Annotation of imach/src/imach.c, revision 1.245
1.245 ! brouard 1: /* $Id: imach.c,v 1.244 2016/09/02 07:17:34 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.245 ! brouard 4: Revision 1.244 2016/09/02 07:17:34 brouard
! 5: *** empty log message ***
! 6:
1.244 brouard 7: Revision 1.243 2016/09/02 06:45:35 brouard
8: *** empty log message ***
9:
1.243 brouard 10: Revision 1.242 2016/08/30 15:01:20 brouard
11: Summary: Fixing a lots
12:
1.242 brouard 13: Revision 1.241 2016/08/29 17:17:25 brouard
14: Summary: gnuplot problem in Back projection to fix
15:
1.241 brouard 16: Revision 1.240 2016/08/29 07:53:18 brouard
17: Summary: Better
18:
1.240 brouard 19: Revision 1.239 2016/08/26 15:51:03 brouard
20: Summary: Improvement in Powell output in order to copy and paste
21:
22: Author:
23:
1.239 brouard 24: Revision 1.238 2016/08/26 14:23:35 brouard
25: Summary: Starting tests of 0.99
26:
1.238 brouard 27: Revision 1.237 2016/08/26 09:20:19 brouard
28: Summary: to valgrind
29:
1.237 brouard 30: Revision 1.236 2016/08/25 10:50:18 brouard
31: *** empty log message ***
32:
1.236 brouard 33: Revision 1.235 2016/08/25 06:59:23 brouard
34: *** empty log message ***
35:
1.235 brouard 36: Revision 1.234 2016/08/23 16:51:20 brouard
37: *** empty log message ***
38:
1.234 brouard 39: Revision 1.233 2016/08/23 07:40:50 brouard
40: Summary: not working
41:
1.233 brouard 42: Revision 1.232 2016/08/22 14:20:21 brouard
43: Summary: not working
44:
1.232 brouard 45: Revision 1.231 2016/08/22 07:17:15 brouard
46: Summary: not working
47:
1.231 brouard 48: Revision 1.230 2016/08/22 06:55:53 brouard
49: Summary: Not working
50:
1.230 brouard 51: Revision 1.229 2016/07/23 09:45:53 brouard
52: Summary: Completing for func too
53:
1.229 brouard 54: Revision 1.228 2016/07/22 17:45:30 brouard
55: Summary: Fixing some arrays, still debugging
56:
1.227 brouard 57: Revision 1.226 2016/07/12 18:42:34 brouard
58: Summary: temp
59:
1.226 brouard 60: Revision 1.225 2016/07/12 08:40:03 brouard
61: Summary: saving but not running
62:
1.225 brouard 63: Revision 1.224 2016/07/01 13:16:01 brouard
64: Summary: Fixes
65:
1.224 brouard 66: Revision 1.223 2016/02/19 09:23:35 brouard
67: Summary: temporary
68:
1.223 brouard 69: Revision 1.222 2016/02/17 08:14:50 brouard
70: Summary: Probably last 0.98 stable version 0.98r6
71:
1.222 brouard 72: Revision 1.221 2016/02/15 23:35:36 brouard
73: Summary: minor bug
74:
1.220 brouard 75: Revision 1.219 2016/02/15 00:48:12 brouard
76: *** empty log message ***
77:
1.219 brouard 78: Revision 1.218 2016/02/12 11:29:23 brouard
79: Summary: 0.99 Back projections
80:
1.218 brouard 81: Revision 1.217 2015/12/23 17:18:31 brouard
82: Summary: Experimental backcast
83:
1.217 brouard 84: Revision 1.216 2015/12/18 17:32:11 brouard
85: Summary: 0.98r4 Warning and status=-2
86:
87: Version 0.98r4 is now:
88: - displaying an error when status is -1, date of interview unknown and date of death known;
89: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
90: Older changes concerning s=-2, dating from 2005 have been supersed.
91:
1.216 brouard 92: Revision 1.215 2015/12/16 08:52:24 brouard
93: Summary: 0.98r4 working
94:
1.215 brouard 95: Revision 1.214 2015/12/16 06:57:54 brouard
96: Summary: temporary not working
97:
1.214 brouard 98: Revision 1.213 2015/12/11 18:22:17 brouard
99: Summary: 0.98r4
100:
1.213 brouard 101: Revision 1.212 2015/11/21 12:47:24 brouard
102: Summary: minor typo
103:
1.212 brouard 104: Revision 1.211 2015/11/21 12:41:11 brouard
105: Summary: 0.98r3 with some graph of projected cross-sectional
106:
107: Author: Nicolas Brouard
108:
1.211 brouard 109: Revision 1.210 2015/11/18 17:41:20 brouard
110: Summary: Start working on projected prevalences
111:
1.210 brouard 112: Revision 1.209 2015/11/17 22:12:03 brouard
113: Summary: Adding ftolpl parameter
114: Author: N Brouard
115:
116: We had difficulties to get smoothed confidence intervals. It was due
117: to the period prevalence which wasn't computed accurately. The inner
118: parameter ftolpl is now an outer parameter of the .imach parameter
119: file after estepm. If ftolpl is small 1.e-4 and estepm too,
120: computation are long.
121:
1.209 brouard 122: Revision 1.208 2015/11/17 14:31:57 brouard
123: Summary: temporary
124:
1.208 brouard 125: Revision 1.207 2015/10/27 17:36:57 brouard
126: *** empty log message ***
127:
1.207 brouard 128: Revision 1.206 2015/10/24 07:14:11 brouard
129: *** empty log message ***
130:
1.206 brouard 131: Revision 1.205 2015/10/23 15:50:53 brouard
132: Summary: 0.98r3 some clarification for graphs on likelihood contributions
133:
1.205 brouard 134: Revision 1.204 2015/10/01 16:20:26 brouard
135: Summary: Some new graphs of contribution to likelihood
136:
1.204 brouard 137: Revision 1.203 2015/09/30 17:45:14 brouard
138: Summary: looking at better estimation of the hessian
139:
140: Also a better criteria for convergence to the period prevalence And
141: therefore adding the number of years needed to converge. (The
142: prevalence in any alive state shold sum to one
143:
1.203 brouard 144: Revision 1.202 2015/09/22 19:45:16 brouard
145: Summary: Adding some overall graph on contribution to likelihood. Might change
146:
1.202 brouard 147: Revision 1.201 2015/09/15 17:34:58 brouard
148: Summary: 0.98r0
149:
150: - Some new graphs like suvival functions
151: - Some bugs fixed like model=1+age+V2.
152:
1.201 brouard 153: Revision 1.200 2015/09/09 16:53:55 brouard
154: Summary: Big bug thanks to Flavia
155:
156: Even model=1+age+V2. did not work anymore
157:
1.200 brouard 158: Revision 1.199 2015/09/07 14:09:23 brouard
159: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
160:
1.199 brouard 161: Revision 1.198 2015/09/03 07:14:39 brouard
162: Summary: 0.98q5 Flavia
163:
1.198 brouard 164: Revision 1.197 2015/09/01 18:24:39 brouard
165: *** empty log message ***
166:
1.197 brouard 167: Revision 1.196 2015/08/18 23:17:52 brouard
168: Summary: 0.98q5
169:
1.196 brouard 170: Revision 1.195 2015/08/18 16:28:39 brouard
171: Summary: Adding a hack for testing purpose
172:
173: After reading the title, ftol and model lines, if the comment line has
174: a q, starting with #q, the answer at the end of the run is quit. It
175: permits to run test files in batch with ctest. The former workaround was
176: $ echo q | imach foo.imach
177:
1.195 brouard 178: Revision 1.194 2015/08/18 13:32:00 brouard
179: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
180:
1.194 brouard 181: Revision 1.193 2015/08/04 07:17:42 brouard
182: Summary: 0.98q4
183:
1.193 brouard 184: Revision 1.192 2015/07/16 16:49:02 brouard
185: Summary: Fixing some outputs
186:
1.192 brouard 187: Revision 1.191 2015/07/14 10:00:33 brouard
188: Summary: Some fixes
189:
1.191 brouard 190: Revision 1.190 2015/05/05 08:51:13 brouard
191: Summary: Adding digits in output parameters (7 digits instead of 6)
192:
193: Fix 1+age+.
194:
1.190 brouard 195: Revision 1.189 2015/04/30 14:45:16 brouard
196: Summary: 0.98q2
197:
1.189 brouard 198: Revision 1.188 2015/04/30 08:27:53 brouard
199: *** empty log message ***
200:
1.188 brouard 201: Revision 1.187 2015/04/29 09:11:15 brouard
202: *** empty log message ***
203:
1.187 brouard 204: Revision 1.186 2015/04/23 12:01:52 brouard
205: Summary: V1*age is working now, version 0.98q1
206:
207: Some codes had been disabled in order to simplify and Vn*age was
208: working in the optimization phase, ie, giving correct MLE parameters,
209: but, as usual, outputs were not correct and program core dumped.
210:
1.186 brouard 211: Revision 1.185 2015/03/11 13:26:42 brouard
212: Summary: Inclusion of compile and links command line for Intel Compiler
213:
1.185 brouard 214: Revision 1.184 2015/03/11 11:52:39 brouard
215: Summary: Back from Windows 8. Intel Compiler
216:
1.184 brouard 217: Revision 1.183 2015/03/10 20:34:32 brouard
218: Summary: 0.98q0, trying with directest, mnbrak fixed
219:
220: We use directest instead of original Powell test; probably no
221: incidence on the results, but better justifications;
222: We fixed Numerical Recipes mnbrak routine which was wrong and gave
223: wrong results.
224:
1.183 brouard 225: Revision 1.182 2015/02/12 08:19:57 brouard
226: Summary: Trying to keep directest which seems simpler and more general
227: Author: Nicolas Brouard
228:
1.182 brouard 229: Revision 1.181 2015/02/11 23:22:24 brouard
230: Summary: Comments on Powell added
231:
232: Author:
233:
1.181 brouard 234: Revision 1.180 2015/02/11 17:33:45 brouard
235: Summary: Finishing move from main to function (hpijx and prevalence_limit)
236:
1.180 brouard 237: Revision 1.179 2015/01/04 09:57:06 brouard
238: Summary: back to OS/X
239:
1.179 brouard 240: Revision 1.178 2015/01/04 09:35:48 brouard
241: *** empty log message ***
242:
1.178 brouard 243: Revision 1.177 2015/01/03 18:40:56 brouard
244: Summary: Still testing ilc32 on OSX
245:
1.177 brouard 246: Revision 1.176 2015/01/03 16:45:04 brouard
247: *** empty log message ***
248:
1.176 brouard 249: Revision 1.175 2015/01/03 16:33:42 brouard
250: *** empty log message ***
251:
1.175 brouard 252: Revision 1.174 2015/01/03 16:15:49 brouard
253: Summary: Still in cross-compilation
254:
1.174 brouard 255: Revision 1.173 2015/01/03 12:06:26 brouard
256: Summary: trying to detect cross-compilation
257:
1.173 brouard 258: Revision 1.172 2014/12/27 12:07:47 brouard
259: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
260:
1.172 brouard 261: Revision 1.171 2014/12/23 13:26:59 brouard
262: Summary: Back from Visual C
263:
264: Still problem with utsname.h on Windows
265:
1.171 brouard 266: Revision 1.170 2014/12/23 11:17:12 brouard
267: Summary: Cleaning some \%% back to %%
268:
269: The escape was mandatory for a specific compiler (which one?), but too many warnings.
270:
1.170 brouard 271: Revision 1.169 2014/12/22 23:08:31 brouard
272: Summary: 0.98p
273:
274: Outputs some informations on compiler used, OS etc. Testing on different platforms.
275:
1.169 brouard 276: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 277: Summary: update
1.169 brouard 278:
1.168 brouard 279: Revision 1.167 2014/12/22 13:50:56 brouard
280: Summary: Testing uname and compiler version and if compiled 32 or 64
281:
282: Testing on Linux 64
283:
1.167 brouard 284: Revision 1.166 2014/12/22 11:40:47 brouard
285: *** empty log message ***
286:
1.166 brouard 287: Revision 1.165 2014/12/16 11:20:36 brouard
288: Summary: After compiling on Visual C
289:
290: * imach.c (Module): Merging 1.61 to 1.162
291:
1.165 brouard 292: Revision 1.164 2014/12/16 10:52:11 brouard
293: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
294:
295: * imach.c (Module): Merging 1.61 to 1.162
296:
1.164 brouard 297: Revision 1.163 2014/12/16 10:30:11 brouard
298: * imach.c (Module): Merging 1.61 to 1.162
299:
1.163 brouard 300: Revision 1.162 2014/09/25 11:43:39 brouard
301: Summary: temporary backup 0.99!
302:
1.162 brouard 303: Revision 1.1 2014/09/16 11:06:58 brouard
304: Summary: With some code (wrong) for nlopt
305:
306: Author:
307:
308: Revision 1.161 2014/09/15 20:41:41 brouard
309: Summary: Problem with macro SQR on Intel compiler
310:
1.161 brouard 311: Revision 1.160 2014/09/02 09:24:05 brouard
312: *** empty log message ***
313:
1.160 brouard 314: Revision 1.159 2014/09/01 10:34:10 brouard
315: Summary: WIN32
316: Author: Brouard
317:
1.159 brouard 318: Revision 1.158 2014/08/27 17:11:51 brouard
319: *** empty log message ***
320:
1.158 brouard 321: Revision 1.157 2014/08/27 16:26:55 brouard
322: Summary: Preparing windows Visual studio version
323: Author: Brouard
324:
325: In order to compile on Visual studio, time.h is now correct and time_t
326: and tm struct should be used. difftime should be used but sometimes I
327: just make the differences in raw time format (time(&now).
328: Trying to suppress #ifdef LINUX
329: Add xdg-open for __linux in order to open default browser.
330:
1.157 brouard 331: Revision 1.156 2014/08/25 20:10:10 brouard
332: *** empty log message ***
333:
1.156 brouard 334: Revision 1.155 2014/08/25 18:32:34 brouard
335: Summary: New compile, minor changes
336: Author: Brouard
337:
1.155 brouard 338: Revision 1.154 2014/06/20 17:32:08 brouard
339: Summary: Outputs now all graphs of convergence to period prevalence
340:
1.154 brouard 341: Revision 1.153 2014/06/20 16:45:46 brouard
342: Summary: If 3 live state, convergence to period prevalence on same graph
343: Author: Brouard
344:
1.153 brouard 345: Revision 1.152 2014/06/18 17:54:09 brouard
346: Summary: open browser, use gnuplot on same dir than imach if not found in the path
347:
1.152 brouard 348: Revision 1.151 2014/06/18 16:43:30 brouard
349: *** empty log message ***
350:
1.151 brouard 351: Revision 1.150 2014/06/18 16:42:35 brouard
352: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
353: Author: brouard
354:
1.150 brouard 355: Revision 1.149 2014/06/18 15:51:14 brouard
356: Summary: Some fixes in parameter files errors
357: Author: Nicolas Brouard
358:
1.149 brouard 359: Revision 1.148 2014/06/17 17:38:48 brouard
360: Summary: Nothing new
361: Author: Brouard
362:
363: Just a new packaging for OS/X version 0.98nS
364:
1.148 brouard 365: Revision 1.147 2014/06/16 10:33:11 brouard
366: *** empty log message ***
367:
1.147 brouard 368: Revision 1.146 2014/06/16 10:20:28 brouard
369: Summary: Merge
370: Author: Brouard
371:
372: Merge, before building revised version.
373:
1.146 brouard 374: Revision 1.145 2014/06/10 21:23:15 brouard
375: Summary: Debugging with valgrind
376: Author: Nicolas Brouard
377:
378: Lot of changes in order to output the results with some covariates
379: After the Edimburgh REVES conference 2014, it seems mandatory to
380: improve the code.
381: No more memory valgrind error but a lot has to be done in order to
382: continue the work of splitting the code into subroutines.
383: Also, decodemodel has been improved. Tricode is still not
384: optimal. nbcode should be improved. Documentation has been added in
385: the source code.
386:
1.144 brouard 387: Revision 1.143 2014/01/26 09:45:38 brouard
388: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
389:
390: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
391: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
392:
1.143 brouard 393: Revision 1.142 2014/01/26 03:57:36 brouard
394: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
395:
396: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
397:
1.142 brouard 398: Revision 1.141 2014/01/26 02:42:01 brouard
399: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
400:
1.141 brouard 401: Revision 1.140 2011/09/02 10:37:54 brouard
402: Summary: times.h is ok with mingw32 now.
403:
1.140 brouard 404: Revision 1.139 2010/06/14 07:50:17 brouard
405: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
406: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
407:
1.139 brouard 408: Revision 1.138 2010/04/30 18:19:40 brouard
409: *** empty log message ***
410:
1.138 brouard 411: Revision 1.137 2010/04/29 18:11:38 brouard
412: (Module): Checking covariates for more complex models
413: than V1+V2. A lot of change to be done. Unstable.
414:
1.137 brouard 415: Revision 1.136 2010/04/26 20:30:53 brouard
416: (Module): merging some libgsl code. Fixing computation
417: of likelione (using inter/intrapolation if mle = 0) in order to
418: get same likelihood as if mle=1.
419: Some cleaning of code and comments added.
420:
1.136 brouard 421: Revision 1.135 2009/10/29 15:33:14 brouard
422: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
423:
1.135 brouard 424: Revision 1.134 2009/10/29 13:18:53 brouard
425: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
426:
1.134 brouard 427: Revision 1.133 2009/07/06 10:21:25 brouard
428: just nforces
429:
1.133 brouard 430: Revision 1.132 2009/07/06 08:22:05 brouard
431: Many tings
432:
1.132 brouard 433: Revision 1.131 2009/06/20 16:22:47 brouard
434: Some dimensions resccaled
435:
1.131 brouard 436: Revision 1.130 2009/05/26 06:44:34 brouard
437: (Module): Max Covariate is now set to 20 instead of 8. A
438: lot of cleaning with variables initialized to 0. Trying to make
439: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
440:
1.130 brouard 441: Revision 1.129 2007/08/31 13:49:27 lievre
442: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
443:
1.129 lievre 444: Revision 1.128 2006/06/30 13:02:05 brouard
445: (Module): Clarifications on computing e.j
446:
1.128 brouard 447: Revision 1.127 2006/04/28 18:11:50 brouard
448: (Module): Yes the sum of survivors was wrong since
449: imach-114 because nhstepm was no more computed in the age
450: loop. Now we define nhstepma in the age loop.
451: (Module): In order to speed up (in case of numerous covariates) we
452: compute health expectancies (without variances) in a first step
453: and then all the health expectancies with variances or standard
454: deviation (needs data from the Hessian matrices) which slows the
455: computation.
456: In the future we should be able to stop the program is only health
457: expectancies and graph are needed without standard deviations.
458:
1.127 brouard 459: Revision 1.126 2006/04/28 17:23:28 brouard
460: (Module): Yes the sum of survivors was wrong since
461: imach-114 because nhstepm was no more computed in the age
462: loop. Now we define nhstepma in the age loop.
463: Version 0.98h
464:
1.126 brouard 465: Revision 1.125 2006/04/04 15:20:31 lievre
466: Errors in calculation of health expectancies. Age was not initialized.
467: Forecasting file added.
468:
469: Revision 1.124 2006/03/22 17:13:53 lievre
470: Parameters are printed with %lf instead of %f (more numbers after the comma).
471: The log-likelihood is printed in the log file
472:
473: Revision 1.123 2006/03/20 10:52:43 brouard
474: * imach.c (Module): <title> changed, corresponds to .htm file
475: name. <head> headers where missing.
476:
477: * imach.c (Module): Weights can have a decimal point as for
478: English (a comma might work with a correct LC_NUMERIC environment,
479: otherwise the weight is truncated).
480: Modification of warning when the covariates values are not 0 or
481: 1.
482: Version 0.98g
483:
484: Revision 1.122 2006/03/20 09:45:41 brouard
485: (Module): Weights can have a decimal point as for
486: English (a comma might work with a correct LC_NUMERIC environment,
487: otherwise the weight is truncated).
488: Modification of warning when the covariates values are not 0 or
489: 1.
