1476: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1477: {
1478: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1479: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1480: */
1481: char *ss; /* pointer */
1.186 brouard 1482: int l1=0, l2=0; /* length counters */
1.126 brouard 1483:
1484: l1 = strlen(path ); /* length of path */
1485: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1486: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1487: if ( ss == NULL ) { /* no directory, so determine current directory */
1488: strcpy( name, path ); /* we got the fullname name because no directory */
1489: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1490: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1491: /* get current working directory */
1492: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1493: #ifdef WIN32
1494: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1495: #else
1496: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1497: #endif
1.126 brouard 1498: return( GLOCK_ERROR_GETCWD );
1499: }
1500: /* got dirc from getcwd*/
1501: printf(" DIRC = %s \n",dirc);
1.205 brouard 1502: } else { /* strip directory from path */
1.126 brouard 1503: ss++; /* after this, the filename */
1504: l2 = strlen( ss ); /* length of filename */
1505: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1506: strcpy( name, ss ); /* save file name */
1507: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1508: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1509: printf(" DIRC2 = %s \n",dirc);
1510: }
1511: /* We add a separator at the end of dirc if not exists */
1512: l1 = strlen( dirc ); /* length of directory */
1513: if( dirc[l1-1] != DIRSEPARATOR ){
1514: dirc[l1] = DIRSEPARATOR;
1515: dirc[l1+1] = 0;
1516: printf(" DIRC3 = %s \n",dirc);
1517: }
1518: ss = strrchr( name, '.' ); /* find last / */
1519: if (ss >0){
1520: ss++;
1521: strcpy(ext,ss); /* save extension */
1522: l1= strlen( name);
1523: l2= strlen(ss)+1;
1524: strncpy( finame, name, l1-l2);
1525: finame[l1-l2]= 0;
1526: }
1527:
1528: return( 0 ); /* we're done */
1529: }
1530:
1531:
1532: /******************************************/
1533:
1534: void replace_back_to_slash(char *s, char*t)
1535: {
1536: int i;
1537: int lg=0;
1538: i=0;
1539: lg=strlen(t);
1540: for(i=0; i<= lg; i++) {
1541: (s[i] = t[i]);
1542: if (t[i]== '\\') s[i]='/';
1543: }
1544: }
1545:
1.132 brouard 1546: char *trimbb(char *out, char *in)
1.137 brouard 1547: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1548: char *s;
1549: s=out;
1550: while (*in != '\0'){
1.137 brouard 1551: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1552: in++;
1553: }
1554: *out++ = *in++;
1555: }
1556: *out='\0';
1557: return s;
1558: }
1559:
1.187 brouard 1560: /* char *substrchaine(char *out, char *in, char *chain) */
1561: /* { */
1562: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1563: /* char *s, *t; */
1564: /* t=in;s=out; */
1565: /* while ((*in != *chain) && (*in != '\0')){ */
1566: /* *out++ = *in++; */
1567: /* } */
1568:
1569: /* /\* *in matches *chain *\/ */
1570: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1571: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1572: /* } */
1573: /* in--; chain--; */
1574: /* while ( (*in != '\0')){ */
1575: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1576: /* *out++ = *in++; */
1577: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1578: /* } */
1579: /* *out='\0'; */
1580: /* out=s; */
1581: /* return out; */
1582: /* } */
1583: char *substrchaine(char *out, char *in, char *chain)
1584: {
1585: /* Substract chain 'chain' from 'in', return and output 'out' */
1586: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1587:
1588: char *strloc;
1589:
1590: strcpy (out, in);
1591: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1592: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1593: if(strloc != NULL){
1594: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1595: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1596: /* strcpy (strloc, strloc +strlen(chain));*/
1597: }
1598: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1599: return out;
1600: }
1601:
1602:
1.145 brouard 1603: char *cutl(char *blocc, char *alocc, char *in, char occ)
1604: {
1.187 brouard 1605: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1606: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1607: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1608: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1609: */
1.160 brouard 1610: char *s, *t;
1.145 brouard 1611: t=in;s=in;
1612: while ((*in != occ) && (*in != '\0')){
1613: *alocc++ = *in++;
1614: }
1615: if( *in == occ){
1616: *(alocc)='\0';
1617: s=++in;
1618: }
1619:
1620: if (s == t) {/* occ not found */
1621: *(alocc-(in-s))='\0';
1622: in=s;
1623: }
1624: while ( *in != '\0'){
1625: *blocc++ = *in++;
1626: }
1627:
1628: *blocc='\0';
1629: return t;
1630: }
1.137 brouard 1631: char *cutv(char *blocc, char *alocc, char *in, char occ)
1632: {
1.187 brouard 1633: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1634: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1635: gives blocc="abcdef2ghi" and alocc="j".
1636: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1637: */
1638: char *s, *t;
1639: t=in;s=in;
1640: while (*in != '\0'){
1641: while( *in == occ){
1642: *blocc++ = *in++;
1643: s=in;
1644: }
1645: *blocc++ = *in++;
1646: }
1647: if (s == t) /* occ not found */
1648: *(blocc-(in-s))='\0';
1649: else
1650: *(blocc-(in-s)-1)='\0';
1651: in=s;
1652: while ( *in != '\0'){
1653: *alocc++ = *in++;
1654: }
1655:
1656: *alocc='\0';
1657: return s;
1658: }
1659:
1.126 brouard 1660: int nbocc(char *s, char occ)
1661: {
1662: int i,j=0;
1663: int lg=20;
1664: i=0;
1665: lg=strlen(s);
1666: for(i=0; i<= lg; i++) {
1.234 brouard 1667: if (s[i] == occ ) j++;
1.126 brouard 1668: }
1669: return j;
1670: }
1671:
1.137 brouard 1672: /* void cutv(char *u,char *v, char*t, char occ) */
1673: /* { */
1674: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1675: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1676: /* gives u="abcdef2ghi" and v="j" *\/ */
1677: /* int i,lg,j,p=0; */
1678: /* i=0; */
1679: /* lg=strlen(t); */
1680: /* for(j=0; j<=lg-1; j++) { */
1681: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1682: /* } */
1.126 brouard 1683:
1.137 brouard 1684: /* for(j=0; j<p; j++) { */
1685: /* (u[j] = t[j]); */
1686: /* } */
1687: /* u[p]='\0'; */
1.126 brouard 1688:
1.137 brouard 1689: /* for(j=0; j<= lg; j++) { */
1690: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1691: /* } */
1692: /* } */
1.126 brouard 1693:
1.160 brouard 1694: #ifdef _WIN32
1695: char * strsep(char **pp, const char *delim)
1696: {
1697: char *p, *q;
1698:
1699: if ((p = *pp) == NULL)
1700: return 0;
1701: if ((q = strpbrk (p, delim)) != NULL)
1702: {
1703: *pp = q + 1;
1704: *q = '\0';
1705: }
1706: else
1707: *pp = 0;
1708: return p;
1709: }
1710: #endif
1711:
1.126 brouard 1712: /********************** nrerror ********************/
1713:
1714: void nrerror(char error_text[])
1715: {
1716: fprintf(stderr,"ERREUR ...\n");
1717: fprintf(stderr,"%s\n",error_text);
1718: exit(EXIT_FAILURE);
1719: }
1720: /*********************** vector *******************/
1721: double *vector(int nl, int nh)
1722: {
1723: double *v;
1724: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1725: if (!v) nrerror("allocation failure in vector");
1726: return v-nl+NR_END;
1727: }
1728:
1729: /************************ free vector ******************/
1730: void free_vector(double*v, int nl, int nh)
1731: {
1732: free((FREE_ARG)(v+nl-NR_END));
1733: }
1734:
1735: /************************ivector *******************************/
1736: int *ivector(long nl,long nh)
1737: {
1738: int *v;
1739: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1740: if (!v) nrerror("allocation failure in ivector");
1741: return v-nl+NR_END;
1742: }
1743:
1744: /******************free ivector **************************/
1745: void free_ivector(int *v, long nl, long nh)
1746: {
1747: free((FREE_ARG)(v+nl-NR_END));
1748: }
1749:
1750: /************************lvector *******************************/
1751: long *lvector(long nl,long nh)
1752: {
1753: long *v;
1754: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1755: if (!v) nrerror("allocation failure in ivector");
1756: return v-nl+NR_END;
1757: }
1758:
1759: /******************free lvector **************************/
1760: void free_lvector(long *v, long nl, long nh)
1761: {
1762: free((FREE_ARG)(v+nl-NR_END));
1763: }
1764:
1765: /******************* imatrix *******************************/
1766: int **imatrix(long nrl, long nrh, long ncl, long nch)
1767: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1768: {
1769: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1770: int **m;
1771:
1772: /* allocate pointers to rows */
1773: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1774: if (!m) nrerror("allocation failure 1 in matrix()");
1775: m += NR_END;
1776: m -= nrl;
1777:
1778:
1779: /* allocate rows and set pointers to them */
1780: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1781: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1782: m[nrl] += NR_END;
1783: m[nrl] -= ncl;
1784:
1785: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1786:
1787: /* return pointer to array of pointers to rows */
1788: return m;
1789: }
1790:
1791: /****************** free_imatrix *************************/
1792: void free_imatrix(m,nrl,nrh,ncl,nch)
1793: int **m;
1794: long nch,ncl,nrh,nrl;
1795: /* free an int matrix allocated by imatrix() */
1796: {
1797: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1798: free((FREE_ARG) (m+nrl-NR_END));
1799: }
1800:
1801: /******************* matrix *******************************/
1802: double **matrix(long nrl, long nrh, long ncl, long nch)
1803: {
1804: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1805: double **m;
1806:
1807: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1808: if (!m) nrerror("allocation failure 1 in matrix()");
1809: m += NR_END;
1810: m -= nrl;
1811:
1812: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1813: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1814: m[nrl] += NR_END;
1815: m[nrl] -= ncl;
1816:
1817: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1818: return m;
1.145 brouard 1819: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1820: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1821: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1822: */
1823: }
1824:
1825: /*************************free matrix ************************/
1826: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1827: {
1828: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1829: free((FREE_ARG)(m+nrl-NR_END));
1830: }
1831:
1832: /******************* ma3x *******************************/
1833: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1834: {
1835: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1836: double ***m;
1837:
1838: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1839: if (!m) nrerror("allocation failure 1 in matrix()");
1840: m += NR_END;
1841: m -= nrl;
1842:
1843: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1844: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1845: m[nrl] += NR_END;
1846: m[nrl] -= ncl;
1847:
1848: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1849:
1850: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1851: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1852: m[nrl][ncl] += NR_END;
1853: m[nrl][ncl] -= nll;
1854: for (j=ncl+1; j<=nch; j++)
1855: m[nrl][j]=m[nrl][j-1]+nlay;
1856:
1857: for (i=nrl+1; i<=nrh; i++) {
1858: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1859: for (j=ncl+1; j<=nch; j++)
1860: m[i][j]=m[i][j-1]+nlay;
1861: }
1862: return m;
1863: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1864: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1865: */
1866: }
1867:
1868: /*************************free ma3x ************************/
1869: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1870: {
1871: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1872: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1873: free((FREE_ARG)(m+nrl-NR_END));
1874: }
1875:
1876: /*************** function subdirf ***********/
1877: char *subdirf(char fileres[])
1878: {
1879: /* Caution optionfilefiname is hidden */
1880: strcpy(tmpout,optionfilefiname);
1881: strcat(tmpout,"/"); /* Add to the right */
1882: strcat(tmpout,fileres);
1883: return tmpout;
1884: }
1885:
1886: /*************** function subdirf2 ***********/
1887: char *subdirf2(char fileres[], char *preop)
1888: {
1.314 brouard 1889: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1890: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1891: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1892: /* Caution optionfilefiname is hidden */
1893: strcpy(tmpout,optionfilefiname);
1894: strcat(tmpout,"/");
1895: strcat(tmpout,preop);
1896: strcat(tmpout,fileres);
1897: return tmpout;
1898: }
1899:
1900: /*************** function subdirf3 ***********/
1901: char *subdirf3(char fileres[], char *preop, char *preop2)
1902: {
1903:
1904: /* Caution optionfilefiname is hidden */
1905: strcpy(tmpout,optionfilefiname);
1906: strcat(tmpout,"/");
1907: strcat(tmpout,preop);
1908: strcat(tmpout,preop2);
1909: strcat(tmpout,fileres);
1910: return tmpout;
1911: }
1.213 brouard 1912:
1913: /*************** function subdirfext ***********/
1914: char *subdirfext(char fileres[], char *preop, char *postop)
1915: {
1916:
1917: strcpy(tmpout,preop);
1918: strcat(tmpout,fileres);
1919: strcat(tmpout,postop);
1920: return tmpout;
1921: }
1.126 brouard 1922:
1.213 brouard 1923: /*************** function subdirfext3 ***********/
1924: char *subdirfext3(char fileres[], char *preop, char *postop)
1925: {
1926:
1927: /* Caution optionfilefiname is hidden */
1928: strcpy(tmpout,optionfilefiname);
1929: strcat(tmpout,"/");
1930: strcat(tmpout,preop);
1931: strcat(tmpout,fileres);
1932: strcat(tmpout,postop);
1933: return tmpout;
1934: }
1935:
1.162 brouard 1936: char *asc_diff_time(long time_sec, char ascdiff[])
1937: {
1938: long sec_left, days, hours, minutes;
1939: days = (time_sec) / (60*60*24);
1940: sec_left = (time_sec) % (60*60*24);
1941: hours = (sec_left) / (60*60) ;
1942: sec_left = (sec_left) %(60*60);
1943: minutes = (sec_left) /60;
1944: sec_left = (sec_left) % (60);
1945: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1946: return ascdiff;
1947: }
1948:
1.126 brouard 1949: /***************** f1dim *************************/
1950: extern int ncom;
1951: extern double *pcom,*xicom;
1952: extern double (*nrfunc)(double []);
1953:
1954: double f1dim(double x)
1955: {
1956: int j;
1957: double f;
1958: double *xt;
1959:
1960: xt=vector(1,ncom);
1961: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1962: f=(*nrfunc)(xt);
1963: free_vector(xt,1,ncom);
1964: return f;
1965: }
1966:
1967: /*****************brent *************************/
1968: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1969: {
1970: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1971: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1972: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1973: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1974: * returned function value.
1975: */
1.126 brouard 1976: int iter;
1977: double a,b,d,etemp;
1.159 brouard 1978: double fu=0,fv,fw,fx;
1.164 brouard 1979: double ftemp=0.;
1.126 brouard 1980: double p,q,r,tol1,tol2,u,v,w,x,xm;
1981: double e=0.0;
1982:
1983: a=(ax < cx ? ax : cx);
1984: b=(ax > cx ? ax : cx);
1985: x=w=v=bx;
1986: fw=fv=fx=(*f)(x);
1987: for (iter=1;iter<=ITMAX;iter++) {
1988: xm=0.5*(a+b);
1989: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1990: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1991: printf(".");fflush(stdout);
1992: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1993: #ifdef DEBUGBRENT
1.126 brouard 1994: 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);
1995: 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);
1996: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1997: #endif
1998: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1999: *xmin=x;
2000: return fx;
2001: }
2002: ftemp=fu;
2003: if (fabs(e) > tol1) {
2004: r=(x-w)*(fx-fv);
2005: q=(x-v)*(fx-fw);
2006: p=(x-v)*q-(x-w)*r;
2007: q=2.0*(q-r);
2008: if (q > 0.0) p = -p;
2009: q=fabs(q);
2010: etemp=e;
2011: e=d;
2012: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2013: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2014: else {
1.224 brouard 2015: d=p/q;
2016: u=x+d;
2017: if (u-a < tol2 || b-u < tol2)
2018: d=SIGN(tol1,xm-x);
1.126 brouard 2019: }
2020: } else {
2021: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2022: }
2023: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2024: fu=(*f)(u);
2025: if (fu <= fx) {
2026: if (u >= x) a=x; else b=x;
2027: SHFT(v,w,x,u)
1.183 brouard 2028: SHFT(fv,fw,fx,fu)
2029: } else {
2030: if (u < x) a=u; else b=u;
2031: if (fu <= fw || w == x) {
1.224 brouard 2032: v=w;
2033: w=u;
2034: fv=fw;
2035: fw=fu;
1.183 brouard 2036: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2037: v=u;
2038: fv=fu;
1.183 brouard 2039: }
2040: }
1.126 brouard 2041: }
2042: nrerror("Too many iterations in brent");
2043: *xmin=x;
2044: return fx;
2045: }
2046:
2047: /****************** mnbrak ***********************/
2048:
2049: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2050: double (*func)(double))
1.183 brouard 2051: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2052: the downhill direction (defined by the function as evaluated at the initial points) and returns
2053: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2054: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2055: */
1.126 brouard 2056: double ulim,u,r,q, dum;
2057: double fu;
1.187 brouard 2058:
2059: double scale=10.;
2060: int iterscale=0;
2061:
2062: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2063: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2064:
2065:
2066: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2067: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2068: /* *bx = *ax - (*ax - *bx)/scale; */
2069: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2070: /* } */
2071:
1.126 brouard 2072: if (*fb > *fa) {
2073: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2074: SHFT(dum,*fb,*fa,dum)
2075: }
1.126 brouard 2076: *cx=(*bx)+GOLD*(*bx-*ax);
2077: *fc=(*func)(*cx);
1.183 brouard 2078: #ifdef DEBUG
1.224 brouard 2079: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2080: 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 2081: #endif
1.224 brouard 2082: 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 2083: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2084: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2085: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2086: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2087: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2088: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2089: fu=(*func)(u);
1.163 brouard 2090: #ifdef DEBUG
2091: /* f(x)=A(x-u)**2+f(u) */
2092: double A, fparabu;
2093: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2094: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2095: 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);
2096: 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 2097: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2098: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2099: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2100: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2101: #endif
1.184 brouard 2102: #ifdef MNBRAKORIGINAL
1.183 brouard 2103: #else
1.191 brouard 2104: /* if (fu > *fc) { */
2105: /* #ifdef DEBUG */
2106: /* printf("mnbrak4 fu > fc \n"); */
2107: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2108: /* #endif */
2109: /* /\* 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 *\\/ *\/ */
2110: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2111: /* dum=u; /\* Shifting c and u *\/ */
2112: /* u = *cx; */
2113: /* *cx = dum; */
2114: /* dum = fu; */
2115: /* fu = *fc; */
2116: /* *fc =dum; */
2117: /* } else { /\* end *\/ */
2118: /* #ifdef DEBUG */
2119: /* printf("mnbrak3 fu < fc \n"); */
2120: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2121: /* #endif */
2122: /* dum=u; /\* Shifting c and u *\/ */
2123: /* u = *cx; */
2124: /* *cx = dum; */
2125: /* dum = fu; */
2126: /* fu = *fc; */
2127: /* *fc =dum; */
2128: /* } */
1.224 brouard 2129: #ifdef DEBUGMNBRAK
2130: double A, fparabu;
2131: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2132: fparabu= *fa - A*(*ax-u)*(*ax-u);
2133: 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);
2134: 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 2135: #endif
1.191 brouard 2136: dum=u; /* Shifting c and u */
2137: u = *cx;
2138: *cx = dum;
2139: dum = fu;
2140: fu = *fc;
2141: *fc =dum;
1.183 brouard 2142: #endif
1.162 brouard 2143: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2144: #ifdef DEBUG
1.224 brouard 2145: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2146: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2147: #endif
1.126 brouard 2148: fu=(*func)(u);
2149: if (fu < *fc) {
1.183 brouard 2150: #ifdef DEBUG
1.224 brouard 2151: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2152: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2153: #endif
2154: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2155: SHFT(*fb,*fc,fu,(*func)(u))
2156: #ifdef DEBUG
2157: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2158: #endif
2159: }
1.162 brouard 2160: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2161: #ifdef DEBUG
1.224 brouard 2162: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2163: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2164: #endif
1.126 brouard 2165: u=ulim;
2166: fu=(*func)(u);
1.183 brouard 2167: } else { /* u could be left to b (if r > q parabola has a maximum) */
2168: #ifdef DEBUG
1.224 brouard 2169: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2170: 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 2171: #endif
1.126 brouard 2172: u=(*cx)+GOLD*(*cx-*bx);
2173: fu=(*func)(u);
1.224 brouard 2174: #ifdef DEBUG
2175: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2176: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2177: #endif
1.183 brouard 2178: } /* end tests */
1.126 brouard 2179: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2180: SHFT(*fa,*fb,*fc,fu)
2181: #ifdef DEBUG
1.224 brouard 2182: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2183: 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 2184: #endif
2185: } /* 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 2186: }
2187:
2188: /*************** linmin ************************/
1.162 brouard 2189: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2190: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2191: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2192: the value of func at the returned location p . This is actually all accomplished by calling the
2193: routines mnbrak and brent .*/
1.126 brouard 2194: int ncom;
2195: double *pcom,*xicom;
2196: double (*nrfunc)(double []);
2197:
1.224 brouard 2198: #ifdef LINMINORIGINAL
1.126 brouard 2199: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2200: #else
2201: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2202: #endif
1.126 brouard 2203: {
2204: double brent(double ax, double bx, double cx,
2205: double (*f)(double), double tol, double *xmin);
2206: double f1dim(double x);
2207: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2208: double *fc, double (*func)(double));
2209: int j;
2210: double xx,xmin,bx,ax;
2211: double fx,fb,fa;
1.187 brouard 2212:
1.203 brouard 2213: #ifdef LINMINORIGINAL
2214: #else
2215: double scale=10., axs, xxs; /* Scale added for infinity */
2216: #endif
2217:
1.126 brouard 2218: ncom=n;
2219: pcom=vector(1,n);
2220: xicom=vector(1,n);
2221: nrfunc=func;
2222: for (j=1;j<=n;j++) {
2223: pcom[j]=p[j];
1.202 brouard 2224: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2225: }
1.187 brouard 2226:
1.203 brouard 2227: #ifdef LINMINORIGINAL
2228: xx=1.;
2229: #else
2230: axs=0.0;
2231: xxs=1.;
2232: do{
2233: xx= xxs;
2234: #endif
1.187 brouard 2235: ax=0.;
2236: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2237: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2238: /* 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)) */
2239: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2240: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2241: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2242: /* 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 2243: #ifdef LINMINORIGINAL
2244: #else
2245: if (fx != fx){
1.224 brouard 2246: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2247: printf("|");
2248: fprintf(ficlog,"|");
1.203 brouard 2249: #ifdef DEBUGLINMIN
1.224 brouard 2250: 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 2251: #endif
2252: }
1.224 brouard 2253: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2254: #endif
2255:
1.191 brouard 2256: #ifdef DEBUGLINMIN
2257: 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 2258: 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 2259: #endif
1.224 brouard 2260: #ifdef LINMINORIGINAL
2261: #else
1.317 ! brouard 2262: if(fb == fx){ /* Flat function in the direction */
! 2263: xmin=xx;
1.224 brouard 2264: *flat=1;
1.317 ! brouard 2265: }else{
1.224 brouard 2266: *flat=0;
2267: #endif
2268: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2269: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2270: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2271: /* fmin = f(p[j] + xmin * xi[j]) */
2272: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2273: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2274: #ifdef DEBUG
1.224 brouard 2275: 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);
2276: 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);
2277: #endif
2278: #ifdef LINMINORIGINAL
2279: #else
2280: }
1.126 brouard 2281: #endif
1.191 brouard 2282: #ifdef DEBUGLINMIN
2283: printf("linmin end ");
1.202 brouard 2284: fprintf(ficlog,"linmin end ");
1.191 brouard 2285: #endif
1.126 brouard 2286: for (j=1;j<=n;j++) {
1.203 brouard 2287: #ifdef LINMINORIGINAL
2288: xi[j] *= xmin;
2289: #else
2290: #ifdef DEBUGLINMIN
2291: if(xxs <1.0)
2292: printf(" before xi[%d]=%12.8f", j,xi[j]);
2293: #endif
2294: 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) */
2295: #ifdef DEBUGLINMIN
2296: if(xxs <1.0)
2297: 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 );
2298: #endif
2299: #endif
1.187 brouard 2300: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2301: }
1.191 brouard 2302: #ifdef DEBUGLINMIN
1.203 brouard 2303: printf("\n");
1.191 brouard 2304: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2305: 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 2306: for (j=1;j<=n;j++) {
1.202 brouard 2307: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2308: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2309: if(j % ncovmodel == 0){
1.191 brouard 2310: printf("\n");
1.202 brouard 2311: fprintf(ficlog,"\n");
2312: }
1.191 brouard 2313: }
1.203 brouard 2314: #else
1.191 brouard 2315: #endif
1.126 brouard 2316: free_vector(xicom,1,n);
2317: free_vector(pcom,1,n);
2318: }
2319:
2320:
2321: /*************** powell ************************/
1.162 brouard 2322: /*
1.317 ! brouard 2323: Minimization of a function func of n variables. Input consists in an initial starting point
! 2324: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
! 2325: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
! 2326: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2327: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2328: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2329: */
1.224 brouard 2330: #ifdef LINMINORIGINAL
2331: #else
2332: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2333: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2334: #endif
1.126 brouard 2335: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2336: double (*func)(double []))
2337: {
1.224 brouard 2338: #ifdef LINMINORIGINAL
2339: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2340: double (*func)(double []));
1.224 brouard 2341: #else
1.241 brouard 2342: void linmin(double p[], double xi[], int n, double *fret,
2343: double (*func)(double []),int *flat);
1.224 brouard 2344: #endif
1.239 brouard 2345: int i,ibig,j,jk,k;
1.126 brouard 2346: double del,t,*pt,*ptt,*xit;
1.181 brouard 2347: double directest;
1.126 brouard 2348: double fp,fptt;
2349: double *xits;
2350: int niterf, itmp;
1.224 brouard 2351: #ifdef LINMINORIGINAL
2352: #else
2353:
2354: flatdir=ivector(1,n);
2355: for (j=1;j<=n;j++) flatdir[j]=0;
2356: #endif
1.126 brouard 2357:
2358: pt=vector(1,n);
2359: ptt=vector(1,n);
2360: xit=vector(1,n);
2361: xits=vector(1,n);
2362: *fret=(*func)(p);
2363: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2364: rcurr_time = time(NULL);
1.126 brouard 2365: for (*iter=1;;++(*iter)) {
1.187 brouard 2366: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2367: ibig=0;
2368: del=0.0;
1.157 brouard 2369: rlast_time=rcurr_time;
2370: /* (void) gettimeofday(&curr_time,&tzp); */
2371: rcurr_time = time(NULL);
2372: curr_time = *localtime(&rcurr_time);
2373: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2374: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2375: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2376: for (i=1;i<=n;i++) {
1.126 brouard 2377: fprintf(ficrespow," %.12lf", p[i]);
2378: }
1.239 brouard 2379: fprintf(ficrespow,"\n");fflush(ficrespow);
2380: printf("\n#model= 1 + age ");
2381: fprintf(ficlog,"\n#model= 1 + age ");
2382: if(nagesqr==1){
1.241 brouard 2383: printf(" + age*age ");
2384: fprintf(ficlog," + age*age ");
1.239 brouard 2385: }
2386: for(j=1;j <=ncovmodel-2;j++){
2387: if(Typevar[j]==0) {
2388: printf(" + V%d ",Tvar[j]);
2389: fprintf(ficlog," + V%d ",Tvar[j]);
2390: }else if(Typevar[j]==1) {
2391: printf(" + V%d*age ",Tvar[j]);
2392: fprintf(ficlog," + V%d*age ",Tvar[j]);
2393: }else if(Typevar[j]==2) {
2394: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2395: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2396: }
2397: }
1.126 brouard 2398: printf("\n");
1.239 brouard 2399: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2400: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2401: fprintf(ficlog,"\n");
1.239 brouard 2402: for(i=1,jk=1; i <=nlstate; i++){
2403: for(k=1; k <=(nlstate+ndeath); k++){
2404: if (k != i) {
2405: printf("%d%d ",i,k);
2406: fprintf(ficlog,"%d%d ",i,k);
2407: for(j=1; j <=ncovmodel; j++){
2408: printf("%12.7f ",p[jk]);
2409: fprintf(ficlog,"%12.7f ",p[jk]);
2410: jk++;
2411: }
2412: printf("\n");
2413: fprintf(ficlog,"\n");
2414: }
2415: }
2416: }
1.241 brouard 2417: if(*iter <=3 && *iter >1){
1.157 brouard 2418: tml = *localtime(&rcurr_time);
2419: strcpy(strcurr,asctime(&tml));
2420: rforecast_time=rcurr_time;
1.126 brouard 2421: itmp = strlen(strcurr);
2422: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2423: strcurr[itmp-1]='\0';
1.162 brouard 2424: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2425: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2426: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2427: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2428: forecast_time = *localtime(&rforecast_time);
2429: strcpy(strfor,asctime(&forecast_time));
2430: itmp = strlen(strfor);
2431: if(strfor[itmp-1]=='\n')
2432: strfor[itmp-1]='\0';
2433: 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);
2434: 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 2435: }
2436: }
1.187 brouard 2437: for (i=1;i<=n;i++) { /* For each direction i */
2438: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2439: fptt=(*fret);
2440: #ifdef DEBUG
1.203 brouard 2441: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2442: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2443: #endif
1.203 brouard 2444: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2445: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2446: #ifdef LINMINORIGINAL
1.188 brouard 2447: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2448: #else
2449: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2450: flatdir[i]=flat; /* Function is vanishing in that direction i */
2451: #endif
2452: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2453: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2454: /* because that direction will be replaced unless the gain del is small */
2455: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2456: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2457: /* with the new direction. */
2458: del=fabs(fptt-(*fret));
2459: ibig=i;
1.126 brouard 2460: }
2461: #ifdef DEBUG
2462: printf("%d %.12e",i,(*fret));
2463: fprintf(ficlog,"%d %.12e",i,(*fret));
2464: for (j=1;j<=n;j++) {
1.224 brouard 2465: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2466: printf(" x(%d)=%.12e",j,xit[j]);
2467: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2468: }
2469: for(j=1;j<=n;j++) {
1.225 brouard 2470: printf(" p(%d)=%.12e",j,p[j]);
2471: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2472: }
2473: printf("\n");
2474: fprintf(ficlog,"\n");
2475: #endif
1.187 brouard 2476: } /* end loop on each direction i */
2477: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2478: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2479: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2480: for(j=1;j<=n;j++) {
1.302 brouard 2481: if(flatdir[j] >0){
2482: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2483: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2484: }
2485: /* printf("\n"); */
2486: /* fprintf(ficlog,"\n"); */
2487: }
1.243 brouard 2488: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2489: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2490: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2491: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2492: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2493: /* decreased of more than 3.84 */
2494: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2495: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2496: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2497:
1.188 brouard 2498: /* Starting the program with initial values given by a former maximization will simply change */
2499: /* the scales of the directions and the directions, because the are reset to canonical directions */
2500: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2501: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2502: #ifdef DEBUG
2503: int k[2],l;
2504: k[0]=1;
2505: k[1]=-1;
2506: printf("Max: %.12e",(*func)(p));
2507: fprintf(ficlog,"Max: %.12e",(*func)(p));
2508: for (j=1;j<=n;j++) {
2509: printf(" %.12e",p[j]);
2510: fprintf(ficlog," %.12e",p[j]);
2511: }
2512: printf("\n");
2513: fprintf(ficlog,"\n");
2514: for(l=0;l<=1;l++) {
2515: for (j=1;j<=n;j++) {
2516: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2517: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2518: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2519: }
2520: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2521: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2522: }
2523: #endif
2524:
1.224 brouard 2525: #ifdef LINMINORIGINAL
2526: #else
2527: free_ivector(flatdir,1,n);
2528: #endif
1.126 brouard 2529: free_vector(xit,1,n);
2530: free_vector(xits,1,n);
2531: free_vector(ptt,1,n);
2532: free_vector(pt,1,n);
2533: return;
1.192 brouard 2534: } /* enough precision */
1.240 brouard 2535: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2536: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2537: ptt[j]=2.0*p[j]-pt[j];
2538: xit[j]=p[j]-pt[j];
2539: pt[j]=p[j];
2540: }
1.181 brouard 2541: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2542: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2543: if (*iter <=4) {
1.225 brouard 2544: #else
2545: #endif
1.224 brouard 2546: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2547: #else
1.161 brouard 2548: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2549: #endif
1.162 brouard 2550: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2551: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2552: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2553: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2554: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2555: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2556: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2557: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2558: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2559: /* Even if f3 <f1, directest can be negative and t >0 */
2560: /* mu² and del² are equal when f3=f1 */
2561: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2562: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2563: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2564: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2565: #ifdef NRCORIGINAL
2566: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2567: #else
2568: 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 2569: t= t- del*SQR(fp-fptt);
1.183 brouard 2570: #endif
1.202 brouard 2571: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2572: #ifdef DEBUG
1.181 brouard 2573: 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);
2574: 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 2575: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2576: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2577: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2578: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2579: 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);
2580: 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);
2581: #endif
1.183 brouard 2582: #ifdef POWELLORIGINAL
2583: if (t < 0.0) { /* Then we use it for new direction */
2584: #else
1.182 brouard 2585: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2586: 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 2587: 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 2588: 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 2589: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2590: }
1.181 brouard 2591: if (directest < 0.0) { /* Then we use it for new direction */
2592: #endif
1.191 brouard 2593: #ifdef DEBUGLINMIN
1.234 brouard 2594: printf("Before linmin in direction P%d-P0\n",n);
2595: for (j=1;j<=n;j++) {
2596: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2597: fprintf(ficlog," Before 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
2604: #ifdef LINMINORIGINAL
1.234 brouard 2605: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2606: #else
1.234 brouard 2607: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2608: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2609: #endif
1.234 brouard 2610:
1.191 brouard 2611: #ifdef DEBUGLINMIN
1.234 brouard 2612: for (j=1;j<=n;j++) {
2613: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2614: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2615: if(j % ncovmodel == 0){
2616: printf("\n");
2617: fprintf(ficlog,"\n");
2618: }
2619: }
1.224 brouard 2620: #endif
1.234 brouard 2621: for (j=1;j<=n;j++) {
2622: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2623: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2624: }
1.224 brouard 2625: #ifdef LINMINORIGINAL
2626: #else
1.234 brouard 2627: for (j=1, flatd=0;j<=n;j++) {
2628: if(flatdir[j]>0)
2629: flatd++;
2630: }
2631: if(flatd >0){
1.255 brouard 2632: printf("%d flat directions: ",flatd);
2633: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2634: for (j=1;j<=n;j++) {
2635: if(flatdir[j]>0){
2636: printf("%d ",j);
2637: fprintf(ficlog,"%d ",j);
2638: }
2639: }
2640: printf("\n");
2641: fprintf(ficlog,"\n");
2642: }
1.191 brouard 2643: #endif
1.234 brouard 2644: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2645: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2646:
1.126 brouard 2647: #ifdef DEBUG
1.234 brouard 2648: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2649: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2650: for(j=1;j<=n;j++){
2651: printf(" %lf",xit[j]);
2652: fprintf(ficlog," %lf",xit[j]);
2653: }
2654: printf("\n");
2655: fprintf(ficlog,"\n");
1.126 brouard 2656: #endif
1.192 brouard 2657: } /* end of t or directest negative */
1.224 brouard 2658: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2659: #else
1.234 brouard 2660: } /* end if (fptt < fp) */
1.192 brouard 2661: #endif
1.225 brouard 2662: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2663: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2664: #else
1.224 brouard 2665: #endif
1.234 brouard 2666: } /* loop iteration */
1.126 brouard 2667: }
1.234 brouard 2668:
1.126 brouard 2669: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2670:
1.235 brouard 2671: 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 2672: {
1.279 brouard 2673: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2674: * (and selected quantitative values in nres)
2675: * by left multiplying the unit
2676: * matrix by transitions matrix until convergence is reached with precision ftolpl
2677: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2678: * Wx is row vector: population in state 1, population in state 2, population dead
2679: * or prevalence in state 1, prevalence in state 2, 0
2680: * newm is the matrix after multiplications, its rows are identical at a factor.
2681: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2682: * Output is prlim.
2683: * Initial matrix pimij
2684: */
1.206 brouard 2685: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2686: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2687: /* 0, 0 , 1} */
2688: /*
2689: * and after some iteration: */
2690: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2691: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2692: /* 0, 0 , 1} */
2693: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2694: /* {0.51571254859325999, 0.4842874514067399, */
2695: /* 0.51326036147820708, 0.48673963852179264} */
2696: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2697:
1.126 brouard 2698: int i, ii,j,k;
1.209 brouard 2699: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2700: /* double **matprod2(); */ /* test */
1.218 brouard 2701: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2702: double **newm;
1.209 brouard 2703: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2704: int ncvloop=0;
1.288 brouard 2705: int first=0;
1.169 brouard 2706:
1.209 brouard 2707: min=vector(1,nlstate);
2708: max=vector(1,nlstate);
2709: meandiff=vector(1,nlstate);
2710:
1.218 brouard 2711: /* Starting with matrix unity */
1.126 brouard 2712: for (ii=1;ii<=nlstate+ndeath;ii++)
2713: for (j=1;j<=nlstate+ndeath;j++){
2714: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2715: }
1.169 brouard 2716:
2717: cov[1]=1.;
2718:
2719: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2720: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2721: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2722: ncvloop++;
1.126 brouard 2723: newm=savm;
2724: /* Covariates have to be included here again */
1.138 brouard 2725: cov[2]=agefin;
1.187 brouard 2726: if(nagesqr==1)
2727: cov[3]= agefin*agefin;;
1.234 brouard 2728: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2729: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2730: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2731: /* 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 2732: }
2733: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2734: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2735: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2736: /* 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 2737: }
1.237 brouard 2738: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2739: if(Dummy[Tvar[Tage[k]]]){
2740: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2741: } else{
1.235 brouard 2742: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2743: }
1.235 brouard 2744: /* 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 2745: }
1.237 brouard 2746: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2747: /* 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 2748: if(Dummy[Tvard[k][1]==0]){
2749: if(Dummy[Tvard[k][2]==0]){
2750: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2751: }else{
2752: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2753: }
2754: }else{
2755: if(Dummy[Tvard[k][2]==0]){
2756: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2757: }else{
2758: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2759: }
2760: }
1.234 brouard 2761: }
1.138 brouard 2762: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2763: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2764: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2765: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2766: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2767: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2768: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2769:
1.126 brouard 2770: savm=oldm;
2771: oldm=newm;
1.209 brouard 2772:
2773: for(j=1; j<=nlstate; j++){
2774: max[j]=0.;
2775: min[j]=1.;
2776: }
2777: for(i=1;i<=nlstate;i++){
2778: sumnew=0;
2779: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2780: for(j=1; j<=nlstate; j++){
2781: prlim[i][j]= newm[i][j]/(1-sumnew);
2782: max[j]=FMAX(max[j],prlim[i][j]);
2783: min[j]=FMIN(min[j],prlim[i][j]);
2784: }
2785: }
2786:
1.126 brouard 2787: maxmax=0.;
1.209 brouard 2788: for(j=1; j<=nlstate; j++){
2789: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2790: maxmax=FMAX(maxmax,meandiff[j]);
2791: /* 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 2792: } /* j loop */
1.203 brouard 2793: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2794: /* 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 2795: if(maxmax < ftolpl){
1.209 brouard 2796: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2797: free_vector(min,1,nlstate);
2798: free_vector(max,1,nlstate);
2799: free_vector(meandiff,1,nlstate);
1.126 brouard 2800: return prlim;
2801: }
1.288 brouard 2802: } /* agefin loop */
1.208 brouard 2803: /* After some age loop it doesn't converge */
1.288 brouard 2804: if(!first){
2805: first=1;
2806: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d). Others in log file only...\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
1.317 ! brouard 2807: 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);
! 2808: }else if (first >=1 && first <10){
! 2809: 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);
! 2810: first++;
! 2811: }else if (first ==10){
! 2812: 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);
! 2813: printf("Warning: the stable prevalence dit not converge. This warning came too often, IMaCh will stop notifying, even in its log file. Look at the graphs to appreciate the non convergence.\n");
! 2814: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
! 2815: first++;
1.288 brouard 2816: }
2817:
1.209 brouard 2818: /* 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); */
2819: free_vector(min,1,nlstate);
2820: free_vector(max,1,nlstate);
2821: free_vector(meandiff,1,nlstate);
1.208 brouard 2822:
1.169 brouard 2823: return prlim; /* should not reach here */
1.126 brouard 2824: }
2825:
1.217 brouard 2826:
2827: /**** Back Prevalence limit (stable or period prevalence) ****************/
2828:
1.218 brouard 2829: /* 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) */
2830: /* 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 2831: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2832: {
1.264 brouard 2833: /* 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 2834: matrix by transitions matrix until convergence is reached with precision ftolpl */
2835: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2836: /* Wx is row vector: population in state 1, population in state 2, population dead */
2837: /* or prevalence in state 1, prevalence in state 2, 0 */
2838: /* newm is the matrix after multiplications, its rows are identical at a factor */
2839: /* Initial matrix pimij */
2840: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2841: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2842: /* 0, 0 , 1} */
2843: /*
2844: * and after some iteration: */
2845: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2846: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2847: /* 0, 0 , 1} */
2848: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2849: /* {0.51571254859325999, 0.4842874514067399, */
2850: /* 0.51326036147820708, 0.48673963852179264} */
2851: /* If we start from prlim again, prlim tends to a constant matrix */
2852:
2853: int i, ii,j,k;
1.247 brouard 2854: int first=0;
1.217 brouard 2855: double *min, *max, *meandiff, maxmax,sumnew=0.;
2856: /* double **matprod2(); */ /* test */
2857: double **out, cov[NCOVMAX+1], **bmij();
2858: double **newm;
1.218 brouard 2859: double **dnewm, **doldm, **dsavm; /* for use */
2860: double **oldm, **savm; /* for use */
2861:
1.217 brouard 2862: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2863: int ncvloop=0;
2864:
2865: min=vector(1,nlstate);
2866: max=vector(1,nlstate);
2867: meandiff=vector(1,nlstate);
2868:
1.266 brouard 2869: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2870: oldm=oldms; savm=savms;
2871:
2872: /* Starting with matrix unity */
2873: for (ii=1;ii<=nlstate+ndeath;ii++)
2874: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2875: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2876: }
2877:
2878: cov[1]=1.;
2879:
2880: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2881: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2882: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2883: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2884: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2885: ncvloop++;
1.218 brouard 2886: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2887: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2888: /* Covariates have to be included here again */
2889: cov[2]=agefin;
2890: if(nagesqr==1)
2891: cov[3]= agefin*agefin;;
1.242 brouard 2892: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2893: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2894: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2895: /* 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 2896: }
2897: /* for (k=1; k<=cptcovn;k++) { */
2898: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2899: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2900: /* /\* 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])]); *\/ */
2901: /* } */
2902: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2903: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2904: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2905: /* 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]); */
2906: }
2907: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2908: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2909: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2910: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2911: for (k=1; k<=cptcovage;k++){ /* For product with age */
2912: if(Dummy[Tvar[Tage[k]]]){
2913: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2914: } else{
2915: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2916: }
2917: /* 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]); */
2918: }
2919: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2920: /* 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]); */
2921: if(Dummy[Tvard[k][1]==0]){
2922: if(Dummy[Tvard[k][2]==0]){
2923: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2924: }else{
2925: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2926: }
2927: }else{
2928: if(Dummy[Tvard[k][2]==0]){
2929: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2930: }else{
2931: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2932: }
2933: }
1.217 brouard 2934: }
2935:
2936: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2937: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2938: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2939: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2940: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2941: /* ij should be linked to the correct index of cov */
2942: /* age and covariate values ij are in 'cov', but we need to pass
2943: * ij for the observed prevalence at age and status and covariate
2944: * number: prevacurrent[(int)agefin][ii][ij]
2945: */
2946: /* 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 *\/ */
2947: /* 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 *\/ */
2948: 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 2949: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2950: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2951: /* for(i=1; i<=nlstate+ndeath; i++) { */
2952: /* printf("%d newm= ",i); */
2953: /* for(j=1;j<=nlstate+ndeath;j++) { */
2954: /* printf("%f ",newm[i][j]); */
2955: /* } */
2956: /* printf("oldm * "); */
2957: /* for(j=1;j<=nlstate+ndeath;j++) { */
2958: /* printf("%f ",oldm[i][j]); */
2959: /* } */
1.268 brouard 2960: /* printf(" bmmij "); */
1.266 brouard 2961: /* for(j=1;j<=nlstate+ndeath;j++) { */
2962: /* printf("%f ",pmmij[i][j]); */
2963: /* } */
2964: /* printf("\n"); */
2965: /* } */
2966: /* } */
1.217 brouard 2967: savm=oldm;
2968: oldm=newm;
1.266 brouard 2969:
1.217 brouard 2970: for(j=1; j<=nlstate; j++){
2971: max[j]=0.;
2972: min[j]=1.;
2973: }
2974: for(j=1; j<=nlstate; j++){
2975: for(i=1;i<=nlstate;i++){
1.234 brouard 2976: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2977: bprlim[i][j]= newm[i][j];
2978: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2979: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2980: }
2981: }
1.218 brouard 2982:
1.217 brouard 2983: maxmax=0.;
2984: for(i=1; i<=nlstate; i++){
2985: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2986: maxmax=FMAX(maxmax,meandiff[i]);
2987: /* 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 2988: } /* i loop */
1.217 brouard 2989: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2990: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2991: if(maxmax < ftolpl){
1.220 brouard 2992: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2993: free_vector(min,1,nlstate);
2994: free_vector(max,1,nlstate);
2995: free_vector(meandiff,1,nlstate);
2996: return bprlim;
2997: }
1.288 brouard 2998: } /* agefin loop */
1.217 brouard 2999: /* After some age loop it doesn't converge */
1.288 brouard 3000: if(!first){
1.247 brouard 3001: first=1;
3002: 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\
3003: 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);
3004: }
3005: 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 3006: 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);
3007: /* 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); */
3008: free_vector(min,1,nlstate);
3009: free_vector(max,1,nlstate);
3010: free_vector(meandiff,1,nlstate);
3011:
3012: return bprlim; /* should not reach here */
3013: }
3014:
1.126 brouard 3015: /*************** transition probabilities ***************/
3016:
3017: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3018: {
1.138 brouard 3019: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3020: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3021: model to the ncovmodel covariates (including constant and age).