490: Version 0.98g
491:
492: Revision 1.121 2006/03/16 17:45:01 lievre
493: * imach.c (Module): Comments concerning covariates added
494:
495: * imach.c (Module): refinements in the computation of lli if
496: status=-2 in order to have more reliable computation if stepm is
497: not 1 month. Version 0.98f
498:
499: Revision 1.120 2006/03/16 15:10:38 lievre
500: (Module): refinements in the computation of lli if
501: status=-2 in order to have more reliable computation if stepm is
502: not 1 month. Version 0.98f
503:
504: Revision 1.119 2006/03/15 17:42:26 brouard
505: (Module): Bug if status = -2, the loglikelihood was
506: computed as likelihood omitting the logarithm. Version O.98e
507:
508: Revision 1.118 2006/03/14 18:20:07 brouard
509: (Module): varevsij Comments added explaining the second
510: table of variances if popbased=1 .
511: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
512: (Module): Function pstamp added
513: (Module): Version 0.98d
514:
515: Revision 1.117 2006/03/14 17:16:22 brouard
516: (Module): varevsij Comments added explaining the second
517: table of variances if popbased=1 .
518: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
519: (Module): Function pstamp added
520: (Module): Version 0.98d
521:
522: Revision 1.116 2006/03/06 10:29:27 brouard
523: (Module): Variance-covariance wrong links and
524: varian-covariance of ej. is needed (Saito).
525:
526: Revision 1.115 2006/02/27 12:17:45 brouard
527: (Module): One freematrix added in mlikeli! 0.98c
528:
529: Revision 1.114 2006/02/26 12:57:58 brouard
530: (Module): Some improvements in processing parameter
531: filename with strsep.
532:
533: Revision 1.113 2006/02/24 14:20:24 brouard
534: (Module): Memory leaks checks with valgrind and:
535: datafile was not closed, some imatrix were not freed and on matrix
536: allocation too.
537:
538: Revision 1.112 2006/01/30 09:55:26 brouard
539: (Module): Back to gnuplot.exe instead of wgnuplot.exe
540:
541: Revision 1.111 2006/01/25 20:38:18 brouard
542: (Module): Lots of cleaning and bugs added (Gompertz)
543: (Module): Comments can be added in data file. Missing date values
544: can be a simple dot '.'.
545:
546: Revision 1.110 2006/01/25 00:51:50 brouard
547: (Module): Lots of cleaning and bugs added (Gompertz)
548:
549: Revision 1.109 2006/01/24 19:37:15 brouard
550: (Module): Comments (lines starting with a #) are allowed in data.
551:
552: Revision 1.108 2006/01/19 18:05:42 lievre
553: Gnuplot problem appeared...
554: To be fixed
555:
556: Revision 1.107 2006/01/19 16:20:37 brouard
557: Test existence of gnuplot in imach path
558:
559: Revision 1.106 2006/01/19 13:24:36 brouard
560: Some cleaning and links added in html output
561:
562: Revision 1.105 2006/01/05 20:23:19 lievre
563: *** empty log message ***
564:
565: Revision 1.104 2005/09/30 16:11:43 lievre
566: (Module): sump fixed, loop imx fixed, and simplifications.
567: (Module): If the status is missing at the last wave but we know
568: that the person is alive, then we can code his/her status as -2
569: (instead of missing=-1 in earlier versions) and his/her
570: contributions to the likelihood is 1 - Prob of dying from last
571: health status (= 1-p13= p11+p12 in the easiest case of somebody in
572: the healthy state at last known wave). Version is 0.98
573:
574: Revision 1.103 2005/09/30 15:54:49 lievre
575: (Module): sump fixed, loop imx fixed, and simplifications.
576:
577: Revision 1.102 2004/09/15 17:31:30 brouard
578: Add the possibility to read data file including tab characters.
579:
580: Revision 1.101 2004/09/15 10:38:38 brouard
581: Fix on curr_time
582:
583: Revision 1.100 2004/07/12 18:29:06 brouard
584: Add version for Mac OS X. Just define UNIX in Makefile
585:
586: Revision 1.99 2004/06/05 08:57:40 brouard
587: *** empty log message ***
588:
589: Revision 1.98 2004/05/16 15:05:56 brouard
590: New version 0.97 . First attempt to estimate force of mortality
591: directly from the data i.e. without the need of knowing the health
592: state at each age, but using a Gompertz model: log u =a + b*age .
593: This is the basic analysis of mortality and should be done before any
594: other analysis, in order to test if the mortality estimated from the
595: cross-longitudinal survey is different from the mortality estimated
596: from other sources like vital statistic data.
597:
598: The same imach parameter file can be used but the option for mle should be -3.
599:
1.133 brouard 600: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 601: former routines in order to include the new code within the former code.
602:
603: The output is very simple: only an estimate of the intercept and of
604: the slope with 95% confident intervals.
605:
606: Current limitations:
607: A) Even if you enter covariates, i.e. with the
608: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
609: B) There is no computation of Life Expectancy nor Life Table.
610:
611: Revision 1.97 2004/02/20 13:25:42 lievre
612: Version 0.96d. Population forecasting command line is (temporarily)
613: suppressed.
614:
615: Revision 1.96 2003/07/15 15:38:55 brouard
616: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
617: rewritten within the same printf. Workaround: many printfs.
618:
619: Revision 1.95 2003/07/08 07:54:34 brouard
620: * imach.c (Repository):
621: (Repository): Using imachwizard code to output a more meaningful covariance
622: matrix (cov(a12,c31) instead of numbers.
623:
624: Revision 1.94 2003/06/27 13:00:02 brouard
625: Just cleaning
626:
627: Revision 1.93 2003/06/25 16:33:55 brouard
628: (Module): On windows (cygwin) function asctime_r doesn't
629: exist so I changed back to asctime which exists.
630: (Module): Version 0.96b
631:
632: Revision 1.92 2003/06/25 16:30:45 brouard
633: (Module): On windows (cygwin) function asctime_r doesn't
634: exist so I changed back to asctime which exists.
635:
636: Revision 1.91 2003/06/25 15:30:29 brouard
637: * imach.c (Repository): Duplicated warning errors corrected.
638: (Repository): Elapsed time after each iteration is now output. It
639: helps to forecast when convergence will be reached. Elapsed time
640: is stamped in powell. We created a new html file for the graphs
641: concerning matrix of covariance. It has extension -cov.htm.
642:
643: Revision 1.90 2003/06/24 12:34:15 brouard
644: (Module): Some bugs corrected for windows. Also, when
645: mle=-1 a template is output in file "or"mypar.txt with the design
646: of the covariance matrix to be input.
647:
648: Revision 1.89 2003/06/24 12:30:52 brouard
649: (Module): Some bugs corrected for windows. Also, when
650: mle=-1 a template is output in file "or"mypar.txt with the design
651: of the covariance matrix to be input.
652:
653: Revision 1.88 2003/06/23 17:54:56 brouard
654: * 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.
655:
656: Revision 1.87 2003/06/18 12:26:01 brouard
657: Version 0.96
658:
659: Revision 1.86 2003/06/17 20:04:08 brouard
660: (Module): Change position of html and gnuplot routines and added
661: routine fileappend.
662:
663: Revision 1.85 2003/06/17 13:12:43 brouard
664: * imach.c (Repository): Check when date of death was earlier that
665: current date of interview. It may happen when the death was just
666: prior to the death. In this case, dh was negative and likelihood
667: was wrong (infinity). We still send an "Error" but patch by
668: assuming that the date of death was just one stepm after the
669: interview.
670: (Repository): Because some people have very long ID (first column)
671: we changed int to long in num[] and we added a new lvector for
672: memory allocation. But we also truncated to 8 characters (left
673: truncation)
674: (Repository): No more line truncation errors.
675:
676: Revision 1.84 2003/06/13 21:44:43 brouard
677: * imach.c (Repository): Replace "freqsummary" at a correct
678: place. It differs from routine "prevalence" which may be called
679: many times. Probs is memory consuming and must be used with
680: parcimony.
681: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
682:
683: Revision 1.83 2003/06/10 13:39:11 lievre
684: *** empty log message ***
685:
686: Revision 1.82 2003/06/05 15:57:20 brouard
687: Add log in imach.c and fullversion number is now printed.
688:
689: */
690: /*
691: Interpolated Markov Chain
692:
693: Short summary of the programme:
694:
1.227 brouard 695: This program computes Healthy Life Expectancies or State-specific
696: (if states aren't health statuses) Expectancies from
697: cross-longitudinal data. Cross-longitudinal data consist in:
698:
699: -1- a first survey ("cross") where individuals from different ages
700: are interviewed on their health status or degree of disability (in
701: the case of a health survey which is our main interest)
702:
703: -2- at least a second wave of interviews ("longitudinal") which
704: measure each change (if any) in individual health status. Health
705: expectancies are computed from the time spent in each health state
706: according to a model. More health states you consider, more time is
707: necessary to reach the Maximum Likelihood of the parameters involved
708: in the model. The simplest model is the multinomial logistic model
709: where pij is the probability to be observed in state j at the second
710: wave conditional to be observed in state i at the first
711: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
712: etc , where 'age' is age and 'sex' is a covariate. If you want to
713: have a more complex model than "constant and age", you should modify
714: the program where the markup *Covariates have to be included here
715: again* invites you to do it. More covariates you add, slower the
1.126 brouard 716: convergence.
717:
718: The advantage of this computer programme, compared to a simple
719: multinomial logistic model, is clear when the delay between waves is not
720: identical for each individual. Also, if a individual missed an
721: intermediate interview, the information is lost, but taken into
722: account using an interpolation or extrapolation.
723:
724: hPijx is the probability to be observed in state i at age x+h
725: conditional to the observed state i at age x. The delay 'h' can be
726: split into an exact number (nh*stepm) of unobserved intermediate
727: states. This elementary transition (by month, quarter,
728: semester or year) is modelled as a multinomial logistic. The hPx
729: matrix is simply the matrix product of nh*stepm elementary matrices
730: and the contribution of each individual to the likelihood is simply
731: hPijx.
732:
733: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 734: of the life expectancies. It also computes the period (stable) prevalence.
735:
736: Back prevalence and projections:
1.227 brouard 737:
738: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
739: double agemaxpar, double ftolpl, int *ncvyearp, double
740: dateprev1,double dateprev2, int firstpass, int lastpass, int
741: mobilavproj)
742:
743: Computes the back prevalence limit for any combination of
744: covariate values k at any age between ageminpar and agemaxpar and
745: returns it in **bprlim. In the loops,
746:
747: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
748: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
749:
750: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 751: Computes for any combination of covariates k and any age between bage and fage
752: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
753: oldm=oldms;savm=savms;
1.227 brouard 754:
755: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 756: Computes the transition matrix starting at age 'age' over
757: 'nhstepm*hstepm*stepm' months (i.e. until
758: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 759: nhstepm*hstepm matrices.
760:
761: Returns p3mat[i][j][h] after calling
762: p3mat[i][j][h]=matprod2(newm,
763: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
764: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
765: oldm);
1.226 brouard 766:
767: Important routines
768:
769: - func (or funcone), computes logit (pij) distinguishing
770: o fixed variables (single or product dummies or quantitative);
771: o varying variables by:
772: (1) wave (single, product dummies, quantitative),
773: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
774: % fixed dummy (treated) or quantitative (not done because time-consuming);
775: % varying dummy (not done) or quantitative (not done);
776: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
777: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
778: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
779: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
780: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 781:
1.226 brouard 782:
783:
1.133 brouard 784: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
785: Institut national d'études démographiques, Paris.
1.126 brouard 786: This software have been partly granted by Euro-REVES, a concerted action
787: from the European Union.
788: It is copyrighted identically to a GNU software product, ie programme and
789: software can be distributed freely for non commercial use. Latest version
790: can be accessed at http://euroreves.ined.fr/imach .
791:
792: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
793: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
794:
795: **********************************************************************/
796: /*
797: main
798: read parameterfile
799: read datafile
800: concatwav
801: freqsummary
802: if (mle >= 1)
803: mlikeli
804: print results files
805: if mle==1
806: computes hessian
807: read end of parameter file: agemin, agemax, bage, fage, estepm
808: begin-prev-date,...
809: open gnuplot file
810: open html file
1.145 brouard 811: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
812: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
813: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
814: freexexit2 possible for memory heap.
815:
816: h Pij x | pij_nom ficrestpij
817: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
818: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
819: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
820:
821: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
822: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
823: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
824: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
825: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
826:
1.126 brouard 827: forecasting if prevfcast==1 prevforecast call prevalence()
828: health expectancies
829: Variance-covariance of DFLE
830: prevalence()
831: movingaverage()
832: varevsij()
833: if popbased==1 varevsij(,popbased)
834: total life expectancies
835: Variance of period (stable) prevalence
836: end
837: */
838:
1.187 brouard 839: /* #define DEBUG */
840: /* #define DEBUGBRENT */
1.203 brouard 841: /* #define DEBUGLINMIN */
842: /* #define DEBUGHESS */
843: #define DEBUGHESSIJ
1.224 brouard 844: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 845: #define POWELL /* Instead of NLOPT */
1.224 brouard 846: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 847: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
848: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 849:
850: #include <math.h>
851: #include <stdio.h>
852: #include <stdlib.h>
853: #include <string.h>
1.226 brouard 854: #include <ctype.h>
1.159 brouard 855:
856: #ifdef _WIN32
857: #include <io.h>
1.172 brouard 858: #include <windows.h>
859: #include <tchar.h>
1.159 brouard 860: #else
1.126 brouard 861: #include <unistd.h>
1.159 brouard 862: #endif
1.126 brouard 863:
864: #include <limits.h>
865: #include <sys/types.h>
1.171 brouard 866:
867: #if defined(__GNUC__)
868: #include <sys/utsname.h> /* Doesn't work on Windows */
869: #endif
870:
1.126 brouard 871: #include <sys/stat.h>
872: #include <errno.h>
1.159 brouard 873: /* extern int errno; */
1.126 brouard 874:
1.157 brouard 875: /* #ifdef LINUX */
876: /* #include <time.h> */
877: /* #include "timeval.h" */
878: /* #else */
879: /* #include <sys/time.h> */
880: /* #endif */
881:
1.126 brouard 882: #include <time.h>
883:
1.136 brouard 884: #ifdef GSL
885: #include <gsl/gsl_errno.h>
886: #include <gsl/gsl_multimin.h>
887: #endif
888:
1.167 brouard 889:
1.162 brouard 890: #ifdef NLOPT
891: #include <nlopt.h>
892: typedef struct {
893: double (* function)(double [] );
894: } myfunc_data ;
895: #endif
896:
1.126 brouard 897: /* #include <libintl.h> */
898: /* #define _(String) gettext (String) */
899:
1.141 brouard 900: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 901:
902: #define GNUPLOTPROGRAM "gnuplot"
903: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
904: #define FILENAMELENGTH 132
905:
906: #define GLOCK_ERROR_NOPATH -1 /* empty path */
907: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
908:
1.144 brouard 909: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
910: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 911:
912: #define NINTERVMAX 8
1.144 brouard 913: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
914: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
915: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 916: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 917: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
918: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 919: #define MAXN 20000
1.144 brouard 920: #define YEARM 12. /**< Number of months per year */
1.218 brouard 921: /* #define AGESUP 130 */
922: #define AGESUP 150
923: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 924: #define AGEBASE 40
1.194 brouard 925: #define AGEOVERFLOW 1.e20
1.164 brouard 926: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 927: #ifdef _WIN32
928: #define DIRSEPARATOR '\\'
929: #define CHARSEPARATOR "\\"
930: #define ODIRSEPARATOR '/'
931: #else
1.126 brouard 932: #define DIRSEPARATOR '/'
933: #define CHARSEPARATOR "/"
934: #define ODIRSEPARATOR '\\'
935: #endif
936:
1.245 ! brouard 937: /* $Id: imach.c,v 1.244 2016/09/02 07:17:34 brouard Exp $ */
1.126 brouard 938: /* $State: Exp $ */
1.196 brouard 939: #include "version.h"
940: char version[]=__IMACH_VERSION__;
1.224 brouard 941: char copyright[]="February 2016,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2018";
1.245 ! brouard 942: char fullversion[]="$Revision: 1.244 $ $Date: 2016/09/02 07:17:34 $";
1.126 brouard 943: char strstart[80];
944: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 945: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 946: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 947: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
948: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
949: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 950: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
951: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 952: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
953: int cptcovprodnoage=0; /**< Number of covariate products without age */
954: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 955: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
956: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 957: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 958: int nsd=0; /**< Total number of single dummy variables (output) */
959: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 960: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 961: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 962: int ntveff=0; /**< ntveff number of effective time varying variables */
963: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 964: int cptcov=0; /* Working variable */
1.218 brouard 965: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 966: int npar=NPARMAX;
967: int nlstate=2; /* Number of live states */
968: int ndeath=1; /* Number of dead states */
1.130 brouard 969: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 970: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 971: int popbased=0;
972:
973: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 974: int maxwav=0; /* Maxim number of waves */
975: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
976: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
977: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 978: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 979: int mle=1, weightopt=0;
1.126 brouard 980: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
981: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
982: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
983: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 984: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 985: int selected(int kvar); /* Is covariate kvar selected for printing results */
986:
1.130 brouard 987: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 988: double **matprod2(); /* test */
1.126 brouard 989: double **oldm, **newm, **savm; /* Working pointers to matrices */
990: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 991: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
992:
1.136 brouard 993: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 994: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 995: FILE *ficlog, *ficrespow;
1.130 brouard 996: int globpr=0; /* Global variable for printing or not */
1.126 brouard 997: double fretone; /* Only one call to likelihood */
1.130 brouard 998: long ipmx=0; /* Number of contributions */
1.126 brouard 999: double sw; /* Sum of weights */
1000: char filerespow[FILENAMELENGTH];
1001: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1002: FILE *ficresilk;
1003: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1004: FILE *ficresprobmorprev;
1005: FILE *fichtm, *fichtmcov; /* Html File */
1006: FILE *ficreseij;
1007: char filerese[FILENAMELENGTH];
1008: FILE *ficresstdeij;
1009: char fileresstde[FILENAMELENGTH];
1010: FILE *ficrescveij;
1011: char filerescve[FILENAMELENGTH];
1012: FILE *ficresvij;
1013: char fileresv[FILENAMELENGTH];
1014: FILE *ficresvpl;
1015: char fileresvpl[FILENAMELENGTH];
1016: char title[MAXLINE];
1.234 brouard 1017: char model[MAXLINE]; /**< The model line */
1.217 brouard 1018: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1019: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1020: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1021: char command[FILENAMELENGTH];
1022: int outcmd=0;
1023:
1.217 brouard 1024: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1025: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1026: char filelog[FILENAMELENGTH]; /* Log file */
1027: char filerest[FILENAMELENGTH];
1028: char fileregp[FILENAMELENGTH];
1029: char popfile[FILENAMELENGTH];
1030:
1031: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1032:
1.157 brouard 1033: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1034: /* struct timezone tzp; */
1035: /* extern int gettimeofday(); */