3022: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3023: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3024: ncth covariate in the global vector x is given by the formula:
3025: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3026: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3027: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3028: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3029: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3030: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3031: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3032: */
3033: double s1, lnpijopii;
1.126 brouard 3034: /*double t34;*/
1.164 brouard 3035: int i,j, nc, ii, jj;
1.126 brouard 3036:
1.223 brouard 3037: for(i=1; i<= nlstate; i++){
3038: for(j=1; j<i;j++){
3039: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3040: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3041: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3042: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3043: }
3044: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3045: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3046: }
3047: for(j=i+1; j<=nlstate+ndeath;j++){
3048: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3049: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3050: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3051: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3052: }
3053: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3054: }
3055: }
1.218 brouard 3056:
1.223 brouard 3057: for(i=1; i<= nlstate; i++){
3058: s1=0;
3059: for(j=1; j<i; j++){
3060: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3061: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3062: }
3063: for(j=i+1; j<=nlstate+ndeath; j++){
3064: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3065: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3066: }
3067: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3068: ps[i][i]=1./(s1+1.);
3069: /* Computing other pijs */
3070: for(j=1; j<i; j++)
3071: ps[i][j]= exp(ps[i][j])*ps[i][i];
3072: for(j=i+1; j<=nlstate+ndeath; j++)
3073: ps[i][j]= exp(ps[i][j])*ps[i][i];
3074: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3075: } /* end i */
1.218 brouard 3076:
1.223 brouard 3077: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3078: for(jj=1; jj<= nlstate+ndeath; jj++){
3079: ps[ii][jj]=0;
3080: ps[ii][ii]=1;
3081: }
3082: }
1.294 brouard 3083:
3084:
1.223 brouard 3085: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3086: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3087: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3088: /* } */
3089: /* printf("\n "); */
3090: /* } */
3091: /* printf("\n ");printf("%lf ",cov[2]);*/
3092: /*
3093: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3094: goto end;*/
1.266 brouard 3095: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3096: }
3097:
1.218 brouard 3098: /*************** backward transition probabilities ***************/
3099:
3100: /* 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 ) */
3101: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3102: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3103: {
1.302 brouard 3104: /* 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 3105: * 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 3106: */
1.218 brouard 3107: int i, ii, j,k;
1.222 brouard 3108:
3109: double **out, **pmij();
3110: double sumnew=0.;
1.218 brouard 3111: double agefin;
1.292 brouard 3112: 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 3113: double **dnewm, **dsavm, **doldm;
3114: double **bbmij;
3115:
1.218 brouard 3116: doldm=ddoldms; /* global pointers */
1.222 brouard 3117: dnewm=ddnewms;
3118: dsavm=ddsavms;
3119:
3120: agefin=cov[2];
1.268 brouard 3121: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3122: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3123: the observed prevalence (with this covariate ij) at beginning of transition */
3124: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3125:
3126: /* P_x */
1.266 brouard 3127: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3128: /* outputs pmmij which is a stochastic matrix in row */
3129:
3130: /* Diag(w_x) */
1.292 brouard 3131: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3132: sumnew=0.;
1.269 brouard 3133: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3134: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3135: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3136: sumnew+=prevacurrent[(int)agefin][ii][ij];
3137: }
3138: if(sumnew >0.01){ /* At least some value in the prevalence */
3139: for (ii=1;ii<=nlstate+ndeath;ii++){
3140: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3141: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3142: }
3143: }else{
3144: for (ii=1;ii<=nlstate+ndeath;ii++){
3145: for (j=1;j<=nlstate+ndeath;j++)
3146: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3147: }
3148: /* if(sumnew <0.9){ */
3149: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3150: /* } */
3151: }
3152: k3=0.0; /* We put the last diagonal to 0 */
3153: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3154: doldm[ii][ii]= k3;
3155: }
3156: /* End doldm, At the end doldm is diag[(w_i)] */
3157:
1.292 brouard 3158: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3159: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3160:
1.292 brouard 3161: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3162: /* 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 3163: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3164: sumnew=0.;
1.222 brouard 3165: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3166: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3167: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3168: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3169: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3170: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3171: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3172: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3173: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3174: /* }else */
1.268 brouard 3175: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3176: } /*End ii */
3177: } /* 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 */
3178:
1.292 brouard 3179: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3180: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3181: /* end bmij */
1.266 brouard 3182: return ps; /*pointer is unchanged */
1.218 brouard 3183: }
1.217 brouard 3184: /*************** transition probabilities ***************/
3185:
1.218 brouard 3186: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3187: {
3188: /* According to parameters values stored in x and the covariate's values stored in cov,
3189: computes the probability to be observed in state j being in state i by appying the
3190: model to the ncovmodel covariates (including constant and age).
3191: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3192: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3193: ncth covariate in the global vector x is given by the formula:
3194: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3195: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3196: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3197: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3198: Outputs ps[i][j] the probability to be observed in j being in j according to
3199: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3200: */
3201: double s1, lnpijopii;
3202: /*double t34;*/
3203: int i,j, nc, ii, jj;
3204:
1.234 brouard 3205: for(i=1; i<= nlstate; i++){
3206: for(j=1; j<i;j++){
3207: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3208: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3209: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3210: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3211: }
3212: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3213: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3214: }
3215: for(j=i+1; j<=nlstate+ndeath;j++){
3216: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3217: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3218: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3219: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3220: }
3221: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3222: }
3223: }
3224:
3225: for(i=1; i<= nlstate; i++){
3226: s1=0;
3227: for(j=1; j<i; j++){
3228: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3229: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3230: }
3231: for(j=i+1; j<=nlstate+ndeath; j++){
3232: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3233: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3234: }
3235: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3236: ps[i][i]=1./(s1+1.);
3237: /* Computing other pijs */
3238: for(j=1; j<i; j++)
3239: ps[i][j]= exp(ps[i][j])*ps[i][i];
3240: for(j=i+1; j<=nlstate+ndeath; j++)
3241: ps[i][j]= exp(ps[i][j])*ps[i][i];
3242: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3243: } /* end i */
3244:
3245: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3246: for(jj=1; jj<= nlstate+ndeath; jj++){
3247: ps[ii][jj]=0;
3248: ps[ii][ii]=1;
3249: }
3250: }
1.296 brouard 3251: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3252: for(jj=1; jj<= nlstate+ndeath; jj++){
3253: s1=0.;
3254: for(ii=1; ii<= nlstate+ndeath; ii++){
3255: s1+=ps[ii][jj];
3256: }
3257: for(ii=1; ii<= nlstate; ii++){
3258: ps[ii][jj]=ps[ii][jj]/s1;
3259: }
3260: }
3261: /* Transposition */
3262: for(jj=1; jj<= nlstate+ndeath; jj++){
3263: for(ii=jj; ii<= nlstate+ndeath; ii++){
3264: s1=ps[ii][jj];
3265: ps[ii][jj]=ps[jj][ii];
3266: ps[jj][ii]=s1;
3267: }
3268: }
3269: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3270: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3271: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3272: /* } */
3273: /* printf("\n "); */
3274: /* } */
3275: /* printf("\n ");printf("%lf ",cov[2]);*/
3276: /*
3277: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3278: goto end;*/
3279: return ps;
1.217 brouard 3280: }
3281:
3282:
1.126 brouard 3283: /**************** Product of 2 matrices ******************/
3284:
1.145 brouard 3285: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3286: {
3287: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3288: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3289: /* in, b, out are matrice of pointers which should have been initialized
3290: before: only the contents of out is modified. The function returns
3291: a pointer to pointers identical to out */
1.145 brouard 3292: int i, j, k;
1.126 brouard 3293: for(i=nrl; i<= nrh; i++)
1.145 brouard 3294: for(k=ncolol; k<=ncoloh; k++){
3295: out[i][k]=0.;
3296: for(j=ncl; j<=nch; j++)
3297: out[i][k] +=in[i][j]*b[j][k];
3298: }
1.126 brouard 3299: return out;
3300: }
3301:
3302:
3303: /************* Higher Matrix Product ***************/
3304:
1.235 brouard 3305: 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 3306: {
1.218 brouard 3307: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3308: 'nhstepm*hstepm*stepm' months (i.e. until
3309: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3310: nhstepm*hstepm matrices.
3311: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3312: (typically every 2 years instead of every month which is too big
3313: for the memory).
3314: Model is determined by parameters x and covariates have to be
3315: included manually here.
3316:
3317: */
3318:
3319: int i, j, d, h, k;
1.131 brouard 3320: double **out, cov[NCOVMAX+1];
1.126 brouard 3321: double **newm;
1.187 brouard 3322: double agexact;
1.214 brouard 3323: double agebegin, ageend;
1.126 brouard 3324:
3325: /* Hstepm could be zero and should return the unit matrix */
3326: for (i=1;i<=nlstate+ndeath;i++)
3327: for (j=1;j<=nlstate+ndeath;j++){
3328: oldm[i][j]=(i==j ? 1.0 : 0.0);
3329: po[i][j][0]=(i==j ? 1.0 : 0.0);
3330: }
3331: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3332: for(h=1; h <=nhstepm; h++){
3333: for(d=1; d <=hstepm; d++){
3334: newm=savm;
3335: /* Covariates have to be included here again */
3336: cov[1]=1.;
1.214 brouard 3337: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3338: cov[2]=agexact;
3339: if(nagesqr==1)
1.227 brouard 3340: cov[3]= agexact*agexact;
1.235 brouard 3341: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3342: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3343: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3344: /* 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)); */
3345: }
3346: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3347: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3348: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3349: /* 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]); */
3350: }
3351: for (k=1; k<=cptcovage;k++){
3352: if(Dummy[Tvar[Tage[k]]]){
3353: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3354: } else{
3355: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3356: }
3357: /* 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]); */
3358: }
3359: for (k=1; k<=cptcovprod;k++){ /* */
3360: /* 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]); */
3361: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3362: }
3363: /* for (k=1; k<=cptcovn;k++) */
3364: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3365: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3366: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3367: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3368: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3369:
3370:
1.126 brouard 3371: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3372: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3373: /* right multiplication of oldm by the current matrix */
1.126 brouard 3374: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3375: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3376: /* if((int)age == 70){ */
3377: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3378: /* for(i=1; i<=nlstate+ndeath; i++) { */
3379: /* printf("%d pmmij ",i); */
3380: /* for(j=1;j<=nlstate+ndeath;j++) { */
3381: /* printf("%f ",pmmij[i][j]); */
3382: /* } */
3383: /* printf(" oldm "); */
3384: /* for(j=1;j<=nlstate+ndeath;j++) { */
3385: /* printf("%f ",oldm[i][j]); */
3386: /* } */
3387: /* printf("\n"); */
3388: /* } */
3389: /* } */
1.126 brouard 3390: savm=oldm;
3391: oldm=newm;
3392: }
3393: for(i=1; i<=nlstate+ndeath; i++)
3394: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3395: po[i][j][h]=newm[i][j];
3396: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3397: }
1.128 brouard 3398: /*printf("h=%d ",h);*/
1.126 brouard 3399: } /* end h */
1.267 brouard 3400: /* printf("\n H=%d \n",h); */
1.126 brouard 3401: return po;
3402: }
3403:
1.217 brouard 3404: /************* Higher Back Matrix Product ***************/
1.218 brouard 3405: /* 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 3406: 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 3407: {
1.266 brouard 3408: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3409: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3410: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3411: nhstepm*hstepm matrices.
3412: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3413: (typically every 2 years instead of every month which is too big
1.217 brouard 3414: for the memory).
1.218 brouard 3415: Model is determined by parameters x and covariates have to be
1.266 brouard 3416: included manually here. Then we use a call to bmij(x and cov)
3417: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3418: */
1.217 brouard 3419:
3420: int i, j, d, h, k;
1.266 brouard 3421: double **out, cov[NCOVMAX+1], **bmij();
3422: double **newm, ***newmm;
1.217 brouard 3423: double agexact;
3424: double agebegin, ageend;
1.222 brouard 3425: double **oldm, **savm;
1.217 brouard 3426:
1.266 brouard 3427: newmm=po; /* To be saved */
3428: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3429: /* Hstepm could be zero and should return the unit matrix */
3430: for (i=1;i<=nlstate+ndeath;i++)
3431: for (j=1;j<=nlstate+ndeath;j++){
3432: oldm[i][j]=(i==j ? 1.0 : 0.0);
3433: po[i][j][0]=(i==j ? 1.0 : 0.0);
3434: }
3435: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3436: for(h=1; h <=nhstepm; h++){
3437: for(d=1; d <=hstepm; d++){
3438: newm=savm;
3439: /* Covariates have to be included here again */
3440: cov[1]=1.;
1.271 brouard 3441: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3442: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3443: cov[2]=agexact;
3444: if(nagesqr==1)
1.222 brouard 3445: cov[3]= agexact*agexact;
1.266 brouard 3446: for (k=1; k<=cptcovn;k++){
3447: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3448: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3449: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3450: /* 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)); */
3451: }
1.267 brouard 3452: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3453: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3454: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3455: /* 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]); */
3456: }
3457: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3458: if(Dummy[Tvar[Tage[k]]]){
3459: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3460: } else{
3461: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3462: }
3463: /* 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]); */
3464: }
3465: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3466: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3467: }
1.217 brouard 3468: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3469: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3470:
1.218 brouard 3471: /* Careful transposed matrix */
1.266 brouard 3472: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3473: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3474: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3475: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3476: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3477: /* if((int)age == 70){ */
3478: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3479: /* for(i=1; i<=nlstate+ndeath; i++) { */
3480: /* printf("%d pmmij ",i); */
3481: /* for(j=1;j<=nlstate+ndeath;j++) { */
3482: /* printf("%f ",pmmij[i][j]); */
3483: /* } */
3484: /* printf(" oldm "); */
3485: /* for(j=1;j<=nlstate+ndeath;j++) { */
3486: /* printf("%f ",oldm[i][j]); */
3487: /* } */
3488: /* printf("\n"); */
3489: /* } */
3490: /* } */
3491: savm=oldm;
3492: oldm=newm;
3493: }
3494: for(i=1; i<=nlstate+ndeath; i++)
3495: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3496: po[i][j][h]=newm[i][j];
1.268 brouard 3497: /* if(h==nhstepm) */
3498: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3499: }
1.268 brouard 3500: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3501: } /* end h */
1.268 brouard 3502: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3503: return po;
3504: }
3505:
3506:
1.162 brouard 3507: #ifdef NLOPT
3508: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3509: double fret;
3510: double *xt;
3511: int j;
3512: myfunc_data *d2 = (myfunc_data *) pd;
3513: /* xt = (p1-1); */
3514: xt=vector(1,n);
3515: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3516:
3517: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3518: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3519: printf("Function = %.12lf ",fret);
3520: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3521: printf("\n");
3522: free_vector(xt,1,n);
3523: return fret;
3524: }
3525: #endif
1.126 brouard 3526:
3527: /*************** log-likelihood *************/
3528: double func( double *x)
3529: {
1.226 brouard 3530: int i, ii, j, k, mi, d, kk;
3531: int ioffset=0;
3532: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3533: double **out;
3534: double lli; /* Individual log likelihood */
3535: int s1, s2;
1.228 brouard 3536: 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 3537: double bbh, survp;
3538: long ipmx;
3539: double agexact;
3540: /*extern weight */
3541: /* We are differentiating ll according to initial status */
3542: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3543: /*for(i=1;i<imx;i++)
3544: printf(" %d\n",s[4][i]);
3545: */
1.162 brouard 3546:
1.226 brouard 3547: ++countcallfunc;
1.162 brouard 3548:
1.226 brouard 3549: cov[1]=1.;
1.126 brouard 3550:
1.226 brouard 3551: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3552: ioffset=0;
1.226 brouard 3553: if(mle==1){
3554: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3555: /* Computes the values of the ncovmodel covariates of the model
3556: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3557: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3558: to be observed in j being in i according to the model.
3559: */
1.243 brouard 3560: ioffset=2+nagesqr ;
1.233 brouard 3561: /* Fixed */
1.234 brouard 3562: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3563: 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)*/
3564: }
1.226 brouard 3565: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3566: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3567: has been calculated etc */
3568: /* For an individual i, wav[i] gives the number of effective waves */
3569: /* We compute the contribution to Likelihood of each effective transition
3570: mw[mi][i] is real wave of the mi th effectve wave */
3571: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3572: s2=s[mw[mi+1][i]][i];
3573: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3574: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3575: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3576: */
3577: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3578: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3579: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3580: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3581: }
3582: for (ii=1;ii<=nlstate+ndeath;ii++)
3583: for (j=1;j<=nlstate+ndeath;j++){
3584: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3585: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3586: }
3587: for(d=0; d<dh[mi][i]; d++){
3588: newm=savm;
3589: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3590: cov[2]=agexact;
3591: if(nagesqr==1)
3592: cov[3]= agexact*agexact; /* Should be changed here */
3593: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3594: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3595: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3596: else
3597: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3598: }
3599: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3600: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3601: savm=oldm;
3602: oldm=newm;
3603: } /* end mult */
3604:
3605: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3606: /* But now since version 0.9 we anticipate for bias at large stepm.
3607: * If stepm is larger than one month (smallest stepm) and if the exact delay
3608: * (in months) between two waves is not a multiple of stepm, we rounded to
3609: * the nearest (and in case of equal distance, to the lowest) interval but now
3610: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3611: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3612: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3613: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3614: * -stepm/2 to stepm/2 .
3615: * For stepm=1 the results are the same as for previous versions of Imach.
3616: * For stepm > 1 the results are less biased than in previous versions.
3617: */
1.234 brouard 3618: s1=s[mw[mi][i]][i];
3619: s2=s[mw[mi+1][i]][i];
3620: bbh=(double)bh[mi][i]/(double)stepm;
3621: /* bias bh is positive if real duration
3622: * is higher than the multiple of stepm and negative otherwise.
3623: */
3624: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3625: if( s2 > nlstate){
3626: /* i.e. if s2 is a death state and if the date of death is known
3627: then the contribution to the likelihood is the probability to
3628: die between last step unit time and current step unit time,
3629: which is also equal to probability to die before dh
3630: minus probability to die before dh-stepm .
3631: In version up to 0.92 likelihood was computed
3632: as if date of death was unknown. Death was treated as any other
3633: health state: the date of the interview describes the actual state
3634: and not the date of a change in health state. The former idea was
3635: to consider that at each interview the state was recorded
3636: (healthy, disable or death) and IMaCh was corrected; but when we
3637: introduced the exact date of death then we should have modified
3638: the contribution of an exact death to the likelihood. This new
3639: contribution is smaller and very dependent of the step unit
3640: stepm. It is no more the probability to die between last interview
3641: and month of death but the probability to survive from last
3642: interview up to one month before death multiplied by the
3643: probability to die within a month. Thanks to Chris
3644: Jackson for correcting this bug. Former versions increased
3645: mortality artificially. The bad side is that we add another loop
3646: which slows down the processing. The difference can be up to 10%
3647: lower mortality.
3648: */
3649: /* If, at the beginning of the maximization mostly, the
3650: cumulative probability or probability to be dead is
3651: constant (ie = 1) over time d, the difference is equal to
3652: 0. out[s1][3] = savm[s1][3]: probability, being at state
3653: s1 at precedent wave, to be dead a month before current
3654: wave is equal to probability, being at state s1 at
3655: precedent wave, to be dead at mont of the current
3656: wave. Then the observed probability (that this person died)
3657: is null according to current estimated parameter. In fact,
3658: it should be very low but not zero otherwise the log go to
3659: infinity.
3660: */
1.183 brouard 3661: /* #ifdef INFINITYORIGINAL */
3662: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3663: /* #else */
3664: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3665: /* lli=log(mytinydouble); */
3666: /* else */
3667: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3668: /* #endif */
1.226 brouard 3669: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3670:
1.226 brouard 3671: } else if ( s2==-1 ) { /* alive */
3672: for (j=1,survp=0. ; j<=nlstate; j++)
3673: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3674: /*survp += out[s1][j]; */
3675: lli= log(survp);
3676: }
3677: else if (s2==-4) {
3678: for (j=3,survp=0. ; j<=nlstate; j++)
3679: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3680: lli= log(survp);
3681: }
3682: else if (s2==-5) {
3683: for (j=1,survp=0. ; j<=2; j++)
3684: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3685: lli= log(survp);
3686: }
3687: else{
3688: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3689: /* 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 */
3690: }
3691: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3692: /*if(lli ==000.0)*/
3693: /*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); */
3694: ipmx +=1;
3695: sw += weight[i];
3696: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3697: /* if (lli < log(mytinydouble)){ */
3698: /* 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); */
3699: /* 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]); */
3700: /* } */
3701: } /* end of wave */
3702: } /* end of individual */
3703: } else if(mle==2){
3704: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3705: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3706: for(mi=1; mi<= wav[i]-1; mi++){
3707: for (ii=1;ii<=nlstate+ndeath;ii++)
3708: for (j=1;j<=nlstate+ndeath;j++){
3709: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3710: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3711: }
3712: for(d=0; d<=dh[mi][i]; d++){
3713: newm=savm;
3714: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3715: cov[2]=agexact;
3716: if(nagesqr==1)
3717: cov[3]= agexact*agexact;
3718: for (kk=1; kk<=cptcovage;kk++) {
3719: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3720: }
3721: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3722: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3723: savm=oldm;
3724: oldm=newm;
3725: } /* end mult */
3726:
3727: s1=s[mw[mi][i]][i];
3728: s2=s[mw[mi+1][i]][i];
3729: bbh=(double)bh[mi][i]/(double)stepm;
3730: 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 */
3731: ipmx +=1;
3732: sw += weight[i];
3733: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3734: } /* end of wave */
3735: } /* end of individual */
3736: } else if(mle==3){ /* exponential inter-extrapolation */
3737: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3738: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3739: for(mi=1; mi<= wav[i]-1; mi++){
3740: for (ii=1;ii<=nlstate+ndeath;ii++)
3741: for (j=1;j<=nlstate+ndeath;j++){
3742: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3743: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3744: }
3745: for(d=0; d<dh[mi][i]; d++){
3746: newm=savm;
3747: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3748: cov[2]=agexact;
3749: if(nagesqr==1)
3750: cov[3]= agexact*agexact;
3751: for (kk=1; kk<=cptcovage;kk++) {
3752: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3753: }
3754: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3755: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3756: savm=oldm;
3757: oldm=newm;
3758: } /* end mult */
3759:
3760: s1=s[mw[mi][i]][i];
3761: s2=s[mw[mi+1][i]][i];
3762: bbh=(double)bh[mi][i]/(double)stepm;
3763: 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 */
3764: ipmx +=1;
3765: sw += weight[i];
3766: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3767: } /* end of wave */
3768: } /* end of individual */
3769: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3770: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3771: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3772: for(mi=1; mi<= wav[i]-1; mi++){
3773: for (ii=1;ii<=nlstate+ndeath;ii++)
3774: for (j=1;j<=nlstate+ndeath;j++){
3775: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3776: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3777: }
3778: for(d=0; d<dh[mi][i]; d++){
3779: newm=savm;
3780: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3781: cov[2]=agexact;
3782: if(nagesqr==1)
3783: cov[3]= agexact*agexact;
3784: for (kk=1; kk<=cptcovage;kk++) {
3785: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3786: }
1.126 brouard 3787:
1.226 brouard 3788: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3789: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3790: savm=oldm;
3791: oldm=newm;
3792: } /* end mult */
3793:
3794: s1=s[mw[mi][i]][i];
3795: s2=s[mw[mi+1][i]][i];
3796: if( s2 > nlstate){
3797: lli=log(out[s1][s2] - savm[s1][s2]);
3798: } else if ( s2==-1 ) { /* alive */
3799: for (j=1,survp=0. ; j<=nlstate; j++)
3800: survp += out[s1][j];
3801: lli= log(survp);
3802: }else{
3803: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3804: }
3805: ipmx +=1;
3806: sw += weight[i];
3807: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3808: /* 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 3809: } /* end of wave */
3810: } /* end of individual */
3811: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3812: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3813: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3814: for(mi=1; mi<= wav[i]-1; mi++){
3815: for (ii=1;ii<=nlstate+ndeath;ii++)
3816: for (j=1;j<=nlstate+ndeath;j++){
3817: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3818: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3819: }
3820: for(d=0; d<dh[mi][i]; d++){
3821: newm=savm;
3822: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3823: cov[2]=agexact;
3824: if(nagesqr==1)
3825: cov[3]= agexact*agexact;
3826: for (kk=1; kk<=cptcovage;kk++) {
3827: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3828: }
1.126 brouard 3829:
1.226 brouard 3830: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3831: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3832: savm=oldm;
3833: oldm=newm;
3834: } /* end mult */
3835:
3836: s1=s[mw[mi][i]][i];
3837: s2=s[mw[mi+1][i]][i];
3838: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3839: ipmx +=1;
3840: sw += weight[i];
3841: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3842: /*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]);*/
3843: } /* end of wave */
3844: } /* end of individual */
3845: } /* End of if */
3846: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3847: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3848: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3849: return -l;
1.126 brouard 3850: }
3851:
3852: /*************** log-likelihood *************/
3853: double funcone( double *x)
3854: {
1.228 brouard 3855: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3856: int i, ii, j, k, mi, d, kk;
1.228 brouard 3857: int ioffset=0;
1.131 brouard 3858: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3859: double **out;
3860: double lli; /* Individual log likelihood */
3861: double llt;
3862: int s1, s2;
1.228 brouard 3863: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3864:
1.126 brouard 3865: double bbh, survp;
1.187 brouard 3866: double agexact;
1.214 brouard 3867: double agebegin, ageend;
1.126 brouard 3868: /*extern weight */
3869: /* We are differentiating ll according to initial status */
3870: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3871: /*for(i=1;i<imx;i++)
3872: printf(" %d\n",s[4][i]);
3873: */
3874: cov[1]=1.;
3875:
3876: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3877: ioffset=0;
3878: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3879: /* ioffset=2+nagesqr+cptcovage; */
3880: ioffset=2+nagesqr;
1.232 brouard 3881: /* Fixed */
1.224 brouard 3882: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3883: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 3884: 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 3885: 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)*/
3886: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3887: /* cov[2+6]=covar[Tvar[6]][i]; */
3888: /* cov[2+6]=covar[2][i]; V2 */
3889: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3890: /* cov[2+7]=covar[Tvar[7]][i]; */
3891: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3892: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3893: /* cov[2+9]=covar[Tvar[9]][i]; */
3894: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3895: }
1.232 brouard 3896: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3897: /* 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?)*\/ */
3898: /* } */
1.231 brouard 3899: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3900: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3901: /* } */
1.225 brouard 3902:
1.233 brouard 3903:
3904: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3905: /* Wave varying (but not age varying) */
3906: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3907: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3908: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3909: }
1.232 brouard 3910: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3911: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3912: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3913: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3914: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3915: /* 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 3916: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3917: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3918: /* /\* 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]); *\/ */
3919: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3920: /* } */
1.126 brouard 3921: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3922: for (j=1;j<=nlstate+ndeath;j++){
3923: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3924: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3925: }
1.214 brouard 3926:
3927: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3928: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3929: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3930: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3931: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3932: and mw[mi+1][i]. dh depends on stepm.*/
3933: newm=savm;
1.247 brouard 3934: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3935: cov[2]=agexact;
3936: if(nagesqr==1)
3937: cov[3]= agexact*agexact;
3938: for (kk=1; kk<=cptcovage;kk++) {
3939: if(!FixedV[Tvar[Tage[kk]]])
3940: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3941: else
3942: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3943: }
3944: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3945: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3946: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3947: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3948: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3949: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3950: savm=oldm;
3951: oldm=newm;
1.126 brouard 3952: } /* end mult */
3953:
3954: s1=s[mw[mi][i]][i];
3955: s2=s[mw[mi+1][i]][i];
1.217 brouard 3956: /* if(s2==-1){ */
1.268 brouard 3957: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3958: /* /\* exit(1); *\/ */
3959: /* } */
1.126 brouard 3960: bbh=(double)bh[mi][i]/(double)stepm;
3961: /* bias is positive if real duration
3962: * is higher than the multiple of stepm and negative otherwise.
3963: */
3964: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3965: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3966: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3967: for (j=1,survp=0. ; j<=nlstate; j++)
3968: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3969: lli= log(survp);
1.126 brouard 3970: }else if (mle==1){
1.242 brouard 3971: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3972: } else if(mle==2){
1.242 brouard 3973: 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 3974: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3975: 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 3976: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3977: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3978: } else{ /* mle=0 back to 1 */
1.242 brouard 3979: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3980: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3981: } /* End of if */
3982: ipmx +=1;
3983: sw += weight[i];
3984: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3985: /*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 3986: if(globpr){
1.246 brouard 3987: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3988: %11.6f %11.6f %11.6f ", \
1.242 brouard 3989: 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 3990: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3991: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3992: llt +=ll[k]*gipmx/gsw;
3993: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3994: }
3995: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3996: }
1.232 brouard 3997: } /* end of wave */
3998: } /* end of individual */
3999: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4000: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4001: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4002: if(globpr==0){ /* First time we count the contributions and weights */
4003: gipmx=ipmx;
4004: gsw=sw;
4005: }
4006: return -l;
1.126 brouard 4007: }
4008:
4009:
4010: /*************** function likelione ***********/
1.292 brouard 4011: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4012: {
4013: /* This routine should help understanding what is done with
4014: the selection of individuals/waves and
4015: to check the exact contribution to the likelihood.
4016: Plotting could be done.
4017: */
4018: int k;
4019:
4020: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4021: strcpy(fileresilk,"ILK_");
1.202 brouard 4022: strcat(fileresilk,fileresu);
1.126 brouard 4023: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4024: printf("Problem with resultfile: %s\n", fileresilk);
4025: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4026: }
1.214 brouard 4027: 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");
4028: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4029: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4030: for(k=1; k<=nlstate; k++)
4031: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4032: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4033: }
4034:
1.292 brouard 4035: *fretone=(*func)(p);
1.126 brouard 4036: if(*globpri !=0){
4037: fclose(ficresilk);
1.205 brouard 4038: if (mle ==0)
4039: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4040: else if(mle >=1)
4041: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4042: 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 4043: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4044:
4045: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4046: 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 4047: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4048: }
1.207 brouard 4049: 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 4050: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4051: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4052: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4053: fflush(fichtm);
1.205 brouard 4054: }
1.126 brouard 4055: return;
4056: }
4057:
4058:
4059: /*********** Maximum Likelihood Estimation ***************/
4060:
4061: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4062: {
1.165 brouard 4063: int i,j, iter=0;
1.126 brouard 4064: double **xi;
4065: double fret;
4066: double fretone; /* Only one call to likelihood */
4067: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4068:
4069: #ifdef NLOPT
4070: int creturn;
4071: nlopt_opt opt;
4072: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4073: double *lb;
4074: double minf; /* the minimum objective value, upon return */
4075: double * p1; /* Shifted parameters from 0 instead of 1 */
4076: myfunc_data dinst, *d = &dinst;
4077: #endif
4078:
4079:
1.126 brouard 4080: xi=matrix(1,npar,1,npar);
4081: for (i=1;i<=npar;i++)
4082: for (j=1;j<=npar;j++)
4083: xi[i][j]=(i==j ? 1.0 : 0.0);
4084: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4085: strcpy(filerespow,"POW_");
1.126 brouard 4086: strcat(filerespow,fileres);
4087: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4088: printf("Problem with resultfile: %s\n", filerespow);
4089: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4090: }
4091: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4092: for (i=1;i<=nlstate;i++)
4093: for(j=1;j<=nlstate+ndeath;j++)
4094: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4095: fprintf(ficrespow,"\n");
1.162 brouard 4096: #ifdef POWELL
1.126 brouard 4097: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4098: #endif
1.126 brouard 4099:
1.162 brouard 4100: #ifdef NLOPT
4101: #ifdef NEWUOA
4102: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4103: #else
4104: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4105: #endif
4106: lb=vector(0,npar-1);
4107: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4108: nlopt_set_lower_bounds(opt, lb);
4109: nlopt_set_initial_step1(opt, 0.1);
4110:
4111: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4112: d->function = func;
4113: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4114: nlopt_set_min_objective(opt, myfunc, d);
4115: nlopt_set_xtol_rel(opt, ftol);
4116: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4117: printf("nlopt failed! %d\n",creturn);
4118: }
4119: else {
4120: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4121: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4122: iter=1; /* not equal */
4123: }
4124: nlopt_destroy(opt);
4125: #endif
1.126 brouard 4126: free_matrix(xi,1,npar,1,npar);
4127: fclose(ficrespow);
1.203 brouard 4128: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4129: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4130: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4131:
4132: }
4133:
4134: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4135: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4136: {
4137: double **a,**y,*x,pd;
1.203 brouard 4138: /* double **hess; */
1.164 brouard 4139: int i, j;
1.126 brouard 4140: int *indx;
4141:
4142: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4143: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4144: void lubksb(double **a, int npar, int *indx, double b[]) ;
4145: void ludcmp(double **a, int npar, int *indx, double *d) ;
4146: double gompertz(double p[]);
1.203 brouard 4147: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4148:
4149: printf("\nCalculation of the hessian matrix. Wait...\n");
4150: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4151: for (i=1;i<=npar;i++){
1.203 brouard 4152: printf("%d-",i);fflush(stdout);
4153: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4154:
4155: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4156:
4157: /* printf(" %f ",p[i]);
4158: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4159: }
4160:
4161: for (i=1;i<=npar;i++) {
4162: for (j=1;j<=npar;j++) {
4163: if (j>i) {
1.203 brouard 4164: printf(".%d-%d",i,j);fflush(stdout);
4165: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4166: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4167:
4168: hess[j][i]=hess[i][j];
4169: /*printf(" %lf ",hess[i][j]);*/
4170: }
4171: }
4172: }
4173: printf("\n");
4174: fprintf(ficlog,"\n");
4175:
4176: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4177: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4178:
4179: a=matrix(1,npar,1,npar);
4180: y=matrix(1,npar,1,npar);
4181: x=vector(1,npar);
4182: indx=ivector(1,npar);
4183: for (i=1;i<=npar;i++)
4184: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4185: ludcmp(a,npar,indx,&pd);
4186:
4187: for (j=1;j<=npar;j++) {
4188: for (i=1;i<=npar;i++) x[i]=0;
4189: x[j]=1;
4190: lubksb(a,npar,indx,x);
4191: for (i=1;i<=npar;i++){
4192: matcov[i][j]=x[i];
4193: }
4194: }
4195:
4196: printf("\n#Hessian matrix#\n");
4197: fprintf(ficlog,"\n#Hessian matrix#\n");
4198: for (i=1;i<=npar;i++) {
4199: for (j=1;j<=npar;j++) {
1.203 brouard 4200: printf("%.6e ",hess[i][j]);
4201: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4202: }
4203: printf("\n");
4204: fprintf(ficlog,"\n");
4205: }
4206:
1.203 brouard 4207: /* printf("\n#Covariance matrix#\n"); */
4208: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4209: /* for (i=1;i<=npar;i++) { */
4210: /* for (j=1;j<=npar;j++) { */
4211: /* printf("%.6e ",matcov[i][j]); */
4212: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4213: /* } */
4214: /* printf("\n"); */
4215: /* fprintf(ficlog,"\n"); */
4216: /* } */
4217:
1.126 brouard 4218: /* Recompute Inverse */
1.203 brouard 4219: /* for (i=1;i<=npar;i++) */
4220: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4221: /* ludcmp(a,npar,indx,&pd); */
4222:
4223: /* printf("\n#Hessian matrix recomputed#\n"); */
4224:
4225: /* for (j=1;j<=npar;j++) { */
4226: /* for (i=1;i<=npar;i++) x[i]=0; */
4227: /* x[j]=1; */
4228: /* lubksb(a,npar,indx,x); */
4229: /* for (i=1;i<=npar;i++){ */
4230: /* y[i][j]=x[i]; */
4231: /* printf("%.3e ",y[i][j]); */
4232: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4233: /* } */
4234: /* printf("\n"); */
4235: /* fprintf(ficlog,"\n"); */
4236: /* } */
4237:
4238: /* Verifying the inverse matrix */
4239: #ifdef DEBUGHESS
4240: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4241:
1.203 brouard 4242: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4243: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4244:
4245: for (j=1;j<=npar;j++) {
4246: for (i=1;i<=npar;i++){
1.203 brouard 4247: printf("%.2f ",y[i][j]);
4248: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4249: }
4250: printf("\n");
4251: fprintf(ficlog,"\n");
4252: }
1.203 brouard 4253: #endif
1.126 brouard 4254:
4255: free_matrix(a,1,npar,1,npar);
4256: free_matrix(y,1,npar,1,npar);
4257: free_vector(x,1,npar);
4258: free_ivector(indx,1,npar);
1.203 brouard 4259: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4260:
4261:
4262: }
4263:
4264: /*************** hessian matrix ****************/
4265: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4266: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4267: int i;
4268: int l=1, lmax=20;
1.203 brouard 4269: double k1,k2, res, fx;
1.132 brouard 4270: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4271: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4272: int k=0,kmax=10;
4273: double l1;
4274:
4275: fx=func(x);
4276: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4277: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4278: l1=pow(10,l);
4279: delts=delt;
4280: for(k=1 ; k <kmax; k=k+1){
4281: delt = delta*(l1*k);
4282: p2[theta]=x[theta] +delt;
1.145 brouard 4283: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4284: p2[theta]=x[theta]-delt;
4285: k2=func(p2)-fx;
4286: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4287: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4288:
1.203 brouard 4289: #ifdef DEBUGHESSII
1.126 brouard 4290: 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);
4291: 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);
4292: #endif
4293: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4294: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4295: k=kmax;
4296: }
4297: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4298: k=kmax; l=lmax*10;
1.126 brouard 4299: }
4300: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4301: delts=delt;
4302: }
1.203 brouard 4303: } /* End loop k */
1.126 brouard 4304: }
4305: delti[theta]=delts;
4306: return res;
4307:
4308: }
4309:
1.203 brouard 4310: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4311: {
4312: int i;
1.164 brouard 4313: int l=1, lmax=20;
1.126 brouard 4314: double k1,k2,k3,k4,res,fx;
1.132 brouard 4315: double p2[MAXPARM+1];
1.203 brouard 4316: int k, kmax=1;
4317: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4318:
4319: int firstime=0;
1.203 brouard 4320:
1.126 brouard 4321: fx=func(x);
1.203 brouard 4322: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4323: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4324: p2[thetai]=x[thetai]+delti[thetai]*k;
4325: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4326: k1=func(p2)-fx;
4327:
1.203 brouard 4328: p2[thetai]=x[thetai]+delti[thetai]*k;
4329: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4330: k2=func(p2)-fx;
4331:
1.203 brouard 4332: p2[thetai]=x[thetai]-delti[thetai]*k;
4333: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4334: k3=func(p2)-fx;
4335:
1.203 brouard 4336: p2[thetai]=x[thetai]-delti[thetai]*k;
4337: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4338: k4=func(p2)-fx;
1.203 brouard 4339: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4340: if(k1*k2*k3*k4 <0.){
1.208 brouard 4341: firstime=1;
1.203 brouard 4342: kmax=kmax+10;
1.208 brouard 4343: }
4344: if(kmax >=10 || firstime ==1){
1.246 brouard 4345: 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);
4346: 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 4347: 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);
4348: 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);
4349: }
4350: #ifdef DEBUGHESSIJ
4351: v1=hess[thetai][thetai];
4352: v2=hess[thetaj][thetaj];
4353: cv12=res;
4354: /* Computing eigen value of Hessian matrix */
4355: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4356: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4357: if ((lc2 <0) || (lc1 <0) ){
4358: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4359: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4360: 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);
4361: 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);
4362: }
1.126 brouard 4363: #endif
4364: }
4365: return res;
4366: }
4367:
1.203 brouard 4368: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4369: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4370: /* { */
4371: /* int i; */
4372: /* int l=1, lmax=20; */
4373: /* double k1,k2,k3,k4,res,fx; */
4374: /* double p2[MAXPARM+1]; */
4375: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4376: /* int k=0,kmax=10; */
4377: /* double l1; */
4378:
4379: /* fx=func(x); */
4380: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4381: /* l1=pow(10,l); */
4382: /* delts=delt; */
4383: /* for(k=1 ; k <kmax; k=k+1){ */
4384: /* delt = delti*(l1*k); */
4385: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4386: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4387: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4388: /* k1=func(p2)-fx; */
4389:
4390: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4391: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4392: /* k2=func(p2)-fx; */
4393:
4394: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4395: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4396: /* k3=func(p2)-fx; */
4397:
4398: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4399: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4400: /* k4=func(p2)-fx; */
4401: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4402: /* #ifdef DEBUGHESSIJ */
4403: /* 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); */
4404: /* 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); */
4405: /* #endif */
4406: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4407: /* k=kmax; */
4408: /* } */
4409: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4410: /* k=kmax; l=lmax*10; */
4411: /* } */
4412: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4413: /* delts=delt; */
4414: /* } */
4415: /* } /\* End loop k *\/ */
4416: /* } */
4417: /* delti[theta]=delts; */
4418: /* return res; */
4419: /* } */
4420:
4421:
1.126 brouard 4422: /************** Inverse of matrix **************/
4423: void ludcmp(double **a, int n, int *indx, double *d)
4424: {
4425: int i,imax,j,k;
4426: double big,dum,sum,temp;
4427: double *vv;
4428:
4429: vv=vector(1,n);
4430: *d=1.0;
4431: for (i=1;i<=n;i++) {
4432: big=0.0;
4433: for (j=1;j<=n;j++)
4434: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4435: if (big == 0.0){
4436: printf(" Singular Hessian matrix at row %d:\n",i);
4437: for (j=1;j<=n;j++) {
4438: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4439: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4440: }
4441: fflush(ficlog);
4442: fclose(ficlog);
4443: nrerror("Singular matrix in routine ludcmp");
4444: }
1.126 brouard 4445: vv[i]=1.0/big;
4446: }
4447: for (j=1;j<=n;j++) {
4448: for (i=1;i<j;i++) {
4449: sum=a[i][j];
4450: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4451: a[i][j]=sum;
4452: }
4453: big=0.0;
4454: for (i=j;i<=n;i++) {
4455: sum=a[i][j];
4456: for (k=1;k<j;k++)
4457: sum -= a[i][k]*a[k][j];
4458: a[i][j]=sum;
4459: if ( (dum=vv[i]*fabs(sum)) >= big) {
4460: big=dum;
4461: imax=i;
4462: }
4463: }
4464: if (j != imax) {
4465: for (k=1;k<=n;k++) {
4466: dum=a[imax][k];
4467: a[imax][k]=a[j][k];
4468: a[j][k]=dum;
4469: }
4470: *d = -(*d);
4471: vv[imax]=vv[j];
4472: }
4473: indx[j]=imax;
4474: if (a[j][j] == 0.0) a[j][j]=TINY;
4475: if (j != n) {
4476: dum=1.0/(a[j][j]);
4477: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4478: }
4479: }
4480: free_vector(vv,1,n); /* Doesn't work */
4481: ;
4482: }
4483:
4484: void lubksb(double **a, int n, int *indx, double b[])
4485: {
4486: int i,ii=0,ip,j;
4487: double sum;
4488:
4489: for (i=1;i<=n;i++) {
4490: ip=indx[i];
4491: sum=b[ip];
4492: b[ip]=b[i];
4493: if (ii)
4494: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4495: else if (sum) ii=i;
4496: b[i]=sum;
4497: }
4498: for (i=n;i>=1;i--) {
4499: sum=b[i];
4500: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4501: b[i]=sum/a[i][i];
4502: }
4503: }
4504:
4505: void pstamp(FILE *fichier)
4506: {
1.196 brouard 4507: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4508: }
4509:
1.297 brouard 4510: void date2dmy(double date,double *day, double *month, double *year){
4511: double yp=0., yp1=0., yp2=0.;
4512:
4513: yp1=modf(date,&yp);/* extracts integral of date in yp and
4514: fractional in yp1 */
4515: *year=yp;
4516: yp2=modf((yp1*12),&yp);
4517: *month=yp;
4518: yp1=modf((yp2*30.5),&yp);
4519: *day=yp;
4520: if(*day==0) *day=1;
4521: if(*month==0) *month=1;
4522: }
4523:
1.253 brouard 4524:
4525:
1.126 brouard 4526: /************ Frequencies ********************/
1.251 brouard 4527: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4528: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4529: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4530: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4531:
1.265 brouard 4532: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4533: int iind=0, iage=0;
4534: int mi; /* Effective wave */
4535: int first;
4536: double ***freq; /* Frequencies */
1.268 brouard 4537: 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 */
4538: 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 4539: double *meanq, *stdq, *idq;
1.226 brouard 4540: double **meanqt;
4541: double *pp, **prop, *posprop, *pospropt;
4542: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4543: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4544: double agebegin, ageend;
4545:
4546: pp=vector(1,nlstate);
1.251 brouard 4547: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4548: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4549: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4550: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4551: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4552: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4553: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4554: meanqt=matrix(1,lastpass,1,nqtveff);
4555: strcpy(fileresp,"P_");
4556: strcat(fileresp,fileresu);
4557: /*strcat(fileresphtm,fileresu);*/
4558: if((ficresp=fopen(fileresp,"w"))==NULL) {
4559: printf("Problem with prevalence resultfile: %s\n", fileresp);
4560: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4561: exit(0);
4562: }
1.240 brouard 4563:
1.226 brouard 4564: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4565: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4566: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4567: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4568: fflush(ficlog);
4569: exit(70);
4570: }
4571: else{
4572: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4573: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4574: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4575: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4576: }
1.237 brouard 4577: 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 4578:
1.226 brouard 4579: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4580: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4581: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4582: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4583: fflush(ficlog);
4584: exit(70);
1.240 brouard 4585: } else{
1.226 brouard 4586: 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 4587: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4588: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4589: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4590: }
1.240 brouard 4591: 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);
4592:
1.253 brouard 4593: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4594: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4595: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4596: j1=0;
1.126 brouard 4597:
1.227 brouard 4598: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4599: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4600: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4601:
4602:
1.226 brouard 4603: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4604: reference=low_education V1=0,V2=0
4605: med_educ V1=1 V2=0,
4606: high_educ V1=0 V2=1
4607: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4608: */
1.249 brouard 4609: dateintsum=0;
4610: k2cpt=0;
4611:
1.253 brouard 4612: if(cptcoveff == 0 )
1.265 brouard 4613: nl=1; /* Constant and age model only */
1.253 brouard 4614: else
4615: nl=2;
1.265 brouard 4616:
4617: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4618: /* Loop on nj=1 or 2 if dummy covariates j!=0
4619: * Loop on j1(1 to 2**cptcoveff) covariate combination
4620: * freq[s1][s2][iage] =0.
4621: * Loop on iind
4622: * ++freq[s1][s2][iage] weighted
4623: * end iind
4624: * if covariate and j!0
4625: * headers Variable on one line
4626: * endif cov j!=0
4627: * header of frequency table by age
4628: * Loop on age
4629: * pp[s1]+=freq[s1][s2][iage] weighted
4630: * pos+=freq[s1][s2][iage] weighted
4631: * Loop on s1 initial state
4632: * fprintf(ficresp
4633: * end s1
4634: * end age
4635: * if j!=0 computes starting values
4636: * end compute starting values
4637: * end j1
4638: * end nl
4639: */
1.253 brouard 4640: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4641: if(nj==1)
4642: j=0; /* First pass for the constant */
1.265 brouard 4643: else{
1.253 brouard 4644: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4645: }
1.251 brouard 4646: first=1;
1.265 brouard 4647: 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 4648: posproptt=0.;
4649: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4650: scanf("%d", i);*/
4651: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4652: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4653: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4654: freq[i][s2][m]=0;
1.251 brouard 4655:
4656: for (i=1; i<=nlstate; i++) {
1.240 brouard 4657: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4658: prop[i][m]=0;
4659: posprop[i]=0;
4660: pospropt[i]=0;
4661: }
1.283 brouard 4662: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4663: idq[z1]=0.;
4664: meanq[z1]=0.;
4665: stdq[z1]=0.;
1.283 brouard 4666: }
4667: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4668: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4669: /* meanqt[m][z1]=0.; */
4670: /* } */
4671: /* } */
1.251 brouard 4672: /* dateintsum=0; */
4673: /* k2cpt=0; */
4674:
1.265 brouard 4675: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4676: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4677: bool=1;
4678: if(j !=0){
4679: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4680: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4681: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4682: /* if(Tvaraff[z1] ==-20){ */
4683: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4684: /* }else if(Tvaraff[z1] ==-10){ */
4685: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4686: /* }else */
4687: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4688: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4689: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4690: /* 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",
4691: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4692: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4693: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4694: } /* Onlyf fixed */
4695: } /* end z1 */
4696: } /* cptcovn > 0 */
4697: } /* end any */
4698: }/* end j==0 */
1.265 brouard 4699: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4700: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4701: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4702: m=mw[mi][iind];
4703: if(j!=0){
4704: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4705: for (z1=1; z1<=cptcoveff; z1++) {
4706: if( Fixed[Tmodelind[z1]]==1){
4707: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4708: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4709: value is -1, we don't select. It differs from the
4710: constant and age model which counts them. */
4711: bool=0; /* not selected */
4712: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4713: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4714: bool=0;
4715: }
4716: }
4717: }
4718: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4719: } /* end j==0 */
4720: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4721: if(bool==1){ /*Selected */
1.251 brouard 4722: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4723: and mw[mi+1][iind]. dh depends on stepm. */
4724: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4725: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4726: if(m >=firstpass && m <=lastpass){
4727: k2=anint[m][iind]+(mint[m][iind]/12.);
4728: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4729: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4730: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4731: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4732: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4733: if (m<lastpass) {
4734: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4735: /* 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]); */
4736: if(s[m][iind]==-1)
4737: 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.));
4738: 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 4739: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4740: if(!isnan(covar[ncovcol+z1][iind])){
4741: idq[z1]=idq[z1]+weight[iind];
4742: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4743: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4744: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4745: }
1.284 brouard 4746: }
1.251 brouard 4747: /* if((int)agev[m][iind] == 55) */
4748: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4749: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4750: 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 4751: }
1.251 brouard 4752: } /* end if between passes */
4753: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4754: dateintsum=dateintsum+k2; /* on all covariates ?*/
4755: k2cpt++;
4756: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4757: }
1.251 brouard 4758: }else{
4759: bool=1;
4760: }/* end bool 2 */
4761: } /* end m */
1.284 brouard 4762: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4763: /* idq[z1]=idq[z1]+weight[iind]; */
4764: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4765: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4766: /* } */
1.251 brouard 4767: } /* end bool */
4768: } /* end iind = 1 to imx */
4769: /* prop[s][age] is feeded for any initial and valid live state as well as
4770: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4771:
4772:
4773: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4774: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4775: pstamp(ficresp);
1.251 brouard 4776: if (cptcoveff>0 && j!=0){
1.265 brouard 4777: pstamp(ficresp);
1.251 brouard 4778: printf( "\n#********** Variable ");
4779: fprintf(ficresp, "\n#********** Variable ");
4780: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4781: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4782: fprintf(ficlog, "\n#********** Variable ");
4783: for (z1=1; z1<=cptcoveff; z1++){
4784: if(!FixedV[Tvaraff[z1]]){
4785: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4786: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4787: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4788: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4789: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4790: }else{
1.251 brouard 4791: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4792: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4793: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4794: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4795: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4796: }
4797: }
4798: printf( "**********\n#");
4799: fprintf(ficresp, "**********\n#");
4800: fprintf(ficresphtm, "**********</h3>\n");
4801: fprintf(ficresphtmfr, "**********</h3>\n");
4802: fprintf(ficlog, "**********\n");
4803: }
1.284 brouard 4804: /*
4805: Printing means of quantitative variables if any
4806: */
4807: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 4808: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 4809: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 4810: if(weightopt==1){
4811: printf(" Weighted mean and standard deviation of");
4812: fprintf(ficlog," Weighted mean and standard deviation of");
4813: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4814: }
1.311 brouard 4815: /* mu = \frac{w x}{\sum w}
4816: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
4817: */
4818: 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]));
4819: 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]));
4820: 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 4821: }
4822: /* for (z1=1; z1<= nqtveff; z1++) { */
4823: /* for(m=1;m<=lastpass;m++){ */
4824: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4825: /* } */
4826: /* } */
1.283 brouard 4827:
1.251 brouard 4828: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4829: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4830: fprintf(ficresp, " Age");
4831: 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 4832: for(i=1; i<=nlstate;i++) {
1.265 brouard 4833: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4834: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4835: }
1.265 brouard 4836: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4837: fprintf(ficresphtm, "\n");
4838:
4839: /* Header of frequency table by age */
4840: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4841: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4842: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4843: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4844: if(s2!=0 && m!=0)
4845: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4846: }
1.226 brouard 4847: }
1.251 brouard 4848: fprintf(ficresphtmfr, "\n");
4849:
4850: /* For each age */
4851: for(iage=iagemin; iage <= iagemax+3; iage++){
4852: fprintf(ficresphtm,"<tr>");
4853: if(iage==iagemax+1){
4854: fprintf(ficlog,"1");
4855: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4856: }else if(iage==iagemax+2){
4857: fprintf(ficlog,"0");
4858: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4859: }else if(iage==iagemax+3){
4860: fprintf(ficlog,"Total");
4861: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4862: }else{
1.240 brouard 4863: if(first==1){
1.251 brouard 4864: first=0;
4865: printf("See log file for details...\n");
4866: }
4867: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4868: fprintf(ficlog,"Age %d", iage);
4869: }
1.265 brouard 4870: for(s1=1; s1 <=nlstate ; s1++){
4871: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4872: pp[s1] += freq[s1][m][iage];
1.251 brouard 4873: }
1.265 brouard 4874: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4875: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4876: pos += freq[s1][m][iage];
4877: if(pp[s1]>=1.e-10){
1.251 brouard 4878: if(first==1){
1.265 brouard 4879: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4880: }
1.265 brouard 4881: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4882: }else{
4883: if(first==1)
1.265 brouard 4884: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4885: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4886: }
4887: }
4888:
1.265 brouard 4889: for(s1=1; s1 <=nlstate ; s1++){
4890: /* posprop[s1]=0; */
4891: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4892: pp[s1] += freq[s1][m][iage];
4893: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4894:
4895: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4896: pos += pp[s1]; /* pos is the total number of transitions until this age */
4897: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4898: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4899: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4900: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4901: }
4902:
4903: /* Writing ficresp */
4904: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4905: if( iage <= iagemax){
4906: fprintf(ficresp," %d",iage);
4907: }
4908: }else if( nj==2){
4909: if( iage <= iagemax){
4910: fprintf(ficresp," %d",iage);
4911: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4912: }
1.240 brouard 4913: }
1.265 brouard 4914: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4915: if(pos>=1.e-5){
1.251 brouard 4916: if(first==1)
1.265 brouard 4917: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4918: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4919: }else{
4920: if(first==1)
1.265 brouard 4921: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4922: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4923: }
4924: if( iage <= iagemax){
4925: if(pos>=1.e-5){
1.265 brouard 4926: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4927: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4928: }else if( nj==2){
4929: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4930: }
4931: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4932: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4933: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4934: } else{
4935: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4936: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4937: }
1.240 brouard 4938: }
1.265 brouard 4939: pospropt[s1] +=posprop[s1];
4940: } /* end loop s1 */
1.251 brouard 4941: /* pospropt=0.; */
1.265 brouard 4942: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4943: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4944: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4945: if(first==1){
1.265 brouard 4946: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4947: }
1.265 brouard 4948: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4949: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4950: }
1.265 brouard 4951: if(s1!=0 && m!=0)
4952: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4953: }
1.265 brouard 4954: } /* end loop s1 */
1.251 brouard 4955: posproptt=0.;
1.265 brouard 4956: for(s1=1; s1 <=nlstate; s1++){
4957: posproptt += pospropt[s1];
1.251 brouard 4958: }
4959: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4960: fprintf(ficresphtm,"</tr>\n");
4961: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4962: if(iage <= iagemax)
4963: fprintf(ficresp,"\n");
1.240 brouard 4964: }
1.251 brouard 4965: if(first==1)
4966: printf("Others in log...\n");
4967: fprintf(ficlog,"\n");
4968: } /* end loop age iage */
1.265 brouard 4969:
1.251 brouard 4970: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4971: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4972: if(posproptt < 1.e-5){
1.265 brouard 4973: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4974: }else{
1.265 brouard 4975: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4976: }
1.226 brouard 4977: }
1.251 brouard 4978: fprintf(ficresphtm,"</tr>\n");
4979: fprintf(ficresphtm,"</table>\n");
4980: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4981: if(posproptt < 1.e-5){
1.251 brouard 4982: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4983: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4984: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4985: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4986: invalidvarcomb[j1]=1;
1.226 brouard 4987: }else{
1.251 brouard 4988: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4989: invalidvarcomb[j1]=0;
1.226 brouard 4990: }
1.251 brouard 4991: fprintf(ficresphtmfr,"</table>\n");
4992: fprintf(ficlog,"\n");
4993: if(j!=0){
4994: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4995: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4996: for(k=1; k <=(nlstate+ndeath); k++){
4997: if (k != i) {
1.265 brouard 4998: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4999: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5000: if(j1==1){ /* All dummy covariates to zero */
5001: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5002: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5003: printf("%d%d ",i,k);
5004: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5005: 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]));
5006: 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]));
5007: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5008: }
1.253 brouard 5009: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5010: for(iage=iagemin; iage <= iagemax+3; iage++){
5011: x[iage]= (double)iage;
5012: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5013: /* 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 5014: }
1.268 brouard 5015: /* Some are not finite, but linreg will ignore these ages */
5016: no=0;
1.253 brouard 5017: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5018: pstart[s1]=b;
5019: pstart[s1-1]=a;
1.252 brouard 5020: }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 */
5021: 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]);
5022: 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 5023: 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 5024: printf("%d%d ",i,k);
5025: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5026: 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 5027: }else{ /* Other cases, like quantitative fixed or varying covariates */
5028: ;
5029: }
5030: /* printf("%12.7f )", param[i][jj][k]); */
5031: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5032: s1++;
1.251 brouard 5033: } /* end jj */
5034: } /* end k!= i */
5035: } /* end k */
1.265 brouard 5036: } /* end i, s1 */
1.251 brouard 5037: } /* end j !=0 */
5038: } /* end selected combination of covariate j1 */
5039: if(j==0){ /* We can estimate starting values from the occurences in each case */
5040: printf("#Freqsummary: Starting values for the constants:\n");
5041: fprintf(ficlog,"\n");
1.265 brouard 5042: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5043: for(k=1; k <=(nlstate+ndeath); k++){
5044: if (k != i) {
5045: printf("%d%d ",i,k);
5046: fprintf(ficlog,"%d%d ",i,k);
5047: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5048: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5049: if(jj==1){ /* Age has to be done */
1.265 brouard 5050: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5051: 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]));
5052: 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 5053: }
5054: /* printf("%12.7f )", param[i][jj][k]); */
5055: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5056: s1++;
1.250 brouard 5057: }
1.251 brouard 5058: printf("\n");
5059: fprintf(ficlog,"\n");
1.250 brouard 5060: }
5061: }
1.284 brouard 5062: } /* end of state i */
1.251 brouard 5063: printf("#Freqsummary\n");
5064: fprintf(ficlog,"\n");
1.265 brouard 5065: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5066: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5067: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5068: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5069: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5070: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5071: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5072: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5073: /* } */
5074: }
1.265 brouard 5075: } /* end loop s1 */
1.251 brouard 5076:
5077: printf("\n");
5078: fprintf(ficlog,"\n");
5079: } /* end j=0 */
1.249 brouard 5080: } /* end j */
1.252 brouard 5081:
1.253 brouard 5082: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5083: for(i=1, jk=1; i <=nlstate; i++){
5084: for(j=1; j <=nlstate+ndeath; j++){
5085: if(j!=i){
5086: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5087: printf("%1d%1d",i,j);
5088: fprintf(ficparo,"%1d%1d",i,j);
5089: for(k=1; k<=ncovmodel;k++){
5090: /* printf(" %lf",param[i][j][k]); */
5091: /* fprintf(ficparo," %lf",param[i][j][k]); */
5092: p[jk]=pstart[jk];
5093: printf(" %f ",pstart[jk]);
5094: fprintf(ficparo," %f ",pstart[jk]);
5095: jk++;
5096: }
5097: printf("\n");
5098: fprintf(ficparo,"\n");
5099: }
5100: }
5101: }
5102: } /* end mle=-2 */
1.226 brouard 5103: dateintmean=dateintsum/k2cpt;
1.296 brouard 5104: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5105:
1.226 brouard 5106: fclose(ficresp);
5107: fclose(ficresphtm);
5108: fclose(ficresphtmfr);
1.283 brouard 5109: free_vector(idq,1,nqfveff);
1.226 brouard 5110: free_vector(meanq,1,nqfveff);
1.284 brouard 5111: free_vector(stdq,1,nqfveff);
1.226 brouard 5112: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5113: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5114: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5115: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5116: free_vector(pospropt,1,nlstate);
5117: free_vector(posprop,1,nlstate);
1.251 brouard 5118: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5119: free_vector(pp,1,nlstate);
5120: /* End of freqsummary */
5121: }
1.126 brouard 5122:
1.268 brouard 5123: /* Simple linear regression */
5124: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5125:
5126: /* y=a+bx regression */
5127: double sumx = 0.0; /* sum of x */
5128: double sumx2 = 0.0; /* sum of x**2 */
5129: double sumxy = 0.0; /* sum of x * y */
5130: double sumy = 0.0; /* sum of y */
5131: double sumy2 = 0.0; /* sum of y**2 */
5132: double sume2 = 0.0; /* sum of square or residuals */
5133: double yhat;
5134:
5135: double denom=0;
5136: int i;
5137: int ne=*no;
5138:
5139: for ( i=ifi, ne=0;i<=ila;i++) {
5140: if(!isfinite(x[i]) || !isfinite(y[i])){
5141: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5142: continue;
5143: }
5144: ne=ne+1;
5145: sumx += x[i];
5146: sumx2 += x[i]*x[i];
5147: sumxy += x[i] * y[i];
5148: sumy += y[i];
5149: sumy2 += y[i]*y[i];
5150: denom = (ne * sumx2 - sumx*sumx);
5151: /* 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); */
5152: }
5153:
5154: denom = (ne * sumx2 - sumx*sumx);
5155: if (denom == 0) {
5156: // vertical, slope m is infinity
5157: *b = INFINITY;
5158: *a = 0;
5159: if (r) *r = 0;
5160: return 1;
5161: }
5162:
5163: *b = (ne * sumxy - sumx * sumy) / denom;
5164: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5165: if (r!=NULL) {
5166: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5167: sqrt((sumx2 - sumx*sumx/ne) *
5168: (sumy2 - sumy*sumy/ne));
5169: }
5170: *no=ne;
5171: for ( i=ifi, ne=0;i<=ila;i++) {
5172: if(!isfinite(x[i]) || !isfinite(y[i])){
5173: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5174: continue;
5175: }
5176: ne=ne+1;
5177: yhat = y[i] - *a -*b* x[i];
5178: sume2 += yhat * yhat ;
5179:
5180: denom = (ne * sumx2 - sumx*sumx);
5181: /* 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); */
5182: }
5183: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5184: *sa= *sb * sqrt(sumx2/ne);
5185:
5186: return 0;
5187: }
5188:
1.126 brouard 5189: /************ Prevalence ********************/
1.227 brouard 5190: 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)
5191: {
5192: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5193: in each health status at the date of interview (if between dateprev1 and dateprev2).
5194: We still use firstpass and lastpass as another selection.
5195: */
1.126 brouard 5196:
1.227 brouard 5197: int i, m, jk, j1, bool, z1,j, iv;
5198: int mi; /* Effective wave */
5199: int iage;
5200: double agebegin, ageend;
5201:
5202: double **prop;
5203: double posprop;
5204: double y2; /* in fractional years */
5205: int iagemin, iagemax;
5206: int first; /** to stop verbosity which is redirected to log file */
5207:
5208: iagemin= (int) agemin;
5209: iagemax= (int) agemax;
5210: /*pp=vector(1,nlstate);*/
1.251 brouard 5211: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5212: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5213: j1=0;
1.222 brouard 5214:
1.227 brouard 5215: /*j=cptcoveff;*/
5216: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5217:
1.288 brouard 5218: first=0;
1.227 brouard 5219: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5220: for (i=1; i<=nlstate; i++)
1.251 brouard 5221: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5222: prop[i][iage]=0.0;
5223: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5224: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5225: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5226:
5227: for (i=1; i<=imx; i++) { /* Each individual */
5228: bool=1;
5229: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5230: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5231: m=mw[mi][i];
5232: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5233: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5234: for (z1=1; z1<=cptcoveff; z1++){
5235: if( Fixed[Tmodelind[z1]]==1){
5236: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5237: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5238: bool=0;
5239: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5240: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5241: bool=0;
5242: }
5243: }
5244: if(bool==1){ /* Otherwise we skip that wave/person */
5245: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5246: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5247: if(m >=firstpass && m <=lastpass){
5248: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5249: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5250: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5251: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5252: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5253: 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);
5254: exit(1);
5255: }
5256: if (s[m][i]>0 && s[m][i]<=nlstate) {
5257: /*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]]);*/
5258: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5259: prop[s[m][i]][iagemax+3] += weight[i];
5260: } /* end valid statuses */
5261: } /* end selection of dates */
5262: } /* end selection of waves */
5263: } /* end bool */
5264: } /* end wave */
5265: } /* end individual */
5266: for(i=iagemin; i <= iagemax+3; i++){
5267: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5268: posprop += prop[jk][i];
5269: }
5270:
5271: for(jk=1; jk <=nlstate ; jk++){
5272: if( i <= iagemax){
5273: if(posprop>=1.e-5){
5274: probs[i][jk][j1]= prop[jk][i]/posprop;
5275: } else{
1.288 brouard 5276: if(!first){
5277: first=1;
1.266 brouard 5278: 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]);
5279: }else{
1.288 brouard 5280: 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 5281: }
5282: }
5283: }
5284: }/* end jk */
5285: }/* end i */
1.222 brouard 5286: /*} *//* end i1 */
1.227 brouard 5287: } /* end j1 */
1.222 brouard 5288:
1.227 brouard 5289: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5290: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5291: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5292: } /* End of prevalence */
1.126 brouard 5293:
5294: /************* Waves Concatenation ***************/
5295:
5296: 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)
5297: {
1.298 brouard 5298: /* 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 5299: Death is a valid wave (if date is known).
5300: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5301: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5302: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5303: */
1.126 brouard 5304:
1.224 brouard 5305: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5306: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5307: double sum=0., jmean=0.;*/
1.224 brouard 5308: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5309: int j, k=0,jk, ju, jl;
5310: double sum=0.;
5311: first=0;
1.214 brouard 5312: firstwo=0;
1.217 brouard 5313: firsthree=0;
1.218 brouard 5314: firstfour=0;
1.164 brouard 5315: jmin=100000;
1.126 brouard 5316: jmax=-1;
5317: jmean=0.;
1.224 brouard 5318:
5319: /* Treating live states */
1.214 brouard 5320: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5321: mi=0; /* First valid wave */
1.227 brouard 5322: mli=0; /* Last valid wave */
1.309 brouard 5323: m=firstpass; /* Loop on waves */
5324: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5325: 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 */
5326: mli=m-1;/* mw[++mi][i]=m-1; */
5327: }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 5328: 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 5329: mli=m;
1.224 brouard 5330: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5331: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5332: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5333: }
1.309 brouard 5334: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5335: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5336: break;
1.224 brouard 5337: #else
1.317 ! brouard 5338: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){ /* no death date and known date of interview, case -2 (vital status unknown is warned later */
1.227 brouard 5339: if(firsthree == 0){
1.302 brouard 5340: 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 5341: firsthree=1;
1.317 ! brouard 5342: }else if(firsthree >=1 && firsthree < 10){
! 5343: 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);
! 5344: firsthree++;
! 5345: }else if(firsthree == 10){
! 5346: printf("Information, too many Information flags: no more reported to log either\n");
! 5347: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
! 5348: firsthree++;
! 5349: }else{
! 5350: firsthree++;
1.227 brouard 5351: }
1.309 brouard 5352: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5353: mli=m;
5354: }
5355: if(s[m][i]==-2){ /* Vital status is really unknown */
5356: nbwarn++;
1.309 brouard 5357: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5358: 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);
5359: 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);
5360: }
5361: break;
5362: }
5363: break;
1.224 brouard 5364: #endif
1.227 brouard 5365: }/* End m >= lastpass */
1.126 brouard 5366: }/* end while */
1.224 brouard 5367:
1.227 brouard 5368: /* 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 5369: /* After last pass */
1.224 brouard 5370: /* Treating death states */
1.214 brouard 5371: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5372: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5373: /* } */
1.126 brouard 5374: mi++; /* Death is another wave */
5375: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5376: /* Only death is a correct wave */
1.126 brouard 5377: mw[mi][i]=m;
1.257 brouard 5378: } /* else not in a death state */
1.224 brouard 5379: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5380: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5381: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5382: 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 5383: nbwarn++;
5384: if(firstfiv==0){
1.309 brouard 5385: 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 5386: firstfiv=1;
5387: }else{
1.309 brouard 5388: 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 5389: }
1.309 brouard 5390: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5391: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5392: nberr++;
5393: if(firstwo==0){
1.309 brouard 5394: 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 5395: firstwo=1;
5396: }
1.309 brouard 5397: 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 5398: }
1.257 brouard 5399: }else{ /* if date of interview is unknown */
1.227 brouard 5400: /* death is known but not confirmed by death status at any wave */
5401: if(firstfour==0){
1.309 brouard 5402: 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 5403: firstfour=1;
5404: }
1.309 brouard 5405: 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 5406: }
1.224 brouard 5407: } /* end if date of death is known */
5408: #endif
1.309 brouard 5409: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5410: /* wav[i]=mw[mi][i]; */
1.126 brouard 5411: if(mi==0){
5412: nbwarn++;
5413: if(first==0){
1.227 brouard 5414: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5415: first=1;
1.126 brouard 5416: }
5417: if(first==1){
1.227 brouard 5418: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5419: }
5420: } /* end mi==0 */
5421: } /* End individuals */
1.214 brouard 5422: /* wav and mw are no more changed */
1.223 brouard 5423:
1.317 ! brouard 5424: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
! 5425: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
! 5426:
! 5427:
1.126 brouard 5428: for(i=1; i<=imx; i++){
5429: for(mi=1; mi<wav[i];mi++){
5430: if (stepm <=0)
1.227 brouard 5431: dh[mi][i]=1;
1.126 brouard 5432: else{
1.260 brouard 5433: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5434: if (agedc[i] < 2*AGESUP) {
5435: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5436: if(j==0) j=1; /* Survives at least one month after exam */
5437: else if(j<0){
5438: nberr++;
5439: 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]);
5440: j=1; /* Temporary Dangerous patch */
5441: 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);
5442: 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]);
5443: 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);
5444: }
5445: k=k+1;
5446: if (j >= jmax){
5447: jmax=j;
5448: ijmax=i;
5449: }
5450: if (j <= jmin){
5451: jmin=j;
5452: ijmin=i;
5453: }
5454: sum=sum+j;
5455: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5456: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5457: }
5458: }
5459: else{
5460: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5461: /* 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 5462:
1.227 brouard 5463: k=k+1;
5464: if (j >= jmax) {
5465: jmax=j;
5466: ijmax=i;
5467: }
5468: else if (j <= jmin){
5469: jmin=j;
5470: ijmin=i;
5471: }
5472: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5473: /*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]);*/
5474: if(j<0){
5475: nberr++;
5476: 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]);
5477: 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]);
5478: }
5479: sum=sum+j;
5480: }
5481: jk= j/stepm;
5482: jl= j -jk*stepm;
5483: ju= j -(jk+1)*stepm;
5484: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5485: if(jl==0){
5486: dh[mi][i]=jk;
5487: bh[mi][i]=0;
5488: }else{ /* We want a negative bias in order to only have interpolation ie
5489: * to avoid the price of an extra matrix product in likelihood */
5490: dh[mi][i]=jk+1;
5491: bh[mi][i]=ju;
5492: }
5493: }else{
5494: if(jl <= -ju){
5495: dh[mi][i]=jk;
5496: bh[mi][i]=jl; /* bias is positive if real duration
5497: * is higher than the multiple of stepm and negative otherwise.
5498: */
5499: }
5500: else{
5501: dh[mi][i]=jk+1;
5502: bh[mi][i]=ju;
5503: }
5504: if(dh[mi][i]==0){
5505: dh[mi][i]=1; /* At least one step */
5506: bh[mi][i]=ju; /* At least one step */
5507: /* 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);*/
5508: }
5509: } /* end if mle */
1.126 brouard 5510: }
5511: } /* end wave */
5512: }
5513: jmean=sum/k;
5514: 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 5515: 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 5516: }
1.126 brouard 5517:
5518: /*********** Tricode ****************************/
1.220 brouard 5519: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5520: {
5521: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5522: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5523: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5524: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5525: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5526: */
1.130 brouard 5527:
1.242 brouard 5528: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5529: int modmaxcovj=0; /* Modality max of covariates j */
5530: int cptcode=0; /* Modality max of covariates j */
5531: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5532:
5533:
1.242 brouard 5534: /* cptcoveff=0; */
5535: /* *cptcov=0; */
1.126 brouard 5536:
1.242 brouard 5537: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5538: for (k=1; k <= maxncov; k++)
5539: for(j=1; j<=2; j++)
5540: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5541:
1.242 brouard 5542: /* Loop on covariates without age and products and no quantitative variable */
5543: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5544: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5545: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5546: switch(Fixed[k]) {
5547: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5548: modmaxcovj=0;
5549: modmincovj=0;
1.242 brouard 5550: 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*/
5551: ij=(int)(covar[Tvar[k]][i]);
5552: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5553: * If product of Vn*Vm, still boolean *:
5554: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5555: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5556: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5557: modality of the nth covariate of individual i. */
5558: if (ij > modmaxcovj)
5559: modmaxcovj=ij;
5560: else if (ij < modmincovj)
5561: modmincovj=ij;
1.287 brouard 5562: if (ij <0 || ij >1 ){
1.311 brouard 5563: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5564: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5565: fflush(ficlog);
5566: exit(1);
1.287 brouard 5567: }
5568: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5569: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5570: exit(1);
5571: }else
5572: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5573: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5574: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5575: /* getting the maximum value of the modality of the covariate
5576: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5577: female ies 1, then modmaxcovj=1.
5578: */
5579: } /* end for loop on individuals i */
5580: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5581: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5582: cptcode=modmaxcovj;
5583: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5584: /*for (i=0; i<=cptcode; i++) {*/
5585: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5586: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5587: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5588: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5589: if( j != -1){
5590: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5591: covariate for which somebody answered excluding
5592: undefined. Usually 2: 0 and 1. */
5593: }
5594: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5595: covariate for which somebody answered including
5596: undefined. Usually 3: -1, 0 and 1. */
5597: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5598: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5599: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5600:
1.242 brouard 5601: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5602: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5603: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5604: /* modmincovj=3; modmaxcovj = 7; */
5605: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5606: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5607: /* defining two dummy variables: variables V1_1 and V1_2.*/
5608: /* nbcode[Tvar[j]][ij]=k; */
5609: /* nbcode[Tvar[j]][1]=0; */
5610: /* nbcode[Tvar[j]][2]=1; */
5611: /* nbcode[Tvar[j]][3]=2; */
5612: /* To be continued (not working yet). */
5613: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5614:
5615: /* 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*/
5616: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5617: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5618: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5619: /*, could be restored in the future */
5620: 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 5621: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5622: break;
5623: }
5624: ij++;
1.287 brouard 5625: 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 5626: cptcode = ij; /* New max modality for covar j */
5627: } /* end of loop on modality i=-1 to 1 or more */
5628: break;
5629: case 1: /* Testing on varying covariate, could be simple and
5630: * should look at waves or product of fixed *
5631: * varying. No time to test -1, assuming 0 and 1 only */
5632: ij=0;
5633: for(i=0; i<=1;i++){
5634: nbcode[Tvar[k]][++ij]=i;
5635: }
5636: break;
5637: default:
5638: break;
5639: } /* end switch */
5640: } /* end dummy test */
1.311 brouard 5641: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5642: 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*/
5643: if(isnan(covar[Tvar[k]][i])){
5644: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5645: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5646: fflush(ficlog);
5647: exit(1);
5648: }
5649: }
5650: }
1.287 brouard 5651: } /* 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 5652:
5653: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5654: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5655: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5656: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5657: 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 */
5658: 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 */
5659: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5660: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5661:
5662: ij=0;
5663: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5664: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5665: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5666: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5667: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5668: /* If product not in single variable we don't print results */
5669: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5670: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5671: 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*/
5672: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5673: 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 */
5674: if(Fixed[k]!=0)
5675: anyvaryingduminmodel=1;
5676: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5677: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5678: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5679: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5680: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5681: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5682: }
5683: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5684: /* ij--; */
5685: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5686: *cptcov=ij; /*Number of total real effective covariates: effective
5687: * because they can be excluded from the model and real
5688: * if in the model but excluded because missing values, but how to get k from ij?*/
5689: for(j=ij+1; j<= cptcovt; j++){
5690: Tvaraff[j]=0;
5691: Tmodelind[j]=0;
5692: }
5693: for(j=ntveff+1; j<= cptcovt; j++){
5694: TmodelInvind[j]=0;
5695: }
5696: /* To be sorted */
5697: ;
5698: }
1.126 brouard 5699:
1.145 brouard 5700:
1.126 brouard 5701: /*********** Health Expectancies ****************/
5702:
1.235 brouard 5703: 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 5704:
5705: {
5706: /* Health expectancies, no variances */
1.164 brouard 5707: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5708: int nhstepma, nstepma; /* Decreasing with age */
5709: double age, agelim, hf;
5710: double ***p3mat;
5711: double eip;
5712:
1.238 brouard 5713: /* pstamp(ficreseij); */
1.126 brouard 5714: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5715: fprintf(ficreseij,"# Age");
5716: for(i=1; i<=nlstate;i++){
5717: for(j=1; j<=nlstate;j++){
5718: fprintf(ficreseij," e%1d%1d ",i,j);
5719: }
5720: fprintf(ficreseij," e%1d. ",i);
5721: }
5722: fprintf(ficreseij,"\n");
5723:
5724:
5725: if(estepm < stepm){
5726: printf ("Problem %d lower than %d\n",estepm, stepm);
5727: }
5728: else hstepm=estepm;
5729: /* We compute the life expectancy from trapezoids spaced every estepm months
5730: * This is mainly to measure the difference between two models: for example
5731: * if stepm=24 months pijx are given only every 2 years and by summing them
5732: * we are calculating an estimate of the Life Expectancy assuming a linear
5733: * progression in between and thus overestimating or underestimating according
5734: * to the curvature of the survival function. If, for the same date, we
5735: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5736: * to compare the new estimate of Life expectancy with the same linear
5737: * hypothesis. A more precise result, taking into account a more precise
5738: * curvature will be obtained if estepm is as small as stepm. */
5739:
5740: /* For example we decided to compute the life expectancy with the smallest unit */
5741: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5742: nhstepm is the number of hstepm from age to agelim
5743: nstepm is the number of stepm from age to agelin.
1.270 brouard 5744: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5745: and note for a fixed period like estepm months */
5746: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5747: survival function given by stepm (the optimization length). Unfortunately it
5748: means that if the survival funtion is printed only each two years of age and if
5749: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5750: results. So we changed our mind and took the option of the best precision.
5751: */
5752: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5753:
5754: agelim=AGESUP;
5755: /* If stepm=6 months */
5756: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5757: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5758:
5759: /* nhstepm age range expressed in number of stepm */
5760: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5761: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5762: /* if (stepm >= YEARM) hstepm=1;*/
5763: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5764: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5765:
5766: for (age=bage; age<=fage; age ++){
5767: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5768: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5769: /* if (stepm >= YEARM) hstepm=1;*/
5770: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5771:
5772: /* If stepm=6 months */
5773: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5774: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5775:
1.235 brouard 5776: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5777:
5778: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5779:
5780: printf("%d|",(int)age);fflush(stdout);
5781: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5782:
5783: /* Computing expectancies */
5784: for(i=1; i<=nlstate;i++)
5785: for(j=1; j<=nlstate;j++)
5786: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5787: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5788:
5789: /* 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]);*/
5790:
5791: }
5792:
5793: fprintf(ficreseij,"%3.0f",age );
5794: for(i=1; i<=nlstate;i++){
5795: eip=0;
5796: for(j=1; j<=nlstate;j++){
5797: eip +=eij[i][j][(int)age];
5798: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5799: }
5800: fprintf(ficreseij,"%9.4f", eip );
5801: }
5802: fprintf(ficreseij,"\n");
5803:
5804: }
5805: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5806: printf("\n");
5807: fprintf(ficlog,"\n");
5808:
5809: }
5810:
1.235 brouard 5811: 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 5812:
5813: {
5814: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5815: to initial status i, ei. .