1036: struct tm tml, *gmtime(), *localtime();
1037:
1038: extern time_t time();
1039:
1040: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1041: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1042: struct tm tm;
1043:
1.126 brouard 1044: char strcurr[80], strfor[80];
1045:
1046: char *endptr;
1047: long lval;
1048: double dval;
1049:
1050: #define NR_END 1
1051: #define FREE_ARG char*
1052: #define FTOL 1.0e-10
1053:
1054: #define NRANSI
1.240 brouard 1055: #define ITMAX 200
1056: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1057:
1058: #define TOL 2.0e-4
1059:
1060: #define CGOLD 0.3819660
1061: #define ZEPS 1.0e-10
1062: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1063:
1064: #define GOLD 1.618034
1065: #define GLIMIT 100.0
1066: #define TINY 1.0e-20
1067:
1068: static double maxarg1,maxarg2;
1069: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1070: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1071:
1072: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1073: #define rint(a) floor(a+0.5)
1.166 brouard 1074: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1075: #define mytinydouble 1.0e-16
1.166 brouard 1076: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1077: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1078: /* static double dsqrarg; */
1079: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1080: static double sqrarg;
1081: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1082: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1083: int agegomp= AGEGOMP;
1084:
1085: int imx;
1086: int stepm=1;
1087: /* Stepm, step in month: minimum step interpolation*/
1088:
1089: int estepm;
1090: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1091:
1092: int m,nb;
1093: long *num;
1.197 brouard 1094: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1095: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1096: covariate for which somebody answered excluding
1097: undefined. Usually 2: 0 and 1. */
1098: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1099: covariate for which somebody answered including
1100: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1101: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1102: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1103: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1104: double *ageexmed,*agecens;
1105: double dateintmean=0;
1106:
1107: double *weight;
1108: int **s; /* Status */
1.141 brouard 1109: double *agedc;
1.145 brouard 1110: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1111: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1112: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1113: double **coqvar; /* Fixed quantitative covariate iqv */
1114: double ***cotvar; /* Time varying covariate itv */
1115: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1116: double idx;
1117: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1118: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1119: /*k 1 2 3 4 5 6 7 8 9 */
1120: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1121: /* Tndvar[k] 1 2 3 4 5 */
1122: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1123: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1124: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1125: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1126: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1127: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1128: /* Tprod[i]=k 4 7 */
1129: /* Tage[i]=k 5 8 */
1130: /* */
1131: /* Type */
1132: /* V 1 2 3 4 5 */
1133: /* F F V V V */
1134: /* D Q D D Q */
1135: /* */
1136: int *TvarsD;
1137: int *TvarsDind;
1138: int *TvarsQ;
1139: int *TvarsQind;
1140:
1.235 brouard 1141: #define MAXRESULTLINES 10
1142: int nresult=0;
1143: int TKresult[MAXRESULTLINES];
1.237 brouard 1144: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1145: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1146: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1147: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1148: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1149: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1150:
1.234 brouard 1151: /* 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 1152: 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 */
1153: 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 */
1154: 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 */
1155: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1156: 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 */
1157: 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 1158: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1159: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1160: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1161: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1162: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1163: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1164: 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 */
1165: 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 */
1166:
1.230 brouard 1167: int *Tvarsel; /**< Selected covariates for output */
1168: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1169: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1170: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1171: 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 1172: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1173: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1174: int *Tage;
1.227 brouard 1175: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1176: 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 1177: 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*/
1178: 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 1179: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1180: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1181: int **Tvard;
1182: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1183: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1184: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1185: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1186: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1187: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1188: double *lsurv, *lpop, *tpop;
1189:
1.231 brouard 1190: #define FD 1; /* Fixed dummy covariate */
1191: #define FQ 2; /* Fixed quantitative covariate */
1192: #define FP 3; /* Fixed product covariate */
1193: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1194: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1195: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1196: #define VD 10; /* Varying dummy covariate */
1197: #define VQ 11; /* Varying quantitative covariate */
1198: #define VP 12; /* Varying product covariate */
1199: #define VPDD 13; /* Varying product dummy*dummy covariate */
1200: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1201: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1202: #define APFD 16; /* Age product * fixed dummy covariate */
1203: #define APFQ 17; /* Age product * fixed quantitative covariate */
1204: #define APVD 18; /* Age product * varying dummy covariate */
1205: #define APVQ 19; /* Age product * varying quantitative covariate */
1206:
1207: #define FTYPE 1; /* Fixed covariate */
1208: #define VTYPE 2; /* Varying covariate (loop in wave) */
1209: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1210:
1211: struct kmodel{
1212: int maintype; /* main type */
1213: int subtype; /* subtype */
1214: };
1215: struct kmodel modell[NCOVMAX];
1216:
1.143 brouard 1217: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1218: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1219:
1220: /**************** split *************************/
1221: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1222: {
1223: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1224: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1225: */
1226: char *ss; /* pointer */
1.186 brouard 1227: int l1=0, l2=0; /* length counters */
1.126 brouard 1228:
1229: l1 = strlen(path ); /* length of path */
1230: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1231: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1232: if ( ss == NULL ) { /* no directory, so determine current directory */
1233: strcpy( name, path ); /* we got the fullname name because no directory */
1234: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1235: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1236: /* get current working directory */
1237: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1238: #ifdef WIN32
1239: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1240: #else
1241: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1242: #endif
1.126 brouard 1243: return( GLOCK_ERROR_GETCWD );
1244: }
1245: /* got dirc from getcwd*/
1246: printf(" DIRC = %s \n",dirc);
1.205 brouard 1247: } else { /* strip directory from path */
1.126 brouard 1248: ss++; /* after this, the filename */
1249: l2 = strlen( ss ); /* length of filename */
1250: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1251: strcpy( name, ss ); /* save file name */
1252: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1253: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1254: printf(" DIRC2 = %s \n",dirc);
1255: }
1256: /* We add a separator at the end of dirc if not exists */
1257: l1 = strlen( dirc ); /* length of directory */
1258: if( dirc[l1-1] != DIRSEPARATOR ){
1259: dirc[l1] = DIRSEPARATOR;
1260: dirc[l1+1] = 0;
1261: printf(" DIRC3 = %s \n",dirc);
1262: }
1263: ss = strrchr( name, '.' ); /* find last / */
1264: if (ss >0){
1265: ss++;
1266: strcpy(ext,ss); /* save extension */
1267: l1= strlen( name);
1268: l2= strlen(ss)+1;
1269: strncpy( finame, name, l1-l2);
1270: finame[l1-l2]= 0;
1271: }
1272:
1273: return( 0 ); /* we're done */
1274: }
1275:
1276:
1277: /******************************************/
1278:
1279: void replace_back_to_slash(char *s, char*t)
1280: {
1281: int i;
1282: int lg=0;
1283: i=0;
1284: lg=strlen(t);
1285: for(i=0; i<= lg; i++) {
1286: (s[i] = t[i]);
1287: if (t[i]== '\\') s[i]='/';
1288: }
1289: }
1290:
1.132 brouard 1291: char *trimbb(char *out, char *in)
1.137 brouard 1292: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1293: char *s;
1294: s=out;
1295: while (*in != '\0'){
1.137 brouard 1296: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1297: in++;
1298: }
1299: *out++ = *in++;
1300: }
1301: *out='\0';
1302: return s;
1303: }
1304:
1.187 brouard 1305: /* char *substrchaine(char *out, char *in, char *chain) */
1306: /* { */
1307: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1308: /* char *s, *t; */
1309: /* t=in;s=out; */
1310: /* while ((*in != *chain) && (*in != '\0')){ */
1311: /* *out++ = *in++; */
1312: /* } */
1313:
1314: /* /\* *in matches *chain *\/ */
1315: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1316: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1317: /* } */
1318: /* in--; chain--; */
1319: /* while ( (*in != '\0')){ */
1320: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1321: /* *out++ = *in++; */
1322: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1323: /* } */
1324: /* *out='\0'; */
1325: /* out=s; */
1326: /* return out; */
1327: /* } */
1328: char *substrchaine(char *out, char *in, char *chain)
1329: {
1330: /* Substract chain 'chain' from 'in', return and output 'out' */
1331: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1332:
1333: char *strloc;
1334:
1335: strcpy (out, in);
1336: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1337: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1338: if(strloc != NULL){
1339: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1340: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1341: /* strcpy (strloc, strloc +strlen(chain));*/
1342: }
1343: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1344: return out;
1345: }
1346:
1347:
1.145 brouard 1348: char *cutl(char *blocc, char *alocc, char *in, char occ)
1349: {
1.187 brouard 1350: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1351: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1352: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1353: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1354: */
1.160 brouard 1355: char *s, *t;
1.145 brouard 1356: t=in;s=in;
1357: while ((*in != occ) && (*in != '\0')){
1358: *alocc++ = *in++;
1359: }
1360: if( *in == occ){
1361: *(alocc)='\0';
1362: s=++in;
1363: }
1364:
1365: if (s == t) {/* occ not found */
1366: *(alocc-(in-s))='\0';
1367: in=s;
1368: }
1369: while ( *in != '\0'){
1370: *blocc++ = *in++;
1371: }
1372:
1373: *blocc='\0';
1374: return t;
1375: }
1.137 brouard 1376: char *cutv(char *blocc, char *alocc, char *in, char occ)
1377: {
1.187 brouard 1378: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1379: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1380: gives blocc="abcdef2ghi" and alocc="j".
1381: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1382: */
1383: char *s, *t;
1384: t=in;s=in;
1385: while (*in != '\0'){
1386: while( *in == occ){
1387: *blocc++ = *in++;
1388: s=in;
1389: }
1390: *blocc++ = *in++;
1391: }
1392: if (s == t) /* occ not found */
1393: *(blocc-(in-s))='\0';
1394: else
1395: *(blocc-(in-s)-1)='\0';
1396: in=s;
1397: while ( *in != '\0'){
1398: *alocc++ = *in++;
1399: }
1400:
1401: *alocc='\0';
1402: return s;
1403: }
1404:
1.126 brouard 1405: int nbocc(char *s, char occ)
1406: {
1407: int i,j=0;
1408: int lg=20;
1409: i=0;
1410: lg=strlen(s);
1411: for(i=0; i<= lg; i++) {
1.234 brouard 1412: if (s[i] == occ ) j++;
1.126 brouard 1413: }
1414: return j;
1415: }
1416:
1.137 brouard 1417: /* void cutv(char *u,char *v, char*t, char occ) */
1418: /* { */
1419: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1420: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1421: /* gives u="abcdef2ghi" and v="j" *\/ */
1422: /* int i,lg,j,p=0; */
1423: /* i=0; */
1424: /* lg=strlen(t); */
1425: /* for(j=0; j<=lg-1; j++) { */
1426: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1427: /* } */
1.126 brouard 1428:
1.137 brouard 1429: /* for(j=0; j<p; j++) { */
1430: /* (u[j] = t[j]); */
1431: /* } */
1432: /* u[p]='\0'; */
1.126 brouard 1433:
1.137 brouard 1434: /* for(j=0; j<= lg; j++) { */
1435: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1436: /* } */
1437: /* } */
1.126 brouard 1438:
1.160 brouard 1439: #ifdef _WIN32
1440: char * strsep(char **pp, const char *delim)
1441: {
1442: char *p, *q;
1443:
1444: if ((p = *pp) == NULL)
1445: return 0;
1446: if ((q = strpbrk (p, delim)) != NULL)
1447: {
1448: *pp = q + 1;
1449: *q = '\0';
1450: }
1451: else
1452: *pp = 0;
1453: return p;
1454: }
1455: #endif
1456:
1.126 brouard 1457: /********************** nrerror ********************/
1458:
1459: void nrerror(char error_text[])
1460: {
1461: fprintf(stderr,"ERREUR ...\n");
1462: fprintf(stderr,"%s\n",error_text);
1463: exit(EXIT_FAILURE);
1464: }
1465: /*********************** vector *******************/
1466: double *vector(int nl, int nh)
1467: {
1468: double *v;
1469: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1470: if (!v) nrerror("allocation failure in vector");
1471: return v-nl+NR_END;
1472: }
1473:
1474: /************************ free vector ******************/
1475: void free_vector(double*v, int nl, int nh)
1476: {
1477: free((FREE_ARG)(v+nl-NR_END));
1478: }
1479:
1480: /************************ivector *******************************/
1481: int *ivector(long nl,long nh)
1482: {
1483: int *v;
1484: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1485: if (!v) nrerror("allocation failure in ivector");
1486: return v-nl+NR_END;
1487: }
1488:
1489: /******************free ivector **************************/
1490: void free_ivector(int *v, long nl, long nh)
1491: {
1492: free((FREE_ARG)(v+nl-NR_END));
1493: }
1494:
1495: /************************lvector *******************************/
1496: long *lvector(long nl,long nh)
1497: {
1498: long *v;
1499: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1500: if (!v) nrerror("allocation failure in ivector");
1501: return v-nl+NR_END;
1502: }
1503:
1504: /******************free lvector **************************/
1505: void free_lvector(long *v, long nl, long nh)
1506: {
1507: free((FREE_ARG)(v+nl-NR_END));
1508: }
1509:
1510: /******************* imatrix *******************************/
1511: int **imatrix(long nrl, long nrh, long ncl, long nch)
1512: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1513: {
1514: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1515: int **m;
1516:
1517: /* allocate pointers to rows */
1518: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1519: if (!m) nrerror("allocation failure 1 in matrix()");
1520: m += NR_END;
1521: m -= nrl;
1522:
1523:
1524: /* allocate rows and set pointers to them */
1525: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1526: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1527: m[nrl] += NR_END;
1528: m[nrl] -= ncl;
1529:
1530: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1531:
1532: /* return pointer to array of pointers to rows */
1533: return m;
1534: }
1535:
1536: /****************** free_imatrix *************************/
1537: void free_imatrix(m,nrl,nrh,ncl,nch)
1538: int **m;
1539: long nch,ncl,nrh,nrl;
1540: /* free an int matrix allocated by imatrix() */
1541: {
1542: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1543: free((FREE_ARG) (m+nrl-NR_END));
1544: }
1545:
1546: /******************* matrix *******************************/
1547: double **matrix(long nrl, long nrh, long ncl, long nch)
1548: {
1549: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1550: double **m;
1551:
1552: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1553: if (!m) nrerror("allocation failure 1 in matrix()");
1554: m += NR_END;
1555: m -= nrl;
1556:
1557: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1558: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1559: m[nrl] += NR_END;
1560: m[nrl] -= ncl;
1561:
1562: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1563: return m;
1.145 brouard 1564: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1565: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1566: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1567: */
1568: }
1569:
1570: /*************************free matrix ************************/
1571: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1572: {
1573: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1574: free((FREE_ARG)(m+nrl-NR_END));
1575: }
1576:
1577: /******************* ma3x *******************************/
1578: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1579: {
1580: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1581: double ***m;
1582:
1583: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1584: if (!m) nrerror("allocation failure 1 in matrix()");
1585: m += NR_END;
1586: m -= nrl;
1587:
1588: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1589: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1590: m[nrl] += NR_END;
1591: m[nrl] -= ncl;
1592:
1593: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1594:
1595: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1596: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1597: m[nrl][ncl] += NR_END;
1598: m[nrl][ncl] -= nll;
1599: for (j=ncl+1; j<=nch; j++)
1600: m[nrl][j]=m[nrl][j-1]+nlay;
1601:
1602: for (i=nrl+1; i<=nrh; i++) {
1603: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1604: for (j=ncl+1; j<=nch; j++)
1605: m[i][j]=m[i][j-1]+nlay;
1606: }
1607: return m;
1608: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1609: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1610: */
1611: }
1612:
1613: /*************************free ma3x ************************/
1614: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1615: {
1616: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1617: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1618: free((FREE_ARG)(m+nrl-NR_END));
1619: }
1620:
1621: /*************** function subdirf ***********/
1622: char *subdirf(char fileres[])
1623: {
1624: /* Caution optionfilefiname is hidden */
1625: strcpy(tmpout,optionfilefiname);
1626: strcat(tmpout,"/"); /* Add to the right */
1627: strcat(tmpout,fileres);
1628: return tmpout;
1629: }
1630:
1631: /*************** function subdirf2 ***********/
1632: char *subdirf2(char fileres[], char *preop)
1633: {
1634:
1635: /* Caution optionfilefiname is hidden */
1636: strcpy(tmpout,optionfilefiname);
1637: strcat(tmpout,"/");
1638: strcat(tmpout,preop);
1639: strcat(tmpout,fileres);
1640: return tmpout;
1641: }
1642:
1643: /*************** function subdirf3 ***********/
1644: char *subdirf3(char fileres[], char *preop, char *preop2)
1645: {
1646:
1647: /* Caution optionfilefiname is hidden */
1648: strcpy(tmpout,optionfilefiname);
1649: strcat(tmpout,"/");
1650: strcat(tmpout,preop);
1651: strcat(tmpout,preop2);
1652: strcat(tmpout,fileres);
1653: return tmpout;
1654: }
1.213 brouard 1655:
1656: /*************** function subdirfext ***********/
1657: char *subdirfext(char fileres[], char *preop, char *postop)
1658: {
1659:
1660: strcpy(tmpout,preop);
1661: strcat(tmpout,fileres);
1662: strcat(tmpout,postop);
1663: return tmpout;
1664: }
1.126 brouard 1665:
1.213 brouard 1666: /*************** function subdirfext3 ***********/
1667: char *subdirfext3(char fileres[], char *preop, char *postop)
1668: {
1669:
1670: /* Caution optionfilefiname is hidden */
1671: strcpy(tmpout,optionfilefiname);
1672: strcat(tmpout,"/");
1673: strcat(tmpout,preop);
1674: strcat(tmpout,fileres);
1675: strcat(tmpout,postop);
1676: return tmpout;
1677: }
1678:
1.162 brouard 1679: char *asc_diff_time(long time_sec, char ascdiff[])
1680: {
1681: long sec_left, days, hours, minutes;
1682: days = (time_sec) / (60*60*24);
1683: sec_left = (time_sec) % (60*60*24);
1684: hours = (sec_left) / (60*60) ;
1685: sec_left = (sec_left) %(60*60);
1686: minutes = (sec_left) /60;
1687: sec_left = (sec_left) % (60);
1688: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1689: return ascdiff;
1690: }
1691:
1.126 brouard 1692: /***************** f1dim *************************/
1693: extern int ncom;
1694: extern double *pcom,*xicom;
1695: extern double (*nrfunc)(double []);
1696:
1697: double f1dim(double x)
1698: {
1699: int j;
1700: double f;
1701: double *xt;
1702:
1703: xt=vector(1,ncom);
1704: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1705: f=(*nrfunc)(xt);
1706: free_vector(xt,1,ncom);
1707: return f;
1708: }
1709:
1710: /*****************brent *************************/
1711: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1712: {
1713: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1714: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1715: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1716: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1717: * returned function value.