1.126 brouard 5816: */
5817: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5818: int nhstepma, nstepma; /* Decreasing with age */
5819: double age, agelim, hf;
5820: double ***p3matp, ***p3matm, ***varhe;
5821: double **dnewm,**doldm;
5822: double *xp, *xm;
5823: double **gp, **gm;
5824: double ***gradg, ***trgradg;
5825: int theta;
5826:
5827: double eip, vip;
5828:
5829: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5830: xp=vector(1,npar);
5831: xm=vector(1,npar);
5832: dnewm=matrix(1,nlstate*nlstate,1,npar);
5833: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5834:
5835: pstamp(ficresstdeij);
5836: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5837: fprintf(ficresstdeij,"# Age");
5838: for(i=1; i<=nlstate;i++){
5839: for(j=1; j<=nlstate;j++)
5840: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5841: fprintf(ficresstdeij," e%1d. ",i);
5842: }
5843: fprintf(ficresstdeij,"\n");
5844:
5845: pstamp(ficrescveij);
5846: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5847: fprintf(ficrescveij,"# Age");
5848: for(i=1; i<=nlstate;i++)
5849: for(j=1; j<=nlstate;j++){
5850: cptj= (j-1)*nlstate+i;
5851: for(i2=1; i2<=nlstate;i2++)
5852: for(j2=1; j2<=nlstate;j2++){
5853: cptj2= (j2-1)*nlstate+i2;
5854: if(cptj2 <= cptj)
5855: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5856: }
5857: }
5858: fprintf(ficrescveij,"\n");
5859:
5860: if(estepm < stepm){
5861: printf ("Problem %d lower than %d\n",estepm, stepm);
5862: }
5863: else hstepm=estepm;
5864: /* We compute the life expectancy from trapezoids spaced every estepm months
5865: * This is mainly to measure the difference between two models: for example
5866: * if stepm=24 months pijx are given only every 2 years and by summing them
5867: * we are calculating an estimate of the Life Expectancy assuming a linear
5868: * progression in between and thus overestimating or underestimating according
5869: * to the curvature of the survival function. If, for the same date, we
5870: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5871: * to compare the new estimate of Life expectancy with the same linear
5872: * hypothesis. A more precise result, taking into account a more precise
5873: * curvature will be obtained if estepm is as small as stepm. */
5874:
5875: /* For example we decided to compute the life expectancy with the smallest unit */
5876: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5877: nhstepm is the number of hstepm from age to agelim
5878: nstepm is the number of stepm from age to agelin.
5879: Look at hpijx to understand the reason of that which relies in memory size
5880: and note for a fixed period like estepm months */
5881: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5882: survival function given by stepm (the optimization length). Unfortunately it
5883: means that if the survival funtion is printed only each two years of age and if
5884: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5885: results. So we changed our mind and took the option of the best precision.
5886: */
5887: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5888:
5889: /* If stepm=6 months */
5890: /* nhstepm age range expressed in number of stepm */
5891: agelim=AGESUP;
5892: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5893: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5894: /* if (stepm >= YEARM) hstepm=1;*/
5895: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5896:
5897: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5898: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5899: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5900: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5901: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5902: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5903:
5904: for (age=bage; age<=fage; age ++){
5905: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5906: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5907: /* if (stepm >= YEARM) hstepm=1;*/
5908: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5909:
1.126 brouard 5910: /* If stepm=6 months */
5911: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5912: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5913:
5914: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5915:
1.126 brouard 5916: /* Computing Variances of health expectancies */
5917: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5918: decrease memory allocation */
5919: for(theta=1; theta <=npar; theta++){
5920: for(i=1; i<=npar; i++){
1.222 brouard 5921: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5922: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5923: }
1.235 brouard 5924: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5925: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5926:
1.126 brouard 5927: for(j=1; j<= nlstate; j++){
1.222 brouard 5928: for(i=1; i<=nlstate; i++){
5929: for(h=0; h<=nhstepm-1; h++){
5930: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5931: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5932: }
5933: }
1.126 brouard 5934: }
1.218 brouard 5935:
1.126 brouard 5936: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5937: for(h=0; h<=nhstepm-1; h++){
5938: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5939: }
1.126 brouard 5940: }/* End theta */
5941:
5942:
5943: for(h=0; h<=nhstepm-1; h++)
5944: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5945: for(theta=1; theta <=npar; theta++)
5946: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5947:
1.218 brouard 5948:
1.222 brouard 5949: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5950: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5951: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5952:
1.222 brouard 5953: printf("%d|",(int)age);fflush(stdout);
5954: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5955: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5956: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5957: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5958: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5959: for(ij=1;ij<=nlstate*nlstate;ij++)
5960: for(ji=1;ji<=nlstate*nlstate;ji++)
5961: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5962: }
5963: }
1.218 brouard 5964:
1.126 brouard 5965: /* Computing expectancies */
1.235 brouard 5966: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5967: for(i=1; i<=nlstate;i++)
5968: for(j=1; j<=nlstate;j++)
1.222 brouard 5969: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5970: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5971:
1.222 brouard 5972: /* 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 5973:
1.222 brouard 5974: }
1.269 brouard 5975:
5976: /* Standard deviation of expectancies ij */
1.126 brouard 5977: fprintf(ficresstdeij,"%3.0f",age );
5978: for(i=1; i<=nlstate;i++){
5979: eip=0.;
5980: vip=0.;
5981: for(j=1; j<=nlstate;j++){
1.222 brouard 5982: eip += eij[i][j][(int)age];
5983: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5984: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5985: 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 5986: }
5987: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5988: }
5989: fprintf(ficresstdeij,"\n");
1.218 brouard 5990:
1.269 brouard 5991: /* Variance of expectancies ij */
1.126 brouard 5992: fprintf(ficrescveij,"%3.0f",age );
5993: for(i=1; i<=nlstate;i++)
5994: for(j=1; j<=nlstate;j++){
1.222 brouard 5995: cptj= (j-1)*nlstate+i;
5996: for(i2=1; i2<=nlstate;i2++)
5997: for(j2=1; j2<=nlstate;j2++){
5998: cptj2= (j2-1)*nlstate+i2;
5999: if(cptj2 <= cptj)
6000: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6001: }
1.126 brouard 6002: }
6003: fprintf(ficrescveij,"\n");
1.218 brouard 6004:
1.126 brouard 6005: }
6006: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6007: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6008: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6009: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6010: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6011: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6012: printf("\n");
6013: fprintf(ficlog,"\n");
1.218 brouard 6014:
1.126 brouard 6015: free_vector(xm,1,npar);
6016: free_vector(xp,1,npar);
6017: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6018: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6019: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6020: }
1.218 brouard 6021:
1.126 brouard 6022: /************ Variance ******************/
1.235 brouard 6023: 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 6024: {
1.279 brouard 6025: /** Variance of health expectancies
6026: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6027: * double **newm;
6028: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6029: */
1.218 brouard 6030:
6031: /* int movingaverage(); */
6032: double **dnewm,**doldm;
6033: double **dnewmp,**doldmp;
6034: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6035: int first=0;
1.218 brouard 6036: int k;
6037: double *xp;
1.279 brouard 6038: double **gp, **gm; /**< for var eij */
6039: double ***gradg, ***trgradg; /**< for var eij */
6040: double **gradgp, **trgradgp; /**< for var p point j */
6041: double *gpp, *gmp; /**< for var p point j */
6042: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6043: double ***p3mat;
6044: double age,agelim, hf;
6045: /* double ***mobaverage; */
6046: int theta;
6047: char digit[4];
6048: char digitp[25];
6049:
6050: char fileresprobmorprev[FILENAMELENGTH];
6051:
6052: if(popbased==1){
6053: if(mobilav!=0)
6054: strcpy(digitp,"-POPULBASED-MOBILAV_");
6055: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6056: }
6057: else
6058: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6059:
1.218 brouard 6060: /* if (mobilav!=0) { */
6061: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6062: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6063: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6064: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6065: /* } */
6066: /* } */
6067:
6068: strcpy(fileresprobmorprev,"PRMORPREV-");
6069: sprintf(digit,"%-d",ij);
6070: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6071: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6072: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6073: strcat(fileresprobmorprev,fileresu);
6074: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6075: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6076: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6077: }
6078: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6079: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6080: pstamp(ficresprobmorprev);
6081: 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 6082: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6083: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6084: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6085: }
6086: for(j=1;j<=cptcoveff;j++)
6087: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6088: fprintf(ficresprobmorprev,"\n");
6089:
1.218 brouard 6090: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6091: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6092: fprintf(ficresprobmorprev," p.%-d SE",j);
6093: for(i=1; i<=nlstate;i++)
6094: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6095: }
6096: fprintf(ficresprobmorprev,"\n");
6097:
6098: fprintf(ficgp,"\n# Routine varevsij");
6099: fprintf(ficgp,"\nunset title \n");
6100: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6101: 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");
6102: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6103:
1.218 brouard 6104: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6105: pstamp(ficresvij);
6106: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6107: if(popbased==1)
6108: 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);
6109: else
6110: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6111: fprintf(ficresvij,"# Age");
6112: for(i=1; i<=nlstate;i++)
6113: for(j=1; j<=nlstate;j++)
6114: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6115: fprintf(ficresvij,"\n");
6116:
6117: xp=vector(1,npar);
6118: dnewm=matrix(1,nlstate,1,npar);
6119: doldm=matrix(1,nlstate,1,nlstate);
6120: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6121: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6122:
6123: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6124: gpp=vector(nlstate+1,nlstate+ndeath);
6125: gmp=vector(nlstate+1,nlstate+ndeath);
6126: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6127:
1.218 brouard 6128: if(estepm < stepm){
6129: printf ("Problem %d lower than %d\n",estepm, stepm);
6130: }
6131: else hstepm=estepm;
6132: /* For example we decided to compute the life expectancy with the smallest unit */
6133: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6134: nhstepm is the number of hstepm from age to agelim
6135: nstepm is the number of stepm from age to agelim.
6136: Look at function hpijx to understand why because of memory size limitations,
6137: we decided (b) to get a life expectancy respecting the most precise curvature of the
6138: survival function given by stepm (the optimization length). Unfortunately it
6139: means that if the survival funtion is printed every two years of age and if
6140: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6141: results. So we changed our mind and took the option of the best precision.
6142: */
6143: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6144: agelim = AGESUP;
6145: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6146: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6147: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6148: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6149: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6150: gp=matrix(0,nhstepm,1,nlstate);
6151: gm=matrix(0,nhstepm,1,nlstate);
6152:
6153:
6154: for(theta=1; theta <=npar; theta++){
6155: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6156: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6157: }
1.279 brouard 6158: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6159: * returns into prlim .
1.288 brouard 6160: */
1.242 brouard 6161: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6162:
6163: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6164: if (popbased==1) {
6165: if(mobilav ==0){
6166: for(i=1; i<=nlstate;i++)
6167: prlim[i][i]=probs[(int)age][i][ij];
6168: }else{ /* mobilav */
6169: for(i=1; i<=nlstate;i++)
6170: prlim[i][i]=mobaverage[(int)age][i][ij];
6171: }
6172: }
1.295 brouard 6173: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6174: */
6175: 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 6176: /**< 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 6177: * at horizon h in state j including mortality.
6178: */
1.218 brouard 6179: for(j=1; j<= nlstate; j++){
6180: for(h=0; h<=nhstepm; h++){
6181: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6182: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6183: }
6184: }
1.279 brouard 6185: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6186: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6187: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6188: */
6189: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6190: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6191: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6192: }
6193:
6194: /* Again with minus shift */
1.218 brouard 6195:
6196: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6197: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6198:
1.242 brouard 6199: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6200:
6201: if (popbased==1) {
6202: if(mobilav ==0){
6203: for(i=1; i<=nlstate;i++)
6204: prlim[i][i]=probs[(int)age][i][ij];
6205: }else{ /* mobilav */
6206: for(i=1; i<=nlstate;i++)
6207: prlim[i][i]=mobaverage[(int)age][i][ij];
6208: }
6209: }
6210:
1.235 brouard 6211: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6212:
6213: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6214: for(h=0; h<=nhstepm; h++){
6215: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6216: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6217: }
6218: }
6219: /* This for computing probability of death (h=1 means
6220: computed over hstepm matrices product = hstepm*stepm months)
6221: as a weighted average of prlim.
6222: */
6223: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6224: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6225: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6226: }
1.279 brouard 6227: /* end shifting computations */
6228:
6229: /**< Computing gradient matrix at horizon h
6230: */
1.218 brouard 6231: for(j=1; j<= nlstate; j++) /* vareij */
6232: for(h=0; h<=nhstepm; h++){
6233: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6234: }
1.279 brouard 6235: /**< Gradient of overall mortality p.3 (or p.j)
6236: */
6237: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6238: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6239: }
6240:
6241: } /* End theta */
1.279 brouard 6242:
6243: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6244: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6245:
6246: for(h=0; h<=nhstepm; h++) /* veij */
6247: for(j=1; j<=nlstate;j++)
6248: for(theta=1; theta <=npar; theta++)
6249: trgradg[h][j][theta]=gradg[h][theta][j];
6250:
6251: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6252: for(theta=1; theta <=npar; theta++)
6253: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6254: /**< as well as its transposed matrix
6255: */
1.218 brouard 6256:
6257: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6258: for(i=1;i<=nlstate;i++)
6259: for(j=1;j<=nlstate;j++)
6260: vareij[i][j][(int)age] =0.;
1.279 brouard 6261:
6262: /* Computing trgradg by matcov by gradg at age and summing over h
6263: * and k (nhstepm) formula 15 of article
6264: * Lievre-Brouard-Heathcote
6265: */
6266:
1.218 brouard 6267: for(h=0;h<=nhstepm;h++){
6268: for(k=0;k<=nhstepm;k++){
6269: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6270: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6271: for(i=1;i<=nlstate;i++)
6272: for(j=1;j<=nlstate;j++)
6273: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6274: }
6275: }
6276:
1.279 brouard 6277: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6278: * p.j overall mortality formula 49 but computed directly because
6279: * we compute the grad (wix pijx) instead of grad (pijx),even if
6280: * wix is independent of theta.
6281: */
1.218 brouard 6282: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6283: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6284: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6285: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6286: varppt[j][i]=doldmp[j][i];
6287: /* end ppptj */
6288: /* x centered again */
6289:
1.242 brouard 6290: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6291:
6292: if (popbased==1) {
6293: if(mobilav ==0){
6294: for(i=1; i<=nlstate;i++)
6295: prlim[i][i]=probs[(int)age][i][ij];
6296: }else{ /* mobilav */
6297: for(i=1; i<=nlstate;i++)
6298: prlim[i][i]=mobaverage[(int)age][i][ij];
6299: }
6300: }
6301:
6302: /* This for computing probability of death (h=1 means
6303: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6304: as a weighted average of prlim.
6305: */
1.235 brouard 6306: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6307: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6308: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6309: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6310: }
6311: /* end probability of death */
6312:
6313: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6314: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6315: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6316: for(i=1; i<=nlstate;i++){
6317: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6318: }
6319: }
6320: fprintf(ficresprobmorprev,"\n");
6321:
6322: fprintf(ficresvij,"%.0f ",age );
6323: for(i=1; i<=nlstate;i++)
6324: for(j=1; j<=nlstate;j++){
6325: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6326: }
6327: fprintf(ficresvij,"\n");
6328: free_matrix(gp,0,nhstepm,1,nlstate);
6329: free_matrix(gm,0,nhstepm,1,nlstate);
6330: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6331: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6332: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6333: } /* End age */
6334: free_vector(gpp,nlstate+1,nlstate+ndeath);
6335: free_vector(gmp,nlstate+1,nlstate+ndeath);
6336: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6337: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6338: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6339: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6340: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6341: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6342: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6343: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6344: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6345: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6346: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6347: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6348: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6349: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6350: 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);
6351: /* 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 6352: */
1.218 brouard 6353: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6354: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6355:
1.218 brouard 6356: free_vector(xp,1,npar);
6357: free_matrix(doldm,1,nlstate,1,nlstate);
6358: free_matrix(dnewm,1,nlstate,1,npar);
6359: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6360: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6361: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6362: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6363: fclose(ficresprobmorprev);
6364: fflush(ficgp);
6365: fflush(fichtm);
6366: } /* end varevsij */
1.126 brouard 6367:
6368: /************ Variance of prevlim ******************/
1.269 brouard 6369: 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 6370: {
1.205 brouard 6371: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6372: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6373:
1.268 brouard 6374: double **dnewmpar,**doldm;
1.126 brouard 6375: int i, j, nhstepm, hstepm;
6376: double *xp;
6377: double *gp, *gm;
6378: double **gradg, **trgradg;
1.208 brouard 6379: double **mgm, **mgp;
1.126 brouard 6380: double age,agelim;
6381: int theta;
6382:
6383: pstamp(ficresvpl);
1.288 brouard 6384: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6385: fprintf(ficresvpl,"# Age ");
6386: if(nresult >=1)
6387: fprintf(ficresvpl," Result# ");
1.126 brouard 6388: for(i=1; i<=nlstate;i++)
6389: fprintf(ficresvpl," %1d-%1d",i,i);
6390: fprintf(ficresvpl,"\n");
6391:
6392: xp=vector(1,npar);
1.268 brouard 6393: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6394: doldm=matrix(1,nlstate,1,nlstate);
6395:
6396: hstepm=1*YEARM; /* Every year of age */
6397: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6398: agelim = AGESUP;
6399: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6400: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6401: if (stepm >= YEARM) hstepm=1;
6402: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6403: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6404: mgp=matrix(1,npar,1,nlstate);
6405: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6406: gp=vector(1,nlstate);
6407: gm=vector(1,nlstate);
6408:
6409: for(theta=1; theta <=npar; theta++){
6410: for(i=1; i<=npar; i++){ /* Computes gradient */
6411: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6412: }
1.288 brouard 6413: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6414: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6415: /* else */
6416: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6417: for(i=1;i<=nlstate;i++){
1.126 brouard 6418: gp[i] = prlim[i][i];
1.208 brouard 6419: mgp[theta][i] = prlim[i][i];
6420: }
1.126 brouard 6421: for(i=1; i<=npar; i++) /* Computes gradient */
6422: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6423: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6424: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6425: /* else */
6426: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6427: for(i=1;i<=nlstate;i++){
1.126 brouard 6428: gm[i] = prlim[i][i];
1.208 brouard 6429: mgm[theta][i] = prlim[i][i];
6430: }
1.126 brouard 6431: for(i=1;i<=nlstate;i++)
6432: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6433: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6434: } /* End theta */
6435:
6436: trgradg =matrix(1,nlstate,1,npar);
6437:
6438: for(j=1; j<=nlstate;j++)
6439: for(theta=1; theta <=npar; theta++)
6440: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6441: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6442: /* printf("\nmgm mgp %d ",(int)age); */
6443: /* for(j=1; j<=nlstate;j++){ */
6444: /* printf(" %d ",j); */
6445: /* for(theta=1; theta <=npar; theta++) */
6446: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6447: /* printf("\n "); */
6448: /* } */
6449: /* } */
6450: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6451: /* printf("\n gradg %d ",(int)age); */
6452: /* for(j=1; j<=nlstate;j++){ */
6453: /* printf("%d ",j); */
6454: /* for(theta=1; theta <=npar; theta++) */
6455: /* printf("%d %lf ",theta,gradg[theta][j]); */
6456: /* printf("\n "); */
6457: /* } */
6458: /* } */
1.126 brouard 6459:
6460: for(i=1;i<=nlstate;i++)
6461: varpl[i][(int)age] =0.;
1.209 brouard 6462: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6463: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6464: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6465: }else{
1.268 brouard 6466: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6467: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6468: }
1.126 brouard 6469: for(i=1;i<=nlstate;i++)
6470: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6471:
6472: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6473: if(nresult >=1)
6474: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6475: for(i=1; i<=nlstate;i++){
1.126 brouard 6476: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6477: /* for(j=1;j<=nlstate;j++) */
6478: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6479: }
1.126 brouard 6480: fprintf(ficresvpl,"\n");
6481: free_vector(gp,1,nlstate);
6482: free_vector(gm,1,nlstate);
1.208 brouard 6483: free_matrix(mgm,1,npar,1,nlstate);
6484: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6485: free_matrix(gradg,1,npar,1,nlstate);
6486: free_matrix(trgradg,1,nlstate,1,npar);
6487: } /* End age */
6488:
6489: free_vector(xp,1,npar);
6490: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6491: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6492:
6493: }
6494:
6495:
6496: /************ Variance of backprevalence limit ******************/
1.269 brouard 6497: 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 6498: {
6499: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6500: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6501:
6502: double **dnewmpar,**doldm;
6503: int i, j, nhstepm, hstepm;
6504: double *xp;
6505: double *gp, *gm;
6506: double **gradg, **trgradg;
6507: double **mgm, **mgp;
6508: double age,agelim;
6509: int theta;
6510:
6511: pstamp(ficresvbl);
6512: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6513: fprintf(ficresvbl,"# Age ");
6514: if(nresult >=1)
6515: fprintf(ficresvbl," Result# ");
6516: for(i=1; i<=nlstate;i++)
6517: fprintf(ficresvbl," %1d-%1d",i,i);
6518: fprintf(ficresvbl,"\n");
6519:
6520: xp=vector(1,npar);
6521: dnewmpar=matrix(1,nlstate,1,npar);
6522: doldm=matrix(1,nlstate,1,nlstate);
6523:
6524: hstepm=1*YEARM; /* Every year of age */
6525: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6526: agelim = AGEINF;
6527: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6528: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6529: if (stepm >= YEARM) hstepm=1;
6530: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6531: gradg=matrix(1,npar,1,nlstate);
6532: mgp=matrix(1,npar,1,nlstate);
6533: mgm=matrix(1,npar,1,nlstate);
6534: gp=vector(1,nlstate);
6535: gm=vector(1,nlstate);
6536:
6537: for(theta=1; theta <=npar; theta++){
6538: for(i=1; i<=npar; i++){ /* Computes gradient */
6539: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6540: }
6541: if(mobilavproj > 0 )
6542: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6543: else
6544: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6545: for(i=1;i<=nlstate;i++){
6546: gp[i] = bprlim[i][i];
6547: mgp[theta][i] = bprlim[i][i];
6548: }
6549: for(i=1; i<=npar; i++) /* Computes gradient */
6550: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6551: if(mobilavproj > 0 )
6552: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6553: else
6554: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6555: for(i=1;i<=nlstate;i++){
6556: gm[i] = bprlim[i][i];
6557: mgm[theta][i] = bprlim[i][i];
6558: }
6559: for(i=1;i<=nlstate;i++)
6560: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6561: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6562: } /* End theta */
6563:
6564: trgradg =matrix(1,nlstate,1,npar);
6565:
6566: for(j=1; j<=nlstate;j++)
6567: for(theta=1; theta <=npar; theta++)
6568: trgradg[j][theta]=gradg[theta][j];
6569: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6570: /* printf("\nmgm mgp %d ",(int)age); */
6571: /* for(j=1; j<=nlstate;j++){ */
6572: /* printf(" %d ",j); */
6573: /* for(theta=1; theta <=npar; theta++) */
6574: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6575: /* printf("\n "); */
6576: /* } */
6577: /* } */
6578: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6579: /* printf("\n gradg %d ",(int)age); */
6580: /* for(j=1; j<=nlstate;j++){ */
6581: /* printf("%d ",j); */
6582: /* for(theta=1; theta <=npar; theta++) */
6583: /* printf("%d %lf ",theta,gradg[theta][j]); */
6584: /* printf("\n "); */
6585: /* } */
6586: /* } */
6587:
6588: for(i=1;i<=nlstate;i++)
6589: varbpl[i][(int)age] =0.;
6590: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6591: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6592: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6593: }else{
6594: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6595: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6596: }
6597: for(i=1;i<=nlstate;i++)
6598: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6599:
6600: fprintf(ficresvbl,"%.0f ",age );
6601: if(nresult >=1)
6602: fprintf(ficresvbl,"%d ",nres );
6603: for(i=1; i<=nlstate;i++)
6604: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6605: fprintf(ficresvbl,"\n");
6606: free_vector(gp,1,nlstate);
6607: free_vector(gm,1,nlstate);
6608: free_matrix(mgm,1,npar,1,nlstate);
6609: free_matrix(mgp,1,npar,1,nlstate);
6610: free_matrix(gradg,1,npar,1,nlstate);
6611: free_matrix(trgradg,1,nlstate,1,npar);
6612: } /* End age */
6613:
6614: free_vector(xp,1,npar);
6615: free_matrix(doldm,1,nlstate,1,npar);
6616: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6617:
6618: }
6619:
6620: /************ Variance of one-step probabilities ******************/
6621: 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 6622: {
6623: int i, j=0, k1, l1, tj;
6624: int k2, l2, j1, z1;
6625: int k=0, l;
6626: int first=1, first1, first2;
6627: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6628: double **dnewm,**doldm;
6629: double *xp;
6630: double *gp, *gm;
6631: double **gradg, **trgradg;
6632: double **mu;
6633: double age, cov[NCOVMAX+1];
6634: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6635: int theta;
6636: char fileresprob[FILENAMELENGTH];
6637: char fileresprobcov[FILENAMELENGTH];
6638: char fileresprobcor[FILENAMELENGTH];
6639: double ***varpij;
6640:
6641: strcpy(fileresprob,"PROB_");
6642: strcat(fileresprob,fileres);
6643: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6644: printf("Problem with resultfile: %s\n", fileresprob);
6645: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6646: }
6647: strcpy(fileresprobcov,"PROBCOV_");
6648: strcat(fileresprobcov,fileresu);
6649: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6650: printf("Problem with resultfile: %s\n", fileresprobcov);
6651: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6652: }
6653: strcpy(fileresprobcor,"PROBCOR_");
6654: strcat(fileresprobcor,fileresu);
6655: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6656: printf("Problem with resultfile: %s\n", fileresprobcor);
6657: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6658: }
6659: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6660: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6661: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6662: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6663: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6664: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6665: pstamp(ficresprob);
6666: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6667: fprintf(ficresprob,"# Age");
6668: pstamp(ficresprobcov);
6669: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6670: fprintf(ficresprobcov,"# Age");
6671: pstamp(ficresprobcor);
6672: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6673: fprintf(ficresprobcor,"# Age");
1.126 brouard 6674:
6675:
1.222 brouard 6676: for(i=1; i<=nlstate;i++)
6677: for(j=1; j<=(nlstate+ndeath);j++){
6678: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6679: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6680: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6681: }
6682: /* fprintf(ficresprob,"\n");
6683: fprintf(ficresprobcov,"\n");
6684: fprintf(ficresprobcor,"\n");
6685: */
6686: xp=vector(1,npar);
6687: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6688: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6689: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6690: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6691: first=1;
6692: fprintf(ficgp,"\n# Routine varprob");
6693: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6694: fprintf(fichtm,"\n");
6695:
1.288 brouard 6696: 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 6697: 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);
6698: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6699: and drawn. It helps understanding how is the covariance between two incidences.\
6700: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6701: 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 6702: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6703: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6704: standard deviations wide on each axis. <br>\
6705: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6706: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6707: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6708:
1.222 brouard 6709: cov[1]=1;
6710: /* tj=cptcoveff; */
1.225 brouard 6711: tj = (int) pow(2,cptcoveff);
1.222 brouard 6712: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6713: j1=0;
1.224 brouard 6714: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6715: if (cptcovn>0) {
6716: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6717: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6718: fprintf(ficresprob, "**********\n#\n");
6719: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6720: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6721: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6722:
1.222 brouard 6723: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6724: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6725: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6726:
6727:
1.222 brouard 6728: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6729: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6730: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6731:
1.222 brouard 6732: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6733: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6734: fprintf(ficresprobcor, "**********\n#");
6735: if(invalidvarcomb[j1]){
6736: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6737: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6738: continue;
6739: }
6740: }
6741: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6742: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6743: gp=vector(1,(nlstate)*(nlstate+ndeath));
6744: gm=vector(1,(nlstate)*(nlstate+ndeath));
6745: for (age=bage; age<=fage; age ++){
6746: cov[2]=age;
6747: if(nagesqr==1)
6748: cov[3]= age*age;
6749: for (k=1; k<=cptcovn;k++) {
6750: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6751: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6752: * 1 1 1 1 1
6753: * 2 2 1 1 1
6754: * 3 1 2 1 1
6755: */
6756: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6757: }
6758: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6759: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6760: for (k=1; k<=cptcovprod;k++)
6761: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6762:
6763:
1.222 brouard 6764: for(theta=1; theta <=npar; theta++){
6765: for(i=1; i<=npar; i++)
6766: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6767:
1.222 brouard 6768: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6769:
1.222 brouard 6770: k=0;
6771: for(i=1; i<= (nlstate); i++){
6772: for(j=1; j<=(nlstate+ndeath);j++){
6773: k=k+1;
6774: gp[k]=pmmij[i][j];
6775: }
6776: }
1.220 brouard 6777:
1.222 brouard 6778: for(i=1; i<=npar; i++)
6779: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6780:
1.222 brouard 6781: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6782: k=0;
6783: for(i=1; i<=(nlstate); i++){
6784: for(j=1; j<=(nlstate+ndeath);j++){
6785: k=k+1;
6786: gm[k]=pmmij[i][j];
6787: }
6788: }
1.220 brouard 6789:
1.222 brouard 6790: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6791: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6792: }
1.126 brouard 6793:
1.222 brouard 6794: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6795: for(theta=1; theta <=npar; theta++)
6796: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6797:
1.222 brouard 6798: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6799: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6800:
1.222 brouard 6801: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6802:
1.222 brouard 6803: k=0;
6804: for(i=1; i<=(nlstate); i++){
6805: for(j=1; j<=(nlstate+ndeath);j++){
6806: k=k+1;
6807: mu[k][(int) age]=pmmij[i][j];
6808: }
6809: }
6810: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6811: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6812: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6813:
1.222 brouard 6814: /*printf("\n%d ",(int)age);
6815: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6816: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6817: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6818: }*/
1.220 brouard 6819:
1.222 brouard 6820: fprintf(ficresprob,"\n%d ",(int)age);
6821: fprintf(ficresprobcov,"\n%d ",(int)age);
6822: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6823:
1.222 brouard 6824: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6825: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6826: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6827: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6828: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6829: }
6830: i=0;
6831: for (k=1; k<=(nlstate);k++){
6832: for (l=1; l<=(nlstate+ndeath);l++){
6833: i++;
6834: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6835: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6836: for (j=1; j<=i;j++){
6837: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6838: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6839: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6840: }
6841: }
6842: }/* end of loop for state */
6843: } /* end of loop for age */
6844: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6845: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6846: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6847: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6848:
6849: /* Confidence intervalle of pij */
6850: /*
6851: fprintf(ficgp,"\nunset parametric;unset label");
6852: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6853: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6854: 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);
6855: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6856: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6857: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6858: */
6859:
6860: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6861: first1=1;first2=2;
6862: for (k2=1; k2<=(nlstate);k2++){
6863: for (l2=1; l2<=(nlstate+ndeath);l2++){
6864: if(l2==k2) continue;
6865: j=(k2-1)*(nlstate+ndeath)+l2;
6866: for (k1=1; k1<=(nlstate);k1++){
6867: for (l1=1; l1<=(nlstate+ndeath);l1++){
6868: if(l1==k1) continue;
6869: i=(k1-1)*(nlstate+ndeath)+l1;
6870: if(i<=j) continue;
6871: for (age=bage; age<=fage; age ++){
6872: if ((int)age %5==0){
6873: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6874: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6875: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6876: mu1=mu[i][(int) age]/stepm*YEARM ;
6877: mu2=mu[j][(int) age]/stepm*YEARM;
6878: c12=cv12/sqrt(v1*v2);
6879: /* Computing eigen value of matrix of covariance */
6880: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6881: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6882: if ((lc2 <0) || (lc1 <0) ){
6883: if(first2==1){
6884: first1=0;
6885: 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);
6886: }
6887: 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);
6888: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6889: /* lc2=fabs(lc2); */
6890: }
1.220 brouard 6891:
1.222 brouard 6892: /* Eigen vectors */
1.280 brouard 6893: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6894: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6895: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6896: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6897: }else
6898: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6899: /*v21=sqrt(1.-v11*v11); *//* error */
6900: v21=(lc1-v1)/cv12*v11;
6901: v12=-v21;
6902: v22=v11;
6903: tnalp=v21/v11;
6904: if(first1==1){
6905: first1=0;
6906: 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);
6907: }
6908: 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);
6909: /*printf(fignu*/
6910: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6911: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6912: if(first==1){
6913: first=0;
6914: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6915: fprintf(ficgp,"\nset parametric;unset label");
6916: 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);
6917: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6918: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6919: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6920: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6921: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6922: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6923: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6924: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6925: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6926: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6927: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6928: 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 6929: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6930: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6931: }else{
6932: first=0;
6933: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6934: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6935: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6936: 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 6937: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6938: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6939: }/* if first */
6940: } /* age mod 5 */
6941: } /* end loop age */
6942: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6943: first=1;
6944: } /*l12 */
6945: } /* k12 */
6946: } /*l1 */
6947: }/* k1 */
6948: } /* loop on combination of covariates j1 */
6949: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6950: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6951: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6952: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6953: free_vector(xp,1,npar);
6954: fclose(ficresprob);
6955: fclose(ficresprobcov);
6956: fclose(ficresprobcor);
6957: fflush(ficgp);
6958: fflush(fichtmcov);
6959: }
1.126 brouard 6960:
6961:
6962: /******************* Printing html file ***********/
1.201 brouard 6963: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6964: int lastpass, int stepm, int weightopt, char model[],\
6965: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6966: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6967: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6968: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6969: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6970:
6971: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6972: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6973: </ul>");
1.237 brouard 6974: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6975: </ul>", model);
1.214 brouard 6976: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6977: 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",
6978: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6979: 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 6980: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6981: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6982: fprintf(fichtm,"\
6983: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6984: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6985: fprintf(fichtm,"\
1.217 brouard 6986: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6987: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6988: fprintf(fichtm,"\
1.288 brouard 6989: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6990: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6991: fprintf(fichtm,"\
1.288 brouard 6992: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6993: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6994: fprintf(fichtm,"\
1.211 brouard 6995: - (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 6996: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6997: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6998: if(prevfcast==1){
6999: fprintf(fichtm,"\
7000: - Prevalence projections by age and states: \
1.201 brouard 7001: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7002: }
1.126 brouard 7003:
7004:
1.225 brouard 7005: m=pow(2,cptcoveff);
1.222 brouard 7006: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7007:
1.317 ! brouard 7008: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7009:
7010: jj1=0;
7011:
7012: fprintf(fichtm," \n<ul>");
7013: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7014: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7015: if(m != 1 && TKresult[nres]!= k1)
7016: continue;
7017: jj1++;
7018: if (cptcovn > 0) {
7019: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#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,"\">");
7027:
7028: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7029: fprintf(fichtm,"************ Results for covariates");
7030: for (cpt=1; cpt<=cptcoveff;cpt++){
7031: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7032: }
7033: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7034: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7035: }
7036: if(invalidvarcomb[k1]){
7037: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7038: continue;
7039: }
7040: fprintf(fichtm,"</a></li>");
7041: } /* cptcovn >0 */
7042: }
1.317 ! brouard 7043: fprintf(fichtm," \n</ul>");
1.264 brouard 7044:
1.222 brouard 7045: jj1=0;
1.237 brouard 7046:
7047: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7048: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7049: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7050: continue;
1.220 brouard 7051:
1.222 brouard 7052: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7053: jj1++;
7054: if (cptcovn > 0) {
1.264 brouard 7055: fprintf(fichtm,"\n<p><a name=\"rescov");
7056: for (cpt=1; cpt<=cptcoveff;cpt++){
7057: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7058: }
7059: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7060: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7061: }
7062: fprintf(fichtm,"\"</a>");
7063:
1.222 brouard 7064: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7065: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7066: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7067: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7068: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7069: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7070: }
1.237 brouard 7071: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7072: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7073: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7074: }
7075:
1.230 brouard 7076: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 7077: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
7078: if(invalidvarcomb[k1]){
7079: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7080: printf("\nCombination (%d) ignored because no cases \n",k1);
7081: continue;
7082: }
7083: }
7084: /* aij, bij */
1.259 brouard 7085: 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 7086: <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 7087: /* Pij */
1.241 brouard 7088: 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> \
7089: <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 7090: /* Quasi-incidences */
7091: 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 7092: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7093: 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 7094: 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> \
7095: <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 7096: /* Survival functions (period) in state j */
7097: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7098: 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 7099: <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 7100: }
7101: /* State specific survival functions (period) */
7102: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7103: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7104: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7105: <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 7106: }
1.288 brouard 7107: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7108: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7109: 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> \
7110: <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 7111: }
1.296 brouard 7112: if(prevbcast==1){
1.288 brouard 7113: /* Backward prevalence in each health state */
1.222 brouard 7114: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7115: 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 7116: <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 7117: }
1.217 brouard 7118: }
1.222 brouard 7119: if(prevfcast==1){
1.288 brouard 7120: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7121: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7122: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) forward prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
7123: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7124: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7125: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7126: }
7127: }
1.296 brouard 7128: if(prevbcast==1){
1.268 brouard 7129: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7130: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7131: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7132: 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 \
7133: account but can visually be appreciated). Or probability to have been in an state %d, knowing that the person was in either state (1 or %d) \
1.314 brouard 7134: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7135: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7136: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7137: }
7138: }
1.220 brouard 7139:
1.222 brouard 7140: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7141: fprintf(fichtm,"\n<br>- Life expectancy by health state (%d) at initial age and its decomposition into health expectancies in each alive state (1 to %d) (or area under each survival functions): <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
7142: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7143: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7144: }
7145: /* } /\* end i1 *\/ */
7146: }/* End k1 */
7147: fprintf(fichtm,"</ul>");
1.126 brouard 7148:
1.222 brouard 7149: fprintf(fichtm,"\
1.126 brouard 7150: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7151: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7152: - 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 7153: But because parameters are usually highly correlated (a higher incidence of disability \
7154: and a higher incidence of recovery can give very close observed transition) it might \
7155: be very useful to look not only at linear confidence intervals estimated from the \
7156: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7157: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7158: covariance matrix of the one-step probabilities. \
7159: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7160:
1.222 brouard 7161: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7162: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7163: fprintf(fichtm,"\
1.126 brouard 7164: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7165: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7166:
1.222 brouard 7167: fprintf(fichtm,"\
1.126 brouard 7168: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7169: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7170: fprintf(fichtm,"\
1.126 brouard 7171: - 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): \
7172: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7173: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7174: fprintf(fichtm,"\
1.126 brouard 7175: - (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): \
7176: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7177: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7178: fprintf(fichtm,"\
1.288 brouard 7179: - 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 7180: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7181: fprintf(fichtm,"\
1.128 brouard 7182: - 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 7183: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7184: fprintf(fichtm,"\
1.288 brouard 7185: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7186: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7187:
7188: /* if(popforecast==1) fprintf(fichtm,"\n */
7189: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7190: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7191: /* <br>",fileres,fileres,fileres,fileres); */
7192: /* else */
7193: /* 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 7194: fflush(fichtm);
1.126 brouard 7195:
1.225 brouard 7196: m=pow(2,cptcoveff);
1.222 brouard 7197: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7198:
1.317 ! brouard 7199: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
! 7200:
! 7201: jj1=0;
! 7202:
! 7203: fprintf(fichtm," \n<ul>");
! 7204: for(nres=1; nres <= nresult; nres++) /* For each resultline */
! 7205: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
! 7206: if(m != 1 && TKresult[nres]!= k1)
! 7207: continue;
! 7208: jj1++;
! 7209: if (cptcovn > 0) {
! 7210: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
! 7211: for (cpt=1; cpt<=cptcoveff;cpt++){
! 7212: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
! 7213: }
! 7214: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
! 7215: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
! 7216: }
! 7217: fprintf(fichtm,"\">");
! 7218:
! 7219: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
! 7220: fprintf(fichtm,"************ Results for covariates");
! 7221: for (cpt=1; cpt<=cptcoveff;cpt++){
! 7222: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
! 7223: }
! 7224: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
! 7225: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
! 7226: }
! 7227: if(invalidvarcomb[k1]){
! 7228: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
! 7229: continue;
! 7230: }
! 7231: fprintf(fichtm,"</a></li>");
! 7232: } /* cptcovn >0 */
! 7233: }
! 7234: fprintf(fichtm," \n</ul>");
! 7235:
1.222 brouard 7236: jj1=0;
1.237 brouard 7237:
1.241 brouard 7238: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7239: for(k1=1; k1<=m;k1++){
1.253 brouard 7240: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7241: continue;
1.222 brouard 7242: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7243: jj1++;
1.126 brouard 7244: if (cptcovn > 0) {
1.317 ! brouard 7245: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
! 7246: for (cpt=1; cpt<=cptcoveff;cpt++){
! 7247: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
! 7248: }
! 7249: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
! 7250: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
! 7251: }
! 7252: fprintf(fichtm,"\"</a>");
! 7253:
1.126 brouard 7254: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 ! brouard 7255: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7256: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 ! brouard 7257: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7258: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 ! brouard 7259: }
1.237 brouard 7260: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7261: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7262: }
7263:
1.126 brouard 7264: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7265:
1.222 brouard 7266: if(invalidvarcomb[k1]){
7267: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7268: continue;
7269: }
1.126 brouard 7270: }
7271: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7272: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7273: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7274: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7275: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7276: }
7277: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7278: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7279: true period expectancies (those weighted with period prevalences are also\
7280: drawn in addition to the population based expectancies computed using\
1.314 brouard 7281: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>",nlstate, subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
7282: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7283: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7284: /* } /\* end i1 *\/ */
7285: }/* End k1 */