1718: */
1.126 brouard 1719: int iter;
1720: double a,b,d,etemp;
1.159 brouard 1721: double fu=0,fv,fw,fx;
1.164 brouard 1722: double ftemp=0.;
1.126 brouard 1723: double p,q,r,tol1,tol2,u,v,w,x,xm;
1724: double e=0.0;
1725:
1726: a=(ax < cx ? ax : cx);
1727: b=(ax > cx ? ax : cx);
1728: x=w=v=bx;
1729: fw=fv=fx=(*f)(x);
1730: for (iter=1;iter<=ITMAX;iter++) {
1731: xm=0.5*(a+b);
1732: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1733: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1734: printf(".");fflush(stdout);
1735: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1736: #ifdef DEBUGBRENT
1.126 brouard 1737: 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);
1738: 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);
1739: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1740: #endif
1741: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1742: *xmin=x;
1743: return fx;
1744: }
1745: ftemp=fu;
1746: if (fabs(e) > tol1) {
1747: r=(x-w)*(fx-fv);
1748: q=(x-v)*(fx-fw);
1749: p=(x-v)*q-(x-w)*r;
1750: q=2.0*(q-r);
1751: if (q > 0.0) p = -p;
1752: q=fabs(q);
1753: etemp=e;
1754: e=d;
1755: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1756: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1757: else {
1.224 brouard 1758: d=p/q;
1759: u=x+d;
1760: if (u-a < tol2 || b-u < tol2)
1761: d=SIGN(tol1,xm-x);
1.126 brouard 1762: }
1763: } else {
1764: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1765: }
1766: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1767: fu=(*f)(u);
1768: if (fu <= fx) {
1769: if (u >= x) a=x; else b=x;
1770: SHFT(v,w,x,u)
1.183 brouard 1771: SHFT(fv,fw,fx,fu)
1772: } else {
1773: if (u < x) a=u; else b=u;
1774: if (fu <= fw || w == x) {
1.224 brouard 1775: v=w;
1776: w=u;
1777: fv=fw;
1778: fw=fu;
1.183 brouard 1779: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1780: v=u;
1781: fv=fu;
1.183 brouard 1782: }
1783: }
1.126 brouard 1784: }
1785: nrerror("Too many iterations in brent");
1786: *xmin=x;
1787: return fx;
1788: }
1789:
1790: /****************** mnbrak ***********************/
1791:
1792: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1793: double (*func)(double))
1.183 brouard 1794: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1795: the downhill direction (defined by the function as evaluated at the initial points) and returns
1796: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1797: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1798: */
1.126 brouard 1799: double ulim,u,r,q, dum;
1800: double fu;
1.187 brouard 1801:
1802: double scale=10.;
1803: int iterscale=0;
1804:
1805: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1806: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1807:
1808:
1809: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1810: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1811: /* *bx = *ax - (*ax - *bx)/scale; */
1812: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1813: /* } */
1814:
1.126 brouard 1815: if (*fb > *fa) {
1816: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1817: SHFT(dum,*fb,*fa,dum)
1818: }
1.126 brouard 1819: *cx=(*bx)+GOLD*(*bx-*ax);
1820: *fc=(*func)(*cx);
1.183 brouard 1821: #ifdef DEBUG
1.224 brouard 1822: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1823: 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 1824: #endif
1.224 brouard 1825: 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 1826: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1827: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1828: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1829: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1830: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1831: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1832: fu=(*func)(u);
1.163 brouard 1833: #ifdef DEBUG
1834: /* f(x)=A(x-u)**2+f(u) */
1835: double A, fparabu;
1836: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1837: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1838: 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);
1839: 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 1840: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1841: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1842: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1843: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1844: #endif
1.184 brouard 1845: #ifdef MNBRAKORIGINAL
1.183 brouard 1846: #else
1.191 brouard 1847: /* if (fu > *fc) { */
1848: /* #ifdef DEBUG */
1849: /* printf("mnbrak4 fu > fc \n"); */
1850: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1851: /* #endif */
1852: /* /\* 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 *\\/ *\/ */
1853: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1854: /* dum=u; /\* Shifting c and u *\/ */
1855: /* u = *cx; */
1856: /* *cx = dum; */
1857: /* dum = fu; */
1858: /* fu = *fc; */
1859: /* *fc =dum; */
1860: /* } else { /\* end *\/ */
1861: /* #ifdef DEBUG */
1862: /* printf("mnbrak3 fu < fc \n"); */
1863: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1864: /* #endif */
1865: /* dum=u; /\* Shifting c and u *\/ */
1866: /* u = *cx; */
1867: /* *cx = dum; */
1868: /* dum = fu; */
1869: /* fu = *fc; */
1870: /* *fc =dum; */
1871: /* } */
1.224 brouard 1872: #ifdef DEBUGMNBRAK
1873: double A, fparabu;
1874: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1875: fparabu= *fa - A*(*ax-u)*(*ax-u);
1876: 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);
1877: 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 1878: #endif
1.191 brouard 1879: dum=u; /* Shifting c and u */
1880: u = *cx;
1881: *cx = dum;
1882: dum = fu;
1883: fu = *fc;
1884: *fc =dum;
1.183 brouard 1885: #endif
1.162 brouard 1886: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1887: #ifdef DEBUG
1.224 brouard 1888: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1889: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1890: #endif
1.126 brouard 1891: fu=(*func)(u);
1892: if (fu < *fc) {
1.183 brouard 1893: #ifdef DEBUG
1.224 brouard 1894: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1895: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1896: #endif
1897: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1898: SHFT(*fb,*fc,fu,(*func)(u))
1899: #ifdef DEBUG
1900: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1901: #endif
1902: }
1.162 brouard 1903: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1904: #ifdef DEBUG
1.224 brouard 1905: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1906: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1907: #endif
1.126 brouard 1908: u=ulim;
1909: fu=(*func)(u);
1.183 brouard 1910: } else { /* u could be left to b (if r > q parabola has a maximum) */
1911: #ifdef DEBUG
1.224 brouard 1912: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1913: 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 1914: #endif
1.126 brouard 1915: u=(*cx)+GOLD*(*cx-*bx);
1916: fu=(*func)(u);
1.224 brouard 1917: #ifdef DEBUG
1918: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1919: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1920: #endif
1.183 brouard 1921: } /* end tests */
1.126 brouard 1922: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1923: SHFT(*fa,*fb,*fc,fu)
1924: #ifdef DEBUG
1.224 brouard 1925: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1926: 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 1927: #endif
1928: } /* 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 1929: }
1930:
1931: /*************** linmin ************************/
1.162 brouard 1932: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1933: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1934: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1935: the value of func at the returned location p . This is actually all accomplished by calling the
1936: routines mnbrak and brent .*/
1.126 brouard 1937: int ncom;
1938: double *pcom,*xicom;
1939: double (*nrfunc)(double []);
1940:
1.224 brouard 1941: #ifdef LINMINORIGINAL
1.126 brouard 1942: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1943: #else
1944: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1945: #endif
1.126 brouard 1946: {
1947: double brent(double ax, double bx, double cx,
1948: double (*f)(double), double tol, double *xmin);
1949: double f1dim(double x);
1950: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1951: double *fc, double (*func)(double));
1952: int j;
1953: double xx,xmin,bx,ax;
1954: double fx,fb,fa;
1.187 brouard 1955:
1.203 brouard 1956: #ifdef LINMINORIGINAL
1957: #else
1958: double scale=10., axs, xxs; /* Scale added for infinity */
1959: #endif
1960:
1.126 brouard 1961: ncom=n;
1962: pcom=vector(1,n);
1963: xicom=vector(1,n);
1964: nrfunc=func;
1965: for (j=1;j<=n;j++) {
1966: pcom[j]=p[j];
1.202 brouard 1967: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1968: }
1.187 brouard 1969:
1.203 brouard 1970: #ifdef LINMINORIGINAL
1971: xx=1.;
1972: #else
1973: axs=0.0;
1974: xxs=1.;
1975: do{
1976: xx= xxs;
1977: #endif
1.187 brouard 1978: ax=0.;
1979: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1980: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1981: /* 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)) */
1982: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1983: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1984: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
1985: /* 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 1986: #ifdef LINMINORIGINAL
1987: #else
1988: if (fx != fx){
1.224 brouard 1989: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
1990: printf("|");
1991: fprintf(ficlog,"|");
1.203 brouard 1992: #ifdef DEBUGLINMIN
1.224 brouard 1993: 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 1994: #endif
1995: }
1.224 brouard 1996: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 1997: #endif
1998:
1.191 brouard 1999: #ifdef DEBUGLINMIN
2000: 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 2001: 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 2002: #endif
1.224 brouard 2003: #ifdef LINMINORIGINAL
2004: #else
2005: if(fb == fx){ /* Flat function in the direction */
2006: xmin=xx;
2007: *flat=1;
2008: }else{
2009: *flat=0;
2010: #endif
2011: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2012: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2013: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2014: /* fmin = f(p[j] + xmin * xi[j]) */
2015: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2016: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2017: #ifdef DEBUG
1.224 brouard 2018: 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);
2019: 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);
2020: #endif
2021: #ifdef LINMINORIGINAL
2022: #else
2023: }
1.126 brouard 2024: #endif
1.191 brouard 2025: #ifdef DEBUGLINMIN
2026: printf("linmin end ");
1.202 brouard 2027: fprintf(ficlog,"linmin end ");
1.191 brouard 2028: #endif
1.126 brouard 2029: for (j=1;j<=n;j++) {
1.203 brouard 2030: #ifdef LINMINORIGINAL
2031: xi[j] *= xmin;
2032: #else
2033: #ifdef DEBUGLINMIN
2034: if(xxs <1.0)
2035: printf(" before xi[%d]=%12.8f", j,xi[j]);
2036: #endif
2037: 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) */
2038: #ifdef DEBUGLINMIN
2039: if(xxs <1.0)
2040: 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 );
2041: #endif
2042: #endif
1.187 brouard 2043: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2044: }
1.191 brouard 2045: #ifdef DEBUGLINMIN
1.203 brouard 2046: printf("\n");
1.191 brouard 2047: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2048: 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 2049: for (j=1;j<=n;j++) {
1.202 brouard 2050: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2051: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2052: if(j % ncovmodel == 0){
1.191 brouard 2053: printf("\n");
1.202 brouard 2054: fprintf(ficlog,"\n");
2055: }
1.191 brouard 2056: }
1.203 brouard 2057: #else
1.191 brouard 2058: #endif
1.126 brouard 2059: free_vector(xicom,1,n);
2060: free_vector(pcom,1,n);
2061: }
2062:
2063:
2064: /*************** powell ************************/
1.162 brouard 2065: /*
2066: Minimization of a function func of n variables. Input consists of an initial starting point
2067: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2068: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2069: such that failure to decrease by more than this amount on one iteration signals doneness. On
2070: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2071: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2072: */
1.224 brouard 2073: #ifdef LINMINORIGINAL
2074: #else
2075: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2076: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2077: #endif
1.126 brouard 2078: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2079: double (*func)(double []))
2080: {
1.224 brouard 2081: #ifdef LINMINORIGINAL
2082: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2083: double (*func)(double []));
1.224 brouard 2084: #else
1.241 brouard 2085: void linmin(double p[], double xi[], int n, double *fret,
2086: double (*func)(double []),int *flat);
1.224 brouard 2087: #endif
1.239 brouard 2088: int i,ibig,j,jk,k;
1.126 brouard 2089: double del,t,*pt,*ptt,*xit;
1.181 brouard 2090: double directest;
1.126 brouard 2091: double fp,fptt;
2092: double *xits;
2093: int niterf, itmp;
1.224 brouard 2094: #ifdef LINMINORIGINAL
2095: #else
2096:
2097: flatdir=ivector(1,n);
2098: for (j=1;j<=n;j++) flatdir[j]=0;
2099: #endif
1.126 brouard 2100:
2101: pt=vector(1,n);
2102: ptt=vector(1,n);
2103: xit=vector(1,n);
2104: xits=vector(1,n);
2105: *fret=(*func)(p);
2106: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2107: rcurr_time = time(NULL);
1.126 brouard 2108: for (*iter=1;;++(*iter)) {
1.187 brouard 2109: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2110: ibig=0;
2111: del=0.0;
1.157 brouard 2112: rlast_time=rcurr_time;
2113: /* (void) gettimeofday(&curr_time,&tzp); */
2114: rcurr_time = time(NULL);
2115: curr_time = *localtime(&rcurr_time);
2116: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2117: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2118: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2119: for (i=1;i<=n;i++) {
1.126 brouard 2120: fprintf(ficrespow," %.12lf", p[i]);
2121: }
1.239 brouard 2122: fprintf(ficrespow,"\n");fflush(ficrespow);
2123: printf("\n#model= 1 + age ");
2124: fprintf(ficlog,"\n#model= 1 + age ");
2125: if(nagesqr==1){
1.241 brouard 2126: printf(" + age*age ");
2127: fprintf(ficlog," + age*age ");
1.239 brouard 2128: }
2129: for(j=1;j <=ncovmodel-2;j++){
2130: if(Typevar[j]==0) {
2131: printf(" + V%d ",Tvar[j]);
2132: fprintf(ficlog," + V%d ",Tvar[j]);
2133: }else if(Typevar[j]==1) {
2134: printf(" + V%d*age ",Tvar[j]);
2135: fprintf(ficlog," + V%d*age ",Tvar[j]);
2136: }else if(Typevar[j]==2) {
2137: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2138: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2139: }
2140: }
1.126 brouard 2141: printf("\n");
1.239 brouard 2142: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2143: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2144: fprintf(ficlog,"\n");
1.239 brouard 2145: for(i=1,jk=1; i <=nlstate; i++){
2146: for(k=1; k <=(nlstate+ndeath); k++){
2147: if (k != i) {
2148: printf("%d%d ",i,k);
2149: fprintf(ficlog,"%d%d ",i,k);
2150: for(j=1; j <=ncovmodel; j++){
2151: printf("%12.7f ",p[jk]);
2152: fprintf(ficlog,"%12.7f ",p[jk]);
2153: jk++;
2154: }
2155: printf("\n");
2156: fprintf(ficlog,"\n");
2157: }
2158: }
2159: }
1.241 brouard 2160: if(*iter <=3 && *iter >1){
1.157 brouard 2161: tml = *localtime(&rcurr_time);
2162: strcpy(strcurr,asctime(&tml));
2163: rforecast_time=rcurr_time;
1.126 brouard 2164: itmp = strlen(strcurr);
2165: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2166: strcurr[itmp-1]='\0';
1.162 brouard 2167: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2168: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2169: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2170: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2171: forecast_time = *localtime(&rforecast_time);
2172: strcpy(strfor,asctime(&forecast_time));
2173: itmp = strlen(strfor);
2174: if(strfor[itmp-1]=='\n')
2175: strfor[itmp-1]='\0';
2176: 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);
2177: 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 2178: }
2179: }
1.187 brouard 2180: for (i=1;i<=n;i++) { /* For each direction i */
2181: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2182: fptt=(*fret);
2183: #ifdef DEBUG
1.203 brouard 2184: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2185: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2186: #endif
1.203 brouard 2187: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2188: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2189: #ifdef LINMINORIGINAL
1.188 brouard 2190: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2191: #else
2192: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2193: flatdir[i]=flat; /* Function is vanishing in that direction i */
2194: #endif
2195: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2196: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2197: /* because that direction will be replaced unless the gain del is small */
2198: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2199: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2200: /* with the new direction. */
2201: del=fabs(fptt-(*fret));
2202: ibig=i;
1.126 brouard 2203: }
2204: #ifdef DEBUG
2205: printf("%d %.12e",i,(*fret));
2206: fprintf(ficlog,"%d %.12e",i,(*fret));
2207: for (j=1;j<=n;j++) {
1.224 brouard 2208: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2209: printf(" x(%d)=%.12e",j,xit[j]);
2210: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2211: }
2212: for(j=1;j<=n;j++) {
1.225 brouard 2213: printf(" p(%d)=%.12e",j,p[j]);
2214: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2215: }
2216: printf("\n");
2217: fprintf(ficlog,"\n");
2218: #endif
1.187 brouard 2219: } /* end loop on each direction i */
2220: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2221: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2222: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2223: for(j=1;j<=n;j++) {
1.225 brouard 2224: if(flatdir[j] >0){
2225: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2226: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2227: }
2228: /* printf("\n"); */
2229: /* fprintf(ficlog,"\n"); */
2230: }
1.243 brouard 2231: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2232: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2233: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2234: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2235: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2236: /* decreased of more than 3.84 */
2237: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2238: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2239: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2240:
1.188 brouard 2241: /* Starting the program with initial values given by a former maximization will simply change */
2242: /* the scales of the directions and the directions, because the are reset to canonical directions */
2243: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2244: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2245: #ifdef DEBUG
2246: int k[2],l;
2247: k[0]=1;
2248: k[1]=-1;
2249: printf("Max: %.12e",(*func)(p));
2250: fprintf(ficlog,"Max: %.12e",(*func)(p));
2251: for (j=1;j<=n;j++) {
2252: printf(" %.12e",p[j]);
2253: fprintf(ficlog," %.12e",p[j]);
2254: }
2255: printf("\n");
2256: fprintf(ficlog,"\n");
2257: for(l=0;l<=1;l++) {
2258: for (j=1;j<=n;j++) {
2259: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2260: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2261: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2262: }
2263: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2264: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2265: }
2266: #endif
2267:
1.224 brouard 2268: #ifdef LINMINORIGINAL
2269: #else
2270: free_ivector(flatdir,1,n);
2271: #endif
1.126 brouard 2272: free_vector(xit,1,n);
2273: free_vector(xits,1,n);
2274: free_vector(ptt,1,n);
2275: free_vector(pt,1,n);
2276: return;
1.192 brouard 2277: } /* enough precision */
1.240 brouard 2278: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2279: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2280: ptt[j]=2.0*p[j]-pt[j];
2281: xit[j]=p[j]-pt[j];
2282: pt[j]=p[j];
2283: }
1.181 brouard 2284: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2285: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2286: if (*iter <=4) {
1.225 brouard 2287: #else
2288: #endif
1.224 brouard 2289: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2290: #else
1.161 brouard 2291: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2292: #endif
1.162 brouard 2293: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2294: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2295: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2296: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2297: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2298: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2299: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2300: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2301: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2302: /* Even if f3 <f1, directest can be negative and t >0 */
2303: /* mu² and del² are equal when f3=f1 */
2304: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2305: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2306: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2307: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2308: #ifdef NRCORIGINAL
2309: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2310: #else
2311: 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 2312: t= t- del*SQR(fp-fptt);
1.183 brouard 2313: #endif
1.202 brouard 2314: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2315: #ifdef DEBUG
1.181 brouard 2316: 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);
2317: 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 2318: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2319: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2320: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2321: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2322: 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);
2323: 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);
2324: #endif
1.183 brouard 2325: #ifdef POWELLORIGINAL
2326: if (t < 0.0) { /* Then we use it for new direction */
2327: #else
1.182 brouard 2328: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2329: 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 2330: 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 2331: 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 2332: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2333: }
1.181 brouard 2334: if (directest < 0.0) { /* Then we use it for new direction */
2335: #endif
1.191 brouard 2336: #ifdef DEBUGLINMIN
1.234 brouard 2337: printf("Before linmin in direction P%d-P0\n",n);
2338: for (j=1;j<=n;j++) {
2339: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2340: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2341: if(j % ncovmodel == 0){
2342: printf("\n");
2343: fprintf(ficlog,"\n");
2344: }
2345: }
1.224 brouard 2346: #endif
2347: #ifdef LINMINORIGINAL
1.234 brouard 2348: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2349: #else
1.234 brouard 2350: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2351: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2352: #endif
1.234 brouard 2353:
1.191 brouard 2354: #ifdef DEBUGLINMIN
1.234 brouard 2355: for (j=1;j<=n;j++) {
2356: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2357: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2358: if(j % ncovmodel == 0){
2359: printf("\n");
2360: fprintf(ficlog,"\n");
2361: }
2362: }
1.224 brouard 2363: #endif
1.234 brouard 2364: for (j=1;j<=n;j++) {
2365: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2366: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2367: }
1.224 brouard 2368: #ifdef LINMINORIGINAL
2369: #else
1.234 brouard 2370: for (j=1, flatd=0;j<=n;j++) {
2371: if(flatdir[j]>0)
2372: flatd++;
2373: }
2374: if(flatd >0){
2375: printf("%d flat directions\n",flatd);
2376: fprintf(ficlog,"%d flat directions\n",flatd);
2377: for (j=1;j<=n;j++) {
2378: if(flatdir[j]>0){
2379: printf("%d ",j);
2380: fprintf(ficlog,"%d ",j);
2381: }
2382: }
2383: printf("\n");
2384: fprintf(ficlog,"\n");
2385: }
1.191 brouard 2386: #endif
1.234 brouard 2387: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2388: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2389:
1.126 brouard 2390: #ifdef DEBUG
1.234 brouard 2391: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2392: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2393: for(j=1;j<=n;j++){
2394: printf(" %lf",xit[j]);
2395: fprintf(ficlog," %lf",xit[j]);
2396: }
2397: printf("\n");
2398: fprintf(ficlog,"\n");
1.126 brouard 2399: #endif
1.192 brouard 2400: } /* end of t or directest negative */
1.224 brouard 2401: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2402: #else
1.234 brouard 2403: } /* end if (fptt < fp) */
1.192 brouard 2404: #endif
1.225 brouard 2405: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2406: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2407: #else
1.224 brouard 2408: #endif
1.234 brouard 2409: } /* loop iteration */
1.126 brouard 2410: }
1.234 brouard 2411:
1.126 brouard 2412: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2413:
1.235 brouard 2414: 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 2415: {
1.235 brouard 2416: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2417: (and selected quantitative values in nres)
2418: by left multiplying the unit
1.234 brouard 2419: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2420: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2421: /* Wx is row vector: population in state 1, population in state 2, population dead */
2422: /* or prevalence in state 1, prevalence in state 2, 0 */
2423: /* newm is the matrix after multiplications, its rows are identical at a factor */
2424: /* Initial matrix pimij */
2425: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2426: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2427: /* 0, 0 , 1} */
2428: /*
2429: * and after some iteration: */
2430: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2431: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2432: /* 0, 0 , 1} */
2433: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2434: /* {0.51571254859325999, 0.4842874514067399, */
2435: /* 0.51326036147820708, 0.48673963852179264} */
2436: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2437:
1.126 brouard 2438: int i, ii,j,k;
1.209 brouard 2439: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2440: /* double **matprod2(); */ /* test */
1.218 brouard 2441: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2442: double **newm;
1.209 brouard 2443: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2444: int ncvloop=0;
1.169 brouard 2445:
1.209 brouard 2446: min=vector(1,nlstate);
2447: max=vector(1,nlstate);
2448: meandiff=vector(1,nlstate);
2449:
1.218 brouard 2450: /* Starting with matrix unity */
1.126 brouard 2451: for (ii=1;ii<=nlstate+ndeath;ii++)
2452: for (j=1;j<=nlstate+ndeath;j++){
2453: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2454: }
1.169 brouard 2455:
2456: cov[1]=1.;
2457:
2458: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2459: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2460: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2461: ncvloop++;
1.126 brouard 2462: newm=savm;
2463: /* Covariates have to be included here again */
1.138 brouard 2464: cov[2]=agefin;
1.187 brouard 2465: if(nagesqr==1)
2466: cov[3]= agefin*agefin;;
1.234 brouard 2467: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2468: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2469: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2470: /* 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 2471: }
2472: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2473: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2474: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2475: /* 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 2476: }
1.237 brouard 2477: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2478: if(Dummy[Tvar[Tage[k]]]){
2479: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2480: } else{
1.235 brouard 2481: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2482: }
1.235 brouard 2483: /* 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 2484: }
1.237 brouard 2485: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2486: /* 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 2487: if(Dummy[Tvard[k][1]==0]){