1.241 brouard 7286: }/* End nres */
1.222 brouard 7287: fprintf(fichtm,"</ul>");
7288: fflush(fichtm);
1.126 brouard 7289: }
7290:
7291: /******************* Gnuplot file **************/
1.296 brouard 7292: 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 7293:
7294: char dirfileres[132],optfileres[132];
1.264 brouard 7295: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7296: 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 7297: int lv=0, vlv=0, kl=0;
1.130 brouard 7298: int ng=0;
1.201 brouard 7299: int vpopbased;
1.223 brouard 7300: int ioffset; /* variable offset for columns */
1.270 brouard 7301: int iyearc=1; /* variable column for year of projection */
7302: int iagec=1; /* variable column for age of projection */
1.235 brouard 7303: int nres=0; /* Index of resultline */
1.266 brouard 7304: int istart=1; /* For starting graphs in projections */
1.219 brouard 7305:
1.126 brouard 7306: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7307: /* printf("Problem with file %s",optionfilegnuplot); */
7308: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7309: /* } */
7310:
7311: /*#ifdef windows */
7312: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7313: /*#endif */
1.225 brouard 7314: m=pow(2,cptcoveff);
1.126 brouard 7315:
1.274 brouard 7316: /* diagram of the model */
7317: fprintf(ficgp,"\n#Diagram of the model \n");
7318: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7319: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7320: 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);
7321:
7322: 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);
7323: fprintf(ficgp,"\n#show arrow\nunset label\n");
7324: 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);
7325: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7326: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7327: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7328: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7329:
1.202 brouard 7330: /* Contribution to likelihood */
7331: /* Plot the probability implied in the likelihood */
1.223 brouard 7332: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7333: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7334: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7335: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7336: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7337: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7338: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7339: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7340: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7341: 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));
7342: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7343: 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));
7344: for (i=1; i<= nlstate ; i ++) {
7345: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7346: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7347: 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);
7348: for (j=2; j<= nlstate+ndeath ; j ++) {
7349: 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);
7350: }
7351: fprintf(ficgp,";\nset out; unset ylabel;\n");
7352: }
7353: /* 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 */
7354: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7355: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7356: fprintf(ficgp,"\nset out;unset log\n");
7357: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7358:
1.126 brouard 7359: strcpy(dirfileres,optionfilefiname);
7360: strcpy(optfileres,"vpl");
1.223 brouard 7361: /* 1eme*/
1.238 brouard 7362: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7363: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7364: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7365: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7366: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7367: continue;
7368: /* We are interested in selected combination by the resultline */
1.246 brouard 7369: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7370: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7371: strcpy(gplotlabel,"(");
1.238 brouard 7372: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7373: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7374: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7375: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7376: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7377: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7378: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7379: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7380: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7381: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7382: }
7383: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7384: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7385: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7386: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7387: }
7388: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7389: /* printf("\n#\n"); */
1.238 brouard 7390: fprintf(ficgp,"\n#\n");
7391: if(invalidvarcomb[k1]){
1.260 brouard 7392: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7393: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7394: continue;
7395: }
1.235 brouard 7396:
1.241 brouard 7397: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7398: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7399: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7400: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7401: 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);
7402: /* 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); */
7403: /* k1-1 error should be nres-1*/
1.238 brouard 7404: for (i=1; i<= nlstate ; i ++) {
7405: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7406: else fprintf(ficgp," %%*lf (%%*lf)");
7407: }
1.288 brouard 7408: 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 7409: for (i=1; i<= nlstate ; i ++) {
7410: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7411: else fprintf(ficgp," %%*lf (%%*lf)");
7412: }
1.260 brouard 7413: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7414: for (i=1; i<= nlstate ; i ++) {
7415: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7416: else fprintf(ficgp," %%*lf (%%*lf)");
7417: }
1.265 brouard 7418: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7419:
7420: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7421: if(cptcoveff ==0){
1.271 brouard 7422: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7423: }else{
7424: kl=0;
7425: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7426: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7427: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7428: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7429: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7430: vlv= nbcode[Tvaraff[k]][lv];
7431: kl++;
7432: /* 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 *\/ */
7433: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7434: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7435: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7436: if(k==cptcoveff){
7437: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7438: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7439: }else{
7440: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7441: kl++;
7442: }
7443: } /* end covariate */
7444: } /* end if no covariate */
7445:
1.296 brouard 7446: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7447: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7448: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7449: if(cptcoveff ==0){
1.245 brouard 7450: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7451: }else{
7452: kl=0;
7453: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7454: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7455: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7456: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7457: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7458: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7459: kl++;
1.238 brouard 7460: /* 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 *\/ */
7461: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7462: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7463: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7464: if(k==cptcoveff){
1.245 brouard 7465: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7466: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7467: }else{
7468: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7469: kl++;
7470: }
7471: } /* end covariate */
7472: } /* end if no covariate */
1.296 brouard 7473: if(prevbcast == 1){
1.268 brouard 7474: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7475: /* k1-1 error should be nres-1*/
7476: for (i=1; i<= nlstate ; i ++) {
7477: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7478: else fprintf(ficgp," %%*lf (%%*lf)");
7479: }
1.271 brouard 7480: 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 7481: for (i=1; i<= nlstate ; i ++) {
7482: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7483: else fprintf(ficgp," %%*lf (%%*lf)");
7484: }
1.276 brouard 7485: 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 7486: for (i=1; i<= nlstate ; i ++) {
7487: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7488: else fprintf(ficgp," %%*lf (%%*lf)");
7489: }
1.274 brouard 7490: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7491: } /* end if backprojcast */
1.296 brouard 7492: } /* end if prevbcast */
1.276 brouard 7493: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7494: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7495: } /* nres */
1.201 brouard 7496: } /* k1 */
7497: } /* cpt */
1.235 brouard 7498:
7499:
1.126 brouard 7500: /*2 eme*/
1.238 brouard 7501: for (k1=1; k1<= m ; k1 ++){
7502: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7503: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7504: continue;
7505: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7506: strcpy(gplotlabel,"(");
1.238 brouard 7507: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7508: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7509: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7510: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7511: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7512: vlv= nbcode[Tvaraff[k]][lv];
7513: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7514: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7515: }
1.237 brouard 7516: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7517: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7518: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7519: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7520: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7521: }
1.264 brouard 7522: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7523: fprintf(ficgp,"\n#\n");
1.223 brouard 7524: if(invalidvarcomb[k1]){
7525: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7526: continue;
7527: }
1.219 brouard 7528:
1.241 brouard 7529: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7530: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7531: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7532: if(vpopbased==0){
1.238 brouard 7533: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7534: }else
1.238 brouard 7535: fprintf(ficgp,"\nreplot ");
7536: for (i=1; i<= nlstate+1 ; i ++) {
7537: k=2*i;
1.261 brouard 7538: 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 7539: for (j=1; j<= nlstate+1 ; j ++) {
7540: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7541: else fprintf(ficgp," %%*lf (%%*lf)");
7542: }
7543: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7544: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7545: 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 7546: for (j=1; j<= nlstate+1 ; j ++) {
7547: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7548: else fprintf(ficgp," %%*lf (%%*lf)");
7549: }
7550: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7551: 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 7552: for (j=1; j<= nlstate+1 ; j ++) {
7553: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7554: else fprintf(ficgp," %%*lf (%%*lf)");
7555: }
7556: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7557: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7558: } /* state */
7559: } /* vpopbased */
1.264 brouard 7560: 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 7561: } /* end nres */
7562: } /* k1 end 2 eme*/
7563:
7564:
7565: /*3eme*/
7566: for (k1=1; k1<= m ; k1 ++){
7567: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7568: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7569: continue;
7570:
7571: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7572: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7573: strcpy(gplotlabel,"(");
1.238 brouard 7574: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7575: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7576: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7577: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7578: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7579: vlv= nbcode[Tvaraff[k]][lv];
7580: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7581: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7582: }
7583: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7584: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7585: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7586: }
1.264 brouard 7587: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7588: fprintf(ficgp,"\n#\n");
7589: if(invalidvarcomb[k1]){
7590: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7591: continue;
7592: }
7593:
7594: /* k=2+nlstate*(2*cpt-2); */
7595: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7596: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7597: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7598: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7599: 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 7600: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7601: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7602: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7603: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7604: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7605: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7606:
1.238 brouard 7607: */
7608: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7609: 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 7610: /* 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 7611:
1.238 brouard 7612: }
1.261 brouard 7613: 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 7614: }
1.264 brouard 7615: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7616: } /* end nres */
7617: } /* end kl 3eme */
1.126 brouard 7618:
1.223 brouard 7619: /* 4eme */
1.201 brouard 7620: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7621: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7622: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7623: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7624: continue;
1.238 brouard 7625: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7626: strcpy(gplotlabel,"(");
1.238 brouard 7627: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7628: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7629: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7630: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7631: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7632: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7633: vlv= nbcode[Tvaraff[k]][lv];
7634: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7635: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7636: }
7637: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7638: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7639: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7640: }
1.264 brouard 7641: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7642: fprintf(ficgp,"\n#\n");
7643: if(invalidvarcomb[k1]){
7644: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7645: continue;
1.223 brouard 7646: }
1.238 brouard 7647:
1.241 brouard 7648: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7649: 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 7650: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7651: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7652: k=3;
7653: for (i=1; i<= nlstate ; i ++){
7654: if(i==1){
7655: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7656: }else{
7657: fprintf(ficgp,", '' ");
7658: }
7659: l=(nlstate+ndeath)*(i-1)+1;
7660: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7661: for (j=2; j<= nlstate+ndeath ; j ++)
7662: fprintf(ficgp,"+$%d",k+l+j-1);
7663: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7664: } /* nlstate */
1.264 brouard 7665: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7666: } /* end cpt state*/
7667: } /* end nres */
7668: } /* end covariate k1 */
7669:
1.220 brouard 7670: /* 5eme */
1.201 brouard 7671: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7672: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7673: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7674: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7675: continue;
1.238 brouard 7676: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7677: strcpy(gplotlabel,"(");
1.238 brouard 7678: 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);
7679: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7680: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7681: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7682: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7683: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7684: vlv= nbcode[Tvaraff[k]][lv];
7685: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7686: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7687: }
7688: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7689: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7690: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7691: }
1.264 brouard 7692: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7693: fprintf(ficgp,"\n#\n");
7694: if(invalidvarcomb[k1]){
7695: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7696: continue;
7697: }
1.227 brouard 7698:
1.241 brouard 7699: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7700: 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 7701: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7702: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7703: k=3;
7704: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7705: if(j==1)
7706: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7707: else
7708: fprintf(ficgp,", '' ");
7709: l=(nlstate+ndeath)*(cpt-1) +j;
7710: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7711: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7712: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7713: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7714: } /* nlstate */
7715: fprintf(ficgp,", '' ");
7716: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7717: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7718: l=(nlstate+ndeath)*(cpt-1) +j;
7719: if(j < nlstate)
7720: fprintf(ficgp,"$%d +",k+l);
7721: else
7722: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7723: }
1.264 brouard 7724: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7725: } /* end cpt state*/
7726: } /* end covariate */
7727: } /* end nres */
1.227 brouard 7728:
1.220 brouard 7729: /* 6eme */
1.202 brouard 7730: /* CV preval stable (period) for each covariate */
1.237 brouard 7731: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7732: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7733: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7734: continue;
1.255 brouard 7735: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7736: strcpy(gplotlabel,"(");
1.288 brouard 7737: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7738: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7739: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7740: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7741: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7742: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7743: vlv= nbcode[Tvaraff[k]][lv];
7744: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7745: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7746: }
1.237 brouard 7747: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7748: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7749: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7750: }
1.264 brouard 7751: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7752: fprintf(ficgp,"\n#\n");
1.223 brouard 7753: if(invalidvarcomb[k1]){
1.227 brouard 7754: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7755: continue;
1.223 brouard 7756: }
1.227 brouard 7757:
1.241 brouard 7758: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7759: 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 7760: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7761: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7762: k=3; /* Offset */
1.255 brouard 7763: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7764: if(i==1)
7765: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7766: else
7767: fprintf(ficgp,", '' ");
1.255 brouard 7768: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7769: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7770: for (j=2; j<= nlstate ; j ++)
7771: fprintf(ficgp,"+$%d",k+l+j-1);
7772: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7773: } /* nlstate */
1.264 brouard 7774: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7775: } /* end cpt state*/
7776: } /* end covariate */
1.227 brouard 7777:
7778:
1.220 brouard 7779: /* 7eme */
1.296 brouard 7780: if(prevbcast == 1){
1.288 brouard 7781: /* CV backward prevalence for each covariate */
1.237 brouard 7782: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7783: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7784: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7785: continue;
1.268 brouard 7786: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7787: strcpy(gplotlabel,"(");
1.288 brouard 7788: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7789: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7790: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7791: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7792: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7793: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7794: vlv= nbcode[Tvaraff[k]][lv];
7795: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7796: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7797: }
1.237 brouard 7798: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7799: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7800: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7801: }
1.264 brouard 7802: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7803: fprintf(ficgp,"\n#\n");
7804: if(invalidvarcomb[k1]){
7805: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7806: continue;
7807: }
7808:
1.241 brouard 7809: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7810: 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 7811: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7812: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7813: k=3; /* Offset */
1.268 brouard 7814: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7815: if(i==1)
7816: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7817: else
7818: fprintf(ficgp,", '' ");
7819: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7820: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7821: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7822: /* 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 7823: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7824: /* for (j=2; j<= nlstate ; j ++) */
7825: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7826: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7827: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7828: } /* nlstate */
1.264 brouard 7829: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7830: } /* end cpt state*/
7831: } /* end covariate */
1.296 brouard 7832: } /* End if prevbcast */
1.218 brouard 7833:
1.223 brouard 7834: /* 8eme */
1.218 brouard 7835: if(prevfcast==1){
1.288 brouard 7836: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7837:
1.237 brouard 7838: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7839: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7840: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7841: continue;
1.211 brouard 7842: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7843: strcpy(gplotlabel,"(");
1.288 brouard 7844: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7845: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7846: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7847: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7848: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7849: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7850: vlv= nbcode[Tvaraff[k]][lv];
7851: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7852: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7853: }
1.237 brouard 7854: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7855: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7856: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7857: }
1.264 brouard 7858: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7859: fprintf(ficgp,"\n#\n");
7860: if(invalidvarcomb[k1]){
7861: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7862: continue;
7863: }
7864:
7865: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7866: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7867: 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 7868: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7869: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7870:
7871: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7872: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7873: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7874: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7875: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7876: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7877: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7878: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7879: if(i==istart){
1.227 brouard 7880: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7881: }else{
7882: fprintf(ficgp,",\\\n '' ");
7883: }
7884: if(cptcoveff ==0){ /* No covariate */
7885: ioffset=2; /* Age is in 2 */
7886: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7887: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7888: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7889: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7890: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7891: if(i==nlstate+1){
1.270 brouard 7892: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7893: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7894: fprintf(ficgp,",\\\n '' ");
7895: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7896: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7897: offyear, \
1.268 brouard 7898: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7899: }else
1.227 brouard 7900: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7901: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7902: }else{ /* more than 2 covariates */
1.270 brouard 7903: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7904: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7905: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7906: iyearc=ioffset-1;
7907: iagec=ioffset;
1.227 brouard 7908: fprintf(ficgp," u %d:(",ioffset);
7909: kl=0;
7910: strcpy(gplotcondition,"(");
7911: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7912: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7913: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7914: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7915: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7916: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7917: kl++;
7918: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7919: kl++;
7920: if(k <cptcoveff && cptcoveff>1)
7921: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7922: }
7923: strcpy(gplotcondition+strlen(gplotcondition),")");
7924: /* 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 *\/ */
7925: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7926: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7927: /* '' 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*/
7928: if(i==nlstate+1){
1.270 brouard 7929: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7930: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7931: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7932: fprintf(ficgp," u %d:(",iagec);
7933: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7934: iyearc, iagec, offyear, \
7935: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7936: /* '' 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 7937: }else{
7938: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7939: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7940: }
7941: } /* end if covariate */
7942: } /* nlstate */
1.264 brouard 7943: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7944: } /* end cpt state*/
7945: } /* end covariate */
7946: } /* End if prevfcast */
1.227 brouard 7947:
1.296 brouard 7948: if(prevbcast==1){
1.268 brouard 7949: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7950:
7951: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7952: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7953: if(m != 1 && TKresult[nres]!= k1)
7954: continue;
7955: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7956: strcpy(gplotlabel,"(");
7957: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7958: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7959: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7960: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7961: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7962: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7963: vlv= nbcode[Tvaraff[k]][lv];
7964: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7965: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7966: }
7967: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7968: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7969: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7970: }
7971: strcpy(gplotlabel+strlen(gplotlabel),")");
7972: fprintf(ficgp,"\n#\n");
7973: if(invalidvarcomb[k1]){
7974: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7975: continue;
7976: }
7977:
7978: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7979: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7980: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7981: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7982: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7983:
7984: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7985: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7986: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7987: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7988: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7989: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7990: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7991: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7992: if(i==istart){
7993: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7994: }else{
7995: fprintf(ficgp,",\\\n '' ");
7996: }
7997: if(cptcoveff ==0){ /* No covariate */
7998: ioffset=2; /* Age is in 2 */
7999: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8000: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8001: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8002: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8003: fprintf(ficgp," u %d:(", ioffset);
8004: if(i==nlstate+1){
1.270 brouard 8005: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8006: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8007: fprintf(ficgp,",\\\n '' ");
8008: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8009: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8010: offbyear, \
8011: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8012: }else
8013: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8014: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8015: }else{ /* more than 2 covariates */
1.270 brouard 8016: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8017: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8018: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8019: iyearc=ioffset-1;
8020: iagec=ioffset;
1.268 brouard 8021: fprintf(ficgp," u %d:(",ioffset);
8022: kl=0;
8023: strcpy(gplotcondition,"(");
8024: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8025: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8026: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8027: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8028: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8029: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8030: kl++;
8031: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8032: kl++;
8033: if(k <cptcoveff && cptcoveff>1)
8034: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8035: }
8036: strcpy(gplotcondition+strlen(gplotcondition),")");
8037: /* 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 *\/ */
8038: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8039: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8040: /* '' 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*/
8041: if(i==nlstate+1){
1.270 brouard 8042: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8043: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8044: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8045: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8046: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8047: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8048: iyearc,iagec,offbyear, \
8049: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8050: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8051: }else{
8052: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8053: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8054: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8055: }
8056: } /* end if covariate */
8057: } /* nlstate */
8058: fprintf(ficgp,"\nset out; unset label;\n");
8059: } /* end cpt state*/
8060: } /* end covariate */
1.296 brouard 8061: } /* End if prevbcast */
1.268 brouard 8062:
1.227 brouard 8063:
1.238 brouard 8064: /* 9eme writing MLE parameters */
8065: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8066: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8067: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8068: for(k=1; k <=(nlstate+ndeath); k++){
8069: if (k != i) {
1.227 brouard 8070: fprintf(ficgp,"# current state %d\n",k);
8071: for(j=1; j <=ncovmodel; j++){
8072: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8073: jk++;
8074: }
8075: fprintf(ficgp,"\n");
1.126 brouard 8076: }
8077: }
1.223 brouard 8078: }
1.187 brouard 8079: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8080:
1.145 brouard 8081: /*goto avoid;*/
1.238 brouard 8082: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8083: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8084: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8085: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8086: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8087: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8088: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8089: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8090: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8091: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8092: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8093: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8094: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8095: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8096: fprintf(ficgp,"#\n");
1.223 brouard 8097: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8098: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8099: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8100: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8101: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8102: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8103: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8104: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8105: continue;
1.264 brouard 8106: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8107: strcpy(gplotlabel,"(");
1.276 brouard 8108: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8109: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8110: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8111: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8112: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8113: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8114: vlv= nbcode[Tvaraff[k]][lv];
8115: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8116: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8117: }
1.237 brouard 8118: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8119: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8120: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8121: }
1.264 brouard 8122: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8123: fprintf(ficgp,"\n#\n");
1.264 brouard 8124: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8125: fprintf(ficgp,"\nset key outside ");
8126: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8127: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8128: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8129: if (ng==1){
8130: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8131: fprintf(ficgp,"\nunset log y");
8132: }else if (ng==2){
8133: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8134: fprintf(ficgp,"\nset log y");
8135: }else if (ng==3){
8136: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8137: fprintf(ficgp,"\nset log y");
8138: }else
8139: fprintf(ficgp,"\nunset title ");
8140: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8141: i=1;
8142: for(k2=1; k2<=nlstate; k2++) {
8143: k3=i;
8144: for(k=1; k<=(nlstate+ndeath); k++) {
8145: if (k != k2){
8146: switch( ng) {
8147: case 1:
8148: if(nagesqr==0)
8149: fprintf(ficgp," p%d+p%d*x",i,i+1);
8150: else /* nagesqr =1 */
8151: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8152: break;
8153: case 2: /* ng=2 */
8154: if(nagesqr==0)
8155: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8156: else /* nagesqr =1 */
8157: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8158: break;
8159: case 3:
8160: if(nagesqr==0)
8161: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8162: else /* nagesqr =1 */
8163: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8164: break;
8165: }
8166: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8167: ijp=1; /* product no age */
8168: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8169: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8170: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8171: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8172: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8173: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8174: if(DummyV[j]==0){
8175: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8176: }else{ /* quantitative */
8177: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8178: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8179: }
8180: ij++;
1.237 brouard 8181: }
1.268 brouard 8182: }
8183: }else if(cptcovprod >0){
8184: if(j==Tprod[ijp]) { /* */
8185: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8186: if(ijp <=cptcovprod) { /* Product */
8187: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8188: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8189: /* 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)]); */
8190: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8191: }else{ /* Vn is dummy and Vm is quanti */
8192: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8193: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8194: }
8195: }else{ /* Vn*Vm Vn is quanti */
8196: if(DummyV[Tvard[ijp][2]]==0){
8197: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8198: }else{ /* Both quanti */
8199: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8200: }
1.237 brouard 8201: }
1.268 brouard 8202: ijp++;
1.237 brouard 8203: }
1.268 brouard 8204: } /* end Tprod */
1.237 brouard 8205: } else{ /* simple covariate */
1.264 brouard 8206: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8207: if(Dummy[j]==0){
8208: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8209: }else{ /* quantitative */
8210: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8211: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8212: }
1.237 brouard 8213: } /* end simple */
8214: } /* end j */
1.223 brouard 8215: }else{
8216: i=i-ncovmodel;
8217: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8218: fprintf(ficgp," (1.");
8219: }
1.227 brouard 8220:
1.223 brouard 8221: if(ng != 1){
8222: fprintf(ficgp,")/(1");
1.227 brouard 8223:
1.264 brouard 8224: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8225: if(nagesqr==0)
1.264 brouard 8226: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8227: else /* nagesqr =1 */
1.264 brouard 8228: 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 8229:
1.223 brouard 8230: ij=1;
8231: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8232: if(cptcovage >0){
8233: if((j-2)==Tage[ij]) { /* Bug valgrind */
8234: if(ij <=cptcovage) { /* Bug valgrind */
8235: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8236: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8237: ij++;
8238: }
8239: }
8240: }else
8241: 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 8242: }
8243: fprintf(ficgp,")");
8244: }
8245: fprintf(ficgp,")");
8246: if(ng ==2)
1.276 brouard 8247: 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 8248: else /* ng= 3 */
1.276 brouard 8249: 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 8250: }else{ /* end ng <> 1 */
8251: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8252: 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 8253: }
8254: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8255: fprintf(ficgp,",");
8256: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8257: fprintf(ficgp,",");
8258: i=i+ncovmodel;
8259: } /* end k */
8260: } /* end k2 */
1.276 brouard 8261: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8262: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8263: } /* end k1 */
1.223 brouard 8264: } /* end ng */
8265: /* avoid: */
8266: fflush(ficgp);
1.126 brouard 8267: } /* end gnuplot */
8268:
8269:
8270: /*************** Moving average **************/
1.219 brouard 8271: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8272: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8273:
1.222 brouard 8274: int i, cpt, cptcod;
8275: int modcovmax =1;
8276: int mobilavrange, mob;
8277: int iage=0;
1.288 brouard 8278: int firstA1=0, firstA2=0;
1.222 brouard 8279:
1.266 brouard 8280: double sum=0., sumr=0.;
1.222 brouard 8281: double age;
1.266 brouard 8282: double *sumnewp, *sumnewm, *sumnewmr;
8283: double *agemingood, *agemaxgood;
8284: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8285:
8286:
1.278 brouard 8287: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8288: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8289:
8290: sumnewp = vector(1,ncovcombmax);
8291: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8292: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8293: agemingood = vector(1,ncovcombmax);
1.266 brouard 8294: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8295: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8296: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8297:
8298: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8299: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8300: sumnewp[cptcod]=0.;
1.266 brouard 8301: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8302: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8303: }
8304: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8305:
1.266 brouard 8306: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8307: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8308: else mobilavrange=mobilav;
8309: for (age=bage; age<=fage; age++)
8310: for (i=1; i<=nlstate;i++)
8311: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8312: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8313: /* We keep the original values on the extreme ages bage, fage and for
8314: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8315: we use a 5 terms etc. until the borders are no more concerned.
8316: */
8317: for (mob=3;mob <=mobilavrange;mob=mob+2){
8318: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8319: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8320: sumnewm[cptcod]=0.;
8321: for (i=1; i<=nlstate;i++){
1.222 brouard 8322: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8323: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8324: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8325: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8326: }
8327: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8328: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8329: } /* end i */
8330: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8331: } /* end cptcod */
1.222 brouard 8332: }/* end age */
8333: }/* end mob */
1.266 brouard 8334: }else{
8335: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8336: return -1;
1.266 brouard 8337: }
8338:
8339: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8340: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8341: if(invalidvarcomb[cptcod]){
8342: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8343: continue;
8344: }
1.219 brouard 8345:
1.266 brouard 8346: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8347: sumnewm[cptcod]=0.;
8348: sumnewmr[cptcod]=0.;
8349: for (i=1; i<=nlstate;i++){
8350: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8351: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8352: }
8353: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8354: agemingoodr[cptcod]=age;
8355: }
8356: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8357: agemingood[cptcod]=age;
8358: }
8359: } /* age */
8360: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8361: sumnewm[cptcod]=0.;
1.266 brouard 8362: sumnewmr[cptcod]=0.;
1.222 brouard 8363: for (i=1; i<=nlstate;i++){
8364: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8365: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8366: }
8367: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8368: agemaxgoodr[cptcod]=age;
1.222 brouard 8369: }
8370: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8371: agemaxgood[cptcod]=age;
8372: }
8373: } /* age */
8374: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8375: /* but they will change */
1.288 brouard 8376: firstA1=0;firstA2=0;
1.266 brouard 8377: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8378: sumnewm[cptcod]=0.;
8379: sumnewmr[cptcod]=0.;
8380: for (i=1; i<=nlstate;i++){
8381: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8382: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8383: }
8384: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8385: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8386: agemaxgoodr[cptcod]=age; /* age min */
8387: for (i=1; i<=nlstate;i++)
8388: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8389: }else{ /* bad we change the value with the values of good ages */
8390: for (i=1; i<=nlstate;i++){
8391: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8392: } /* i */
8393: } /* end bad */
8394: }else{
8395: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8396: agemaxgood[cptcod]=age;
8397: }else{ /* bad we change the value with the values of good ages */
8398: for (i=1; i<=nlstate;i++){
8399: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8400: } /* i */
8401: } /* end bad */
8402: }/* end else */
8403: sum=0.;sumr=0.;
8404: for (i=1; i<=nlstate;i++){
8405: sum+=mobaverage[(int)age][i][cptcod];
8406: sumr+=probs[(int)age][i][cptcod];
8407: }
8408: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8409: if(!firstA1){
8410: firstA1=1;
8411: 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);
8412: }
8413: 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 8414: } /* end bad */
8415: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8416: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8417: if(!firstA2){
8418: firstA2=1;
8419: 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);
8420: }
8421: 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 8422: } /* end bad */
8423: }/* age */
1.266 brouard 8424:
8425: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8426: sumnewm[cptcod]=0.;
1.266 brouard 8427: sumnewmr[cptcod]=0.;
1.222 brouard 8428: for (i=1; i<=nlstate;i++){
8429: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8430: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8431: }
8432: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8433: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8434: agemingoodr[cptcod]=age;
8435: for (i=1; i<=nlstate;i++)
8436: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8437: }else{ /* bad we change the value with the values of good ages */
8438: for (i=1; i<=nlstate;i++){
8439: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8440: } /* i */
8441: } /* end bad */
8442: }else{
8443: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8444: agemingood[cptcod]=age;
8445: }else{ /* bad */
8446: for (i=1; i<=nlstate;i++){
8447: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8448: } /* i */
8449: } /* end bad */
8450: }/* end else */
8451: sum=0.;sumr=0.;
8452: for (i=1; i<=nlstate;i++){
8453: sum+=mobaverage[(int)age][i][cptcod];
8454: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8455: }
1.266 brouard 8456: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8457: 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 8458: } /* end bad */
8459: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8460: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8461: 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 8462: } /* end bad */
8463: }/* age */
1.266 brouard 8464:
1.222 brouard 8465:
8466: for (age=bage; age<=fage; age++){
1.235 brouard 8467: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8468: sumnewp[cptcod]=0.;
8469: sumnewm[cptcod]=0.;
8470: for (i=1; i<=nlstate;i++){
8471: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8472: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8473: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8474: }
8475: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8476: }
8477: /* printf("\n"); */
8478: /* } */
1.266 brouard 8479:
1.222 brouard 8480: /* brutal averaging */
1.266 brouard 8481: /* for (i=1; i<=nlstate;i++){ */
8482: /* for (age=1; age<=bage; age++){ */
8483: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8484: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8485: /* } */
8486: /* for (age=fage; age<=AGESUP; age++){ */
8487: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8488: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8489: /* } */
8490: /* } /\* end i status *\/ */
8491: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8492: /* for (age=1; age<=AGESUP; age++){ */
8493: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8494: /* mobaverage[(int)age][i][cptcod]=0.; */
8495: /* } */
8496: /* } */
1.222 brouard 8497: }/* end cptcod */
1.266 brouard 8498: free_vector(agemaxgoodr,1, ncovcombmax);
8499: free_vector(agemaxgood,1, ncovcombmax);
8500: free_vector(agemingood,1, ncovcombmax);
8501: free_vector(agemingoodr,1, ncovcombmax);
8502: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8503: free_vector(sumnewm,1, ncovcombmax);
8504: free_vector(sumnewp,1, ncovcombmax);
8505: return 0;
8506: }/* End movingaverage */
1.218 brouard 8507:
1.126 brouard 8508:
1.296 brouard 8509:
1.126 brouard 8510: /************** Forecasting ******************/
1.296 brouard 8511: /* 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)*/
8512: 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){
8513: /* dateintemean, mean date of interviews
8514: dateprojd, year, month, day of starting projection
8515: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8516: agemin, agemax range of age
8517: dateprev1 dateprev2 range of dates during which prevalence is computed
8518: */
1.296 brouard 8519: /* double anprojd, mprojd, jprojd; */
8520: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8521: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8522: double agec; /* generic age */
1.296 brouard 8523: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8524: double *popeffectif,*popcount;
8525: double ***p3mat;
1.218 brouard 8526: /* double ***mobaverage; */
1.126 brouard 8527: char fileresf[FILENAMELENGTH];
8528:
8529: agelim=AGESUP;
1.211 brouard 8530: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8531: in each health status at the date of interview (if between dateprev1 and dateprev2).
8532: We still use firstpass and lastpass as another selection.
8533: */
1.214 brouard 8534: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8535: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8536:
1.201 brouard 8537: strcpy(fileresf,"F_");
8538: strcat(fileresf,fileresu);
1.126 brouard 8539: if((ficresf=fopen(fileresf,"w"))==NULL) {
8540: printf("Problem with forecast resultfile: %s\n", fileresf);
8541: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8542: }
1.235 brouard 8543: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8544: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8545:
1.225 brouard 8546: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8547:
8548:
8549: stepsize=(int) (stepm+YEARM-1)/YEARM;
8550: if (stepm<=12) stepsize=1;
8551: if(estepm < stepm){
8552: printf ("Problem %d lower than %d\n",estepm, stepm);
8553: }
1.270 brouard 8554: else{
8555: hstepm=estepm;
8556: }
8557: if(estepm > stepm){ /* Yes every two year */
8558: stepsize=2;
8559: }
1.296 brouard 8560: hstepm=hstepm/stepm;
1.126 brouard 8561:
1.296 brouard 8562:
8563: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8564: /* fractional in yp1 *\/ */
8565: /* aintmean=yp; */
8566: /* yp2=modf((yp1*12),&yp); */
8567: /* mintmean=yp; */
8568: /* yp1=modf((yp2*30.5),&yp); */
8569: /* jintmean=yp; */
8570: /* if(jintmean==0) jintmean=1; */
8571: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8572:
1.296 brouard 8573:
8574: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8575: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8576: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8577: i1=pow(2,cptcoveff);
1.126 brouard 8578: if (cptcovn < 1){i1=1;}
8579:
1.296 brouard 8580: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8581:
8582: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8583:
1.126 brouard 8584: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8585: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8586: for(k=1; k<=i1;k++){
1.253 brouard 8587: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8588: continue;
1.227 brouard 8589: if(invalidvarcomb[k]){
8590: printf("\nCombination (%d) projection ignored because no cases \n",k);
8591: continue;
8592: }
8593: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8594: for(j=1;j<=cptcoveff;j++) {
8595: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8596: }
1.235 brouard 8597: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8598: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8599: }
1.227 brouard 8600: fprintf(ficresf," yearproj age");
8601: for(j=1; j<=nlstate+ndeath;j++){
8602: for(i=1; i<=nlstate;i++)
8603: fprintf(ficresf," p%d%d",i,j);
8604: fprintf(ficresf," wp.%d",j);
8605: }
1.296 brouard 8606: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8607: fprintf(ficresf,"\n");
1.296 brouard 8608: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8609: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8610: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8611: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8612: nhstepm = nhstepm/hstepm;
8613: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8614: oldm=oldms;savm=savms;
1.268 brouard 8615: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8616: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8617: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8618: for (h=0; h<=nhstepm; h++){
8619: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8620: break;
8621: }
8622: }
8623: fprintf(ficresf,"\n");
8624: for(j=1;j<=cptcoveff;j++)
8625: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8626: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8627:
8628: for(j=1; j<=nlstate+ndeath;j++) {
8629: ppij=0.;
8630: for(i=1; i<=nlstate;i++) {
1.278 brouard 8631: if (mobilav>=1)
8632: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8633: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8634: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8635: }
1.268 brouard 8636: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8637: } /* end i */
8638: fprintf(ficresf," %.3f", ppij);
8639: }/* end j */
1.227 brouard 8640: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8641: } /* end agec */
1.266 brouard 8642: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8643: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8644: } /* end yearp */
8645: } /* end k */
1.219 brouard 8646:
1.126 brouard 8647: fclose(ficresf);
1.215 brouard 8648: printf("End of Computing forecasting \n");
8649: fprintf(ficlog,"End of Computing forecasting\n");
8650:
1.126 brouard 8651: }
8652:
1.269 brouard 8653: /************** Back Forecasting ******************/
1.296 brouard 8654: /* 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){ */
8655: 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){
8656: /* back1, year, month, day of starting backprojection
1.267 brouard 8657: agemin, agemax range of age
8658: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8659: anback2 year of end of backprojection (same day and month as back1).
8660: prevacurrent and prev are prevalences.
1.267 brouard 8661: */
8662: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8663: double agec; /* generic age */
1.302 brouard 8664: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8665: double *popeffectif,*popcount;
8666: double ***p3mat;
8667: /* double ***mobaverage; */
8668: char fileresfb[FILENAMELENGTH];
8669:
1.268 brouard 8670: agelim=AGEINF;
1.267 brouard 8671: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8672: in each health status at the date of interview (if between dateprev1 and dateprev2).
8673: We still use firstpass and lastpass as another selection.