2488: if(Dummy[Tvard[k][2]==0]){
2489: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2490: }else{
2491: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2492: }
2493: }else{
2494: if(Dummy[Tvard[k][2]==0]){
2495: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2496: }else{
2497: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2498: }
2499: }
1.234 brouard 2500: }
1.138 brouard 2501: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2502: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2503: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2504: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2505: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2506: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2507: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2508:
1.126 brouard 2509: savm=oldm;
2510: oldm=newm;
1.209 brouard 2511:
2512: for(j=1; j<=nlstate; j++){
2513: max[j]=0.;
2514: min[j]=1.;
2515: }
2516: for(i=1;i<=nlstate;i++){
2517: sumnew=0;
2518: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2519: for(j=1; j<=nlstate; j++){
2520: prlim[i][j]= newm[i][j]/(1-sumnew);
2521: max[j]=FMAX(max[j],prlim[i][j]);
2522: min[j]=FMIN(min[j],prlim[i][j]);
2523: }
2524: }
2525:
1.126 brouard 2526: maxmax=0.;
1.209 brouard 2527: for(j=1; j<=nlstate; j++){
2528: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2529: maxmax=FMAX(maxmax,meandiff[j]);
2530: /* 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 2531: } /* j loop */
1.203 brouard 2532: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2533: /* 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 2534: if(maxmax < ftolpl){
1.209 brouard 2535: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2536: free_vector(min,1,nlstate);
2537: free_vector(max,1,nlstate);
2538: free_vector(meandiff,1,nlstate);
1.126 brouard 2539: return prlim;
2540: }
1.169 brouard 2541: } /* age loop */
1.208 brouard 2542: /* After some age loop it doesn't converge */
1.209 brouard 2543: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.208 brouard 2544: Earliest 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);
1.209 brouard 2545: /* 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); */
2546: free_vector(min,1,nlstate);
2547: free_vector(max,1,nlstate);
2548: free_vector(meandiff,1,nlstate);
1.208 brouard 2549:
1.169 brouard 2550: return prlim; /* should not reach here */
1.126 brouard 2551: }
2552:
1.217 brouard 2553:
2554: /**** Back Prevalence limit (stable or period prevalence) ****************/
2555:
1.218 brouard 2556: /* 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) */
2557: /* 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 2558: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2559: {
1.218 brouard 2560: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2561: matrix by transitions matrix until convergence is reached with precision ftolpl */
2562: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2563: /* Wx is row vector: population in state 1, population in state 2, population dead */
2564: /* or prevalence in state 1, prevalence in state 2, 0 */
2565: /* newm is the matrix after multiplications, its rows are identical at a factor */
2566: /* Initial matrix pimij */
2567: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2568: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2569: /* 0, 0 , 1} */
2570: /*
2571: * and after some iteration: */
2572: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2573: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2574: /* 0, 0 , 1} */
2575: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2576: /* {0.51571254859325999, 0.4842874514067399, */
2577: /* 0.51326036147820708, 0.48673963852179264} */
2578: /* If we start from prlim again, prlim tends to a constant matrix */
2579:
2580: int i, ii,j,k;
2581: double *min, *max, *meandiff, maxmax,sumnew=0.;
2582: /* double **matprod2(); */ /* test */
2583: double **out, cov[NCOVMAX+1], **bmij();
2584: double **newm;
1.218 brouard 2585: double **dnewm, **doldm, **dsavm; /* for use */
2586: double **oldm, **savm; /* for use */
2587:
1.217 brouard 2588: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2589: int ncvloop=0;
2590:
2591: min=vector(1,nlstate);
2592: max=vector(1,nlstate);
2593: meandiff=vector(1,nlstate);
2594:
1.218 brouard 2595: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2596: oldm=oldms; savm=savms;
2597:
2598: /* Starting with matrix unity */
2599: for (ii=1;ii<=nlstate+ndeath;ii++)
2600: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2601: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2602: }
2603:
2604: cov[1]=1.;
2605:
2606: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2607: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2608: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2609: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2610: ncvloop++;
1.218 brouard 2611: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2612: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2613: /* Covariates have to be included here again */
2614: cov[2]=agefin;
2615: if(nagesqr==1)
2616: cov[3]= agefin*agefin;;
1.242 brouard 2617: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2618: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2619: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2620: /* printf("bprevalim 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)); */
2621: }
2622: /* for (k=1; k<=cptcovn;k++) { */
2623: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2624: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2625: /* /\* 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])]); *\/ */
2626: /* } */
2627: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2628: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2629: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2630: /* 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]); */
2631: }
2632: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2633: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2634: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2635: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2636: for (k=1; k<=cptcovage;k++){ /* For product with age */
2637: if(Dummy[Tvar[Tage[k]]]){
2638: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2639: } else{
2640: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2641: }
2642: /* 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]); */
2643: }
2644: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2645: /* 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]); */
2646: if(Dummy[Tvard[k][1]==0]){
2647: if(Dummy[Tvard[k][2]==0]){
2648: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2649: }else{
2650: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2651: }
2652: }else{
2653: if(Dummy[Tvard[k][2]==0]){
2654: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2655: }else{
2656: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2657: }
2658: }
1.217 brouard 2659: }
2660:
2661: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2662: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2663: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2664: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2665: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2666: /* ij should be linked to the correct index of cov */
2667: /* age and covariate values ij are in 'cov', but we need to pass
2668: * ij for the observed prevalence at age and status and covariate
2669: * number: prevacurrent[(int)agefin][ii][ij]
2670: */
2671: /* 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 *\/ */
2672: /* 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 *\/ */
2673: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.217 brouard 2674: savm=oldm;
2675: oldm=newm;
2676: for(j=1; j<=nlstate; j++){
2677: max[j]=0.;
2678: min[j]=1.;
2679: }
2680: for(j=1; j<=nlstate; j++){
2681: for(i=1;i<=nlstate;i++){
1.234 brouard 2682: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2683: bprlim[i][j]= newm[i][j];
2684: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2685: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2686: }
2687: }
1.218 brouard 2688:
1.217 brouard 2689: maxmax=0.;
2690: for(i=1; i<=nlstate; i++){
2691: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2692: maxmax=FMAX(maxmax,meandiff[i]);
2693: /* 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); */
2694: } /* j loop */
2695: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2696: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2697: if(maxmax < ftolpl){
1.220 brouard 2698: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2699: free_vector(min,1,nlstate);
2700: free_vector(max,1,nlstate);
2701: free_vector(meandiff,1,nlstate);
2702: return bprlim;
2703: }
2704: } /* age loop */
2705: /* After some age loop it doesn't converge */
2706: 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'. \n\
2707: 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);
2708: /* 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); */
2709: free_vector(min,1,nlstate);
2710: free_vector(max,1,nlstate);
2711: free_vector(meandiff,1,nlstate);
2712:
2713: return bprlim; /* should not reach here */
2714: }
2715:
1.126 brouard 2716: /*************** transition probabilities ***************/
2717:
2718: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2719: {
1.138 brouard 2720: /* According to parameters values stored in x and the covariate's values stored in cov,
2721: computes the probability to be observed in state j being in state i by appying the
2722: model to the ncovmodel covariates (including constant and age).
2723: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2724: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2725: ncth covariate in the global vector x is given by the formula:
2726: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2727: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2728: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2729: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2730: Outputs ps[i][j] the probability to be observed in j being in j according to
2731: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2732: */
2733: double s1, lnpijopii;
1.126 brouard 2734: /*double t34;*/
1.164 brouard 2735: int i,j, nc, ii, jj;
1.126 brouard 2736:
1.223 brouard 2737: for(i=1; i<= nlstate; i++){
2738: for(j=1; j<i;j++){
2739: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2740: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2741: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2742: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2743: }
2744: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2745: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2746: }
2747: for(j=i+1; j<=nlstate+ndeath;j++){
2748: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2749: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2750: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2751: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2752: }
2753: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2754: }
2755: }
1.218 brouard 2756:
1.223 brouard 2757: for(i=1; i<= nlstate; i++){
2758: s1=0;
2759: for(j=1; j<i; j++){
2760: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2761: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2762: }
2763: for(j=i+1; j<=nlstate+ndeath; j++){
2764: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2765: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2766: }
2767: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2768: ps[i][i]=1./(s1+1.);
2769: /* Computing other pijs */
2770: for(j=1; j<i; j++)
2771: ps[i][j]= exp(ps[i][j])*ps[i][i];
2772: for(j=i+1; j<=nlstate+ndeath; j++)
2773: ps[i][j]= exp(ps[i][j])*ps[i][i];
2774: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2775: } /* end i */
1.218 brouard 2776:
1.223 brouard 2777: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2778: for(jj=1; jj<= nlstate+ndeath; jj++){
2779: ps[ii][jj]=0;
2780: ps[ii][ii]=1;
2781: }
2782: }
1.218 brouard 2783:
2784:
1.223 brouard 2785: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2786: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2787: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2788: /* } */
2789: /* printf("\n "); */
2790: /* } */
2791: /* printf("\n ");printf("%lf ",cov[2]);*/
2792: /*
2793: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2794: goto end;*/
1.223 brouard 2795: return ps;
1.126 brouard 2796: }
2797:
1.218 brouard 2798: /*************** backward transition probabilities ***************/
2799:
2800: /* 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 ) */
2801: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2802: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2803: {
1.222 brouard 2804: /* Computes the backward probability at age agefin and covariate ij
2805: * and returns in **ps as well as **bmij.
2806: */
1.218 brouard 2807: int i, ii, j,k;
1.222 brouard 2808:
2809: double **out, **pmij();
2810: double sumnew=0.;
1.218 brouard 2811: double agefin;
1.222 brouard 2812:
2813: double **dnewm, **dsavm, **doldm;
2814: double **bbmij;
2815:
1.218 brouard 2816: doldm=ddoldms; /* global pointers */
1.222 brouard 2817: dnewm=ddnewms;
2818: dsavm=ddsavms;
2819:
2820: agefin=cov[2];
2821: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2822: the observed prevalence (with this covariate ij) */
2823: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2824: /* We do have the matrix Px in savm and we need pij */
2825: for (j=1;j<=nlstate+ndeath;j++){
2826: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2827: for (ii=1;ii<=nlstate;ii++){
2828: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2829: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2830: for (ii=1;ii<=nlstate+ndeath;ii++){
2831: if(sumnew >= 1.e-10){
2832: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2833: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2834: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2835: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2836: /* }else */
2837: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2838: }else{
1.242 brouard 2839: ;
2840: /* printf("ii=%d, i=%d, doldm=%lf dsavm=%lf, probs=%lf, sumnew=%lf,agefin=%d\n",ii,j,doldm[ii][j],dsavm[ii][j],prevacurrent[(int)agefin][ii][ij],sumnew, (int)agefin); */
1.222 brouard 2841: }
2842: } /*End ii */
2843: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2844: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2845: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2846: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2847: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2848: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2849: /* left Product of this matrix by diag matrix of prevalences (savm) */
2850: for (j=1;j<=nlstate+ndeath;j++){
2851: for (ii=1;ii<=nlstate+ndeath;ii++){
2852: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2853: }
2854: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2855: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2856: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2857: /* end bmij */
2858: return ps;
1.218 brouard 2859: }
1.217 brouard 2860: /*************** transition probabilities ***************/
2861:
1.218 brouard 2862: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2863: {
2864: /* According to parameters values stored in x and the covariate's values stored in cov,
2865: computes the probability to be observed in state j being in state i by appying the
2866: model to the ncovmodel covariates (including constant and age).
2867: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2868: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2869: ncth covariate in the global vector x is given by the formula:
2870: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2871: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2872: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2873: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2874: Outputs ps[i][j] the probability to be observed in j being in j according to
2875: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2876: */
2877: double s1, lnpijopii;
2878: /*double t34;*/
2879: int i,j, nc, ii, jj;
2880:
1.234 brouard 2881: for(i=1; i<= nlstate; i++){
2882: for(j=1; j<i;j++){
2883: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2884: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2885: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2886: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2887: }
2888: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2889: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2890: }
2891: for(j=i+1; j<=nlstate+ndeath;j++){
2892: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2893: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2894: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2895: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2896: }
2897: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2898: }
2899: }
2900:
2901: for(i=1; i<= nlstate; i++){
2902: s1=0;
2903: for(j=1; j<i; j++){
2904: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2905: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2906: }
2907: for(j=i+1; j<=nlstate+ndeath; j++){
2908: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2909: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2910: }
2911: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2912: ps[i][i]=1./(s1+1.);
2913: /* Computing other pijs */
2914: for(j=1; j<i; j++)
2915: ps[i][j]= exp(ps[i][j])*ps[i][i];
2916: for(j=i+1; j<=nlstate+ndeath; j++)
2917: ps[i][j]= exp(ps[i][j])*ps[i][i];
2918: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2919: } /* end i */
2920:
2921: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2922: for(jj=1; jj<= nlstate+ndeath; jj++){
2923: ps[ii][jj]=0;
2924: ps[ii][ii]=1;
2925: }
2926: }
2927: /* Added for backcast */ /* Transposed matrix too */
2928: for(jj=1; jj<= nlstate+ndeath; jj++){
2929: s1=0.;
2930: for(ii=1; ii<= nlstate+ndeath; ii++){
2931: s1+=ps[ii][jj];
2932: }
2933: for(ii=1; ii<= nlstate; ii++){
2934: ps[ii][jj]=ps[ii][jj]/s1;
2935: }
2936: }
2937: /* Transposition */
2938: for(jj=1; jj<= nlstate+ndeath; jj++){
2939: for(ii=jj; ii<= nlstate+ndeath; ii++){
2940: s1=ps[ii][jj];
2941: ps[ii][jj]=ps[jj][ii];
2942: ps[jj][ii]=s1;
2943: }
2944: }
2945: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2946: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2947: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2948: /* } */
2949: /* printf("\n "); */
2950: /* } */
2951: /* printf("\n ");printf("%lf ",cov[2]);*/
2952: /*
2953: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2954: goto end;*/
2955: return ps;
1.217 brouard 2956: }
2957:
2958:
1.126 brouard 2959: /**************** Product of 2 matrices ******************/
2960:
1.145 brouard 2961: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2962: {
2963: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2964: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2965: /* in, b, out are matrice of pointers which should have been initialized
2966: before: only the contents of out is modified. The function returns
2967: a pointer to pointers identical to out */
1.145 brouard 2968: int i, j, k;
1.126 brouard 2969: for(i=nrl; i<= nrh; i++)
1.145 brouard 2970: for(k=ncolol; k<=ncoloh; k++){
2971: out[i][k]=0.;
2972: for(j=ncl; j<=nch; j++)
2973: out[i][k] +=in[i][j]*b[j][k];
2974: }
1.126 brouard 2975: return out;
2976: }
2977:
2978:
2979: /************* Higher Matrix Product ***************/
2980:
1.235 brouard 2981: 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 2982: {
1.218 brouard 2983: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 2984: 'nhstepm*hstepm*stepm' months (i.e. until
2985: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
2986: nhstepm*hstepm matrices.
2987: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
2988: (typically every 2 years instead of every month which is too big
2989: for the memory).
2990: Model is determined by parameters x and covariates have to be
2991: included manually here.
2992:
2993: */
2994:
2995: int i, j, d, h, k;
1.131 brouard 2996: double **out, cov[NCOVMAX+1];
1.126 brouard 2997: double **newm;
1.187 brouard 2998: double agexact;
1.214 brouard 2999: double agebegin, ageend;
1.126 brouard 3000:
3001: /* Hstepm could be zero and should return the unit matrix */
3002: for (i=1;i<=nlstate+ndeath;i++)
3003: for (j=1;j<=nlstate+ndeath;j++){
3004: oldm[i][j]=(i==j ? 1.0 : 0.0);
3005: po[i][j][0]=(i==j ? 1.0 : 0.0);
3006: }
3007: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3008: for(h=1; h <=nhstepm; h++){
3009: for(d=1; d <=hstepm; d++){
3010: newm=savm;
3011: /* Covariates have to be included here again */
3012: cov[1]=1.;
1.214 brouard 3013: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3014: cov[2]=agexact;
3015: if(nagesqr==1)
1.227 brouard 3016: cov[3]= agexact*agexact;
1.235 brouard 3017: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3018: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3019: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3020: /* 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)); */
3021: }
3022: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3023: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3024: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3025: /* 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]); */
3026: }
3027: for (k=1; k<=cptcovage;k++){
3028: if(Dummy[Tvar[Tage[k]]]){
3029: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3030: } else{
3031: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3032: }
3033: /* 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]); */
3034: }
3035: for (k=1; k<=cptcovprod;k++){ /* */
3036: /* 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]); */
3037: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3038: }
3039: /* for (k=1; k<=cptcovn;k++) */
3040: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3041: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3042: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3043: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3044: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3045:
3046:
1.126 brouard 3047: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3048: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3049: /* right multiplication of oldm by the current matrix */
1.126 brouard 3050: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3051: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3052: /* if((int)age == 70){ */
3053: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3054: /* for(i=1; i<=nlstate+ndeath; i++) { */
3055: /* printf("%d pmmij ",i); */
3056: /* for(j=1;j<=nlstate+ndeath;j++) { */
3057: /* printf("%f ",pmmij[i][j]); */
3058: /* } */
3059: /* printf(" oldm "); */
3060: /* for(j=1;j<=nlstate+ndeath;j++) { */
3061: /* printf("%f ",oldm[i][j]); */
3062: /* } */
3063: /* printf("\n"); */
3064: /* } */
3065: /* } */
1.126 brouard 3066: savm=oldm;
3067: oldm=newm;
3068: }
3069: for(i=1; i<=nlstate+ndeath; i++)
3070: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3071: po[i][j][h]=newm[i][j];
3072: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3073: }
1.128 brouard 3074: /*printf("h=%d ",h);*/
1.126 brouard 3075: } /* end h */
1.218 brouard 3076: /* printf("\n H=%d \n",h); */
1.126 brouard 3077: return po;
3078: }
3079:
1.217 brouard 3080: /************* Higher Back Matrix Product ***************/
1.218 brouard 3081: /* 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 ) */
1.222 brouard 3082: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3083: {
1.218 brouard 3084: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3085: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3086: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3087: nhstepm*hstepm matrices.
3088: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3089: (typically every 2 years instead of every month which is too big
1.217 brouard 3090: for the memory).
1.218 brouard 3091: Model is determined by parameters x and covariates have to be
3092: included manually here.
1.217 brouard 3093:
1.222 brouard 3094: */
1.217 brouard 3095:
3096: int i, j, d, h, k;
3097: double **out, cov[NCOVMAX+1];
3098: double **newm;
3099: double agexact;
3100: double agebegin, ageend;
1.222 brouard 3101: double **oldm, **savm;
1.217 brouard 3102:
1.222 brouard 3103: oldm=oldms;savm=savms;
1.217 brouard 3104: /* Hstepm could be zero and should return the unit matrix */
3105: for (i=1;i<=nlstate+ndeath;i++)
3106: for (j=1;j<=nlstate+ndeath;j++){
3107: oldm[i][j]=(i==j ? 1.0 : 0.0);
3108: po[i][j][0]=(i==j ? 1.0 : 0.0);
3109: }
3110: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3111: for(h=1; h <=nhstepm; h++){
3112: for(d=1; d <=hstepm; d++){
3113: newm=savm;
3114: /* Covariates have to be included here again */
3115: cov[1]=1.;
3116: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3117: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3118: cov[2]=agexact;
3119: if(nagesqr==1)
1.222 brouard 3120: cov[3]= agexact*agexact;
1.218 brouard 3121: for (k=1; k<=cptcovn;k++)
1.222 brouard 3122: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3123: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3124: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3125: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3126: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3127: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3128: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3129: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3130: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
1.218 brouard 3131:
3132:
1.217 brouard 3133: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3134: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3135: /* Careful transposed matrix */
1.222 brouard 3136: /* age is in cov[2] */
1.218 brouard 3137: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3138: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3139: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3140: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3141: /* if((int)age == 70){ */
3142: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3143: /* for(i=1; i<=nlstate+ndeath; i++) { */
3144: /* printf("%d pmmij ",i); */
3145: /* for(j=1;j<=nlstate+ndeath;j++) { */
3146: /* printf("%f ",pmmij[i][j]); */
3147: /* } */
3148: /* printf(" oldm "); */
3149: /* for(j=1;j<=nlstate+ndeath;j++) { */
3150: /* printf("%f ",oldm[i][j]); */
3151: /* } */
3152: /* printf("\n"); */
3153: /* } */
3154: /* } */
3155: savm=oldm;
3156: oldm=newm;
3157: }
3158: for(i=1; i<=nlstate+ndeath; i++)
3159: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3160: po[i][j][h]=newm[i][j];
3161: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3162: }
3163: /*printf("h=%d ",h);*/
3164: } /* end h */
1.222 brouard 3165: /* printf("\n H=%d \n",h); */
1.217 brouard 3166: return po;
3167: }
3168:
3169:
1.162 brouard 3170: #ifdef NLOPT
3171: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3172: double fret;
3173: double *xt;
3174: int j;
3175: myfunc_data *d2 = (myfunc_data *) pd;
3176: /* xt = (p1-1); */
3177: xt=vector(1,n);
3178: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3179:
3180: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3181: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3182: printf("Function = %.12lf ",fret);
3183: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3184: printf("\n");
3185: free_vector(xt,1,n);
3186: return fret;
3187: }
3188: #endif
1.126 brouard 3189:
3190: /*************** log-likelihood *************/
3191: double func( double *x)
3192: {
1.226 brouard 3193: int i, ii, j, k, mi, d, kk;
3194: int ioffset=0;
3195: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3196: double **out;
3197: double lli; /* Individual log likelihood */
3198: int s1, s2;
1.228 brouard 3199: 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 3200: double bbh, survp;
3201: long ipmx;
3202: double agexact;
3203: /*extern weight */
3204: /* We are differentiating ll according to initial status */
3205: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3206: /*for(i=1;i<imx;i++)
3207: printf(" %d\n",s[4][i]);
3208: */
1.162 brouard 3209:
1.226 brouard 3210: ++countcallfunc;
1.162 brouard 3211:
1.226 brouard 3212: cov[1]=1.;
1.126 brouard 3213:
1.226 brouard 3214: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3215: ioffset=0;
1.226 brouard 3216: if(mle==1){
3217: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3218: /* Computes the values of the ncovmodel covariates of the model
3219: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3220: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3221: to be observed in j being in i according to the model.
3222: */
1.243 brouard 3223: ioffset=2+nagesqr ;
1.233 brouard 3224: /* Fixed */
1.234 brouard 3225: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3226: 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)*/
3227: }
1.226 brouard 3228: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3229: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3230: has been calculated etc */
3231: /* For an individual i, wav[i] gives the number of effective waves */
3232: /* We compute the contribution to Likelihood of each effective transition
3233: mw[mi][i] is real wave of the mi th effectve wave */
3234: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3235: s2=s[mw[mi+1][i]][i];
3236: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3237: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3238: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3239: */
3240: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3241: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3242: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3243: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3244: }
3245: for (ii=1;ii<=nlstate+ndeath;ii++)
3246: for (j=1;j<=nlstate+ndeath;j++){
3247: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3248: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3249: }
3250: for(d=0; d<dh[mi][i]; d++){
3251: newm=savm;
3252: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3253: cov[2]=agexact;
3254: if(nagesqr==1)
3255: cov[3]= agexact*agexact; /* Should be changed here */
3256: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3257: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3258: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3259: else
3260: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3261: }
3262: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3263: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3264: savm=oldm;
3265: oldm=newm;
3266: } /* end mult */
3267:
3268: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3269: /* But now since version 0.9 we anticipate for bias at large stepm.
3270: * If stepm is larger than one month (smallest stepm) and if the exact delay
3271: * (in months) between two waves is not a multiple of stepm, we rounded to
3272: * the nearest (and in case of equal distance, to the lowest) interval but now
3273: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3274: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3275: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3276: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3277: * -stepm/2 to stepm/2 .
3278: * For stepm=1 the results are the same as for previous versions of Imach.
3279: * For stepm > 1 the results are less biased than in previous versions.
3280: */
1.234 brouard 3281: s1=s[mw[mi][i]][i];
3282: s2=s[mw[mi+1][i]][i];
3283: bbh=(double)bh[mi][i]/(double)stepm;
3284: /* bias bh is positive if real duration
3285: * is higher than the multiple of stepm and negative otherwise.
3286: */
3287: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3288: if( s2 > nlstate){
3289: /* i.e. if s2 is a death state and if the date of death is known
3290: then the contribution to the likelihood is the probability to
3291: die between last step unit time and current step unit time,
3292: which is also equal to probability to die before dh
3293: minus probability to die before dh-stepm .
3294: In version up to 0.92 likelihood was computed
3295: as if date of death was unknown. Death was treated as any other
3296: health state: the date of the interview describes the actual state
3297: and not the date of a change in health state. The former idea was
3298: to consider that at each interview the state was recorded
3299: (healthy, disable or death) and IMaCh was corrected; but when we
3300: introduced the exact date of death then we should have modified
3301: the contribution of an exact death to the likelihood. This new
3302: contribution is smaller and very dependent of the step unit
3303: stepm. It is no more the probability to die between last interview
3304: and month of death but the probability to survive from last
3305: interview up to one month before death multiplied by the
3306: probability to die within a month. Thanks to Chris
3307: Jackson for correcting this bug. Former versions increased
3308: mortality artificially. The bad side is that we add another loop
3309: which slows down the processing. The difference can be up to 10%
3310: lower mortality.
3311: */
3312: /* If, at the beginning of the maximization mostly, the
3313: cumulative probability or probability to be dead is
3314: constant (ie = 1) over time d, the difference is equal to
3315: 0. out[s1][3] = savm[s1][3]: probability, being at state
3316: s1 at precedent wave, to be dead a month before current
3317: wave is equal to probability, being at state s1 at
3318: precedent wave, to be dead at mont of the current
3319: wave. Then the observed probability (that this person died)
3320: is null according to current estimated parameter. In fact,
3321: it should be very low but not zero otherwise the log go to
3322: infinity.
3323: */
1.183 brouard 3324: /* #ifdef INFINITYORIGINAL */
3325: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3326: /* #else */
3327: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3328: /* lli=log(mytinydouble); */
3329: /* else */
3330: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3331: /* #endif */
1.226 brouard 3332: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3333:
1.226 brouard 3334: } else if ( s2==-1 ) { /* alive */
3335: for (j=1,survp=0. ; j<=nlstate; j++)
3336: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3337: /*survp += out[s1][j]; */
3338: lli= log(survp);
3339: }
3340: else if (s2==-4) {
3341: for (j=3,survp=0. ; j<=nlstate; j++)
3342: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3343: lli= log(survp);
3344: }
3345: else if (s2==-5) {
3346: for (j=1,survp=0. ; j<=2; j++)
3347: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3348: lli= log(survp);
3349: }
3350: else{
3351: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3352: /* 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 */
3353: }
3354: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3355: /*if(lli ==000.0)*/
3356: /*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); */
3357: ipmx +=1;
3358: sw += weight[i];
3359: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3360: /* if (lli < log(mytinydouble)){ */
3361: /* 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); */
3362: /* 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]); */
3363: /* } */
3364: } /* end of wave */
3365: } /* end of individual */
3366: } else if(mle==2){
3367: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3368: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3369: for(mi=1; mi<= wav[i]-1; mi++){
3370: for (ii=1;ii<=nlstate+ndeath;ii++)
3371: for (j=1;j<=nlstate+ndeath;j++){
3372: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3373: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3374: }
3375: for(d=0; d<=dh[mi][i]; d++){
3376: newm=savm;
3377: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3378: cov[2]=agexact;
3379: if(nagesqr==1)
3380: cov[3]= agexact*agexact;
3381: for (kk=1; kk<=cptcovage;kk++) {
3382: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3383: }
3384: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3385: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3386: savm=oldm;
3387: oldm=newm;
3388: } /* end mult */
3389:
3390: s1=s[mw[mi][i]][i];
3391: s2=s[mw[mi+1][i]][i];
3392: bbh=(double)bh[mi][i]/(double)stepm;
3393: 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 */
3394: ipmx +=1;
3395: sw += weight[i];
3396: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3397: } /* end of wave */
3398: } /* end of individual */
3399: } else if(mle==3){ /* exponential inter-extrapolation */
3400: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3401: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3402: for(mi=1; mi<= wav[i]-1; mi++){
3403: for (ii=1;ii<=nlstate+ndeath;ii++)
3404: for (j=1;j<=nlstate+ndeath;j++){
3405: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3406: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3407: }
3408: for(d=0; d<dh[mi][i]; d++){
3409: newm=savm;
3410: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3411: cov[2]=agexact;
3412: if(nagesqr==1)
3413: cov[3]= agexact*agexact;
3414: for (kk=1; kk<=cptcovage;kk++) {
3415: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3416: }
3417: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3418: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3419: savm=oldm;
3420: oldm=newm;
3421: } /* end mult */
3422:
3423: s1=s[mw[mi][i]][i];
3424: s2=s[mw[mi+1][i]][i];
3425: bbh=(double)bh[mi][i]/(double)stepm;
3426: 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 */
3427: ipmx +=1;
3428: sw += weight[i];
3429: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3430: } /* end of wave */
3431: } /* end of individual */
3432: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3433: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3434: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3435: for(mi=1; mi<= wav[i]-1; mi++){
3436: for (ii=1;ii<=nlstate+ndeath;ii++)
3437: for (j=1;j<=nlstate+ndeath;j++){
3438: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3439: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3440: }
3441: for(d=0; d<dh[mi][i]; d++){
3442: newm=savm;
3443: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3444: cov[2]=agexact;
3445: if(nagesqr==1)
3446: cov[3]= agexact*agexact;
3447: for (kk=1; kk<=cptcovage;kk++) {
3448: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3449: }
1.126 brouard 3450:
1.226 brouard 3451: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3452: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3453: savm=oldm;
3454: oldm=newm;
3455: } /* end mult */
3456:
3457: s1=s[mw[mi][i]][i];
3458: s2=s[mw[mi+1][i]][i];
3459: if( s2 > nlstate){
3460: lli=log(out[s1][s2] - savm[s1][s2]);
3461: } else if ( s2==-1 ) { /* alive */
3462: for (j=1,survp=0. ; j<=nlstate; j++)
3463: survp += out[s1][j];
3464: lli= log(survp);
3465: }else{
3466: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3467: }
3468: ipmx +=1;
3469: sw += weight[i];
3470: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3471: /* 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 3472: } /* end of wave */
3473: } /* end of individual */
3474: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3475: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3476: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3477: for(mi=1; mi<= wav[i]-1; mi++){
3478: for (ii=1;ii<=nlstate+ndeath;ii++)
3479: for (j=1;j<=nlstate+ndeath;j++){
3480: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3481: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3482: }
3483: for(d=0; d<dh[mi][i]; d++){
3484: newm=savm;
3485: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3486: cov[2]=agexact;
3487: if(nagesqr==1)
3488: cov[3]= agexact*agexact;
3489: for (kk=1; kk<=cptcovage;kk++) {
3490: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3491: }
1.126 brouard 3492:
1.226 brouard 3493: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3494: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3495: savm=oldm;
3496: oldm=newm;
3497: } /* end mult */
3498:
3499: s1=s[mw[mi][i]][i];
3500: s2=s[mw[mi+1][i]][i];
3501: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3502: ipmx +=1;
3503: sw += weight[i];
3504: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3505: /*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]);*/
3506: } /* end of wave */
3507: } /* end of individual */
3508: } /* End of if */
3509: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3510: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3511: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3512: return -l;
1.126 brouard 3513: }
3514:
3515: /*************** log-likelihood *************/
3516: double funcone( double *x)
3517: {
1.228 brouard 3518: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3519: int i, ii, j, k, mi, d, kk;
1.228 brouard 3520: int ioffset=0;
1.131 brouard 3521: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3522: double **out;
3523: double lli; /* Individual log likelihood */
3524: double llt;
3525: int s1, s2;
1.228 brouard 3526: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3527:
1.126 brouard 3528: double bbh, survp;
1.187 brouard 3529: double agexact;
1.214 brouard 3530: double agebegin, ageend;
1.126 brouard 3531: /*extern weight */
3532: /* We are differentiating ll according to initial status */
3533: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3534: /*for(i=1;i<imx;i++)
3535: printf(" %d\n",s[4][i]);
3536: */
3537: cov[1]=1.;
3538:
3539: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3540: ioffset=0;
3541: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3542: /* ioffset=2+nagesqr+cptcovage; */
3543: ioffset=2+nagesqr;
1.232 brouard 3544: /* Fixed */
1.224 brouard 3545: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3546: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3547: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3548: 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)*/
3549: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3550: /* cov[2+6]=covar[Tvar[6]][i]; */
3551: /* cov[2+6]=covar[2][i]; V2 */
3552: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3553: /* cov[2+7]=covar[Tvar[7]][i]; */
3554: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3555: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3556: /* cov[2+9]=covar[Tvar[9]][i]; */
3557: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3558: }
1.232 brouard 3559: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3560: /* 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?)*\/ */
3561: /* } */
1.231 brouard 3562: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3563: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3564: /* } */
1.225 brouard 3565:
1.233 brouard 3566:
3567: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3568: /* Wave varying (but not age varying) */
3569: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3570: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3571: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3572: }
1.232 brouard 3573: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3574: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3575: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3576: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3577: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3578: /* 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 3579: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3580: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3581: /* /\* 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]); *\/ */
3582: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3583: /* } */
1.126 brouard 3584: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3585: for (j=1;j<=nlstate+ndeath;j++){
3586: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3587: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3588: }
1.214 brouard 3589:
3590: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3591: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3592: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.242 brouard 3593: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3594: and mw[mi+1][i]. dh depends on stepm.*/
3595: newm=savm;
3596: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3597: cov[2]=agexact;
3598: if(nagesqr==1)
3599: cov[3]= agexact*agexact;
3600: for (kk=1; kk<=cptcovage;kk++) {
3601: if(!FixedV[Tvar[Tage[kk]]])
3602: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3603: else
3604: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3605: }
3606: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3607: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3608: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3609: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3610: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3611: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3612: savm=oldm;
3613: oldm=newm;
1.126 brouard 3614: } /* end mult */
3615:
3616: s1=s[mw[mi][i]][i];
3617: s2=s[mw[mi+1][i]][i];
1.217 brouard 3618: /* if(s2==-1){ */
3619: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3620: /* /\* exit(1); *\/ */
3621: /* } */
1.126 brouard 3622: bbh=(double)bh[mi][i]/(double)stepm;
3623: /* bias is positive if real duration
3624: * is higher than the multiple of stepm and negative otherwise.
3625: */
3626: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3627: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3628: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3629: for (j=1,survp=0. ; j<=nlstate; j++)
3630: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3631: lli= log(survp);
1.126 brouard 3632: }else if (mle==1){
1.242 brouard 3633: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3634: } else if(mle==2){
1.242 brouard 3635: 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 3636: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3637: 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 3638: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3639: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3640: } else{ /* mle=0 back to 1 */
1.242 brouard 3641: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3642: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3643: } /* End of if */
3644: ipmx +=1;
3645: sw += weight[i];
3646: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3647: /*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 3648: if(globpr){
1.242 brouard 3649: fprintf(ficresilk,"%9ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3650: %11.6f %11.6f %11.6f ", \
1.242 brouard 3651: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3652: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3653: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3654: llt +=ll[k]*gipmx/gsw;
3655: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3656: }
3657: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3658: }
1.232 brouard 3659: } /* end of wave */
3660: } /* end of individual */
3661: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3662: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3663: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3664: if(globpr==0){ /* First time we count the contributions and weights */
3665: gipmx=ipmx;
3666: gsw=sw;
3667: }
3668: return -l;
1.126 brouard 3669: }
3670:
3671:
3672: /*************** function likelione ***********/
3673: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3674: {
3675: /* This routine should help understanding what is done with
3676: the selection of individuals/waves and
3677: to check the exact contribution to the likelihood.
3678: Plotting could be done.