8674: */
8675: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8676: /* firstpass, lastpass, stepm, weightopt, model); */
8677:
8678: /*Do we need to compute prevalence again?*/
8679:
8680: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8681:
8682: strcpy(fileresfb,"FB_");
8683: strcat(fileresfb,fileresu);
8684: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8685: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8686: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8687: }
8688: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8689: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8690:
8691: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8692:
8693:
8694: stepsize=(int) (stepm+YEARM-1)/YEARM;
8695: if (stepm<=12) stepsize=1;
8696: if(estepm < stepm){
8697: printf ("Problem %d lower than %d\n",estepm, stepm);
8698: }
1.270 brouard 8699: else{
8700: hstepm=estepm;
8701: }
8702: if(estepm >= stepm){ /* Yes every two year */
8703: stepsize=2;
8704: }
1.267 brouard 8705:
8706: hstepm=hstepm/stepm;
1.296 brouard 8707: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8708: /* fractional in yp1 *\/ */
8709: /* aintmean=yp; */
8710: /* yp2=modf((yp1*12),&yp); */
8711: /* mintmean=yp; */
8712: /* yp1=modf((yp2*30.5),&yp); */
8713: /* jintmean=yp; */
8714: /* if(jintmean==0) jintmean=1; */
8715: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8716:
8717: i1=pow(2,cptcoveff);
8718: if (cptcovn < 1){i1=1;}
8719:
1.296 brouard 8720: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8721: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8722:
8723: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8724:
8725: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8726: for(k=1; k<=i1;k++){
8727: if(i1 != 1 && TKresult[nres]!= k)
8728: continue;
8729: if(invalidvarcomb[k]){
8730: printf("\nCombination (%d) projection ignored because no cases \n",k);
8731: continue;
8732: }
1.268 brouard 8733: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8734: for(j=1;j<=cptcoveff;j++) {
8735: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8736: }
8737: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8738: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8739: }
8740: fprintf(ficresfb," yearbproj age");
8741: for(j=1; j<=nlstate+ndeath;j++){
8742: for(i=1; i<=nlstate;i++)
1.268 brouard 8743: fprintf(ficresfb," b%d%d",i,j);
8744: fprintf(ficresfb," b.%d",j);
1.267 brouard 8745: }
1.296 brouard 8746: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8747: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8748: fprintf(ficresfb,"\n");
1.296 brouard 8749: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8750: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8751: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8752: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8753: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8754: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8755: nhstepm = nhstepm/hstepm;
8756: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8757: oldm=oldms;savm=savms;
1.268 brouard 8758: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8759: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8760: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8761: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8762: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8763: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8764: for (h=0; h<=nhstepm; h++){
1.268 brouard 8765: if (h*hstepm/YEARM*stepm ==-yearp) {
8766: break;
8767: }
8768: }
8769: fprintf(ficresfb,"\n");
8770: for(j=1;j<=cptcoveff;j++)
8771: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8772: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8773: for(i=1; i<=nlstate+ndeath;i++) {
8774: ppij=0.;ppi=0.;
8775: for(j=1; j<=nlstate;j++) {
8776: /* if (mobilav==1) */
1.269 brouard 8777: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8778: ppi=ppi+prevacurrent[(int)agec][j][k];
8779: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8780: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8781: /* else { */
8782: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8783: /* } */
1.268 brouard 8784: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8785: } /* end j */
8786: if(ppi <0.99){
8787: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8788: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8789: }
8790: fprintf(ficresfb," %.3f", ppij);
8791: }/* end j */
1.267 brouard 8792: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8793: } /* end agec */
8794: } /* end yearp */
8795: } /* end k */
1.217 brouard 8796:
1.267 brouard 8797: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8798:
1.267 brouard 8799: fclose(ficresfb);
8800: printf("End of Computing Back forecasting \n");
8801: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8802:
1.267 brouard 8803: }
1.217 brouard 8804:
1.269 brouard 8805: /* Variance of prevalence limit: varprlim */
8806: 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 8807: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8808:
8809: char fileresvpl[FILENAMELENGTH];
8810: FILE *ficresvpl;
8811: double **oldm, **savm;
8812: double **varpl; /* Variances of prevalence limits by age */
8813: int i1, k, nres, j ;
8814:
8815: strcpy(fileresvpl,"VPL_");
8816: strcat(fileresvpl,fileresu);
8817: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8818: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8819: exit(0);
8820: }
1.288 brouard 8821: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8822: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8823:
8824: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8825: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8826:
8827: i1=pow(2,cptcoveff);
8828: if (cptcovn < 1){i1=1;}
8829:
8830: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8831: for(k=1; k<=i1;k++){
8832: if(i1 != 1 && TKresult[nres]!= k)
8833: continue;
8834: fprintf(ficresvpl,"\n#****** ");
8835: printf("\n#****** ");
8836: fprintf(ficlog,"\n#****** ");
8837: for(j=1;j<=cptcoveff;j++) {
8838: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8839: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8840: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8841: }
8842: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8843: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8844: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8845: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8846: }
8847: fprintf(ficresvpl,"******\n");
8848: printf("******\n");
8849: fprintf(ficlog,"******\n");
8850:
8851: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8852: oldm=oldms;savm=savms;
8853: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8854: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8855: /*}*/
8856: }
8857:
8858: fclose(ficresvpl);
1.288 brouard 8859: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8860: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8861:
8862: }
8863: /* Variance of back prevalence: varbprlim */
8864: 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){
8865: /*------- Variance of back (stable) prevalence------*/
8866:
8867: char fileresvbl[FILENAMELENGTH];
8868: FILE *ficresvbl;
8869:
8870: double **oldm, **savm;
8871: double **varbpl; /* Variances of back prevalence limits by age */
8872: int i1, k, nres, j ;
8873:
8874: strcpy(fileresvbl,"VBL_");
8875: strcat(fileresvbl,fileresu);
8876: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8877: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8878: exit(0);
8879: }
8880: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8881: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8882:
8883:
8884: i1=pow(2,cptcoveff);
8885: if (cptcovn < 1){i1=1;}
8886:
8887: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8888: for(k=1; k<=i1;k++){
8889: if(i1 != 1 && TKresult[nres]!= k)
8890: continue;
8891: fprintf(ficresvbl,"\n#****** ");
8892: printf("\n#****** ");
8893: fprintf(ficlog,"\n#****** ");
8894: for(j=1;j<=cptcoveff;j++) {
8895: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8896: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8897: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8898: }
8899: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8900: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8901: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8902: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8903: }
8904: fprintf(ficresvbl,"******\n");
8905: printf("******\n");
8906: fprintf(ficlog,"******\n");
8907:
8908: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8909: oldm=oldms;savm=savms;
8910:
8911: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8912: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8913: /*}*/
8914: }
8915:
8916: fclose(ficresvbl);
8917: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8918: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8919:
8920: } /* End of varbprlim */
8921:
1.126 brouard 8922: /************** Forecasting *****not tested NB*************/
1.227 brouard 8923: /* 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 8924:
1.227 brouard 8925: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8926: /* int *popage; */
8927: /* double calagedatem, agelim, kk1, kk2; */
8928: /* double *popeffectif,*popcount; */
8929: /* double ***p3mat,***tabpop,***tabpopprev; */
8930: /* /\* double ***mobaverage; *\/ */
8931: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8932:
1.227 brouard 8933: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8934: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8935: /* agelim=AGESUP; */
8936: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8937:
1.227 brouard 8938: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8939:
8940:
1.227 brouard 8941: /* strcpy(filerespop,"POP_"); */
8942: /* strcat(filerespop,fileresu); */
8943: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8944: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8945: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8946: /* } */
8947: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8948: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8949:
1.227 brouard 8950: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8951:
1.227 brouard 8952: /* /\* if (mobilav!=0) { *\/ */
8953: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8954: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8955: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8956: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8957: /* /\* } *\/ */
8958: /* /\* } *\/ */
1.126 brouard 8959:
1.227 brouard 8960: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8961: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8962:
1.227 brouard 8963: /* agelim=AGESUP; */
1.126 brouard 8964:
1.227 brouard 8965: /* hstepm=1; */
8966: /* hstepm=hstepm/stepm; */
1.218 brouard 8967:
1.227 brouard 8968: /* if (popforecast==1) { */
8969: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8970: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8971: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8972: /* } */
8973: /* popage=ivector(0,AGESUP); */
8974: /* popeffectif=vector(0,AGESUP); */
8975: /* popcount=vector(0,AGESUP); */
1.126 brouard 8976:
1.227 brouard 8977: /* i=1; */
8978: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8979:
1.227 brouard 8980: /* imx=i; */
8981: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8982: /* } */
1.218 brouard 8983:
1.227 brouard 8984: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8985: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8986: /* k=k+1; */
8987: /* fprintf(ficrespop,"\n#******"); */
8988: /* for(j=1;j<=cptcoveff;j++) { */
8989: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8990: /* } */
8991: /* fprintf(ficrespop,"******\n"); */
8992: /* fprintf(ficrespop,"# Age"); */
8993: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8994: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8995:
1.227 brouard 8996: /* for (cpt=0; cpt<=0;cpt++) { */
8997: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8998:
1.227 brouard 8999: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9000: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9001: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9002:
1.227 brouard 9003: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9004: /* oldm=oldms;savm=savms; */
9005: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9006:
1.227 brouard 9007: /* for (h=0; h<=nhstepm; h++){ */
9008: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9009: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9010: /* } */
9011: /* for(j=1; j<=nlstate+ndeath;j++) { */
9012: /* kk1=0.;kk2=0; */
9013: /* for(i=1; i<=nlstate;i++) { */
9014: /* if (mobilav==1) */
9015: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9016: /* else { */
9017: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9018: /* } */
9019: /* } */
9020: /* if (h==(int)(calagedatem+12*cpt)){ */
9021: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9022: /* /\*fprintf(ficrespop," %.3f", kk1); */
9023: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9024: /* } */
9025: /* } */
9026: /* for(i=1; i<=nlstate;i++){ */
9027: /* kk1=0.; */
9028: /* for(j=1; j<=nlstate;j++){ */
9029: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9030: /* } */
9031: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9032: /* } */
1.218 brouard 9033:
1.227 brouard 9034: /* if (h==(int)(calagedatem+12*cpt)) */
9035: /* for(j=1; j<=nlstate;j++) */
9036: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9037: /* } */
9038: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9039: /* } */
9040: /* } */
1.218 brouard 9041:
1.227 brouard 9042: /* /\******\/ */
1.218 brouard 9043:
1.227 brouard 9044: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9045: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9046: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9047: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9048: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9049:
1.227 brouard 9050: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9051: /* oldm=oldms;savm=savms; */
9052: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9053: /* for (h=0; h<=nhstepm; h++){ */
9054: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9055: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9056: /* } */
9057: /* for(j=1; j<=nlstate+ndeath;j++) { */
9058: /* kk1=0.;kk2=0; */
9059: /* for(i=1; i<=nlstate;i++) { */
9060: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9061: /* } */
9062: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9063: /* } */
9064: /* } */
9065: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9066: /* } */
9067: /* } */
9068: /* } */
9069: /* } */
1.218 brouard 9070:
1.227 brouard 9071: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9072:
1.227 brouard 9073: /* if (popforecast==1) { */
9074: /* free_ivector(popage,0,AGESUP); */
9075: /* free_vector(popeffectif,0,AGESUP); */
9076: /* free_vector(popcount,0,AGESUP); */
9077: /* } */
9078: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9079: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9080: /* fclose(ficrespop); */
9081: /* } /\* End of popforecast *\/ */
1.218 brouard 9082:
1.126 brouard 9083: int fileappend(FILE *fichier, char *optionfich)
9084: {
9085: if((fichier=fopen(optionfich,"a"))==NULL) {
9086: printf("Problem with file: %s\n", optionfich);
9087: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9088: return (0);
9089: }
9090: fflush(fichier);
9091: return (1);
9092: }
9093:
9094:
9095: /**************** function prwizard **********************/
9096: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9097: {
9098:
9099: /* Wizard to print covariance matrix template */
9100:
1.164 brouard 9101: char ca[32], cb[32];
9102: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9103: int numlinepar;
9104:
9105: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9106: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9107: for(i=1; i <=nlstate; i++){
9108: jj=0;
9109: for(j=1; j <=nlstate+ndeath; j++){
9110: if(j==i) continue;
9111: jj++;
9112: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9113: printf("%1d%1d",i,j);
9114: fprintf(ficparo,"%1d%1d",i,j);
9115: for(k=1; k<=ncovmodel;k++){
9116: /* printf(" %lf",param[i][j][k]); */
9117: /* fprintf(ficparo," %lf",param[i][j][k]); */
9118: printf(" 0.");
9119: fprintf(ficparo," 0.");
9120: }
9121: printf("\n");
9122: fprintf(ficparo,"\n");
9123: }
9124: }
9125: printf("# Scales (for hessian or gradient estimation)\n");
9126: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9127: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9128: for(i=1; i <=nlstate; i++){
9129: jj=0;
9130: for(j=1; j <=nlstate+ndeath; j++){
9131: if(j==i) continue;
9132: jj++;
9133: fprintf(ficparo,"%1d%1d",i,j);
9134: printf("%1d%1d",i,j);
9135: fflush(stdout);
9136: for(k=1; k<=ncovmodel;k++){
9137: /* printf(" %le",delti3[i][j][k]); */
9138: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9139: printf(" 0.");
9140: fprintf(ficparo," 0.");
9141: }
9142: numlinepar++;
9143: printf("\n");
9144: fprintf(ficparo,"\n");
9145: }
9146: }
9147: printf("# Covariance matrix\n");
9148: /* # 121 Var(a12)\n\ */
9149: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9150: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9151: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9152: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9153: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9154: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9155: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9156: fflush(stdout);
9157: fprintf(ficparo,"# Covariance matrix\n");
9158: /* # 121 Var(a12)\n\ */
9159: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9160: /* # ...\n\ */
9161: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9162:
9163: for(itimes=1;itimes<=2;itimes++){
9164: jj=0;
9165: for(i=1; i <=nlstate; i++){
9166: for(j=1; j <=nlstate+ndeath; j++){
9167: if(j==i) continue;
9168: for(k=1; k<=ncovmodel;k++){
9169: jj++;
9170: ca[0]= k+'a'-1;ca[1]='\0';
9171: if(itimes==1){
9172: printf("#%1d%1d%d",i,j,k);
9173: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9174: }else{
9175: printf("%1d%1d%d",i,j,k);
9176: fprintf(ficparo,"%1d%1d%d",i,j,k);
9177: /* printf(" %.5le",matcov[i][j]); */
9178: }
9179: ll=0;
9180: for(li=1;li <=nlstate; li++){
9181: for(lj=1;lj <=nlstate+ndeath; lj++){
9182: if(lj==li) continue;
9183: for(lk=1;lk<=ncovmodel;lk++){
9184: ll++;
9185: if(ll<=jj){
9186: cb[0]= lk +'a'-1;cb[1]='\0';
9187: if(ll<jj){
9188: if(itimes==1){
9189: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9190: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9191: }else{
9192: printf(" 0.");
9193: fprintf(ficparo," 0.");
9194: }
9195: }else{
9196: if(itimes==1){
9197: printf(" Var(%s%1d%1d)",ca,i,j);
9198: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9199: }else{
9200: printf(" 0.");
9201: fprintf(ficparo," 0.");
9202: }
9203: }
9204: }
9205: } /* end lk */
9206: } /* end lj */
9207: } /* end li */
9208: printf("\n");
9209: fprintf(ficparo,"\n");
9210: numlinepar++;
9211: } /* end k*/
9212: } /*end j */
9213: } /* end i */
9214: } /* end itimes */
9215:
9216: } /* end of prwizard */
9217: /******************* Gompertz Likelihood ******************************/
9218: double gompertz(double x[])
9219: {
1.302 brouard 9220: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9221: int i,n=0; /* n is the size of the sample */
9222:
1.220 brouard 9223: for (i=1;i<=imx ; i++) {
1.126 brouard 9224: sump=sump+weight[i];
9225: /* sump=sump+1;*/
9226: num=num+1;
9227: }
1.302 brouard 9228: L=0.0;
9229: /* agegomp=AGEGOMP; */
1.126 brouard 9230: /* for (i=0; i<=imx; i++)
9231: 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]);*/
9232:
1.302 brouard 9233: for (i=1;i<=imx ; i++) {
9234: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9235: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9236: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9237: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9238: * +
9239: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9240: */
9241: if (wav[i] > 1 || agedc[i] < AGESUP) {
9242: if (cens[i] == 1){
9243: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9244: } else if (cens[i] == 0){
1.126 brouard 9245: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9246: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9247: } else
9248: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9249: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9250: L=L+A*weight[i];
1.126 brouard 9251: /* 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 9252: }
9253: }
1.126 brouard 9254:
1.302 brouard 9255: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9256:
9257: return -2*L*num/sump;
9258: }
9259:
1.136 brouard 9260: #ifdef GSL
9261: /******************* Gompertz_f Likelihood ******************************/
9262: double gompertz_f(const gsl_vector *v, void *params)
9263: {
1.302 brouard 9264: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9265: double *x= (double *) v->data;
9266: int i,n=0; /* n is the size of the sample */
9267:
9268: for (i=0;i<=imx-1 ; i++) {
9269: sump=sump+weight[i];
9270: /* sump=sump+1;*/
9271: num=num+1;
9272: }
9273:
9274:
9275: /* for (i=0; i<=imx; i++)
9276: 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]);*/
9277: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9278: for (i=1;i<=imx ; i++)
9279: {
9280: if (cens[i] == 1 && wav[i]>1)
9281: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9282:
9283: if (cens[i] == 0 && wav[i]>1)
9284: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9285: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9286:
9287: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9288: if (wav[i] > 1 ) { /* ??? */
9289: LL=LL+A*weight[i];
9290: /* 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]);*/
9291: }
9292: }
9293:
9294: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9295: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9296:
9297: return -2*LL*num/sump;
9298: }
9299: #endif
9300:
1.126 brouard 9301: /******************* Printing html file ***********/
1.201 brouard 9302: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9303: int lastpass, int stepm, int weightopt, char model[],\
9304: int imx, double p[],double **matcov,double agemortsup){
9305: int i,k;
9306:
9307: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9308: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9309: for (i=1;i<=2;i++)
9310: 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 9311: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9312: fprintf(fichtm,"</ul>");
9313:
9314: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9315:
9316: 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>");
9317:
9318: for (k=agegomp;k<(agemortsup-2);k++)
9319: 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]);
9320:
9321:
9322: fflush(fichtm);
9323: }
9324:
9325: /******************* Gnuplot file **************/
1.201 brouard 9326: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9327:
9328: char dirfileres[132],optfileres[132];
1.164 brouard 9329:
1.126 brouard 9330: int ng;
9331:
9332:
9333: /*#ifdef windows */
9334: fprintf(ficgp,"cd \"%s\" \n",pathc);
9335: /*#endif */
9336:
9337:
9338: strcpy(dirfileres,optionfilefiname);
9339: strcpy(optfileres,"vpl");
1.199 brouard 9340: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9341: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9342: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9343: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9344: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9345:
9346: }
9347:
1.136 brouard 9348: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9349: {
1.126 brouard 9350:
1.136 brouard 9351: /*-------- data file ----------*/
9352: FILE *fic;
9353: char dummy[]=" ";
1.240 brouard 9354: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9355: int lstra;
1.136 brouard 9356: int linei, month, year,iout;
1.302 brouard 9357: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9358: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9359: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9360: char *stratrunc;
1.223 brouard 9361:
1.240 brouard 9362: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9363: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9364:
1.240 brouard 9365: for(v=1; v <=ncovcol;v++){
9366: DummyV[v]=0;
9367: FixedV[v]=0;
9368: }
9369: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9370: DummyV[v]=1;
9371: FixedV[v]=0;
9372: }
9373: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9374: DummyV[v]=0;
9375: FixedV[v]=1;
9376: }
9377: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9378: DummyV[v]=1;
9379: FixedV[v]=1;
9380: }
9381: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9382: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9383: 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]);
9384: }
1.126 brouard 9385:
1.136 brouard 9386: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9387: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9388: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9389: }
1.126 brouard 9390:
1.302 brouard 9391: /* Is it a BOM UTF-8 Windows file? */
9392: /* First data line */
9393: linei=0;
9394: while(fgets(line, MAXLINE, fic)) {
9395: noffset=0;
9396: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9397: {
9398: noffset=noffset+3;
9399: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9400: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9401: fflush(ficlog); return 1;
9402: }
9403: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9404: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9405: {
9406: noffset=noffset+2;
1.304 brouard 9407: 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);
9408: 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 9409: fflush(ficlog); return 1;
9410: }
9411: else if( line[0] == 0 && line[1] == 0)
9412: {
9413: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9414: noffset=noffset+4;
1.304 brouard 9415: 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);
9416: 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 9417: fflush(ficlog); return 1;
9418: }
9419: } else{
9420: ;/*printf(" Not a BOM file\n");*/
9421: }
9422: /* If line starts with a # it is a comment */
9423: if (line[noffset] == '#') {
9424: linei=linei+1;
9425: break;
9426: }else{
9427: break;
9428: }
9429: }
9430: fclose(fic);
9431: if((fic=fopen(datafile,"r"))==NULL) {
9432: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9433: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9434: }
9435: /* Not a Bom file */
9436:
1.136 brouard 9437: i=1;
9438: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9439: linei=linei+1;
9440: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9441: if(line[j] == '\t')
9442: line[j] = ' ';
9443: }
9444: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9445: ;
9446: };
9447: line[j+1]=0; /* Trims blanks at end of line */
9448: if(line[0]=='#'){
9449: fprintf(ficlog,"Comment line\n%s\n",line);
9450: printf("Comment line\n%s\n",line);
9451: continue;
9452: }
9453: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9454: strcpy(line, linetmp);
1.223 brouard 9455:
9456: /* Loops on waves */
9457: for (j=maxwav;j>=1;j--){
9458: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9459: cutv(stra, strb, line, ' ');
9460: if(strb[0]=='.') { /* Missing value */
9461: lval=-1;
9462: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9463: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9464: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9465: 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);
9466: 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);
9467: return 1;
9468: }
9469: }else{
9470: errno=0;
9471: /* what_kind_of_number(strb); */
9472: dval=strtod(strb,&endptr);
9473: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9474: /* if(strb != endptr && *endptr == '\0') */
9475: /* dval=dlval; */
9476: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9477: if( strb[0]=='\0' || (*endptr != '\0')){
9478: 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);
9479: 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);
9480: return 1;
9481: }
9482: cotqvar[j][iv][i]=dval;
9483: cotvar[j][ntv+iv][i]=dval;
9484: }
9485: strcpy(line,stra);
1.223 brouard 9486: }/* end loop ntqv */
1.225 brouard 9487:
1.223 brouard 9488: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9489: cutv(stra, strb, line, ' ');
9490: if(strb[0]=='.') { /* Missing value */
9491: lval=-1;
9492: }else{
9493: errno=0;
9494: lval=strtol(strb,&endptr,10);
9495: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9496: if( strb[0]=='\0' || (*endptr != '\0')){
9497: 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);
9498: 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);
9499: return 1;
9500: }
9501: }
9502: if(lval <-1 || lval >1){
9503: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9504: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9505: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9506: For example, for multinomial values like 1, 2 and 3,\n \
9507: build V1=0 V2=0 for the reference value (1),\n \
9508: V1=1 V2=0 for (2) \n \
1.223 brouard 9509: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9510: output of IMaCh is often meaningless.\n \
1.223 brouard 9511: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9512: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9513: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9514: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9515: For example, for multinomial values like 1, 2 and 3,\n \
9516: build V1=0 V2=0 for the reference value (1),\n \
9517: V1=1 V2=0 for (2) \n \
1.223 brouard 9518: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9519: output of IMaCh is often meaningless.\n \
1.223 brouard 9520: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9521: return 1;
9522: }
9523: cotvar[j][iv][i]=(double)(lval);
9524: strcpy(line,stra);
1.223 brouard 9525: }/* end loop ntv */
1.225 brouard 9526:
1.223 brouard 9527: /* Statuses at wave */
1.137 brouard 9528: cutv(stra, strb, line, ' ');
1.223 brouard 9529: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9530: lval=-1;
1.136 brouard 9531: }else{
1.238 brouard 9532: errno=0;
9533: lval=strtol(strb,&endptr,10);
9534: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9535: if( strb[0]=='\0' || (*endptr != '\0')){
9536: 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);
9537: 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);
9538: return 1;
9539: }
1.136 brouard 9540: }
1.225 brouard 9541:
1.136 brouard 9542: s[j][i]=lval;
1.225 brouard 9543:
1.223 brouard 9544: /* Date of Interview */
1.136 brouard 9545: strcpy(line,stra);
9546: cutv(stra, strb,line,' ');
1.169 brouard 9547: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9548: }
1.169 brouard 9549: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9550: month=99;
9551: year=9999;
1.136 brouard 9552: }else{
1.225 brouard 9553: 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);
9554: 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);
9555: return 1;
1.136 brouard 9556: }
9557: anint[j][i]= (double) year;
1.302 brouard 9558: mint[j][i]= (double)month;
9559: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9560: /* 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]); */
9561: /* 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]); */
9562: /* } */
1.136 brouard 9563: strcpy(line,stra);
1.223 brouard 9564: } /* End loop on waves */
1.225 brouard 9565:
1.223 brouard 9566: /* Date of death */
1.136 brouard 9567: cutv(stra, strb,line,' ');
1.169 brouard 9568: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9569: }
1.169 brouard 9570: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9571: month=99;
9572: year=9999;
9573: }else{
1.141 brouard 9574: 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 9575: 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);
9576: return 1;
1.136 brouard 9577: }
9578: andc[i]=(double) year;
9579: moisdc[i]=(double) month;
9580: strcpy(line,stra);
9581:
1.223 brouard 9582: /* Date of birth */
1.136 brouard 9583: cutv(stra, strb,line,' ');
1.169 brouard 9584: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9585: }
1.169 brouard 9586: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9587: month=99;
9588: year=9999;
9589: }else{
1.141 brouard 9590: 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);
9591: 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 9592: return 1;
1.136 brouard 9593: }
9594: if (year==9999) {
1.141 brouard 9595: 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);
9596: 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 9597: return 1;
9598:
1.136 brouard 9599: }
9600: annais[i]=(double)(year);
1.302 brouard 9601: moisnais[i]=(double)(month);
9602: for (j=1;j<=maxwav;j++){
9603: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9604: 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]);
9605: 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]);
9606: }
9607: }
9608:
1.136 brouard 9609: strcpy(line,stra);
1.225 brouard 9610:
1.223 brouard 9611: /* Sample weight */
1.136 brouard 9612: cutv(stra, strb,line,' ');
9613: errno=0;
9614: dval=strtod(strb,&endptr);
9615: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9616: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9617: 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 9618: fflush(ficlog);
9619: return 1;
9620: }
9621: weight[i]=dval;
9622: strcpy(line,stra);
1.225 brouard 9623:
1.223 brouard 9624: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9625: cutv(stra, strb, line, ' ');
9626: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9627: lval=-1;
1.311 brouard 9628: coqvar[iv][i]=NAN;
9629: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9630: }else{
1.225 brouard 9631: errno=0;
9632: /* what_kind_of_number(strb); */
9633: dval=strtod(strb,&endptr);
9634: /* if(strb != endptr && *endptr == '\0') */
9635: /* dval=dlval; */
9636: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9637: if( strb[0]=='\0' || (*endptr != '\0')){
9638: 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);
9639: 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);
9640: return 1;
9641: }
9642: coqvar[iv][i]=dval;
1.226 brouard 9643: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9644: }
9645: strcpy(line,stra);
9646: }/* end loop nqv */
1.136 brouard 9647:
1.223 brouard 9648: /* Covariate values */
1.136 brouard 9649: for (j=ncovcol;j>=1;j--){
9650: cutv(stra, strb,line,' ');
1.223 brouard 9651: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9652: lval=-1;
1.136 brouard 9653: }else{
1.225 brouard 9654: errno=0;
9655: lval=strtol(strb,&endptr,10);
9656: if( strb[0]=='\0' || (*endptr != '\0')){
9657: 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);
9658: 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);
9659: return 1;
9660: }
1.136 brouard 9661: }
9662: if(lval <-1 || lval >1){
1.225 brouard 9663: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9664: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9665: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9666: For example, for multinomial values like 1, 2 and 3,\n \
9667: build V1=0 V2=0 for the reference value (1),\n \
9668: V1=1 V2=0 for (2) \n \
1.136 brouard 9669: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9670: output of IMaCh is often meaningless.\n \
1.136 brouard 9671: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9672: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9673: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9674: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9675: For example, for multinomial values like 1, 2 and 3,\n \
9676: build V1=0 V2=0 for the reference value (1),\n \
9677: V1=1 V2=0 for (2) \n \
1.136 brouard 9678: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9679: output of IMaCh is often meaningless.\n \
1.136 brouard 9680: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9681: return 1;
1.136 brouard 9682: }
9683: covar[j][i]=(double)(lval);
9684: strcpy(line,stra);
9685: }
9686: lstra=strlen(stra);
1.225 brouard 9687:
1.136 brouard 9688: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9689: stratrunc = &(stra[lstra-9]);
9690: num[i]=atol(stratrunc);
9691: }
9692: else
9693: num[i]=atol(stra);
9694: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9695: 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;}*/
9696:
9697: i=i+1;
9698: } /* End loop reading data */
1.225 brouard 9699:
1.136 brouard 9700: *imax=i-1; /* Number of individuals */
9701: fclose(fic);
1.225 brouard 9702:
1.136 brouard 9703: return (0);
1.164 brouard 9704: /* endread: */
1.225 brouard 9705: printf("Exiting readdata: ");
9706: fclose(fic);
9707: return (1);
1.223 brouard 9708: }
1.126 brouard 9709:
1.234 brouard 9710: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9711: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9712: while (*p2 == ' ')
1.234 brouard 9713: p2++;
9714: /* while ((*p1++ = *p2++) !=0) */
9715: /* ; */
9716: /* do */
9717: /* while (*p2 == ' ') */
9718: /* p2++; */
9719: /* while (*p1++ == *p2++); */
9720: *stri=p2;
1.145 brouard 9721: }
9722:
1.235 brouard 9723: int decoderesult ( char resultline[], int nres)
1.230 brouard 9724: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9725: {
1.235 brouard 9726: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9727: char resultsav[MAXLINE];
1.234 brouard 9728: int resultmodel[MAXLINE];
9729: int modelresult[MAXLINE];
1.230 brouard 9730: char stra[80], strb[80], strc[80], strd[80],stre[80];
9731:
1.234 brouard 9732: removefirstspace(&resultline);
1.230 brouard 9733:
9734: if (strstr(resultline,"v") !=0){
9735: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9736: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9737: return 1;
9738: }
9739: trimbb(resultsav, resultline);
9740: if (strlen(resultsav) >1){
9741: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9742: }
1.253 brouard 9743: if(j == 0){ /* Resultline but no = */
9744: TKresult[nres]=0; /* Combination for the nresult and the model */
9745: return (0);
9746: }
1.234 brouard 9747: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.310 brouard 9748: 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);
9749: 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 9750: }
9751: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9752: if(nbocc(resultsav,'=') >1){
9753: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
1.310 brouard 9754: resultsav= V4=1 V5=25.1 V3=0 stra= V5=25.1 V3=0 strb= V4=1 */
1.234 brouard 9755: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9756: }else
9757: cutl(strc,strd,resultsav,'=');
1.230 brouard 9758: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9759:
1.230 brouard 9760: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9761: Tvarsel[k]=atoi(strc);
9762: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9763: /* cptcovsel++; */
9764: if (nbocc(stra,'=') >0)
9765: strcpy(resultsav,stra); /* and analyzes it */
9766: }
1.235 brouard 9767: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9768: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9769: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9770: match=0;
1.236 brouard 9771: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9772: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9773: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9774: match=1;
9775: break;
9776: }
9777: }
9778: if(match == 0){
1.310 brouard 9779: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9780: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9781: return 1;
1.234 brouard 9782: }
9783: }
9784: }
1.235 brouard 9785: /* Checking for missing or useless values in comparison of current model needs */
9786: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9787: match=0;
1.235 brouard 9788: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9789: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9790: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9791: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9792: ++match;
9793: }
9794: }
9795: }
9796: if(match == 0){
9797: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 9798: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9799: return 1;
1.234 brouard 9800: }else if(match > 1){
9801: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 9802: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9803: return 1;
1.234 brouard 9804: }
9805: }
1.235 brouard 9806:
1.234 brouard 9807: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9808: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9809: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9810: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9811: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9812: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9813: /* 1 0 0 0 */
9814: /* 2 1 0 0 */
9815: /* 3 0 1 0 */
9816: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9817: /* 5 0 0 1 */
9818: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9819: /* 7 0 1 1 */
9820: /* 8 1 1 1 */
1.237 brouard 9821: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9822: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9823: /* V5*age V5 known which value for nres? */
9824: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9825: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9826: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9827: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9828: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9829: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9830: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9831: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9832: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9833: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9834: k4++;;
9835: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9836: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9837: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9838: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9839: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9840: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9841: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9842: k4q++;;
9843: }
9844: }
1.234 brouard 9845:
1.235 brouard 9846: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9847: return (0);
9848: }
1.235 brouard 9849:
1.230 brouard 9850: int decodemodel( char model[], int lastobs)
9851: /**< This routine decodes the model and returns:
1.224 brouard 9852: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9853: * - nagesqr = 1 if age*age in the model, otherwise 0.
9854: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9855: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9856: * - cptcovage number of covariates with age*products =2
9857: * - cptcovs number of simple covariates
9858: * - 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
9859: * which is a new column after the 9 (ncovcol) variables.
9860: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9861: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9862: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9863: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9864: */
1.136 brouard 9865: {
1.238 brouard 9866: int i, j, k, ks, v;
1.227 brouard 9867: int j1, k1, k2, k3, k4;
1.136 brouard 9868: char modelsav[80];
1.145 brouard 9869: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9870: char *strpt;
1.136 brouard 9871:
1.145 brouard 9872: /*removespace(model);*/
1.136 brouard 9873: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9874: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9875: if (strstr(model,"AGE") !=0){
1.192 brouard 9876: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9877: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9878: return 1;
9879: }
1.141 brouard 9880: if (strstr(model,"v") !=0){
9881: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9882: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9883: return 1;
9884: }
1.187 brouard 9885: strcpy(modelsav,model);
9886: if ((strpt=strstr(model,"age*age")) !=0){
9887: printf(" strpt=%s, model=%s\n",strpt, model);
9888: if(strpt != model){
1.234 brouard 9889: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9890: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9891: corresponding column of parameters.\n",model);
1.234 brouard 9892: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9893: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9894: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9895: return 1;
1.225 brouard 9896: }
1.187 brouard 9897: nagesqr=1;
9898: if (strstr(model,"+age*age") !=0)
1.234 brouard 9899: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9900: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9901: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9902: else
1.234 brouard 9903: substrchaine(modelsav, model, "age*age");
1.187 brouard 9904: }else
9905: nagesqr=0;
9906: if (strlen(modelsav) >1){
9907: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9908: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9909: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9910: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9911: * cst, age and age*age
9912: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9913: /* including age products which are counted in cptcovage.
9914: * but the covariates which are products must be treated
9915: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9916: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9917: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9918:
9919:
1.187 brouard 9920: /* Design
9921: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9922: * < ncovcol=8 >
9923: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9924: * k= 1 2 3 4 5 6 7 8
9925: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9926: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9927: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9928: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9929: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9930: * Tage[++cptcovage]=k
9931: * if products, new covar are created after ncovcol with k1
9932: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9933: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9934: * 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
9935: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9936: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9937: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9938: * < ncovcol=8 >
9939: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9940: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9941: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9942: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9943: * p Tprod[1]@2={ 6, 5}
9944: *p Tvard[1][1]@4= {7, 8, 5, 6}
9945: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9946: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9947: *How to reorganize?
9948: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9949: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9950: * {2, 1, 4, 8, 5, 6, 3, 7}
9951: * Struct []
9952: */
1.225 brouard 9953:
1.187 brouard 9954: /* This loop fills the array Tvar from the string 'model'.*/
9955: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9956: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9957: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9958: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9959: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9960: /* k=1 Tvar[1]=2 (from V2) */
9961: /* k=5 Tvar[5] */
9962: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9963: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9964: /* } */
1.198 brouard 9965: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9966: /*
9967: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9968: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9969: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9970: }
1.187 brouard 9971: cptcovage=0;
9972: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9973: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9974: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9975: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9976: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9977: /*scanf("%d",i);*/
9978: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9979: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9980: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9981: /* covar is not filled and then is empty */
9982: cptcovprod--;
9983: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9984: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9985: Typevar[k]=1; /* 1 for age product */
9986: cptcovage++; /* Sums the number of covariates which include age as a product */
9987: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9988: /*printf("stre=%s ", stre);*/
9989: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9990: cptcovprod--;
9991: cutl(stre,strb,strc,'V');
9992: Tvar[k]=atoi(stre);
9993: Typevar[k]=1; /* 1 for age product */
9994: cptcovage++;
9995: Tage[cptcovage]=k;
9996: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9997: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9998: cptcovn++;
9999: cptcovprodnoage++;k1++;
10000: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10001: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10002: because this model-covariate is a construction we invent a new column
10003: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
10004: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
10005: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
10006: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10007: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10008: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
10009: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
10010: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
10011: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
10012: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10013: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10014: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10015: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10016: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10017: for (i=1; i<=lastobs;i++){
10018: /* Computes the new covariate which is a product of
10019: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10020: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10021: }
10022: } /* End age is not in the model */
10023: } /* End if model includes a product */
10024: else { /* no more sum */
10025: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10026: /* scanf("%d",i);*/
10027: cutl(strd,strc,strb,'V');
10028: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10029: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10030: Tvar[k]=atoi(strd);
10031: Typevar[k]=0; /* 0 for simple covariates */
10032: }
10033: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10034: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10035: scanf("%d",i);*/
1.187 brouard 10036: } /* end of loop + on total covariates */
10037: } /* end if strlen(modelsave == 0) age*age might exist */
10038: } /* end if strlen(model == 0) */
1.136 brouard 10039:
10040: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10041: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10042:
1.136 brouard 10043: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10044: printf("cptcovprod=%d ", cptcovprod);
10045: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10046: scanf("%d ",i);*/
10047:
10048:
1.230 brouard 10049: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10050: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10051: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10052: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10053: k = 1 2 3 4 5 6 7 8 9
10054: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
10055: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 10056: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10057: Dummy[k] 1 0 0 0 3 1 1 2 3
10058: Tmodelind[combination of covar]=k;
1.225 brouard 10059: */
10060: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10061: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10062: /* 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 10063: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 10064: printf("Model=%s\n\
10065: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10066: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10067: 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);
10068: fprintf(ficlog,"Model=%s\n\
10069: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10070: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10071: 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 10072: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10073: 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 */
10074: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10075: Fixed[k]= 0;
10076: Dummy[k]= 0;
1.225 brouard 10077: ncoveff++;
1.232 brouard 10078: ncovf++;
1.234 brouard 10079: nsd++;
10080: modell[k].maintype= FTYPE;
10081: TvarsD[nsd]=Tvar[k];
10082: TvarsDind[nsd]=k;
10083: TvarF[ncovf]=Tvar[k];
10084: TvarFind[ncovf]=k;
10085: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10086: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10087: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10088: Fixed[k]= 0;
10089: Dummy[k]= 0;
10090: ncoveff++;
10091: ncovf++;
10092: modell[k].maintype= FTYPE;
10093: TvarF[ncovf]=Tvar[k];
10094: TvarFind[ncovf]=k;
1.230 brouard 10095: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10096: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10097: }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 10098: Fixed[k]= 0;
10099: Dummy[k]= 1;
1.230 brouard 10100: nqfveff++;
1.234 brouard 10101: modell[k].maintype= FTYPE;
10102: modell[k].subtype= FQ;
10103: nsq++;
10104: TvarsQ[nsq]=Tvar[k];
10105: TvarsQind[nsq]=k;
1.232 brouard 10106: ncovf++;
1.234 brouard 10107: TvarF[ncovf]=Tvar[k];
10108: TvarFind[ncovf]=k;
1.231 brouard 10109: 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 10110: 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 10111: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10112: Fixed[k]= 1;
10113: Dummy[k]= 0;
1.225 brouard 10114: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10115: modell[k].maintype= VTYPE;
10116: modell[k].subtype= VD;
10117: nsd++;
10118: TvarsD[nsd]=Tvar[k];
10119: TvarsDind[nsd]=k;
10120: ncovv++; /* Only simple time varying variables */
10121: TvarV[ncovv]=Tvar[k];
1.242 brouard 10122: 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 10123: 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 */
10124: 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 10125: 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);
10126: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10127: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10128: Fixed[k]= 1;
10129: Dummy[k]= 1;
10130: nqtveff++;
10131: modell[k].maintype= VTYPE;
10132: modell[k].subtype= VQ;
10133: ncovv++; /* Only simple time varying variables */
10134: nsq++;
10135: TvarsQ[nsq]=Tvar[k];
10136: TvarsQind[nsq]=k;
10137: TvarV[ncovv]=Tvar[k];
1.242 brouard 10138: 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 10139: 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 */
10140: 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 10141: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10142: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10143: 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 10144: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10145: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10146: ncova++;
10147: TvarA[ncova]=Tvar[k];
10148: TvarAind[ncova]=k;
1.231 brouard 10149: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10150: Fixed[k]= 2;
10151: Dummy[k]= 2;
10152: modell[k].maintype= ATYPE;
10153: modell[k].subtype= APFD;
10154: /* ncoveff++; */
1.227 brouard 10155: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10156: Fixed[k]= 2;
10157: Dummy[k]= 3;
10158: modell[k].maintype= ATYPE;
10159: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10160: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10161: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10162: Fixed[k]= 3;
10163: Dummy[k]= 2;
10164: modell[k].maintype= ATYPE;
10165: modell[k].subtype= APVD; /* Product age * varying dummy */
10166: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10167: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10168: Fixed[k]= 3;
10169: Dummy[k]= 3;
10170: modell[k].maintype= ATYPE;
10171: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10172: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10173: }
10174: }else if (Typevar[k] == 2) { /* product without age */
10175: k1=Tposprod[k];
10176: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10177: if(Tvard[k1][2] <=ncovcol){
10178: Fixed[k]= 1;
10179: Dummy[k]= 0;
10180: modell[k].maintype= FTYPE;
10181: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10182: ncovf++; /* Fixed variables without age */
10183: TvarF[ncovf]=Tvar[k];
10184: TvarFind[ncovf]=k;
10185: }else if(Tvard[k1][2] <=ncovcol+nqv){
10186: Fixed[k]= 0; /* or 2 ?*/
10187: Dummy[k]= 1;
10188: modell[k].maintype= FTYPE;
10189: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10190: ncovf++; /* Varying variables without age */
10191: TvarF[ncovf]=Tvar[k];
10192: TvarFind[ncovf]=k;
10193: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10194: Fixed[k]= 1;
10195: Dummy[k]= 0;
10196: modell[k].maintype= VTYPE;
10197: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10198: ncovv++; /* Varying variables without age */
10199: TvarV[ncovv]=Tvar[k];
10200: TvarVind[ncovv]=k;
10201: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10202: Fixed[k]= 1;
10203: Dummy[k]= 1;
10204: modell[k].maintype= VTYPE;
10205: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10206: ncovv++; /* Varying variables without age */
10207: TvarV[ncovv]=Tvar[k];
10208: TvarVind[ncovv]=k;
10209: }
1.227 brouard 10210: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10211: if(Tvard[k1][2] <=ncovcol){
10212: Fixed[k]= 0; /* or 2 ?*/
10213: Dummy[k]= 1;
10214: modell[k].maintype= FTYPE;
10215: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10216: ncovf++; /* Fixed variables without age */
10217: TvarF[ncovf]=Tvar[k];
10218: TvarFind[ncovf]=k;
10219: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10220: Fixed[k]= 1;
10221: Dummy[k]= 1;
10222: modell[k].maintype= VTYPE;
10223: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10224: ncovv++; /* Varying variables without age */
10225: TvarV[ncovv]=Tvar[k];
10226: TvarVind[ncovv]=k;
10227: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10228: Fixed[k]= 1;
10229: Dummy[k]= 1;
10230: modell[k].maintype= VTYPE;
10231: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10232: ncovv++; /* Varying variables without age */
10233: TvarV[ncovv]=Tvar[k];
10234: TvarVind[ncovv]=k;
10235: ncovv++; /* Varying variables without age */
10236: TvarV[ncovv]=Tvar[k];
10237: TvarVind[ncovv]=k;
10238: }
1.227 brouard 10239: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10240: if(Tvard[k1][2] <=ncovcol){
10241: Fixed[k]= 1;
10242: Dummy[k]= 1;
10243: modell[k].maintype= VTYPE;
10244: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10245: ncovv++; /* Varying variables without age */
10246: TvarV[ncovv]=Tvar[k];
10247: TvarVind[ncovv]=k;
10248: }else if(Tvard[k1][2] <=ncovcol+nqv){
10249: Fixed[k]= 1;
10250: Dummy[k]= 1;
10251: modell[k].maintype= VTYPE;
10252: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10253: ncovv++; /* Varying variables without age */
10254: TvarV[ncovv]=Tvar[k];
10255: TvarVind[ncovv]=k;
10256: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10257: Fixed[k]= 1;
10258: Dummy[k]= 0;
10259: modell[k].maintype= VTYPE;
10260: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10261: ncovv++; /* Varying variables without age */
10262: TvarV[ncovv]=Tvar[k];
10263: TvarVind[ncovv]=k;
10264: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10265: Fixed[k]= 1;
10266: Dummy[k]= 1;
10267: modell[k].maintype= VTYPE;
10268: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10269: ncovv++; /* Varying variables without age */
10270: TvarV[ncovv]=Tvar[k];
10271: TvarVind[ncovv]=k;
10272: }
1.227 brouard 10273: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10274: if(Tvard[k1][2] <=ncovcol){
10275: Fixed[k]= 1;
10276: Dummy[k]= 1;
10277: modell[k].maintype= VTYPE;
10278: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10279: ncovv++; /* Varying variables without age */
10280: TvarV[ncovv]=Tvar[k];
10281: TvarVind[ncovv]=k;
10282: }else if(Tvard[k1][2] <=ncovcol+nqv){
10283: Fixed[k]= 1;
10284: Dummy[k]= 1;
10285: modell[k].maintype= VTYPE;
10286: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10287: ncovv++; /* Varying variables without age */
10288: TvarV[ncovv]=Tvar[k];
10289: TvarVind[ncovv]=k;
10290: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10291: Fixed[k]= 1;
10292: Dummy[k]= 1;
10293: modell[k].maintype= VTYPE;
10294: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10295: ncovv++; /* Varying variables without age */
10296: TvarV[ncovv]=Tvar[k];
10297: TvarVind[ncovv]=k;
10298: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10299: Fixed[k]= 1;
10300: Dummy[k]= 1;
10301: modell[k].maintype= VTYPE;
10302: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10303: ncovv++; /* Varying variables without age */
10304: TvarV[ncovv]=Tvar[k];
10305: TvarVind[ncovv]=k;
10306: }
1.227 brouard 10307: }else{
1.240 brouard 10308: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10309: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10310: } /*end k1*/
1.225 brouard 10311: }else{
1.226 brouard 10312: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10313: 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 10314: }
1.227 brouard 10315: 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 10316: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10317: 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]);
10318: }
10319: /* Searching for doublons in the model */
10320: for(k1=1; k1<= cptcovt;k1++){
10321: for(k2=1; k2 <k1;k2++){
1.285 brouard 10322: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10323: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10324: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10325: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10326: 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]);
10327: 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 10328: return(1);
10329: }
10330: }else if (Typevar[k1] ==2){
10331: k3=Tposprod[k1];
10332: k4=Tposprod[k2];
10333: 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])) ){
10334: 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]]);
10335: 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);
10336: return(1);
10337: }
10338: }
1.227 brouard 10339: }
10340: }
1.225 brouard 10341: }
10342: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10343: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10344: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10345: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10346: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10347: /*endread:*/
1.225 brouard 10348: printf("Exiting decodemodel: ");
10349: return (1);
1.136 brouard 10350: }
10351:
1.169 brouard 10352: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10353: {/* Check ages at death */
1.136 brouard 10354: int i, m;
1.218 brouard 10355: int firstone=0;
10356:
1.136 brouard 10357: for (i=1; i<=imx; i++) {
10358: for(m=2; (m<= maxwav); m++) {
10359: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10360: anint[m][i]=9999;
1.216 brouard 10361: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10362: s[m][i]=-1;
1.136 brouard 10363: }
10364: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10365: *nberr = *nberr + 1;
1.218 brouard 10366: if(firstone == 0){
10367: firstone=1;
1.260 brouard 10368: 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 10369: }
1.262 brouard 10370: 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 10371: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10372: }
10373: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10374: (*nberr)++;
1.259 brouard 10375: 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 10376: 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 10377: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10378: }
10379: }
10380: }
10381:
10382: for (i=1; i<=imx; i++) {
10383: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10384: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10385: 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 10386: if (s[m][i] >= nlstate+1) {
1.169 brouard 10387: if(agedc[i]>0){
10388: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10389: agev[m][i]=agedc[i];
1.214 brouard 10390: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10391: }else {
1.136 brouard 10392: if ((int)andc[i]!=9999){
10393: nbwarn++;
10394: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10395: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10396: agev[m][i]=-1;
10397: }
10398: }
1.169 brouard 10399: } /* agedc > 0 */
1.214 brouard 10400: } /* end if */
1.136 brouard 10401: else if(s[m][i] !=9){ /* Standard case, age in fractional
10402: years but with the precision of a month */
10403: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10404: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10405: agev[m][i]=1;
10406: else if(agev[m][i] < *agemin){
10407: *agemin=agev[m][i];
10408: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10409: }
10410: else if(agev[m][i] >*agemax){
10411: *agemax=agev[m][i];
1.156 brouard 10412: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10413: }
10414: /*agev[m][i]=anint[m][i]-annais[i];*/
10415: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10416: } /* en if 9*/
1.136 brouard 10417: else { /* =9 */
1.214 brouard 10418: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10419: agev[m][i]=1;
10420: s[m][i]=-1;
10421: }
10422: }
1.214 brouard 10423: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10424: agev[m][i]=1;
1.214 brouard 10425: else{
10426: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10427: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10428: agev[m][i]=0;
10429: }
10430: } /* End for lastpass */
10431: }
1.136 brouard 10432:
10433: for (i=1; i<=imx; i++) {
10434: for(m=firstpass; (m<=lastpass); m++){
10435: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10436: (*nberr)++;
1.136 brouard 10437: 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);
10438: 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);
10439: return 1;
10440: }
10441: }
10442: }
10443:
10444: /*for (i=1; i<=imx; i++){
10445: for (m=firstpass; (m<lastpass); m++){
10446: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10447: }
10448:
10449: }*/
10450:
10451:
1.139 brouard 10452: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10453: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10454:
10455: return (0);
1.164 brouard 10456: /* endread:*/
1.136 brouard 10457: printf("Exiting calandcheckages: ");
10458: return (1);
10459: }
10460:
1.172 brouard 10461: #if defined(_MSC_VER)
10462: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10463: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10464: //#include "stdafx.h"
10465: //#include <stdio.h>
10466: //#include <tchar.h>
10467: //#include <windows.h>
10468: //#include <iostream>
10469: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10470:
10471: LPFN_ISWOW64PROCESS fnIsWow64Process;
10472:
10473: BOOL IsWow64()
10474: {
10475: BOOL bIsWow64 = FALSE;
10476:
10477: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10478: // (HANDLE, PBOOL);
10479:
10480: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10481:
10482: HMODULE module = GetModuleHandle(_T("kernel32"));
10483: const char funcName[] = "IsWow64Process";
10484: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10485: GetProcAddress(module, funcName);
10486:
10487: if (NULL != fnIsWow64Process)
10488: {
10489: if (!fnIsWow64Process(GetCurrentProcess(),
10490: &bIsWow64))
10491: //throw std::exception("Unknown error");
10492: printf("Unknown error\n");
10493: }
10494: return bIsWow64 != FALSE;
10495: }
10496: #endif
1.177 brouard 10497:
1.191 brouard 10498: void syscompilerinfo(int logged)
1.292 brouard 10499: {
10500: #include <stdint.h>
10501:
10502: /* #include "syscompilerinfo.h"*/
1.185 brouard 10503: /* command line Intel compiler 32bit windows, XP compatible:*/
10504: /* /GS /W3 /Gy
10505: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10506: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10507: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10508: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10509: */
10510: /* 64 bits */
1.185 brouard 10511: /*
10512: /GS /W3 /Gy
10513: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10514: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10515: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10516: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10517: /* Optimization are useless and O3 is slower than O2 */
10518: /*
10519: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10520: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10521: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10522: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10523: */
1.186 brouard 10524: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10525: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10526: /PDB:"visual studio
10527: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10528: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10529: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10530: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10531: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10532: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10533: uiAccess='false'"
10534: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10535: /NOLOGO /TLBID:1
10536: */
1.292 brouard 10537:
10538:
1.177 brouard 10539: #if defined __INTEL_COMPILER
1.178 brouard 10540: #if defined(__GNUC__)
10541: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10542: #endif
1.177 brouard 10543: #elif defined(__GNUC__)
1.179 brouard 10544: #ifndef __APPLE__
1.174 brouard 10545: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10546: #endif
1.177 brouard 10547: struct utsname sysInfo;
1.178 brouard 10548: int cross = CROSS;
10549: if (cross){
10550: printf("Cross-");
1.191 brouard 10551: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10552: }
1.174 brouard 10553: #endif
10554:
1.191 brouard 10555: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10556: #if defined(__clang__)
1.191 brouard 10557: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10558: #endif
10559: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10560: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10561: #endif
10562: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10563: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10564: #endif
10565: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10566: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10567: #endif
10568: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10569: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10570: #endif
10571: #if defined(_MSC_VER)
1.191 brouard 10572: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10573: #endif
10574: #if defined(__PGI)
1.191 brouard 10575: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10576: #endif
10577: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10578: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10579: #endif
1.191 brouard 10580: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10581:
1.167 brouard 10582: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10583: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10584: // Windows (x64 and x86)
1.191 brouard 10585: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10586: #elif __unix__ // all unices, not all compilers
10587: // Unix
1.191 brouard 10588: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10589: #elif __linux__
10590: // linux
1.191 brouard 10591: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10592: #elif __APPLE__
1.174 brouard 10593: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10594: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10595: #endif
10596:
10597: /* __MINGW32__ */
10598: /* __CYGWIN__ */
10599: /* __MINGW64__ */
10600: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10601: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10602: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10603: /* _WIN64 // Defined for applications for Win64. */
10604: /* _M_X64 // Defined for compilations that target x64 processors. */
10605: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10606:
1.167 brouard 10607: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10608: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10609: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10610: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10611: #else
1.191 brouard 10612: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10613: #endif
10614:
1.169 brouard 10615: #if defined(__GNUC__)
10616: # if defined(__GNUC_PATCHLEVEL__)
10617: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10618: + __GNUC_MINOR__ * 100 \
10619: + __GNUC_PATCHLEVEL__)
10620: # else
10621: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10622: + __GNUC_MINOR__ * 100)
10623: # endif
1.174 brouard 10624: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10625: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10626:
10627: if (uname(&sysInfo) != -1) {
10628: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10629: 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 10630: }
10631: else
10632: perror("uname() error");
1.179 brouard 10633: //#ifndef __INTEL_COMPILER
10634: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10635: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10636: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10637: #endif
1.169 brouard 10638: #endif
1.172 brouard 10639:
1.286 brouard 10640: // void main ()
1.172 brouard 10641: // {
1.169 brouard 10642: #if defined(_MSC_VER)
1.174 brouard 10643: if (IsWow64()){
1.191 brouard 10644: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10645: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10646: }
10647: else{
1.191 brouard 10648: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10649: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10650: }
1.172 brouard 10651: // printf("\nPress Enter to continue...");
10652: // getchar();
10653: // }
10654:
1.169 brouard 10655: #endif
10656:
1.167 brouard 10657:
1.219 brouard 10658: }
1.136 brouard 10659:
1.219 brouard 10660: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10661: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10662: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10663: /* double ftolpl = 1.e-10; */
1.180 brouard 10664: double age, agebase, agelim;
1.203 brouard 10665: double tot;
1.180 brouard 10666:
1.202 brouard 10667: strcpy(filerespl,"PL_");
10668: strcat(filerespl,fileresu);
10669: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10670: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10671: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10672: }
1.288 brouard 10673: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10674: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10675: pstamp(ficrespl);
1.288 brouard 10676: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10677: fprintf(ficrespl,"#Age ");
10678: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10679: fprintf(ficrespl,"\n");
1.180 brouard 10680:
1.219 brouard 10681: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10682:
1.219 brouard 10683: agebase=ageminpar;
10684: agelim=agemaxpar;
1.180 brouard 10685:
1.227 brouard 10686: /* i1=pow(2,ncoveff); */
1.234 brouard 10687: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10688: if (cptcovn < 1){i1=1;}
1.180 brouard 10689:
1.238 brouard 10690: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10691: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10692: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10693: continue;
1.235 brouard 10694:
1.238 brouard 10695: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10696: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10697: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10698: /* k=k+1; */
10699: /* to clean */
10700: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10701: fprintf(ficrespl,"#******");
10702: printf("#******");
10703: fprintf(ficlog,"#******");
10704: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10705: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10706: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10707: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10708: }
10709: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10710: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10711: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10712: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10713: }
10714: fprintf(ficrespl,"******\n");
10715: printf("******\n");
10716: fprintf(ficlog,"******\n");
10717: if(invalidvarcomb[k]){
10718: printf("\nCombination (%d) ignored because no case \n",k);
10719: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10720: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10721: continue;
10722: }
1.219 brouard 10723:
1.238 brouard 10724: fprintf(ficrespl,"#Age ");
10725: for(j=1;j<=cptcoveff;j++) {
10726: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10727: }
10728: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10729: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10730:
1.238 brouard 10731: for (age=agebase; age<=agelim; age++){
10732: /* for (age=agebase; age<=agebase; age++){ */
10733: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10734: fprintf(ficrespl,"%.0f ",age );
10735: for(j=1;j<=cptcoveff;j++)
10736: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10737: tot=0.;
10738: for(i=1; i<=nlstate;i++){
10739: tot += prlim[i][i];
10740: fprintf(ficrespl," %.5f", prlim[i][i]);
10741: }
10742: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10743: } /* Age */
10744: /* was end of cptcod */
10745: } /* cptcov */
10746: } /* nres */
1.219 brouard 10747: return 0;
1.180 brouard 10748: }
10749:
1.218 brouard 10750: 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 10751: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10752:
10753: /* Computes the back prevalence limit for any combination of covariate values
10754: * at any age between ageminpar and agemaxpar
10755: */
1.235 brouard 10756: int i, j, k, i1, nres=0 ;
1.217 brouard 10757: /* double ftolpl = 1.e-10; */
10758: double age, agebase, agelim;
10759: double tot;
1.218 brouard 10760: /* double ***mobaverage; */
10761: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10762:
10763: strcpy(fileresplb,"PLB_");
10764: strcat(fileresplb,fileresu);
10765: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10766: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10767: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10768: }
1.288 brouard 10769: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10770: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10771: pstamp(ficresplb);
1.288 brouard 10772: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10773: fprintf(ficresplb,"#Age ");
10774: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10775: fprintf(ficresplb,"\n");
10776:
1.218 brouard 10777:
10778: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10779:
10780: agebase=ageminpar;
10781: agelim=agemaxpar;
10782:
10783:
1.227 brouard 10784: i1=pow(2,cptcoveff);
1.218 brouard 10785: if (cptcovn < 1){i1=1;}
1.227 brouard 10786:
1.238 brouard 10787: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10788: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10789: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10790: continue;
10791: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10792: fprintf(ficresplb,"#******");
10793: printf("#******");
10794: fprintf(ficlog,"#******");
10795: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10796: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10797: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10798: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10799: }
10800: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10801: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10802: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10803: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10804: }
10805: fprintf(ficresplb,"******\n");
10806: printf("******\n");
10807: fprintf(ficlog,"******\n");
10808: if(invalidvarcomb[k]){
10809: printf("\nCombination (%d) ignored because no cases \n",k);
10810: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10811: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10812: continue;
10813: }
1.218 brouard 10814:
1.238 brouard 10815: fprintf(ficresplb,"#Age ");
10816: for(j=1;j<=cptcoveff;j++) {
10817: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10818: }
10819: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10820: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10821:
10822:
1.238 brouard 10823: for (age=agebase; age<=agelim; age++){
10824: /* for (age=agebase; age<=agebase; age++){ */
10825: if(mobilavproj > 0){
10826: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10827: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10828: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10829: }else if (mobilavproj == 0){
10830: 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);
10831: 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);
10832: exit(1);
10833: }else{
10834: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10835: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10836: /* printf("TOTOT\n"); */
10837: /* exit(1); */
1.238 brouard 10838: }
10839: fprintf(ficresplb,"%.0f ",age );
10840: for(j=1;j<=cptcoveff;j++)
10841: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10842: tot=0.;
10843: for(i=1; i<=nlstate;i++){
10844: tot += bprlim[i][i];
10845: fprintf(ficresplb," %.5f", bprlim[i][i]);
10846: }
10847: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10848: } /* Age */
10849: /* was end of cptcod */
1.255 brouard 10850: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10851: } /* end of any combination */
10852: } /* end of nres */
1.218 brouard 10853: /* hBijx(p, bage, fage); */
10854: /* fclose(ficrespijb); */
10855:
10856: return 0;
1.217 brouard 10857: }
1.218 brouard 10858:
1.180 brouard 10859: int hPijx(double *p, int bage, int fage){
10860: /*------------- h Pij x at various ages ------------*/
10861:
10862: int stepsize;
10863: int agelim;
10864: int hstepm;
10865: int nhstepm;
1.235 brouard 10866: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10867:
10868: double agedeb;
10869: double ***p3mat;
10870:
1.201 brouard 10871: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10872: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10873: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10874: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10875: }
10876: printf("Computing pij: result on file '%s' \n", filerespij);
10877: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10878:
10879: stepsize=(int) (stepm+YEARM-1)/YEARM;
10880: /*if (stepm<=24) stepsize=2;*/
10881:
10882: agelim=AGESUP;
10883: hstepm=stepsize*YEARM; /* Every year of age */
10884: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10885:
1.180 brouard 10886: /* hstepm=1; aff par mois*/
10887: pstamp(ficrespij);
10888: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10889: i1= pow(2,cptcoveff);
1.218 brouard 10890: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10891: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10892: /* k=k+1; */
1.235 brouard 10893: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10894: for(k=1; k<=i1;k++){
1.253 brouard 10895: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10896: continue;
1.183 brouard 10897: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10898: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10899: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10900: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10901: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10902: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10903: }
1.183 brouard 10904: fprintf(ficrespij,"******\n");
10905:
10906: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10907: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10908: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10909:
10910: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10911:
1.183 brouard 10912: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10913: oldm=oldms;savm=savms;
1.235 brouard 10914: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10915: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10916: for(i=1; i<=nlstate;i++)
10917: for(j=1; j<=nlstate+ndeath;j++)
10918: fprintf(ficrespij," %1d-%1d",i,j);
10919: fprintf(ficrespij,"\n");
10920: for (h=0; h<=nhstepm; h++){
10921: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10922: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10923: for(i=1; i<=nlstate;i++)
10924: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10925: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10926: fprintf(ficrespij,"\n");
10927: }
1.183 brouard 10928: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10929: fprintf(ficrespij,"\n");
10930: }
1.180 brouard 10931: /*}*/
10932: }
1.218 brouard 10933: return 0;
1.180 brouard 10934: }
1.218 brouard 10935:
10936: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10937: /*------------- h Bij x at various ages ------------*/
10938:
10939: int stepsize;
1.218 brouard 10940: /* int agelim; */
10941: int ageminl;
1.217 brouard 10942: int hstepm;
10943: int nhstepm;
1.238 brouard 10944: int h, i, i1, j, k, nres;
1.218 brouard 10945:
1.217 brouard 10946: double agedeb;
10947: double ***p3mat;
1.218 brouard 10948:
10949: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10950: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10951: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10952: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10953: }
10954: printf("Computing pij back: result on file '%s' \n", filerespijb);
10955: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10956:
10957: stepsize=(int) (stepm+YEARM-1)/YEARM;
10958: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10959:
1.218 brouard 10960: /* agelim=AGESUP; */
1.289 brouard 10961: ageminl=AGEINF; /* was 30 */
1.218 brouard 10962: hstepm=stepsize*YEARM; /* Every year of age */
10963: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10964:
10965: /* hstepm=1; aff par mois*/
10966: pstamp(ficrespijb);
1.255 brouard 10967: 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 10968: i1= pow(2,cptcoveff);
1.218 brouard 10969: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10970: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10971: /* k=k+1; */
1.238 brouard 10972: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10973: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10974: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10975: continue;
10976: fprintf(ficrespijb,"\n#****** ");
10977: for(j=1;j<=cptcoveff;j++)
10978: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10979: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10980: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10981: }
10982: fprintf(ficrespijb,"******\n");
1.264 brouard 10983: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10984: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10985: continue;
10986: }
10987:
10988: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10989: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10990: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10991: 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 */
10992: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10993:
10994: /* nhstepm=nhstepm*YEARM; aff par mois*/
10995:
1.266 brouard 10996: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10997: /* and memory limitations if stepm is small */
10998:
1.238 brouard 10999: /* oldm=oldms;savm=savms; */
11000: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 11001: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 11002: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11003: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11004: for(i=1; i<=nlstate;i++)
11005: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11006: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11007: fprintf(ficrespijb,"\n");
1.238 brouard 11008: for (h=0; h<=nhstepm; h++){
11009: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11010: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11011: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11012: for(i=1; i<=nlstate;i++)
11013: for(j=1; j<=nlstate+ndeath;j++)
11014: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
11015: fprintf(ficrespijb,"\n");
11016: }
11017: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11018: fprintf(ficrespijb,"\n");
11019: } /* end age deb */
11020: } /* end combination */
11021: } /* end nres */
1.218 brouard 11022: return 0;
11023: } /* hBijx */
1.217 brouard 11024:
1.180 brouard 11025:
1.136 brouard 11026: /***********************************************/
11027: /**************** Main Program *****************/
11028: /***********************************************/
11029:
11030: int main(int argc, char *argv[])
11031: {
11032: #ifdef GSL
11033: const gsl_multimin_fminimizer_type *T;
11034: size_t iteri = 0, it;
11035: int rval = GSL_CONTINUE;
11036: int status = GSL_SUCCESS;
11037: double ssval;
11038: #endif
11039: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11040: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11041: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11042: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11043: int jj, ll, li, lj, lk;
1.136 brouard 11044: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11045: int num_filled;
1.136 brouard 11046: int itimes;
11047: int NDIM=2;
11048: int vpopbased=0;
1.235 brouard 11049: int nres=0;
1.258 brouard 11050: int endishere=0;
1.277 brouard 11051: int noffset=0;
1.274 brouard 11052: int ncurrv=0; /* Temporary variable */
11053:
1.164 brouard 11054: char ca[32], cb[32];
1.136 brouard 11055: /* FILE *fichtm; *//* Html File */
11056: /* FILE *ficgp;*/ /*Gnuplot File */
11057: struct stat info;
1.191 brouard 11058: double agedeb=0.;
1.194 brouard 11059:
11060: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11061: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11062:
1.165 brouard 11063: double fret;
1.191 brouard 11064: double dum=0.; /* Dummy variable */
1.136 brouard 11065: double ***p3mat;
1.218 brouard 11066: /* double ***mobaverage; */
1.164 brouard 11067:
11068: char line[MAXLINE];
1.197 brouard 11069: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11070:
1.234 brouard 11071: char modeltemp[MAXLINE];
1.230 brouard 11072: char resultline[MAXLINE];
11073:
1.136 brouard 11074: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11075: char *tok, *val; /* pathtot */
1.290 brouard 11076: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11077: int c, h , cpt, c2;
1.191 brouard 11078: int jl=0;
11079: int i1, j1, jk, stepsize=0;
1.194 brouard 11080: int count=0;
11081:
1.164 brouard 11082: int *tab;
1.136 brouard 11083: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11084: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11085: /* double anprojf, mprojf, jprojf; */
11086: /* double jintmean,mintmean,aintmean; */
11087: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11088: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11089: double yrfproj= 10.0; /* Number of years of forward projections */
11090: double yrbproj= 10.0; /* Number of years of backward projections */
11091: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11092: int mobilav=0,popforecast=0;
1.191 brouard 11093: int hstepm=0, nhstepm=0;
1.136 brouard 11094: int agemortsup;
11095: float sumlpop=0.;
11096: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11097: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11098:
1.191 brouard 11099: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11100: double ftolpl=FTOL;
11101: double **prlim;
1.217 brouard 11102: double **bprlim;
1.317 ! brouard 11103: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
! 11104: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11105: double ***paramstart; /* Matrix of starting parameter values */
11106: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11107: double **matcov; /* Matrix of covariance */
1.203 brouard 11108: double **hess; /* Hessian matrix */
1.136 brouard 11109: double ***delti3; /* Scale */
11110: double *delti; /* Scale */
11111: double ***eij, ***vareij;
11112: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11113:
1.136 brouard 11114: double *epj, vepp;
1.164 brouard 11115:
1.273 brouard 11116: double dateprev1, dateprev2;
1.296 brouard 11117: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11118: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11119:
1.217 brouard 11120:
1.136 brouard 11121: double **ximort;
1.145 brouard 11122: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11123: int *dcwave;
11124:
1.164 brouard 11125: char z[1]="c";
1.136 brouard 11126:
11127: /*char *strt;*/
11128: char strtend[80];
1.126 brouard 11129:
1.164 brouard 11130:
1.126 brouard 11131: /* setlocale (LC_ALL, ""); */
11132: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11133: /* textdomain (PACKAGE); */
11134: /* setlocale (LC_CTYPE, ""); */
11135: /* setlocale (LC_MESSAGES, ""); */
11136:
11137: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11138: rstart_time = time(NULL);
11139: /* (void) gettimeofday(&start_time,&tzp);*/
11140: start_time = *localtime(&rstart_time);
1.126 brouard 11141: curr_time=start_time;
1.157 brouard 11142: /*tml = *localtime(&start_time.tm_sec);*/
11143: /* strcpy(strstart,asctime(&tml)); */
11144: strcpy(strstart,asctime(&start_time));
1.126 brouard 11145:
11146: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11147: /* tp.tm_sec = tp.tm_sec +86400; */
11148: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11149: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11150: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11151: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11152: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11153: /* strt=asctime(&tmg); */
11154: /* printf("Time(after) =%s",strstart); */
11155: /* (void) time (&time_value);
11156: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11157: * tm = *localtime(&time_value);
11158: * strstart=asctime(&tm);
11159: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11160: */
11161:
11162: nberr=0; /* Number of errors and warnings */
11163: nbwarn=0;
1.184 brouard 11164: #ifdef WIN32
11165: _getcwd(pathcd, size);
11166: #else
1.126 brouard 11167: getcwd(pathcd, size);
1.184 brouard 11168: #endif
1.191 brouard 11169: syscompilerinfo(0);
1.196 brouard 11170: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11171: if(argc <=1){
11172: printf("\nEnter the parameter file name: ");
1.205 brouard 11173: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11174: printf("ERROR Empty parameter file name\n");
11175: goto end;
11176: }
1.126 brouard 11177: i=strlen(pathr);
11178: if(pathr[i-1]=='\n')
11179: pathr[i-1]='\0';
1.156 brouard 11180: i=strlen(pathr);
1.205 brouard 11181: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11182: pathr[i-1]='\0';
1.205 brouard 11183: }
11184: i=strlen(pathr);
11185: if( i==0 ){
11186: printf("ERROR Empty parameter file name\n");
11187: goto end;
11188: }
11189: for (tok = pathr; tok != NULL; ){
1.126 brouard 11190: printf("Pathr |%s|\n",pathr);
11191: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11192: printf("val= |%s| pathr=%s\n",val,pathr);
11193: strcpy (pathtot, val);
11194: if(pathr[0] == '\0') break; /* Dirty */
11195: }
11196: }
1.281 brouard 11197: else if (argc<=2){
11198: strcpy(pathtot,argv[1]);
11199: }
1.126 brouard 11200: else{
11201: strcpy(pathtot,argv[1]);
1.281 brouard 11202: strcpy(z,argv[2]);
11203: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11204: }
11205: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11206: /*cygwin_split_path(pathtot,path,optionfile);
11207: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11208: /* cutv(path,optionfile,pathtot,'\\');*/
11209:
11210: /* Split argv[0], imach program to get pathimach */
11211: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11212: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11213: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11214: /* strcpy(pathimach,argv[0]); */
11215: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11216: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11217: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11218: #ifdef WIN32
11219: _chdir(path); /* Can be a relative path */
11220: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11221: #else
1.126 brouard 11222: chdir(path); /* Can be a relative path */
1.184 brouard 11223: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11224: #endif
11225: printf("Current directory %s!\n",pathcd);
1.126 brouard 11226: strcpy(command,"mkdir ");
11227: strcat(command,optionfilefiname);
11228: if((outcmd=system(command)) != 0){
1.169 brouard 11229: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11230: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11231: /* fclose(ficlog); */
11232: /* exit(1); */
11233: }
11234: /* if((imk=mkdir(optionfilefiname))<0){ */
11235: /* perror("mkdir"); */
11236: /* } */
11237:
11238: /*-------- arguments in the command line --------*/
11239:
1.186 brouard 11240: /* Main Log file */
1.126 brouard 11241: strcat(filelog, optionfilefiname);
11242: strcat(filelog,".log"); /* */
11243: if((ficlog=fopen(filelog,"w"))==NULL) {
11244: printf("Problem with logfile %s\n",filelog);
11245: goto end;
11246: }
11247: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11248: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11249: fprintf(ficlog,"\nEnter the parameter file name: \n");
11250: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11251: path=%s \n\
11252: optionfile=%s\n\
11253: optionfilext=%s\n\
1.156 brouard 11254: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11255:
1.197 brouard 11256: syscompilerinfo(1);
1.167 brouard 11257:
1.126 brouard 11258: printf("Local time (at start):%s",strstart);
11259: fprintf(ficlog,"Local time (at start): %s",strstart);
11260: fflush(ficlog);
11261: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11262: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11263:
11264: /* */
11265: strcpy(fileres,"r");
11266: strcat(fileres, optionfilefiname);
1.201 brouard 11267: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11268: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11269: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11270:
1.186 brouard 11271: /* Main ---------arguments file --------*/
1.126 brouard 11272:
11273: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11274: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11275: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11276: fflush(ficlog);
1.149 brouard 11277: /* goto end; */
11278: exit(70);
1.126 brouard 11279: }
11280:
11281: strcpy(filereso,"o");
1.201 brouard 11282: strcat(filereso,fileresu);
1.126 brouard 11283: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11284: printf("Problem with Output resultfile: %s\n", filereso);
11285: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11286: fflush(ficlog);
11287: goto end;
11288: }
1.278 brouard 11289: /*-------- Rewriting parameter file ----------*/
11290: strcpy(rfileres,"r"); /* "Rparameterfile */
11291: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11292: strcat(rfileres,"."); /* */
11293: strcat(rfileres,optionfilext); /* Other files have txt extension */
11294: if((ficres =fopen(rfileres,"w"))==NULL) {
11295: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11296: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11297: fflush(ficlog);
11298: goto end;
11299: }
11300: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11301:
1.278 brouard 11302:
1.126 brouard 11303: /* Reads comments: lines beginning with '#' */
11304: numlinepar=0;
1.277 brouard 11305: /* Is it a BOM UTF-8 Windows file? */
11306: /* First parameter line */
1.197 brouard 11307: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11308: noffset=0;
11309: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11310: {
11311: noffset=noffset+3;
11312: printf("# File is an UTF8 Bom.\n"); // 0xBF
11313: }
1.302 brouard 11314: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11315: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11316: {
11317: noffset=noffset+2;
11318: printf("# File is an UTF16BE BOM file\n");
11319: }
11320: else if( line[0] == 0 && line[1] == 0)
11321: {
11322: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11323: noffset=noffset+4;
11324: printf("# File is an UTF16BE BOM file\n");
11325: }
11326: } else{
11327: ;/*printf(" Not a BOM file\n");*/
11328: }
11329:
1.197 brouard 11330: /* If line starts with a # it is a comment */
1.277 brouard 11331: if (line[noffset] == '#') {
1.197 brouard 11332: numlinepar++;
11333: fputs(line,stdout);
11334: fputs(line,ficparo);
1.278 brouard 11335: fputs(line,ficres);
1.197 brouard 11336: fputs(line,ficlog);
11337: continue;
11338: }else
11339: break;
11340: }
11341: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11342: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11343: if (num_filled != 5) {
11344: printf("Should be 5 parameters\n");
1.283 brouard 11345: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11346: }
1.126 brouard 11347: numlinepar++;
1.197 brouard 11348: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11349: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11350: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11351: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11352: }
11353: /* Second parameter line */
11354: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11355: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11356: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11357: if (line[0] == '#') {
11358: numlinepar++;
1.283 brouard 11359: printf("%s",line);
11360: fprintf(ficres,"%s",line);
11361: fprintf(ficparo,"%s",line);
11362: fprintf(ficlog,"%s",line);
1.197 brouard 11363: continue;
11364: }else
11365: break;
11366: }
1.223 brouard 11367: 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", \
11368: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11369: if (num_filled != 11) {
11370: 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 11371: printf("but line=%s\n",line);
1.283 brouard 11372: 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");
11373: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11374: }
1.286 brouard 11375: if( lastpass > maxwav){
11376: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11377: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11378: fflush(ficlog);
11379: goto end;
11380: }
11381: 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 11382: 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 11383: 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 11384: 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 11385: }
1.203 brouard 11386: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11387: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11388: /* Third parameter line */
11389: while(fgets(line, MAXLINE, ficpar)) {
11390: /* If line starts with a # it is a comment */
11391: if (line[0] == '#') {
11392: numlinepar++;
1.283 brouard 11393: printf("%s",line);
11394: fprintf(ficres,"%s",line);
11395: fprintf(ficparo,"%s",line);
11396: fprintf(ficlog,"%s",line);
1.197 brouard 11397: continue;
11398: }else
11399: break;
11400: }
1.201 brouard 11401: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11402: if (num_filled != 1){
1.302 brouard 11403: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11404: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11405: model[0]='\0';
11406: goto end;
11407: }
11408: else{
11409: if (model[0]=='+'){
11410: for(i=1; i<=strlen(model);i++)
11411: modeltemp[i-1]=model[i];
1.201 brouard 11412: strcpy(model,modeltemp);
1.197 brouard 11413: }
11414: }
1.199 brouard 11415: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11416: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11417: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11418: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11419: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11420: }
11421: /* 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); */
11422: /* numlinepar=numlinepar+3; /\* In general *\/ */
11423: /* 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 11424: /* 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); */
11425: /* 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 11426: fflush(ficlog);
1.190 brouard 11427: /* if(model[0]=='#'|| model[0]== '\0'){ */
11428: if(model[0]=='#'){
1.279 brouard 11429: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11430: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11431: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11432: if(mle != -1){
1.279 brouard 11433: 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 11434: exit(1);
11435: }
11436: }
1.126 brouard 11437: while((c=getc(ficpar))=='#' && c!= EOF){
11438: ungetc(c,ficpar);
11439: fgets(line, MAXLINE, ficpar);
11440: numlinepar++;
1.195 brouard 11441: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11442: z[0]=line[1];
11443: }
11444: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11445: fputs(line, stdout);
11446: //puts(line);
1.126 brouard 11447: fputs(line,ficparo);
11448: fputs(line,ficlog);
11449: }
11450: ungetc(c,ficpar);
11451:
11452:
1.290 brouard 11453: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11454: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11455: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11456: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11457: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11458: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11459: v1+v2*age+v2*v3 makes cptcovn = 3
11460: */
11461: if (strlen(model)>1)
1.187 brouard 11462: 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 11463: else
1.187 brouard 11464: ncovmodel=2; /* Constant and age */
1.133 brouard 11465: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11466: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11467: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11468: 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);
11469: 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);
11470: fflush(stdout);
11471: fclose (ficlog);
11472: goto end;
11473: }
1.126 brouard 11474: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11475: delti=delti3[1][1];
11476: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11477: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11478: /* We could also provide initial parameters values giving by simple logistic regression
11479: * only one way, that is without matrix product. We will have nlstate maximizations */
11480: /* for(i=1;i<nlstate;i++){ */
11481: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11482: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11483: /* } */
1.126 brouard 11484: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11485: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11486: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11487: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11488: fclose (ficparo);
11489: fclose (ficlog);
11490: goto end;
11491: exit(0);
1.220 brouard 11492: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11493: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11494: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11495: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11496: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11497: matcov=matrix(1,npar,1,npar);
1.203 brouard 11498: hess=matrix(1,npar,1,npar);
1.220 brouard 11499: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11500: /* Read guessed parameters */
1.126 brouard 11501: /* Reads comments: lines beginning with '#' */
11502: while((c=getc(ficpar))=='#' && c!= EOF){
11503: ungetc(c,ficpar);
11504: fgets(line, MAXLINE, ficpar);
11505: numlinepar++;
1.141 brouard 11506: fputs(line,stdout);
1.126 brouard 11507: fputs(line,ficparo);
11508: fputs(line,ficlog);
11509: }
11510: ungetc(c,ficpar);
11511:
11512: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11513: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11514: for(i=1; i <=nlstate; i++){
1.234 brouard 11515: j=0;
1.126 brouard 11516: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11517: if(jj==i) continue;
11518: j++;
1.292 brouard 11519: while((c=getc(ficpar))=='#' && c!= EOF){
11520: ungetc(c,ficpar);
11521: fgets(line, MAXLINE, ficpar);
11522: numlinepar++;
11523: fputs(line,stdout);
11524: fputs(line,ficparo);
11525: fputs(line,ficlog);
11526: }
11527: ungetc(c,ficpar);
1.234 brouard 11528: fscanf(ficpar,"%1d%1d",&i1,&j1);
11529: if ((i1 != i) || (j1 != jj)){
11530: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11531: It might be a problem of design; if ncovcol and the model are correct\n \
11532: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11533: exit(1);
11534: }
11535: fprintf(ficparo,"%1d%1d",i1,j1);
11536: if(mle==1)
11537: printf("%1d%1d",i,jj);
11538: fprintf(ficlog,"%1d%1d",i,jj);
11539: for(k=1; k<=ncovmodel;k++){
11540: fscanf(ficpar," %lf",¶m[i][j][k]);
11541: if(mle==1){
11542: printf(" %lf",param[i][j][k]);
11543: fprintf(ficlog," %lf",param[i][j][k]);
11544: }
11545: else
11546: fprintf(ficlog," %lf",param[i][j][k]);
11547: fprintf(ficparo," %lf",param[i][j][k]);
11548: }
11549: fscanf(ficpar,"\n");
11550: numlinepar++;
11551: if(mle==1)
11552: printf("\n");
11553: fprintf(ficlog,"\n");
11554: fprintf(ficparo,"\n");
1.126 brouard 11555: }
11556: }
11557: fflush(ficlog);
1.234 brouard 11558:
1.251 brouard 11559: /* Reads parameters values */
1.126 brouard 11560: p=param[1][1];
1.251 brouard 11561: pstart=paramstart[1][1];
1.126 brouard 11562:
11563: /* Reads comments: lines beginning with '#' */
11564: while((c=getc(ficpar))=='#' && c!= EOF){
11565: ungetc(c,ficpar);
11566: fgets(line, MAXLINE, ficpar);
11567: numlinepar++;
1.141 brouard 11568: fputs(line,stdout);
1.126 brouard 11569: fputs(line,ficparo);
11570: fputs(line,ficlog);
11571: }
11572: ungetc(c,ficpar);
11573:
11574: for(i=1; i <=nlstate; i++){
11575: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11576: fscanf(ficpar,"%1d%1d",&i1,&j1);
11577: if ( (i1-i) * (j1-j) != 0){
11578: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11579: exit(1);
11580: }
11581: printf("%1d%1d",i,j);
11582: fprintf(ficparo,"%1d%1d",i1,j1);
11583: fprintf(ficlog,"%1d%1d",i1,j1);
11584: for(k=1; k<=ncovmodel;k++){
11585: fscanf(ficpar,"%le",&delti3[i][j][k]);
11586: printf(" %le",delti3[i][j][k]);
11587: fprintf(ficparo," %le",delti3[i][j][k]);
11588: fprintf(ficlog," %le",delti3[i][j][k]);
11589: }
11590: fscanf(ficpar,"\n");
11591: numlinepar++;
11592: printf("\n");
11593: fprintf(ficparo,"\n");
11594: fprintf(ficlog,"\n");
1.126 brouard 11595: }
11596: }
11597: fflush(ficlog);
1.234 brouard 11598:
1.145 brouard 11599: /* Reads covariance matrix */
1.126 brouard 11600: delti=delti3[1][1];
1.220 brouard 11601:
11602:
1.126 brouard 11603: /* 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 11604:
1.126 brouard 11605: /* Reads comments: lines beginning with '#' */
11606: while((c=getc(ficpar))=='#' && c!= EOF){
11607: ungetc(c,ficpar);
11608: fgets(line, MAXLINE, ficpar);
11609: numlinepar++;
1.141 brouard 11610: fputs(line,stdout);
1.126 brouard 11611: fputs(line,ficparo);
11612: fputs(line,ficlog);
11613: }
11614: ungetc(c,ficpar);
1.220 brouard 11615:
1.126 brouard 11616: matcov=matrix(1,npar,1,npar);
1.203 brouard 11617: hess=matrix(1,npar,1,npar);
1.131 brouard 11618: for(i=1; i <=npar; i++)
11619: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11620:
1.194 brouard 11621: /* Scans npar lines */
1.126 brouard 11622: for(i=1; i <=npar; i++){
1.226 brouard 11623: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11624: if(count != 3){
1.226 brouard 11625: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11626: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11627: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11628: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11629: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11630: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11631: exit(1);
1.220 brouard 11632: }else{
1.226 brouard 11633: if(mle==1)
11634: printf("%1d%1d%d",i1,j1,jk);
11635: }
11636: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11637: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11638: for(j=1; j <=i; j++){
1.226 brouard 11639: fscanf(ficpar," %le",&matcov[i][j]);
11640: if(mle==1){
11641: printf(" %.5le",matcov[i][j]);
11642: }
11643: fprintf(ficlog," %.5le",matcov[i][j]);
11644: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11645: }
11646: fscanf(ficpar,"\n");
11647: numlinepar++;
11648: if(mle==1)
1.220 brouard 11649: printf("\n");
1.126 brouard 11650: fprintf(ficlog,"\n");
11651: fprintf(ficparo,"\n");
11652: }
1.194 brouard 11653: /* End of read covariance matrix npar lines */
1.126 brouard 11654: for(i=1; i <=npar; i++)
11655: for(j=i+1;j<=npar;j++)
1.226 brouard 11656: matcov[i][j]=matcov[j][i];
1.126 brouard 11657:
11658: if(mle==1)
11659: printf("\n");
11660: fprintf(ficlog,"\n");
11661:
11662: fflush(ficlog);
11663:
11664: } /* End of mle != -3 */
1.218 brouard 11665:
1.186 brouard 11666: /* Main data
11667: */
1.290 brouard 11668: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11669: /* num=lvector(1,n); */
11670: /* moisnais=vector(1,n); */
11671: /* annais=vector(1,n); */
11672: /* moisdc=vector(1,n); */
11673: /* andc=vector(1,n); */
11674: /* weight=vector(1,n); */
11675: /* agedc=vector(1,n); */
11676: /* cod=ivector(1,n); */
11677: /* for(i=1;i<=n;i++){ */
11678: num=lvector(firstobs,lastobs);
11679: moisnais=vector(firstobs,lastobs);
11680: annais=vector(firstobs,lastobs);
11681: moisdc=vector(firstobs,lastobs);
11682: andc=vector(firstobs,lastobs);
11683: weight=vector(firstobs,lastobs);
11684: agedc=vector(firstobs,lastobs);
11685: cod=ivector(firstobs,lastobs);
11686: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11687: num[i]=0;
11688: moisnais[i]=0;
11689: annais[i]=0;
11690: moisdc[i]=0;
11691: andc[i]=0;
11692: agedc[i]=0;
11693: cod[i]=0;
11694: weight[i]=1.0; /* Equal weights, 1 by default */
11695: }
1.290 brouard 11696: mint=matrix(1,maxwav,firstobs,lastobs);
11697: anint=matrix(1,maxwav,firstobs,lastobs);
11698: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11699: tab=ivector(1,NCOVMAX);
1.144 brouard 11700: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11701: 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 11702:
1.136 brouard 11703: /* Reads data from file datafile */
11704: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11705: goto end;
11706:
11707: /* Calculation of the number of parameters from char model */
1.234 brouard 11708: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11709: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11710: k=3 V4 Tvar[k=3]= 4 (from V4)
11711: k=2 V1 Tvar[k=2]= 1 (from V1)
11712: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11713: */
11714:
11715: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11716: TvarsDind=ivector(1,NCOVMAX); /* */
11717: TvarsD=ivector(1,NCOVMAX); /* */
11718: TvarsQind=ivector(1,NCOVMAX); /* */
11719: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11720: TvarF=ivector(1,NCOVMAX); /* */
11721: TvarFind=ivector(1,NCOVMAX); /* */
11722: TvarV=ivector(1,NCOVMAX); /* */
11723: TvarVind=ivector(1,NCOVMAX); /* */
11724: TvarA=ivector(1,NCOVMAX); /* */
11725: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11726: TvarFD=ivector(1,NCOVMAX); /* */
11727: TvarFDind=ivector(1,NCOVMAX); /* */
11728: TvarFQ=ivector(1,NCOVMAX); /* */
11729: TvarFQind=ivector(1,NCOVMAX); /* */
11730: TvarVD=ivector(1,NCOVMAX); /* */
11731: TvarVDind=ivector(1,NCOVMAX); /* */
11732: TvarVQ=ivector(1,NCOVMAX); /* */
11733: TvarVQind=ivector(1,NCOVMAX); /* */
11734:
1.230 brouard 11735: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11736: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11737: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11738: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11739: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11740: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11741: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11742: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11743: */
11744: /* For model-covariate k tells which data-covariate to use but
11745: because this model-covariate is a construction we invent a new column
11746: ncovcol + k1
11747: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11748: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11749: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11750: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11751: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11752: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11753: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11754: */
1.145 brouard 11755: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11756: 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 11757: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11758: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11759: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11760: 4 covariates (3 plus signs)
11761: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11762: */
1.230 brouard 11763: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11764: * individual dummy, fixed or varying:
11765: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11766: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11767: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11768: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11769: * Tmodelind[1]@9={9,0,3,2,}*/
11770: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11771: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11772: * individual quantitative, fixed or varying:
11773: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11774: * 3, 1, 0, 0, 0, 0, 0, 0},
11775: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11776: /* Main decodemodel */
11777:
1.187 brouard 11778:
1.223 brouard 11779: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11780: goto end;
11781:
1.137 brouard 11782: if((double)(lastobs-imx)/(double)imx > 1.10){
11783: nbwarn++;
11784: 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);
11785: 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);
11786: }
1.136 brouard 11787: /* if(mle==1){*/
1.137 brouard 11788: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11789: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11790: }
11791:
11792: /*-calculation of age at interview from date of interview and age at death -*/
11793: agev=matrix(1,maxwav,1,imx);
11794:
11795: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11796: goto end;
11797:
1.126 brouard 11798:
1.136 brouard 11799: agegomp=(int)agemin;
1.290 brouard 11800: free_vector(moisnais,firstobs,lastobs);
11801: free_vector(annais,firstobs,lastobs);
1.126 brouard 11802: /* free_matrix(mint,1,maxwav,1,n);
11803: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11804: /* free_vector(moisdc,1,n); */
11805: /* free_vector(andc,1,n); */
1.145 brouard 11806: /* */
11807:
1.126 brouard 11808: wav=ivector(1,imx);
1.214 brouard 11809: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11810: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11811: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11812: 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.*/
11813: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11814: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11815:
11816: /* Concatenates waves */
1.214 brouard 11817: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11818: Death is a valid wave (if date is known).