3679: */
3680: int k;
3681:
3682: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3683: strcpy(fileresilk,"ILK_");
1.202 brouard 3684: strcat(fileresilk,fileresu);
1.126 brouard 3685: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3686: printf("Problem with resultfile: %s\n", fileresilk);
3687: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3688: }
1.214 brouard 3689: 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");
3690: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3691: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3692: for(k=1; k<=nlstate; k++)
3693: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3694: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3695: }
3696:
3697: *fretone=(*funcone)(p);
3698: if(*globpri !=0){
3699: fclose(ficresilk);
1.205 brouard 3700: if (mle ==0)
3701: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3702: else if(mle >=1)
3703: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3704: 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));
1.207 brouard 3705:
1.208 brouard 3706:
3707: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3708: 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 3709: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3710: }
1.207 brouard 3711: 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 3712: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3713: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3714: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3715: fflush(fichtm);
1.205 brouard 3716: }
1.126 brouard 3717: return;
3718: }
3719:
3720:
3721: /*********** Maximum Likelihood Estimation ***************/
3722:
3723: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3724: {
1.165 brouard 3725: int i,j, iter=0;
1.126 brouard 3726: double **xi;
3727: double fret;
3728: double fretone; /* Only one call to likelihood */
3729: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3730:
3731: #ifdef NLOPT
3732: int creturn;
3733: nlopt_opt opt;
3734: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3735: double *lb;
3736: double minf; /* the minimum objective value, upon return */
3737: double * p1; /* Shifted parameters from 0 instead of 1 */
3738: myfunc_data dinst, *d = &dinst;
3739: #endif
3740:
3741:
1.126 brouard 3742: xi=matrix(1,npar,1,npar);
3743: for (i=1;i<=npar;i++)
3744: for (j=1;j<=npar;j++)
3745: xi[i][j]=(i==j ? 1.0 : 0.0);
3746: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3747: strcpy(filerespow,"POW_");
1.126 brouard 3748: strcat(filerespow,fileres);
3749: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3750: printf("Problem with resultfile: %s\n", filerespow);
3751: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3752: }
3753: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3754: for (i=1;i<=nlstate;i++)
3755: for(j=1;j<=nlstate+ndeath;j++)
3756: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3757: fprintf(ficrespow,"\n");
1.162 brouard 3758: #ifdef POWELL
1.126 brouard 3759: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3760: #endif
1.126 brouard 3761:
1.162 brouard 3762: #ifdef NLOPT
3763: #ifdef NEWUOA
3764: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3765: #else
3766: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3767: #endif
3768: lb=vector(0,npar-1);
3769: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3770: nlopt_set_lower_bounds(opt, lb);
3771: nlopt_set_initial_step1(opt, 0.1);
3772:
3773: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3774: d->function = func;
3775: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3776: nlopt_set_min_objective(opt, myfunc, d);
3777: nlopt_set_xtol_rel(opt, ftol);
3778: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3779: printf("nlopt failed! %d\n",creturn);
3780: }
3781: else {
3782: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3783: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3784: iter=1; /* not equal */
3785: }
3786: nlopt_destroy(opt);
3787: #endif
1.126 brouard 3788: free_matrix(xi,1,npar,1,npar);
3789: fclose(ficrespow);
1.203 brouard 3790: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3791: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3792: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3793:
3794: }
3795:
3796: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3797: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3798: {
3799: double **a,**y,*x,pd;
1.203 brouard 3800: /* double **hess; */
1.164 brouard 3801: int i, j;
1.126 brouard 3802: int *indx;
3803:
3804: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3805: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3806: void lubksb(double **a, int npar, int *indx, double b[]) ;
3807: void ludcmp(double **a, int npar, int *indx, double *d) ;
3808: double gompertz(double p[]);
1.203 brouard 3809: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3810:
3811: printf("\nCalculation of the hessian matrix. Wait...\n");
3812: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3813: for (i=1;i<=npar;i++){
1.203 brouard 3814: printf("%d-",i);fflush(stdout);
3815: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3816:
3817: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3818:
3819: /* printf(" %f ",p[i]);
3820: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3821: }
3822:
3823: for (i=1;i<=npar;i++) {
3824: for (j=1;j<=npar;j++) {
3825: if (j>i) {
1.203 brouard 3826: printf(".%d-%d",i,j);fflush(stdout);
3827: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3828: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3829:
3830: hess[j][i]=hess[i][j];
3831: /*printf(" %lf ",hess[i][j]);*/
3832: }
3833: }
3834: }
3835: printf("\n");
3836: fprintf(ficlog,"\n");
3837:
3838: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3839: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3840:
3841: a=matrix(1,npar,1,npar);
3842: y=matrix(1,npar,1,npar);
3843: x=vector(1,npar);
3844: indx=ivector(1,npar);
3845: for (i=1;i<=npar;i++)
3846: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3847: ludcmp(a,npar,indx,&pd);
3848:
3849: for (j=1;j<=npar;j++) {
3850: for (i=1;i<=npar;i++) x[i]=0;
3851: x[j]=1;
3852: lubksb(a,npar,indx,x);
3853: for (i=1;i<=npar;i++){
3854: matcov[i][j]=x[i];
3855: }
3856: }
3857:
3858: printf("\n#Hessian matrix#\n");
3859: fprintf(ficlog,"\n#Hessian matrix#\n");
3860: for (i=1;i<=npar;i++) {
3861: for (j=1;j<=npar;j++) {
1.203 brouard 3862: printf("%.6e ",hess[i][j]);
3863: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3864: }
3865: printf("\n");
3866: fprintf(ficlog,"\n");
3867: }
3868:
1.203 brouard 3869: /* printf("\n#Covariance matrix#\n"); */
3870: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3871: /* for (i=1;i<=npar;i++) { */
3872: /* for (j=1;j<=npar;j++) { */
3873: /* printf("%.6e ",matcov[i][j]); */
3874: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3875: /* } */
3876: /* printf("\n"); */
3877: /* fprintf(ficlog,"\n"); */
3878: /* } */
3879:
1.126 brouard 3880: /* Recompute Inverse */
1.203 brouard 3881: /* for (i=1;i<=npar;i++) */
3882: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3883: /* ludcmp(a,npar,indx,&pd); */
3884:
3885: /* printf("\n#Hessian matrix recomputed#\n"); */
3886:
3887: /* for (j=1;j<=npar;j++) { */
3888: /* for (i=1;i<=npar;i++) x[i]=0; */
3889: /* x[j]=1; */
3890: /* lubksb(a,npar,indx,x); */
3891: /* for (i=1;i<=npar;i++){ */
3892: /* y[i][j]=x[i]; */
3893: /* printf("%.3e ",y[i][j]); */
3894: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3895: /* } */
3896: /* printf("\n"); */
3897: /* fprintf(ficlog,"\n"); */
3898: /* } */
3899:
3900: /* Verifying the inverse matrix */
3901: #ifdef DEBUGHESS
3902: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3903:
1.203 brouard 3904: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3905: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3906:
3907: for (j=1;j<=npar;j++) {
3908: for (i=1;i<=npar;i++){
1.203 brouard 3909: printf("%.2f ",y[i][j]);
3910: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3911: }
3912: printf("\n");
3913: fprintf(ficlog,"\n");
3914: }
1.203 brouard 3915: #endif
1.126 brouard 3916:
3917: free_matrix(a,1,npar,1,npar);
3918: free_matrix(y,1,npar,1,npar);
3919: free_vector(x,1,npar);
3920: free_ivector(indx,1,npar);
1.203 brouard 3921: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3922:
3923:
3924: }
3925:
3926: /*************** hessian matrix ****************/
3927: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3928: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3929: int i;
3930: int l=1, lmax=20;
1.203 brouard 3931: double k1,k2, res, fx;
1.132 brouard 3932: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3933: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3934: int k=0,kmax=10;
3935: double l1;
3936:
3937: fx=func(x);
3938: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3939: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3940: l1=pow(10,l);
3941: delts=delt;
3942: for(k=1 ; k <kmax; k=k+1){
3943: delt = delta*(l1*k);
3944: p2[theta]=x[theta] +delt;
1.145 brouard 3945: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3946: p2[theta]=x[theta]-delt;
3947: k2=func(p2)-fx;
3948: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3949: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3950:
1.203 brouard 3951: #ifdef DEBUGHESSII
1.126 brouard 3952: 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);
3953: 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);
3954: #endif
3955: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3956: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3957: k=kmax;
3958: }
3959: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3960: k=kmax; l=lmax*10;
1.126 brouard 3961: }
3962: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3963: delts=delt;
3964: }
1.203 brouard 3965: } /* End loop k */
1.126 brouard 3966: }
3967: delti[theta]=delts;
3968: return res;
3969:
3970: }
3971:
1.203 brouard 3972: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3973: {
3974: int i;
1.164 brouard 3975: int l=1, lmax=20;
1.126 brouard 3976: double k1,k2,k3,k4,res,fx;
1.132 brouard 3977: double p2[MAXPARM+1];
1.203 brouard 3978: int k, kmax=1;
3979: double v1, v2, cv12, lc1, lc2;
1.208 brouard 3980:
3981: int firstime=0;
1.203 brouard 3982:
1.126 brouard 3983: fx=func(x);
1.203 brouard 3984: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 3985: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 3986: p2[thetai]=x[thetai]+delti[thetai]*k;
3987: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3988: k1=func(p2)-fx;
3989:
1.203 brouard 3990: p2[thetai]=x[thetai]+delti[thetai]*k;
3991: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3992: k2=func(p2)-fx;
3993:
1.203 brouard 3994: p2[thetai]=x[thetai]-delti[thetai]*k;
3995: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3996: k3=func(p2)-fx;
3997:
1.203 brouard 3998: p2[thetai]=x[thetai]-delti[thetai]*k;
3999: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4000: k4=func(p2)-fx;
1.203 brouard 4001: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4002: if(k1*k2*k3*k4 <0.){
1.208 brouard 4003: firstime=1;
1.203 brouard 4004: kmax=kmax+10;
1.208 brouard 4005: }
4006: if(kmax >=10 || firstime ==1){
1.218 brouard 4007: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
4008: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you may increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 4009: 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);
4010: 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);
4011: }
4012: #ifdef DEBUGHESSIJ
4013: v1=hess[thetai][thetai];
4014: v2=hess[thetaj][thetaj];
4015: cv12=res;
4016: /* Computing eigen value of Hessian matrix */
4017: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4018: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4019: if ((lc2 <0) || (lc1 <0) ){
4020: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4021: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4022: 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);
4023: 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);
4024: }
1.126 brouard 4025: #endif
4026: }
4027: return res;
4028: }
4029:
1.203 brouard 4030: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4031: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4032: /* { */
4033: /* int i; */
4034: /* int l=1, lmax=20; */
4035: /* double k1,k2,k3,k4,res,fx; */
4036: /* double p2[MAXPARM+1]; */
4037: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4038: /* int k=0,kmax=10; */
4039: /* double l1; */
4040:
4041: /* fx=func(x); */
4042: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4043: /* l1=pow(10,l); */
4044: /* delts=delt; */
4045: /* for(k=1 ; k <kmax; k=k+1){ */
4046: /* delt = delti*(l1*k); */
4047: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4048: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4049: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4050: /* k1=func(p2)-fx; */
4051:
4052: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4053: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4054: /* k2=func(p2)-fx; */
4055:
4056: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4057: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4058: /* k3=func(p2)-fx; */
4059:
4060: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4061: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4062: /* k4=func(p2)-fx; */
4063: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4064: /* #ifdef DEBUGHESSIJ */
4065: /* 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); */
4066: /* 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); */
4067: /* #endif */
4068: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4069: /* k=kmax; */
4070: /* } */
4071: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4072: /* k=kmax; l=lmax*10; */
4073: /* } */
4074: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4075: /* delts=delt; */
4076: /* } */
4077: /* } /\* End loop k *\/ */
4078: /* } */
4079: /* delti[theta]=delts; */
4080: /* return res; */
4081: /* } */
4082:
4083:
1.126 brouard 4084: /************** Inverse of matrix **************/
4085: void ludcmp(double **a, int n, int *indx, double *d)
4086: {
4087: int i,imax,j,k;
4088: double big,dum,sum,temp;
4089: double *vv;
4090:
4091: vv=vector(1,n);
4092: *d=1.0;
4093: for (i=1;i<=n;i++) {
4094: big=0.0;
4095: for (j=1;j<=n;j++)
4096: if ((temp=fabs(a[i][j])) > big) big=temp;
4097: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4098: vv[i]=1.0/big;
4099: }
4100: for (j=1;j<=n;j++) {
4101: for (i=1;i<j;i++) {
4102: sum=a[i][j];
4103: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4104: a[i][j]=sum;
4105: }
4106: big=0.0;
4107: for (i=j;i<=n;i++) {
4108: sum=a[i][j];
4109: for (k=1;k<j;k++)
4110: sum -= a[i][k]*a[k][j];
4111: a[i][j]=sum;
4112: if ( (dum=vv[i]*fabs(sum)) >= big) {
4113: big=dum;
4114: imax=i;
4115: }
4116: }
4117: if (j != imax) {
4118: for (k=1;k<=n;k++) {
4119: dum=a[imax][k];
4120: a[imax][k]=a[j][k];
4121: a[j][k]=dum;
4122: }
4123: *d = -(*d);
4124: vv[imax]=vv[j];
4125: }
4126: indx[j]=imax;
4127: if (a[j][j] == 0.0) a[j][j]=TINY;
4128: if (j != n) {
4129: dum=1.0/(a[j][j]);
4130: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4131: }
4132: }
4133: free_vector(vv,1,n); /* Doesn't work */
4134: ;
4135: }
4136:
4137: void lubksb(double **a, int n, int *indx, double b[])
4138: {
4139: int i,ii=0,ip,j;
4140: double sum;
4141:
4142: for (i=1;i<=n;i++) {
4143: ip=indx[i];
4144: sum=b[ip];
4145: b[ip]=b[i];
4146: if (ii)
4147: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4148: else if (sum) ii=i;
4149: b[i]=sum;
4150: }
4151: for (i=n;i>=1;i--) {
4152: sum=b[i];
4153: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4154: b[i]=sum/a[i][i];
4155: }
4156: }
4157:
4158: void pstamp(FILE *fichier)
4159: {
1.196 brouard 4160: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4161: }
4162:
4163: /************ Frequencies ********************/
1.226 brouard 4164: void freqsummary(char fileres[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
4165: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4166: int firstpass, int lastpass, int stepm, int weightopt, char model[])
4167: { /* Some frequencies */
4168:
1.227 brouard 4169: int i, m, jk, j1, bool, z1,j, k, iv;
1.226 brouard 4170: int iind=0, iage=0;
4171: int mi; /* Effective wave */
4172: int first;
4173: double ***freq; /* Frequencies */
4174: double *meanq;
4175: double **meanqt;
4176: double *pp, **prop, *posprop, *pospropt;
4177: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4178: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4179: double agebegin, ageend;
4180:
4181: pp=vector(1,nlstate);
4182: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4183: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4184: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4185: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4186: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4187: meanqt=matrix(1,lastpass,1,nqtveff);
4188: strcpy(fileresp,"P_");
4189: strcat(fileresp,fileresu);
4190: /*strcat(fileresphtm,fileresu);*/
4191: if((ficresp=fopen(fileresp,"w"))==NULL) {
4192: printf("Problem with prevalence resultfile: %s\n", fileresp);
4193: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4194: exit(0);
4195: }
1.240 brouard 4196:
1.226 brouard 4197: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4198: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4199: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4200: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4201: fflush(ficlog);
4202: exit(70);
4203: }
4204: else{
4205: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4206: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4207: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4208: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4209: }
1.237 brouard 4210: 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 4211:
1.226 brouard 4212: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4213: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4214: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4215: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4216: fflush(ficlog);
4217: exit(70);
1.240 brouard 4218: } else{
1.226 brouard 4219: 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 4220: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4221: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4222: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4223: }
1.240 brouard 4224: 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);
4225:
1.226 brouard 4226: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4227: j1=0;
1.126 brouard 4228:
1.227 brouard 4229: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4230: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4231: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4232:
1.226 brouard 4233: first=1;
1.240 brouard 4234:
1.226 brouard 4235: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4236: reference=low_education V1=0,V2=0
4237: med_educ V1=1 V2=0,
4238: high_educ V1=0 V2=1
4239: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4240: */
1.240 brouard 4241:
1.227 brouard 4242: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on covariates combination in order of model, excluding quantitatives V4=0, V3=0 for example, fixed or varying covariates */
1.226 brouard 4243: posproptt=0.;
4244: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4245: scanf("%d", i);*/
4246: for (i=-5; i<=nlstate+ndeath; i++)
4247: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4248: for(m=iagemin; m <= iagemax+3; m++)
4249: freq[i][jk][m]=0;
4250:
1.226 brouard 4251: for (i=1; i<=nlstate; i++) {
4252: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4253: prop[i][m]=0;
1.226 brouard 4254: posprop[i]=0;
4255: pospropt[i]=0;
4256: }
1.227 brouard 4257: /* for (z1=1; z1<= nqfveff; z1++) { */
4258: /* meanq[z1]+=0.; */
4259: /* for(m=1;m<=lastpass;m++){ */
4260: /* meanqt[m][z1]=0.; */
4261: /* } */
4262: /* } */
1.240 brouard 4263:
1.226 brouard 4264: dateintsum=0;
4265: k2cpt=0;
1.227 brouard 4266: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4267: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4268: bool=1;
1.227 brouard 4269: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4270: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4271: /* for (z1=1; z1<= nqfveff; z1++) { */
4272: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4273: /* } */
1.234 brouard 4274: for (z1=1; z1<=cptcoveff; z1++) {
4275: /* if(Tvaraff[z1] ==-20){ */
4276: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4277: /* }else if(Tvaraff[z1] ==-10){ */
4278: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4279: /* }else */
4280: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){
4281: /* Tests if this individual iind responded to j1 (V4=1 V3=0) */
4282: bool=0;
4283: /* 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",
4284: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4285: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4286: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4287: } /* Onlyf fixed */
4288: } /* end z1 */
4289: } /* cptcovn > 0 */
1.227 brouard 4290: } /* end any */
4291: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4292: /* for(m=firstpass; m<=lastpass; m++){ */
4293: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4294: m=mw[mi][iind];
4295: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4296: for (z1=1; z1<=cptcoveff; z1++) {
4297: if( Fixed[Tmodelind[z1]]==1){
4298: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4299: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4300: bool=0;
4301: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4302: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4303: bool=0;
4304: }
4305: }
4306: }
4307: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4308: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4309: if(bool==1){
4310: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4311: and mw[mi+1][iind]. dh depends on stepm. */
4312: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4313: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4314: if(m >=firstpass && m <=lastpass){
4315: k2=anint[m][iind]+(mint[m][iind]/12.);
4316: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4317: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4318: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4319: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4320: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4321: if (m<lastpass) {
4322: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4323: /* 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]); */
4324: if(s[m][iind]==-1)
4325: 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.));
4326: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4327: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4328: 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 */
4329: }
4330: } /* end if between passes */
4331: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99)) {
4332: dateintsum=dateintsum+k2;
4333: k2cpt++;
4334: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4335: }
4336: } /* end bool 2 */
4337: } /* end m */
1.226 brouard 4338: } /* end bool */
4339: } /* end iind = 1 to imx */
4340: /* prop[s][age] is feeded for any initial and valid live state as well as
4341: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4342:
4343:
1.226 brouard 4344: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4345: pstamp(ficresp);
1.240 brouard 4346: if (cptcoveff>0){
1.226 brouard 4347: fprintf(ficresp, "\n#********** Variable ");
4348: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4349: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4350: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4351: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4352: if(DummyV[z1]){
4353: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4354: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4355: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4356: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4357: }else{
4358: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4359: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4360: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4361: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4362: }
1.226 brouard 4363: }
4364: fprintf(ficresp, "**********\n#");
4365: fprintf(ficresphtm, "**********</h3>\n");
4366: fprintf(ficresphtmfr, "**********</h3>\n");
4367: fprintf(ficlog, "**********\n");
4368: }
4369: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4370: for(i=1; i<=nlstate;i++) {
1.240 brouard 4371: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4372: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4373: }
4374: fprintf(ficresp, "\n");
4375: fprintf(ficresphtm, "\n");
1.240 brouard 4376:
1.226 brouard 4377: /* Header of frequency table by age */
4378: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4379: fprintf(ficresphtmfr,"<th>Age</th> ");
4380: for(jk=-1; jk <=nlstate+ndeath; jk++){
4381: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4382: if(jk!=0 && m!=0)
4383: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4384: }
4385: }
4386: fprintf(ficresphtmfr, "\n");
1.240 brouard 4387:
1.226 brouard 4388: /* For each age */
4389: for(iage=iagemin; iage <= iagemax+3; iage++){
4390: fprintf(ficresphtm,"<tr>");
4391: if(iage==iagemax+1){
1.240 brouard 4392: fprintf(ficlog,"1");
4393: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4394: }else if(iage==iagemax+2){
1.240 brouard 4395: fprintf(ficlog,"0");
4396: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4397: }else if(iage==iagemax+3){
1.240 brouard 4398: fprintf(ficlog,"Total");
4399: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4400: }else{
1.240 brouard 4401: if(first==1){
4402: first=0;
4403: printf("See log file for details...\n");
4404: }
4405: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4406: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4407: }
4408: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4409: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4410: pp[jk] += freq[jk][m][iage];
1.226 brouard 4411: }
4412: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4413: for(m=-1, pos=0; m <=0 ; m++)
4414: pos += freq[jk][m][iage];
4415: if(pp[jk]>=1.e-10){
4416: if(first==1){
4417: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4418: }
4419: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4420: }else{
4421: if(first==1)
4422: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4423: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4424: }
1.226 brouard 4425: }
1.240 brouard 4426:
1.226 brouard 4427: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4428: /* posprop[jk]=0; */
4429: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4430: pp[jk] += freq[jk][m][iage];
1.226 brouard 4431: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4432:
1.226 brouard 4433: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4434: pos += pp[jk]; /* pos is the total number of transitions until this age */
4435: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4436: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4437: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4438: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4439: }
4440: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4441: if(pos>=1.e-5){
4442: if(first==1)
4443: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4444: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4445: }else{
4446: if(first==1)
4447: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4448: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4449: }
4450: if( iage <= iagemax){
4451: if(pos>=1.e-5){
4452: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4453: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4454: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4455: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4456: }
4457: else{
4458: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4459: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4460: }
4461: }
4462: pospropt[jk] +=posprop[jk];
1.226 brouard 4463: } /* end loop jk */
4464: /* pospropt=0.; */
4465: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4466: for(m=-1; m <=nlstate+ndeath; m++){
4467: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4468: if(first==1){
4469: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4470: }
4471: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4472: }
4473: if(jk!=0 && m!=0)
4474: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4475: }
1.226 brouard 4476: } /* end loop jk */
4477: posproptt=0.;
4478: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4479: posproptt += pospropt[jk];
1.226 brouard 4480: }
4481: fprintf(ficresphtmfr,"</tr>\n ");
4482: if(iage <= iagemax){
1.240 brouard 4483: fprintf(ficresp,"\n");
4484: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4485: }
4486: if(first==1)
1.240 brouard 4487: printf("Others in log...\n");
1.226 brouard 4488: fprintf(ficlog,"\n");
4489: } /* end loop age iage */
4490: fprintf(ficresphtm,"<tr><th>Tot</th>");
4491: for(jk=1; jk <=nlstate ; jk++){
4492: if(posproptt < 1.e-5){
1.240 brouard 4493: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4494: }else{
1.240 brouard 4495: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4496: }
4497: }
4498: fprintf(ficresphtm,"</tr>\n");
4499: fprintf(ficresphtm,"</table>\n");
4500: fprintf(ficresphtmfr,"</table>\n");
4501: if(posproptt < 1.e-5){
4502: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4503: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4504: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4505: invalidvarcomb[j1]=1;
4506: }else{
4507: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4508: invalidvarcomb[j1]=0;
4509: }
4510: fprintf(ficresphtmfr,"</table>\n");
4511: } /* end selected combination of covariate j1 */
4512: dateintmean=dateintsum/k2cpt;
1.240 brouard 4513:
1.226 brouard 4514: fclose(ficresp);
4515: fclose(ficresphtm);
4516: fclose(ficresphtmfr);
4517: free_vector(meanq,1,nqfveff);
4518: free_matrix(meanqt,1,lastpass,1,nqtveff);
4519: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4520: free_vector(pospropt,1,nlstate);
4521: free_vector(posprop,1,nlstate);
4522: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4523: free_vector(pp,1,nlstate);
4524: /* End of freqsummary */
4525: }
1.126 brouard 4526:
4527: /************ Prevalence ********************/
1.227 brouard 4528: 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)
4529: {
4530: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4531: in each health status at the date of interview (if between dateprev1 and dateprev2).
4532: We still use firstpass and lastpass as another selection.