11819: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11820: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11821: and mw[mi+1][i]. dh depends on stepm.
11822: */
11823:
1.126 brouard 11824: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11825: /* Concatenates waves */
1.145 brouard 11826:
1.290 brouard 11827: free_vector(moisdc,firstobs,lastobs);
11828: free_vector(andc,firstobs,lastobs);
1.215 brouard 11829:
1.126 brouard 11830: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11831: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11832: ncodemax[1]=1;
1.145 brouard 11833: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11834: cptcoveff=0;
1.220 brouard 11835: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11836: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11837: }
11838:
11839: ncovcombmax=pow(2,cptcoveff);
11840: invalidvarcomb=ivector(1, ncovcombmax);
11841: for(i=1;i<ncovcombmax;i++)
11842: invalidvarcomb[i]=0;
11843:
1.211 brouard 11844: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11845: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11846: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11847:
1.200 brouard 11848: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11849: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11850: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11851: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11852: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11853: * (currently 0 or 1) in the data.
11854: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11855: * corresponding modality (h,j).
11856: */
11857:
1.145 brouard 11858: h=0;
11859: /*if (cptcovn > 0) */
1.126 brouard 11860: m=pow(2,cptcoveff);
11861:
1.144 brouard 11862: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11863: * For k=4 covariates, h goes from 1 to m=2**k
11864: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11865: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11866: * h\k 1 2 3 4
1.143 brouard 11867: *______________________________
11868: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11869: * 2 2 1 1 1
11870: * 3 i=2 1 2 1 1
11871: * 4 2 2 1 1
11872: * 5 i=3 1 i=2 1 2 1
11873: * 6 2 1 2 1
11874: * 7 i=4 1 2 2 1
11875: * 8 2 2 2 1
1.197 brouard 11876: * 9 i=5 1 i=3 1 i=2 1 2
11877: * 10 2 1 1 2
11878: * 11 i=6 1 2 1 2
11879: * 12 2 2 1 2
11880: * 13 i=7 1 i=4 1 2 2
11881: * 14 2 1 2 2
11882: * 15 i=8 1 2 2 2
11883: * 16 2 2 2 2
1.143 brouard 11884: */
1.212 brouard 11885: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11886: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11887: * and the value of each covariate?
11888: * V1=1, V2=1, V3=2, V4=1 ?
11889: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11890: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11891: * In order to get the real value in the data, we use nbcode
11892: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11893: * We are keeping this crazy system in order to be able (in the future?)
11894: * to have more than 2 values (0 or 1) for a covariate.
11895: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11896: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11897: * bbbbbbbb
11898: * 76543210
11899: * h-1 00000101 (6-1=5)
1.219 brouard 11900: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11901: * &
11902: * 1 00000001 (1)
1.219 brouard 11903: * 00000000 = 1 & ((h-1) >> (k-1))
11904: * +1= 00000001 =1
1.211 brouard 11905: *
11906: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11907: * h' 1101 =2^3+2^2+0x2^1+2^0
11908: * >>k' 11
11909: * & 00000001
11910: * = 00000001
11911: * +1 = 00000010=2 = codtabm(14,3)
11912: * Reverse h=6 and m=16?
11913: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11914: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11915: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11916: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11917: * V3=decodtabm(14,3,2**4)=2
11918: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11919: *(h-1) >> (j-1) 0011 =13 >> 2
11920: * &1 000000001
11921: * = 000000001
11922: * +1= 000000010 =2
11923: * 2211
11924: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11925: * V3=2
1.220 brouard 11926: * codtabm and decodtabm are identical
1.211 brouard 11927: */
11928:
1.145 brouard 11929:
11930: free_ivector(Ndum,-1,NCOVMAX);
11931:
11932:
1.126 brouard 11933:
1.186 brouard 11934: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11935: strcpy(optionfilegnuplot,optionfilefiname);
11936: if(mle==-3)
1.201 brouard 11937: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11938: strcat(optionfilegnuplot,".gp");
11939:
11940: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11941: printf("Problem with file %s",optionfilegnuplot);
11942: }
11943: else{
1.204 brouard 11944: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11945: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11946: //fprintf(ficgp,"set missing 'NaNq'\n");
11947: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11948: }
11949: /* fclose(ficgp);*/
1.186 brouard 11950:
11951:
11952: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11953:
11954: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11955: if(mle==-3)
1.201 brouard 11956: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11957: strcat(optionfilehtm,".htm");
11958: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11959: printf("Problem with %s \n",optionfilehtm);
11960: exit(0);
1.126 brouard 11961: }
11962:
11963: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11964: strcat(optionfilehtmcov,"-cov.htm");
11965: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11966: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11967: }
11968: else{
11969: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11970: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11971: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11972: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11973: }
11974:
1.213 brouard 11975: 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 11976: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11977: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11978: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11979: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11980: \n\
11981: <hr size=\"2\" color=\"#EC5E5E\">\
11982: <ul><li><h4>Parameter files</h4>\n\
11983: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11984: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11985: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11986: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11987: - Date and time at start: %s</ul>\n",\
11988: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11989: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11990: fileres,fileres,\
11991: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11992: fflush(fichtm);
11993:
11994: strcpy(pathr,path);
11995: strcat(pathr,optionfilefiname);
1.184 brouard 11996: #ifdef WIN32
11997: _chdir(optionfilefiname); /* Move to directory named optionfile */
11998: #else
1.126 brouard 11999: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12000: #endif
12001:
1.126 brouard 12002:
1.220 brouard 12003: /* Calculates basic frequencies. Computes observed prevalence at single age
12004: and for any valid combination of covariates
1.126 brouard 12005: and prints on file fileres'p'. */
1.251 brouard 12006: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12007: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12008:
12009: fprintf(fichtm,"\n");
1.286 brouard 12010: 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 12011: ftol, stepm);
12012: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12013: ncurrv=1;
12014: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12015: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12016: ncurrv=i;
12017: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12018: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12019: ncurrv=i;
12020: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12021: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12022: ncurrv=i;
12023: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12024: 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", \
12025: nlstate, ndeath, maxwav, mle, weightopt);
12026:
12027: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12028: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12029:
12030:
1.317 ! brouard 12031: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12032: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12033: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12034: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12035: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12036: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12037: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12038: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12039: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12040:
1.126 brouard 12041: /* For Powell, parameters are in a vector p[] starting at p[1]
12042: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12043: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12044:
12045: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12046: /* For mortality only */
1.126 brouard 12047: if (mle==-3){
1.136 brouard 12048: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12049: for(i=1;i<=NDIM;i++)
12050: for(j=1;j<=NDIM;j++)
12051: ximort[i][j]=0.;
1.186 brouard 12052: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12053: cens=ivector(firstobs,lastobs);
12054: ageexmed=vector(firstobs,lastobs);
12055: agecens=vector(firstobs,lastobs);
12056: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12057:
1.126 brouard 12058: for (i=1; i<=imx; i++){
12059: dcwave[i]=-1;
12060: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12061: if (s[m][i]>nlstate) {
12062: dcwave[i]=m;
12063: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12064: break;
12065: }
1.126 brouard 12066: }
1.226 brouard 12067:
1.126 brouard 12068: for (i=1; i<=imx; i++) {
12069: if (wav[i]>0){
1.226 brouard 12070: ageexmed[i]=agev[mw[1][i]][i];
12071: j=wav[i];
12072: agecens[i]=1.;
12073:
12074: if (ageexmed[i]> 1 && wav[i] > 0){
12075: agecens[i]=agev[mw[j][i]][i];
12076: cens[i]= 1;
12077: }else if (ageexmed[i]< 1)
12078: cens[i]= -1;
12079: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12080: cens[i]=0 ;
1.126 brouard 12081: }
12082: else cens[i]=-1;
12083: }
12084:
12085: for (i=1;i<=NDIM;i++) {
12086: for (j=1;j<=NDIM;j++)
1.226 brouard 12087: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12088: }
12089:
1.302 brouard 12090: p[1]=0.0268; p[NDIM]=0.083;
12091: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12092:
12093:
1.136 brouard 12094: #ifdef GSL
12095: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12096: #else
1.126 brouard 12097: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12098: #endif
1.201 brouard 12099: strcpy(filerespow,"POW-MORT_");
12100: strcat(filerespow,fileresu);
1.126 brouard 12101: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12102: printf("Problem with resultfile: %s\n", filerespow);
12103: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12104: }
1.136 brouard 12105: #ifdef GSL
12106: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12107: #else
1.126 brouard 12108: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12109: #endif
1.126 brouard 12110: /* for (i=1;i<=nlstate;i++)
12111: for(j=1;j<=nlstate+ndeath;j++)
12112: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12113: */
12114: fprintf(ficrespow,"\n");
1.136 brouard 12115: #ifdef GSL
12116: /* gsl starts here */
12117: T = gsl_multimin_fminimizer_nmsimplex;
12118: gsl_multimin_fminimizer *sfm = NULL;
12119: gsl_vector *ss, *x;
12120: gsl_multimin_function minex_func;
12121:
12122: /* Initial vertex size vector */
12123: ss = gsl_vector_alloc (NDIM);
12124:
12125: if (ss == NULL){
12126: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12127: }
12128: /* Set all step sizes to 1 */
12129: gsl_vector_set_all (ss, 0.001);
12130:
12131: /* Starting point */
1.126 brouard 12132:
1.136 brouard 12133: x = gsl_vector_alloc (NDIM);
12134:
12135: if (x == NULL){
12136: gsl_vector_free(ss);
12137: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12138: }
12139:
12140: /* Initialize method and iterate */
12141: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12142: /* gsl_vector_set(x, 0, 0.0268); */
12143: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12144: gsl_vector_set(x, 0, p[1]);
12145: gsl_vector_set(x, 1, p[2]);
12146:
12147: minex_func.f = &gompertz_f;
12148: minex_func.n = NDIM;
12149: minex_func.params = (void *)&p; /* ??? */
12150:
12151: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12152: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12153:
12154: printf("Iterations beginning .....\n\n");
12155: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12156:
12157: iteri=0;
12158: while (rval == GSL_CONTINUE){
12159: iteri++;
12160: status = gsl_multimin_fminimizer_iterate(sfm);
12161:
12162: if (status) printf("error: %s\n", gsl_strerror (status));
12163: fflush(0);
12164:
12165: if (status)
12166: break;
12167:
12168: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12169: ssval = gsl_multimin_fminimizer_size (sfm);
12170:
12171: if (rval == GSL_SUCCESS)
12172: printf ("converged to a local maximum at\n");
12173:
12174: printf("%5d ", iteri);
12175: for (it = 0; it < NDIM; it++){
12176: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12177: }
12178: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12179: }
12180:
12181: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12182:
12183: gsl_vector_free(x); /* initial values */
12184: gsl_vector_free(ss); /* inital step size */
12185: for (it=0; it<NDIM; it++){
12186: p[it+1]=gsl_vector_get(sfm->x,it);
12187: fprintf(ficrespow," %.12lf", p[it]);
12188: }
12189: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12190: #endif
12191: #ifdef POWELL
12192: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12193: #endif
1.126 brouard 12194: fclose(ficrespow);
12195:
1.203 brouard 12196: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12197:
12198: for(i=1; i <=NDIM; i++)
12199: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12200: matcov[i][j]=matcov[j][i];
1.126 brouard 12201:
12202: printf("\nCovariance matrix\n ");
1.203 brouard 12203: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12204: for(i=1; i <=NDIM; i++) {
12205: for(j=1;j<=NDIM;j++){
1.220 brouard 12206: printf("%f ",matcov[i][j]);
12207: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12208: }
1.203 brouard 12209: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12210: }
12211:
12212: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12213: for (i=1;i<=NDIM;i++) {
1.126 brouard 12214: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12215: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12216: }
1.302 brouard 12217: lsurv=vector(agegomp,AGESUP);
12218: lpop=vector(agegomp,AGESUP);
12219: tpop=vector(agegomp,AGESUP);
1.126 brouard 12220: lsurv[agegomp]=100000;
12221:
12222: for (k=agegomp;k<=AGESUP;k++) {
12223: agemortsup=k;
12224: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12225: }
12226:
12227: for (k=agegomp;k<agemortsup;k++)
12228: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12229:
12230: for (k=agegomp;k<agemortsup;k++){
12231: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12232: sumlpop=sumlpop+lpop[k];
12233: }
12234:
12235: tpop[agegomp]=sumlpop;
12236: for (k=agegomp;k<(agemortsup-3);k++){
12237: /* tpop[k+1]=2;*/
12238: tpop[k+1]=tpop[k]-lpop[k];
12239: }
12240:
12241:
12242: printf("\nAge lx qx dx Lx Tx e(x)\n");
12243: for (k=agegomp;k<(agemortsup-2);k++)
12244: 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]);
12245:
12246:
12247: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12248: ageminpar=50;
12249: agemaxpar=100;
1.194 brouard 12250: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12251: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12252: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12253: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12254: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12255: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12256: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12257: }else{
12258: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12259: 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 12260: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12261: }
1.201 brouard 12262: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12263: stepm, weightopt,\
12264: model,imx,p,matcov,agemortsup);
12265:
1.302 brouard 12266: free_vector(lsurv,agegomp,AGESUP);
12267: free_vector(lpop,agegomp,AGESUP);
12268: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12269: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12270: free_ivector(dcwave,firstobs,lastobs);
12271: free_vector(agecens,firstobs,lastobs);
12272: free_vector(ageexmed,firstobs,lastobs);
12273: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12274: #ifdef GSL
1.136 brouard 12275: #endif
1.186 brouard 12276: } /* Endof if mle==-3 mortality only */
1.205 brouard 12277: /* Standard */
12278: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12279: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12280: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12281: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12282: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12283: for (k=1; k<=npar;k++)
12284: printf(" %d %8.5f",k,p[k]);
12285: printf("\n");
1.205 brouard 12286: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12287: /* mlikeli uses func not funcone */
1.247 brouard 12288: /* for(i=1;i<nlstate;i++){ */
12289: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12290: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12291: /* } */
1.205 brouard 12292: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12293: }
12294: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12295: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12296: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12297: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12298: }
12299: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12300: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12301: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12302: for (k=1; k<=npar;k++)
12303: printf(" %d %8.5f",k,p[k]);
12304: printf("\n");
12305:
12306: /*--------- results files --------------*/
1.283 brouard 12307: /* 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 12308:
12309:
12310: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12311: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12312: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12313: for(i=1,jk=1; i <=nlstate; i++){
12314: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12315: if (k != i) {
12316: printf("%d%d ",i,k);
12317: fprintf(ficlog,"%d%d ",i,k);
12318: fprintf(ficres,"%1d%1d ",i,k);
12319: for(j=1; j <=ncovmodel; j++){
12320: printf("%12.7f ",p[jk]);
12321: fprintf(ficlog,"%12.7f ",p[jk]);
12322: fprintf(ficres,"%12.7f ",p[jk]);
12323: jk++;
12324: }
12325: printf("\n");
12326: fprintf(ficlog,"\n");
12327: fprintf(ficres,"\n");
12328: }
1.126 brouard 12329: }
12330: }
1.203 brouard 12331: if(mle != 0){
12332: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12333: ftolhess=ftol; /* Usually correct */
1.203 brouard 12334: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12335: 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");
12336: 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");
12337: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12338: for(k=1; k <=(nlstate+ndeath); k++){
12339: if (k != i) {
12340: printf("%d%d ",i,k);
12341: fprintf(ficlog,"%d%d ",i,k);
12342: for(j=1; j <=ncovmodel; j++){
12343: 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]));
12344: 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]));
12345: jk++;
12346: }
12347: printf("\n");
12348: fprintf(ficlog,"\n");
12349: }
12350: }
1.193 brouard 12351: }
1.203 brouard 12352: } /* end of hesscov and Wald tests */
1.225 brouard 12353:
1.203 brouard 12354: /* */
1.126 brouard 12355: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12356: printf("# Scales (for hessian or gradient estimation)\n");
12357: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12358: for(i=1,jk=1; i <=nlstate; i++){
12359: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12360: if (j!=i) {
12361: fprintf(ficres,"%1d%1d",i,j);
12362: printf("%1d%1d",i,j);
12363: fprintf(ficlog,"%1d%1d",i,j);
12364: for(k=1; k<=ncovmodel;k++){
12365: printf(" %.5e",delti[jk]);
12366: fprintf(ficlog," %.5e",delti[jk]);
12367: fprintf(ficres," %.5e",delti[jk]);
12368: jk++;
12369: }
12370: printf("\n");
12371: fprintf(ficlog,"\n");
12372: fprintf(ficres,"\n");
12373: }
1.126 brouard 12374: }
12375: }
12376:
12377: 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 12378: if(mle >= 1) /* To big for the screen */
1.126 brouard 12379: 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");
12380: 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");
12381: /* # 121 Var(a12)\n\ */
12382: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12383: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12384: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12385: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12386: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12387: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12388: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12389:
12390:
12391: /* Just to have a covariance matrix which will be more understandable
12392: even is we still don't want to manage dictionary of variables
12393: */
12394: for(itimes=1;itimes<=2;itimes++){
12395: jj=0;
12396: for(i=1; i <=nlstate; i++){
1.225 brouard 12397: for(j=1; j <=nlstate+ndeath; j++){
12398: if(j==i) continue;
12399: for(k=1; k<=ncovmodel;k++){
12400: jj++;
12401: ca[0]= k+'a'-1;ca[1]='\0';
12402: if(itimes==1){
12403: if(mle>=1)
12404: printf("#%1d%1d%d",i,j,k);
12405: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12406: fprintf(ficres,"#%1d%1d%d",i,j,k);
12407: }else{
12408: if(mle>=1)
12409: printf("%1d%1d%d",i,j,k);
12410: fprintf(ficlog,"%1d%1d%d",i,j,k);
12411: fprintf(ficres,"%1d%1d%d",i,j,k);
12412: }
12413: ll=0;
12414: for(li=1;li <=nlstate; li++){
12415: for(lj=1;lj <=nlstate+ndeath; lj++){
12416: if(lj==li) continue;
12417: for(lk=1;lk<=ncovmodel;lk++){
12418: ll++;
12419: if(ll<=jj){
12420: cb[0]= lk +'a'-1;cb[1]='\0';
12421: if(ll<jj){
12422: if(itimes==1){
12423: if(mle>=1)
12424: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12425: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12426: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12427: }else{
12428: if(mle>=1)
12429: printf(" %.5e",matcov[jj][ll]);
12430: fprintf(ficlog," %.5e",matcov[jj][ll]);
12431: fprintf(ficres," %.5e",matcov[jj][ll]);
12432: }
12433: }else{
12434: if(itimes==1){
12435: if(mle>=1)
12436: printf(" Var(%s%1d%1d)",ca,i,j);
12437: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12438: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12439: }else{
12440: if(mle>=1)
12441: printf(" %.7e",matcov[jj][ll]);
12442: fprintf(ficlog," %.7e",matcov[jj][ll]);
12443: fprintf(ficres," %.7e",matcov[jj][ll]);
12444: }
12445: }
12446: }
12447: } /* end lk */
12448: } /* end lj */
12449: } /* end li */
12450: if(mle>=1)
12451: printf("\n");
12452: fprintf(ficlog,"\n");
12453: fprintf(ficres,"\n");
12454: numlinepar++;
12455: } /* end k*/
12456: } /*end j */
1.126 brouard 12457: } /* end i */
12458: } /* end itimes */
12459:
12460: fflush(ficlog);
12461: fflush(ficres);
1.225 brouard 12462: while(fgets(line, MAXLINE, ficpar)) {
12463: /* If line starts with a # it is a comment */
12464: if (line[0] == '#') {
12465: numlinepar++;
12466: fputs(line,stdout);
12467: fputs(line,ficparo);
12468: fputs(line,ficlog);
1.299 brouard 12469: fputs(line,ficres);
1.225 brouard 12470: continue;
12471: }else
12472: break;
12473: }
12474:
1.209 brouard 12475: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12476: /* ungetc(c,ficpar); */
12477: /* fgets(line, MAXLINE, ficpar); */
12478: /* fputs(line,stdout); */
12479: /* fputs(line,ficparo); */
12480: /* } */
12481: /* ungetc(c,ficpar); */
1.126 brouard 12482:
12483: estepm=0;
1.209 brouard 12484: 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 12485:
12486: if (num_filled != 6) {
12487: 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);
12488: 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);
12489: goto end;
12490: }
12491: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12492: }
12493: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12494: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12495:
1.209 brouard 12496: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12497: if (estepm==0 || estepm < stepm) estepm=stepm;
12498: if (fage <= 2) {
12499: bage = ageminpar;
12500: fage = agemaxpar;
12501: }
12502:
12503: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12504: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12505: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12506:
1.186 brouard 12507: /* Other stuffs, more or less useful */
1.254 brouard 12508: while(fgets(line, MAXLINE, ficpar)) {
12509: /* If line starts with a # it is a comment */
12510: if (line[0] == '#') {
12511: numlinepar++;
12512: fputs(line,stdout);
12513: fputs(line,ficparo);
12514: fputs(line,ficlog);
1.299 brouard 12515: fputs(line,ficres);
1.254 brouard 12516: continue;
12517: }else
12518: break;
12519: }
12520:
12521: 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){
12522:
12523: if (num_filled != 7) {
12524: 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);
12525: 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);
12526: goto end;
12527: }
12528: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12529: 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);
12530: 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);
12531: 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 12532: }
1.254 brouard 12533:
12534: while(fgets(line, MAXLINE, ficpar)) {
12535: /* If line starts with a # it is a comment */
12536: if (line[0] == '#') {
12537: numlinepar++;
12538: fputs(line,stdout);
12539: fputs(line,ficparo);
12540: fputs(line,ficlog);
1.299 brouard 12541: fputs(line,ficres);
1.254 brouard 12542: continue;
12543: }else
12544: break;
1.126 brouard 12545: }
12546:
12547:
12548: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12549: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12550:
1.254 brouard 12551: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12552: if (num_filled != 1) {
12553: 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);
12554: 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);
12555: goto end;
12556: }
12557: printf("pop_based=%d\n",popbased);
12558: fprintf(ficlog,"pop_based=%d\n",popbased);
12559: fprintf(ficparo,"pop_based=%d\n",popbased);
12560: fprintf(ficres,"pop_based=%d\n",popbased);
12561: }
12562:
1.258 brouard 12563: /* Results */
1.307 brouard 12564: endishere=0;
1.258 brouard 12565: nresult=0;
1.308 brouard 12566: parameterline=0;
1.258 brouard 12567: do{
12568: if(!fgets(line, MAXLINE, ficpar)){
12569: endishere=1;
1.308 brouard 12570: parameterline=15;
1.258 brouard 12571: }else if (line[0] == '#') {
12572: /* If line starts with a # it is a comment */
1.254 brouard 12573: numlinepar++;
12574: fputs(line,stdout);
12575: fputs(line,ficparo);
12576: fputs(line,ficlog);
1.299 brouard 12577: fputs(line,ficres);
1.254 brouard 12578: continue;
1.258 brouard 12579: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12580: parameterline=11;
1.296 brouard 12581: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12582: parameterline=12;
1.307 brouard 12583: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12584: parameterline=13;
1.307 brouard 12585: }
1.258 brouard 12586: else{
12587: parameterline=14;
1.254 brouard 12588: }
1.308 brouard 12589: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12590: case 11:
1.296 brouard 12591: 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)){
12592: 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 12593: 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);
12594: 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);
12595: 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);
12596: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12597: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12598: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12599: prvforecast = 1;
12600: }
12601: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12602: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12603: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12604: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12605: prvforecast = 2;
12606: }
12607: else {
12608: 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);
12609: 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);
12610: goto end;
1.258 brouard 12611: }
1.254 brouard 12612: break;
1.258 brouard 12613: case 12:
1.296 brouard 12614: 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)){
12615: 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);
12616: 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);
12617: 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);
12618: 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);
12619: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12620: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12621: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12622: prvbackcast = 1;
12623: }
12624: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12625: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12626: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12627: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12628: prvbackcast = 2;
12629: }
12630: else {
12631: 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);
12632: 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);
12633: goto end;
1.258 brouard 12634: }
1.230 brouard 12635: break;
1.258 brouard 12636: case 13:
1.307 brouard 12637: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12638: nresult++; /* Sum of resultlines */
12639: printf("Result %d: result:%s\n",nresult, resultline);
12640: if(nresult > MAXRESULTLINES){
12641: 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);
12642: 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);
12643: goto end;
12644: }
1.310 brouard 12645: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 12646: fprintf(ficparo,"result: %s\n",resultline);
12647: fprintf(ficres,"result: %s\n",resultline);
12648: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12649: } else
12650: goto end;
1.307 brouard 12651: break;
12652: case 14:
12653: printf("Error: Unknown command '%s'\n",line);
12654: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 12655: if(line[0] == ' ' || line[0] == '\n'){
12656: printf("It should not be an empty line '%s'\n",line);
12657: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
12658: }
1.307 brouard 12659: if(ncovmodel >=2 && nresult==0 ){
12660: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12661: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12662: }
1.307 brouard 12663: /* goto end; */
12664: break;
1.308 brouard 12665: case 15:
12666: printf("End of resultlines.\n");
12667: fprintf(ficlog,"End of resultlines.\n");
12668: break;
12669: default: /* parameterline =0 */
1.307 brouard 12670: nresult=1;
12671: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12672: } /* End switch parameterline */
12673: }while(endishere==0); /* End do */
1.126 brouard 12674:
1.230 brouard 12675: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12676: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12677:
12678: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12679: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12680: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12681: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12682: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12683: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12684: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12685: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12686: }else{
1.270 brouard 12687: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12688: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12689: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12690: if(prvforecast==1){
12691: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12692: jprojd=jproj1;
12693: mprojd=mproj1;
12694: anprojd=anproj1;
12695: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12696: jprojf=jproj2;
12697: mprojf=mproj2;
12698: anprojf=anproj2;
12699: } else if(prvforecast == 2){
12700: dateprojd=dateintmean;
12701: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12702: dateprojf=dateintmean+yrfproj;
12703: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12704: }
12705: if(prvbackcast==1){
12706: datebackd=(jback1+12*mback1+365*anback1)/365;
12707: jbackd=jback1;
12708: mbackd=mback1;
12709: anbackd=anback1;
12710: datebackf=(jback2+12*mback2+365*anback2)/365;
12711: jbackf=jback2;
12712: mbackf=mback2;
12713: anbackf=anback2;
12714: } else if(prvbackcast == 2){
12715: datebackd=dateintmean;
12716: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12717: datebackf=dateintmean-yrbproj;
12718: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12719: }
12720:
12721: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12722: }
12723: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12724: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12725: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12726:
1.225 brouard 12727: /*------------ free_vector -------------*/
12728: /* chdir(path); */
1.220 brouard 12729:
1.215 brouard 12730: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12731: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12732: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12733: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12734: free_lvector(num,firstobs,lastobs);
12735: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12736: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12737: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12738: fclose(ficparo);
12739: fclose(ficres);
1.220 brouard 12740:
12741:
1.186 brouard 12742: /* Other results (useful)*/
1.220 brouard 12743:
12744:
1.126 brouard 12745: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12746: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12747: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12748: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12749: fclose(ficrespl);
12750:
12751: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12752: /*#include "hpijx.h"*/
12753: hPijx(p, bage, fage);
1.145 brouard 12754: fclose(ficrespij);
1.227 brouard 12755:
1.220 brouard 12756: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12757: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12758: k=1;
1.126 brouard 12759: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12760:
1.269 brouard 12761: /* Prevalence for each covariate combination in probs[age][status][cov] */
12762: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12763: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12764: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12765: for(k=1;k<=ncovcombmax;k++)
12766: probs[i][j][k]=0.;
1.269 brouard 12767: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12768: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12769: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12770: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12771: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12772: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12773: for(k=1;k<=ncovcombmax;k++)
12774: mobaverages[i][j][k]=0.;
1.219 brouard 12775: mobaverage=mobaverages;
12776: if (mobilav!=0) {
1.235 brouard 12777: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12778: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12779: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12780: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12781: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12782: }
1.269 brouard 12783: } else if (mobilavproj !=0) {
1.235 brouard 12784: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12785: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12786: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12787: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12788: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12789: }
1.269 brouard 12790: }else{
12791: printf("Internal error moving average\n");
12792: fflush(stdout);
12793: exit(1);
1.219 brouard 12794: }
12795: }/* end if moving average */
1.227 brouard 12796:
1.126 brouard 12797: /*---------- Forecasting ------------------*/
1.296 brouard 12798: if(prevfcast==1){
12799: /* /\* if(stepm ==1){*\/ */
12800: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12801: /*This done previously after freqsummary.*/
12802: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12803: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12804:
12805: /* } else if (prvforecast==2){ */
12806: /* /\* if(stepm ==1){*\/ */
12807: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12808: /* } */
12809: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12810: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12811: }
1.269 brouard 12812:
1.296 brouard 12813: /* Prevbcasting */
12814: if(prevbcast==1){
1.219 brouard 12815: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12816: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12817: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12818:
12819: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12820:
12821: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12822:
1.219 brouard 12823: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12824: fclose(ficresplb);
12825:
1.222 brouard 12826: hBijx(p, bage, fage, mobaverage);
12827: fclose(ficrespijb);
1.219 brouard 12828:
1.296 brouard 12829: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12830: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12831: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12832: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12833: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12834: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12835:
12836:
1.269 brouard 12837: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12838:
12839:
1.269 brouard 12840: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12841: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12842: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12843: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12844: } /* end Prevbcasting */
1.268 brouard 12845:
1.186 brouard 12846:
12847: /* ------ Other prevalence ratios------------ */
1.126 brouard 12848:
1.215 brouard 12849: free_ivector(wav,1,imx);
12850: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12851: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12852: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12853:
12854:
1.127 brouard 12855: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12856:
1.201 brouard 12857: strcpy(filerese,"E_");
12858: strcat(filerese,fileresu);
1.126 brouard 12859: if((ficreseij=fopen(filerese,"w"))==NULL) {
12860: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12861: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12862: }
1.208 brouard 12863: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12864: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12865:
12866: pstamp(ficreseij);
1.219 brouard 12867:
1.235 brouard 12868: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12869: if (cptcovn < 1){i1=1;}
12870:
12871: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12872: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12873: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12874: continue;
1.219 brouard 12875: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12876: printf("\n#****** ");
1.225 brouard 12877: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12878: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12879: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12880: }
12881: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12882: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12883: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12884: }
12885: fprintf(ficreseij,"******\n");
1.235 brouard 12886: printf("******\n");
1.219 brouard 12887:
12888: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12889: oldm=oldms;savm=savms;
1.235 brouard 12890: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12891:
1.219 brouard 12892: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12893: }
12894: fclose(ficreseij);
1.208 brouard 12895: printf("done evsij\n");fflush(stdout);
12896: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12897:
1.218 brouard 12898:
1.227 brouard 12899: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12900:
1.201 brouard 12901: strcpy(filerest,"T_");
12902: strcat(filerest,fileresu);
1.127 brouard 12903: if((ficrest=fopen(filerest,"w"))==NULL) {
12904: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12905: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12906: }
1.208 brouard 12907: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12908: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12909: strcpy(fileresstde,"STDE_");
12910: strcat(fileresstde,fileresu);
1.126 brouard 12911: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12912: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12913: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12914: }
1.227 brouard 12915: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12916: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12917:
1.201 brouard 12918: strcpy(filerescve,"CVE_");
12919: strcat(filerescve,fileresu);
1.126 brouard 12920: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12921: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12922: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12923: }
1.227 brouard 12924: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12925: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12926:
1.201 brouard 12927: strcpy(fileresv,"V_");
12928: strcat(fileresv,fileresu);
1.126 brouard 12929: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12930: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12931: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12932: }
1.227 brouard 12933: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12934: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12935:
1.235 brouard 12936: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12937: if (cptcovn < 1){i1=1;}
12938:
12939: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12940: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12941: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12942: continue;
1.242 brouard 12943: printf("\n#****** Result for:");
12944: fprintf(ficrest,"\n#****** Result for:");
12945: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12946: for(j=1;j<=cptcoveff;j++){
12947: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12948: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12949: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12950: }
1.235 brouard 12951: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12952: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12953: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12954: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12955: }
1.208 brouard 12956: fprintf(ficrest,"******\n");
1.227 brouard 12957: fprintf(ficlog,"******\n");
12958: printf("******\n");
1.208 brouard 12959:
12960: fprintf(ficresstdeij,"\n#****** ");
12961: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12962: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12963: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12964: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12965: }
1.235 brouard 12966: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12967: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12968: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12969: }
1.208 brouard 12970: fprintf(ficresstdeij,"******\n");
12971: fprintf(ficrescveij,"******\n");
12972:
12973: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12974: /* pstamp(ficresvij); */
1.225 brouard 12975: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12976: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12977: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12978: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12979: }
1.208 brouard 12980: fprintf(ficresvij,"******\n");
12981:
12982: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12983: oldm=oldms;savm=savms;
1.235 brouard 12984: printf(" cvevsij ");
12985: fprintf(ficlog, " cvevsij ");
12986: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12987: printf(" end cvevsij \n ");
12988: fprintf(ficlog, " end cvevsij \n ");
12989:
12990: /*
12991: */
12992: /* goto endfree; */
12993:
12994: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12995: pstamp(ficrest);
12996:
1.269 brouard 12997: epj=vector(1,nlstate+1);
1.208 brouard 12998: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12999: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13000: cptcod= 0; /* To be deleted */
13001: printf("varevsij vpopbased=%d \n",vpopbased);
13002: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13003: 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 13004: 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 ");
13005: if(vpopbased==1)
13006: 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);
13007: else
1.288 brouard 13008: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13009: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13010: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13011: fprintf(ficrest,"\n");
13012: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13013: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13014: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13015: for(age=bage; age <=fage ;age++){
1.235 brouard 13016: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13017: if (vpopbased==1) {
13018: if(mobilav ==0){
13019: for(i=1; i<=nlstate;i++)
13020: prlim[i][i]=probs[(int)age][i][k];
13021: }else{ /* mobilav */
13022: for(i=1; i<=nlstate;i++)
13023: prlim[i][i]=mobaverage[(int)age][i][k];
13024: }
13025: }
1.219 brouard 13026:
1.227 brouard 13027: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13028: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13029: /* printf(" age %4.0f ",age); */
13030: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13031: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13032: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13033: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13034: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13035: }
13036: epj[nlstate+1] +=epj[j];
13037: }
13038: /* printf(" age %4.0f \n",age); */
1.219 brouard 13039:
1.227 brouard 13040: for(i=1, vepp=0.;i <=nlstate;i++)
13041: for(j=1;j <=nlstate;j++)
13042: vepp += vareij[i][j][(int)age];
13043: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13044: for(j=1;j <=nlstate;j++){
13045: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13046: }
13047: fprintf(ficrest,"\n");
13048: }
1.208 brouard 13049: } /* End vpopbased */
1.269 brouard 13050: free_vector(epj,1,nlstate+1);
1.208 brouard 13051: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13052: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13053: printf("done selection\n");fflush(stdout);
13054: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13055:
1.235 brouard 13056: } /* End k selection */
1.227 brouard 13057:
13058: printf("done State-specific expectancies\n");fflush(stdout);
13059: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13060:
1.288 brouard 13061: /* variance-covariance of forward period prevalence*/
1.269 brouard 13062: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13063:
1.227 brouard 13064:
1.290 brouard 13065: free_vector(weight,firstobs,lastobs);
1.227 brouard 13066: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13067: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13068: free_matrix(anint,1,maxwav,firstobs,lastobs);
13069: free_matrix(mint,1,maxwav,firstobs,lastobs);
13070: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13071: free_ivector(tab,1,NCOVMAX);
13072: fclose(ficresstdeij);
13073: fclose(ficrescveij);
13074: fclose(ficresvij);
13075: fclose(ficrest);
13076: fclose(ficpar);
13077:
13078:
1.126 brouard 13079: /*---------- End : free ----------------*/
1.219 brouard 13080: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13081: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13082: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13083: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13084: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13085: } /* mle==-3 arrives here for freeing */
1.227 brouard 13086: /* endfree:*/
13087: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13088: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13089: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13090: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13091: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13092: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13093: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13094: free_matrix(matcov,1,npar,1,npar);
13095: free_matrix(hess,1,npar,1,npar);
13096: /*free_vector(delti,1,npar);*/
13097: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13098: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13099: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13100: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13101:
13102: free_ivector(ncodemax,1,NCOVMAX);
13103: free_ivector(ncodemaxwundef,1,NCOVMAX);
13104: free_ivector(Dummy,-1,NCOVMAX);
13105: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13106: free_ivector(DummyV,1,NCOVMAX);
13107: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13108: free_ivector(Typevar,-1,NCOVMAX);
13109: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13110: free_ivector(TvarsQ,1,NCOVMAX);
13111: free_ivector(TvarsQind,1,NCOVMAX);
13112: free_ivector(TvarsD,1,NCOVMAX);
13113: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13114: free_ivector(TvarFD,1,NCOVMAX);
13115: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13116: free_ivector(TvarF,1,NCOVMAX);
13117: free_ivector(TvarFind,1,NCOVMAX);
13118: free_ivector(TvarV,1,NCOVMAX);
13119: free_ivector(TvarVind,1,NCOVMAX);
13120: free_ivector(TvarA,1,NCOVMAX);
13121: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13122: free_ivector(TvarFQ,1,NCOVMAX);
13123: free_ivector(TvarFQind,1,NCOVMAX);
13124: free_ivector(TvarVD,1,NCOVMAX);
13125: free_ivector(TvarVDind,1,NCOVMAX);
13126: free_ivector(TvarVQ,1,NCOVMAX);
13127: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13128: free_ivector(Tvarsel,1,NCOVMAX);
13129: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13130: free_ivector(Tposprod,1,NCOVMAX);
13131: free_ivector(Tprod,1,NCOVMAX);
13132: free_ivector(Tvaraff,1,NCOVMAX);
13133: free_ivector(invalidvarcomb,1,ncovcombmax);
13134: free_ivector(Tage,1,NCOVMAX);
13135: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13136: free_ivector(TmodelInvind,1,NCOVMAX);
13137: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13138:
13139: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13140: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13141: fflush(fichtm);
13142: fflush(ficgp);
13143:
1.227 brouard 13144:
1.126 brouard 13145: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13146: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13147: 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 13148: }else{
13149: printf("End of Imach\n");
13150: fprintf(ficlog,"End of Imach\n");
13151: }
13152: printf("See log file on %s\n",filelog);
13153: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13154: /*(void) gettimeofday(&end_time,&tzp);*/
13155: rend_time = time(NULL);
13156: end_time = *localtime(&rend_time);
13157: /* tml = *localtime(&end_time.tm_sec); */
13158: strcpy(strtend,asctime(&end_time));
1.126 brouard 13159: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13160: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13161: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13162:
1.157 brouard 13163: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13164: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13165: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13166: /* printf("Total time was %d uSec.\n", total_usecs);*/
13167: /* if(fileappend(fichtm,optionfilehtm)){ */
13168: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13169: fclose(fichtm);
13170: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13171: fclose(fichtmcov);
13172: fclose(ficgp);
13173: fclose(ficlog);
13174: /*------ End -----------*/
1.227 brouard 13175:
1.281 brouard 13176:
13177: /* Executes gnuplot */
1.227 brouard 13178:
13179: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13180: #ifdef WIN32
1.227 brouard 13181: if (_chdir(pathcd) != 0)
13182: printf("Can't move to directory %s!\n",path);
13183: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13184: #else
1.227 brouard 13185: if(chdir(pathcd) != 0)
13186: printf("Can't move to directory %s!\n", path);
13187: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13188: #endif
1.126 brouard 13189: printf("Current directory %s!\n",pathcd);
13190: /*strcat(plotcmd,CHARSEPARATOR);*/
13191: sprintf(plotcmd,"gnuplot");
1.157 brouard 13192: #ifdef _WIN32
1.126 brouard 13193: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13194: #endif
13195: if(!stat(plotcmd,&info)){
1.158 brouard 13196: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13197: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13198: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13199: }else
13200: strcpy(pplotcmd,plotcmd);
1.157 brouard 13201: #ifdef __unix
1.126 brouard 13202: strcpy(plotcmd,GNUPLOTPROGRAM);
13203: if(!stat(plotcmd,&info)){
1.158 brouard 13204: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13205: }else
13206: strcpy(pplotcmd,plotcmd);
13207: #endif
13208: }else
13209: strcpy(pplotcmd,plotcmd);
13210:
13211: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13212: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13213: strcpy(pplotcmd,plotcmd);
1.227 brouard 13214:
1.126 brouard 13215: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13216: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13217: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13218: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13219: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13220: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13221: strcpy(plotcmd,pplotcmd);
13222: }
1.126 brouard 13223: }
1.158 brouard 13224: printf(" Successful, please wait...");
1.126 brouard 13225: while (z[0] != 'q') {
13226: /* chdir(path); */
1.154 brouard 13227: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13228: scanf("%s",z);
13229: /* if (z[0] == 'c') system("./imach"); */
13230: if (z[0] == 'e') {
1.158 brouard 13231: #ifdef __APPLE__
1.152 brouard 13232: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13233: #elif __linux
13234: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13235: #else
1.152 brouard 13236: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13237: #endif
13238: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13239: system(pplotcmd);
1.126 brouard 13240: }
13241: else if (z[0] == 'g') system(plotcmd);
13242: else if (z[0] == 'q') exit(0);
13243: }
1.227 brouard 13244: end:
1.126 brouard 13245: while (z[0] != 'q') {
1.195 brouard 13246: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13247: scanf("%s",z);
13248: }
1.283 brouard 13249: printf("End\n");
1.282 brouard 13250: exit(0);
1.126 brouard 13251: }
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