4533: */
1.126 brouard 4534:
1.227 brouard 4535: int i, m, jk, j1, bool, z1,j, iv;
4536: int mi; /* Effective wave */
4537: int iage;
4538: double agebegin, ageend;
4539:
4540: double **prop;
4541: double posprop;
4542: double y2; /* in fractional years */
4543: int iagemin, iagemax;
4544: int first; /** to stop verbosity which is redirected to log file */
4545:
4546: iagemin= (int) agemin;
4547: iagemax= (int) agemax;
4548: /*pp=vector(1,nlstate);*/
4549: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4550: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4551: j1=0;
1.222 brouard 4552:
1.227 brouard 4553: /*j=cptcoveff;*/
4554: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4555:
1.227 brouard 4556: first=1;
4557: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4558: for (i=1; i<=nlstate; i++)
4559: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4560: prop[i][iage]=0.0;
4561: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4562: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4563: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4564:
4565: for (i=1; i<=imx; i++) { /* Each individual */
4566: bool=1;
4567: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4568: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4569: m=mw[mi][i];
4570: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4571: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4572: for (z1=1; z1<=cptcoveff; z1++){
4573: if( Fixed[Tmodelind[z1]]==1){
4574: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4575: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4576: bool=0;
4577: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4578: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4579: bool=0;
4580: }
4581: }
4582: if(bool==1){ /* Otherwise we skip that wave/person */
4583: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4584: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4585: if(m >=firstpass && m <=lastpass){
4586: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4587: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4588: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4589: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4590: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4591: 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);
4592: exit(1);
4593: }
4594: if (s[m][i]>0 && s[m][i]<=nlstate) {
4595: /*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]]);*/
4596: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4597: prop[s[m][i]][iagemax+3] += weight[i];
4598: } /* end valid statuses */
4599: } /* end selection of dates */
4600: } /* end selection of waves */
4601: } /* end bool */
4602: } /* end wave */
4603: } /* end individual */
4604: for(i=iagemin; i <= iagemax+3; i++){
4605: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4606: posprop += prop[jk][i];
4607: }
4608:
4609: for(jk=1; jk <=nlstate ; jk++){
4610: if( i <= iagemax){
4611: if(posprop>=1.e-5){
4612: probs[i][jk][j1]= prop[jk][i]/posprop;
4613: } else{
4614: if(first==1){
4615: first=0;
4616: printf("Warning Observed prevalence probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,j1,probs[i][jk][j1]);
4617: }
4618: }
4619: }
4620: }/* end jk */
4621: }/* end i */
1.222 brouard 4622: /*} *//* end i1 */
1.227 brouard 4623: } /* end j1 */
1.222 brouard 4624:
1.227 brouard 4625: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4626: /*free_vector(pp,1,nlstate);*/
4627: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4628: } /* End of prevalence */
1.126 brouard 4629:
4630: /************* Waves Concatenation ***************/
4631:
4632: 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)
4633: {
4634: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4635: Death is a valid wave (if date is known).
4636: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4637: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4638: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4639: */
1.126 brouard 4640:
1.224 brouard 4641: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4642: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4643: double sum=0., jmean=0.;*/
1.224 brouard 4644: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4645: int j, k=0,jk, ju, jl;
4646: double sum=0.;
4647: first=0;
1.214 brouard 4648: firstwo=0;
1.217 brouard 4649: firsthree=0;
1.218 brouard 4650: firstfour=0;
1.164 brouard 4651: jmin=100000;
1.126 brouard 4652: jmax=-1;
4653: jmean=0.;
1.224 brouard 4654:
4655: /* Treating live states */
1.214 brouard 4656: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4657: mi=0; /* First valid wave */
1.227 brouard 4658: mli=0; /* Last valid wave */
1.126 brouard 4659: m=firstpass;
1.214 brouard 4660: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4661: 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 */
4662: mli=m-1;/* mw[++mi][i]=m-1; */
4663: }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 */
4664: mw[++mi][i]=m;
4665: mli=m;
1.224 brouard 4666: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4667: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4668: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4669: }
1.227 brouard 4670: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4671: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4672: break;
1.224 brouard 4673: #else
1.227 brouard 4674: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4675: if(firsthree == 0){
4676: 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 pi. .\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);
4677: firsthree=1;
4678: }
4679: 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 pi. .\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);
4680: mw[++mi][i]=m;
4681: mli=m;
4682: }
4683: if(s[m][i]==-2){ /* Vital status is really unknown */
4684: nbwarn++;
4685: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4686: 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);
4687: 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);
4688: }
4689: break;
4690: }
4691: break;
1.224 brouard 4692: #endif
1.227 brouard 4693: }/* End m >= lastpass */
1.126 brouard 4694: }/* end while */
1.224 brouard 4695:
1.227 brouard 4696: /* 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 4697: /* After last pass */
1.224 brouard 4698: /* Treating death states */
1.214 brouard 4699: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4700: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4701: /* } */
1.126 brouard 4702: mi++; /* Death is another wave */
4703: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4704: /* Only death is a correct wave */
1.126 brouard 4705: mw[mi][i]=m;
1.224 brouard 4706: }
4707: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4708: else if ((int) andc[i] != 9999) { /* Status is negative. A death occured after lastpass, we can't take it into account because of potential bias */
1.216 brouard 4709: /* m++; */
4710: /* mi++; */
4711: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4712: /* mw[mi][i]=m; */
1.218 brouard 4713: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4714: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* death occured before last wave and status should have been death instead of -1 */
4715: nbwarn++;
4716: if(firstfiv==0){
4717: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
4718: firstfiv=1;
4719: }else{
4720: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d interviewed at %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
4721: }
4722: }else{ /* Death occured afer last wave potential bias */
4723: nberr++;
4724: if(firstwo==0){
4725: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4726: firstwo=1;
4727: }
4728: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4729: }
1.218 brouard 4730: }else{ /* end date of interview is known */
1.227 brouard 4731: /* death is known but not confirmed by death status at any wave */
4732: if(firstfour==0){
4733: printf("Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4734: firstfour=1;
4735: }
4736: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.214 brouard 4737: }
1.224 brouard 4738: } /* end if date of death is known */
4739: #endif
4740: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4741: /* wav[i]=mw[mi][i]; */
1.126 brouard 4742: if(mi==0){
4743: nbwarn++;
4744: if(first==0){
1.227 brouard 4745: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4746: first=1;
1.126 brouard 4747: }
4748: if(first==1){
1.227 brouard 4749: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4750: }
4751: } /* end mi==0 */
4752: } /* End individuals */
1.214 brouard 4753: /* wav and mw are no more changed */
1.223 brouard 4754:
1.214 brouard 4755:
1.126 brouard 4756: for(i=1; i<=imx; i++){
4757: for(mi=1; mi<wav[i];mi++){
4758: if (stepm <=0)
1.227 brouard 4759: dh[mi][i]=1;
1.126 brouard 4760: else{
1.227 brouard 4761: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4762: if (agedc[i] < 2*AGESUP) {
4763: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4764: if(j==0) j=1; /* Survives at least one month after exam */
4765: else if(j<0){
4766: nberr++;
4767: 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]);
4768: j=1; /* Temporary Dangerous patch */
4769: 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);
4770: 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]);
4771: 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);
4772: }
4773: k=k+1;
4774: if (j >= jmax){
4775: jmax=j;
4776: ijmax=i;
4777: }
4778: if (j <= jmin){
4779: jmin=j;
4780: ijmin=i;
4781: }
4782: sum=sum+j;
4783: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4784: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4785: }
4786: }
4787: else{
4788: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4789: /* 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 4790:
1.227 brouard 4791: k=k+1;
4792: if (j >= jmax) {
4793: jmax=j;
4794: ijmax=i;
4795: }
4796: else if (j <= jmin){
4797: jmin=j;
4798: ijmin=i;
4799: }
4800: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4801: /*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]);*/
4802: if(j<0){
4803: nberr++;
4804: 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]);
4805: 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]);
4806: }
4807: sum=sum+j;
4808: }
4809: jk= j/stepm;
4810: jl= j -jk*stepm;
4811: ju= j -(jk+1)*stepm;
4812: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4813: if(jl==0){
4814: dh[mi][i]=jk;
4815: bh[mi][i]=0;
4816: }else{ /* We want a negative bias in order to only have interpolation ie
4817: * to avoid the price of an extra matrix product in likelihood */
4818: dh[mi][i]=jk+1;
4819: bh[mi][i]=ju;
4820: }
4821: }else{
4822: if(jl <= -ju){
4823: dh[mi][i]=jk;
4824: bh[mi][i]=jl; /* bias is positive if real duration
4825: * is higher than the multiple of stepm and negative otherwise.
4826: */
4827: }
4828: else{
4829: dh[mi][i]=jk+1;
4830: bh[mi][i]=ju;
4831: }
4832: if(dh[mi][i]==0){
4833: dh[mi][i]=1; /* At least one step */
4834: bh[mi][i]=ju; /* At least one step */
4835: /* 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);*/
4836: }
4837: } /* end if mle */
1.126 brouard 4838: }
4839: } /* end wave */
4840: }
4841: jmean=sum/k;
4842: 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 4843: 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 4844: }
1.126 brouard 4845:
4846: /*********** Tricode ****************************/
1.220 brouard 4847: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4848: {
4849: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4850: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4851: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4852: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4853: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4854: */
1.130 brouard 4855:
1.242 brouard 4856: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
4857: int modmaxcovj=0; /* Modality max of covariates j */
4858: int cptcode=0; /* Modality max of covariates j */
4859: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4860:
4861:
1.242 brouard 4862: /* cptcoveff=0; */
4863: /* *cptcov=0; */
1.126 brouard 4864:
1.242 brouard 4865: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4866:
1.242 brouard 4867: /* Loop on covariates without age and products and no quantitative variable */
4868: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
4869: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4870: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4871: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4872: switch(Fixed[k]) {
4873: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
4874: 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*/
4875: ij=(int)(covar[Tvar[k]][i]);
4876: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4877: * If product of Vn*Vm, still boolean *:
4878: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4879: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4880: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4881: modality of the nth covariate of individual i. */
4882: if (ij > modmaxcovj)
4883: modmaxcovj=ij;
4884: else if (ij < modmincovj)
4885: modmincovj=ij;
4886: if ((ij < -1) && (ij > NCOVMAX)){
4887: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
4888: exit(1);
4889: }else
4890: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
4891: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
4892: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
4893: /* getting the maximum value of the modality of the covariate
4894: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
4895: female ies 1, then modmaxcovj=1.
4896: */
4897: } /* end for loop on individuals i */
4898: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4899: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4900: cptcode=modmaxcovj;
4901: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
4902: /*for (i=0; i<=cptcode; i++) {*/
4903: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
4904: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4905: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4906: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
4907: if( j != -1){
4908: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
4909: covariate for which somebody answered excluding
4910: undefined. Usually 2: 0 and 1. */
4911: }
4912: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
4913: covariate for which somebody answered including
4914: undefined. Usually 3: -1, 0 and 1. */
4915: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
4916: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
4917: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 4918:
1.242 brouard 4919: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
4920: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
4921: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
4922: /* modmincovj=3; modmaxcovj = 7; */
4923: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
4924: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
4925: /* defining two dummy variables: variables V1_1 and V1_2.*/
4926: /* nbcode[Tvar[j]][ij]=k; */
4927: /* nbcode[Tvar[j]][1]=0; */
4928: /* nbcode[Tvar[j]][2]=1; */
4929: /* nbcode[Tvar[j]][3]=2; */
4930: /* To be continued (not working yet). */
4931: ij=0; /* ij is similar to i but can jump over null modalities */
4932: 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*/
4933: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
4934: break;
4935: }
4936: ij++;
4937: 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*/
4938: cptcode = ij; /* New max modality for covar j */
4939: } /* end of loop on modality i=-1 to 1 or more */
4940: break;
4941: case 1: /* Testing on varying covariate, could be simple and
4942: * should look at waves or product of fixed *
4943: * varying. No time to test -1, assuming 0 and 1 only */
4944: ij=0;
4945: for(i=0; i<=1;i++){
4946: nbcode[Tvar[k]][++ij]=i;
4947: }
4948: break;
4949: default:
4950: break;
4951: } /* end switch */
4952: } /* end dummy test */
4953:
4954: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
4955: /* /\*recode from 0 *\/ */
4956: /* k is a modality. If we have model=V1+V1*sex */
4957: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
4958: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
4959: /* } */
4960: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
4961: /* if (ij > ncodemax[j]) { */
4962: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4963: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4964: /* break; */
4965: /* } */
4966: /* } /\* end of loop on modality k *\/ */
4967: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
4968:
4969: for (k=-1; k< maxncov; k++) Ndum[k]=0;
4970: /* Look at fixed dummy (single or product) covariates to check empty modalities */
4971: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
4972: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
4973: 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 */
4974: 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 */
4975: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
4976: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
4977:
4978: ij=0;
4979: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
4980: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
4981: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
4982: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
4983: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
4984: /* If product not in single variable we don't print results */
4985: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
4986: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
4987: 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*/
4988: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
4989: 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 */
4990: if(Fixed[k]!=0)
4991: anyvaryingduminmodel=1;
4992: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
4993: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
4994: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
4995: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
4996: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
4997: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
4998: }
4999: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5000: /* ij--; */
5001: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5002: *cptcov=ij; /*Number of total real effective covariates: effective
5003: * because they can be excluded from the model and real
5004: * if in the model but excluded because missing values, but how to get k from ij?*/
5005: for(j=ij+1; j<= cptcovt; j++){
5006: Tvaraff[j]=0;
5007: Tmodelind[j]=0;
5008: }
5009: for(j=ntveff+1; j<= cptcovt; j++){
5010: TmodelInvind[j]=0;
5011: }
5012: /* To be sorted */
5013: ;
5014: }
1.126 brouard 5015:
1.145 brouard 5016:
1.126 brouard 5017: /*********** Health Expectancies ****************/
5018:
1.235 brouard 5019: 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 5020:
5021: {
5022: /* Health expectancies, no variances */
1.164 brouard 5023: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5024: int nhstepma, nstepma; /* Decreasing with age */
5025: double age, agelim, hf;
5026: double ***p3mat;
5027: double eip;
5028:
1.238 brouard 5029: /* pstamp(ficreseij); */
1.126 brouard 5030: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5031: fprintf(ficreseij,"# Age");
5032: for(i=1; i<=nlstate;i++){
5033: for(j=1; j<=nlstate;j++){
5034: fprintf(ficreseij," e%1d%1d ",i,j);
5035: }
5036: fprintf(ficreseij," e%1d. ",i);
5037: }
5038: fprintf(ficreseij,"\n");
5039:
5040:
5041: if(estepm < stepm){
5042: printf ("Problem %d lower than %d\n",estepm, stepm);
5043: }
5044: else hstepm=estepm;
5045: /* We compute the life expectancy from trapezoids spaced every estepm months
5046: * This is mainly to measure the difference between two models: for example
5047: * if stepm=24 months pijx are given only every 2 years and by summing them
5048: * we are calculating an estimate of the Life Expectancy assuming a linear
5049: * progression in between and thus overestimating or underestimating according
5050: * to the curvature of the survival function. If, for the same date, we
5051: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5052: * to compare the new estimate of Life expectancy with the same linear
5053: * hypothesis. A more precise result, taking into account a more precise
5054: * curvature will be obtained if estepm is as small as stepm. */
5055:
5056: /* For example we decided to compute the life expectancy with the smallest unit */
5057: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5058: nhstepm is the number of hstepm from age to agelim
5059: nstepm is the number of stepm from age to agelin.
5060: Look at hpijx to understand the reason of that which relies in memory size
5061: and note for a fixed period like estepm months */
5062: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5063: survival function given by stepm (the optimization length). Unfortunately it
5064: means that if the survival funtion is printed only each two years of age and if
5065: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5066: results. So we changed our mind and took the option of the best precision.
5067: */
5068: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5069:
5070: agelim=AGESUP;
5071: /* If stepm=6 months */
5072: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5073: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5074:
5075: /* nhstepm age range expressed in number of stepm */
5076: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5077: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5078: /* if (stepm >= YEARM) hstepm=1;*/
5079: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5080: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5081:
5082: for (age=bage; age<=fage; age ++){
5083: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5084: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5085: /* if (stepm >= YEARM) hstepm=1;*/
5086: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5087:
5088: /* If stepm=6 months */
5089: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5090: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5091:
1.235 brouard 5092: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5093:
5094: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5095:
5096: printf("%d|",(int)age);fflush(stdout);
5097: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5098:
5099: /* Computing expectancies */
5100: for(i=1; i<=nlstate;i++)
5101: for(j=1; j<=nlstate;j++)
5102: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5103: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5104:
5105: /* 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]);*/
5106:
5107: }
5108:
5109: fprintf(ficreseij,"%3.0f",age );
5110: for(i=1; i<=nlstate;i++){
5111: eip=0;
5112: for(j=1; j<=nlstate;j++){
5113: eip +=eij[i][j][(int)age];
5114: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5115: }
5116: fprintf(ficreseij,"%9.4f", eip );
5117: }
5118: fprintf(ficreseij,"\n");
5119:
5120: }
5121: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5122: printf("\n");
5123: fprintf(ficlog,"\n");
5124:
5125: }
5126:
1.235 brouard 5127: 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 5128:
5129: {
5130: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5131: to initial status i, ei. .
1.126 brouard 5132: */
5133: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5134: int nhstepma, nstepma; /* Decreasing with age */
5135: double age, agelim, hf;
5136: double ***p3matp, ***p3matm, ***varhe;
5137: double **dnewm,**doldm;
5138: double *xp, *xm;
5139: double **gp, **gm;
5140: double ***gradg, ***trgradg;
5141: int theta;
5142:
5143: double eip, vip;
5144:
5145: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5146: xp=vector(1,npar);
5147: xm=vector(1,npar);
5148: dnewm=matrix(1,nlstate*nlstate,1,npar);
5149: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5150:
5151: pstamp(ficresstdeij);
5152: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5153: fprintf(ficresstdeij,"# Age");
5154: for(i=1; i<=nlstate;i++){
5155: for(j=1; j<=nlstate;j++)
5156: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5157: fprintf(ficresstdeij," e%1d. ",i);
5158: }
5159: fprintf(ficresstdeij,"\n");
5160:
5161: pstamp(ficrescveij);
5162: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5163: fprintf(ficrescveij,"# Age");
5164: for(i=1; i<=nlstate;i++)
5165: for(j=1; j<=nlstate;j++){
5166: cptj= (j-1)*nlstate+i;
5167: for(i2=1; i2<=nlstate;i2++)
5168: for(j2=1; j2<=nlstate;j2++){
5169: cptj2= (j2-1)*nlstate+i2;
5170: if(cptj2 <= cptj)
5171: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5172: }
5173: }
5174: fprintf(ficrescveij,"\n");
5175:
5176: if(estepm < stepm){
5177: printf ("Problem %d lower than %d\n",estepm, stepm);
5178: }
5179: else hstepm=estepm;
5180: /* We compute the life expectancy from trapezoids spaced every estepm months
5181: * This is mainly to measure the difference between two models: for example
5182: * if stepm=24 months pijx are given only every 2 years and by summing them
5183: * we are calculating an estimate of the Life Expectancy assuming a linear
5184: * progression in between and thus overestimating or underestimating according
5185: * to the curvature of the survival function. If, for the same date, we
5186: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5187: * to compare the new estimate of Life expectancy with the same linear
5188: * hypothesis. A more precise result, taking into account a more precise
5189: * curvature will be obtained if estepm is as small as stepm. */
5190:
5191: /* For example we decided to compute the life expectancy with the smallest unit */
5192: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5193: nhstepm is the number of hstepm from age to agelim
5194: nstepm is the number of stepm from age to agelin.
5195: Look at hpijx to understand the reason of that which relies in memory size
5196: and note for a fixed period like estepm months */
5197: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5198: survival function given by stepm (the optimization length). Unfortunately it
5199: means that if the survival funtion is printed only each two years of age and if
5200: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5201: results. So we changed our mind and took the option of the best precision.
5202: */
5203: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5204:
5205: /* If stepm=6 months */
5206: /* nhstepm age range expressed in number of stepm */
5207: agelim=AGESUP;
5208: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5209: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5210: /* if (stepm >= YEARM) hstepm=1;*/
5211: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5212:
5213: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5214: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5215: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5216: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5217: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5218: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5219:
5220: for (age=bage; age<=fage; age ++){
5221: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5222: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5223: /* if (stepm >= YEARM) hstepm=1;*/
5224: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5225:
1.126 brouard 5226: /* If stepm=6 months */
5227: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5228: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5229:
5230: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5231:
1.126 brouard 5232: /* Computing Variances of health expectancies */
5233: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5234: decrease memory allocation */
5235: for(theta=1; theta <=npar; theta++){
5236: for(i=1; i<=npar; i++){
1.222 brouard 5237: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5238: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5239: }
1.235 brouard 5240: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5241: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5242:
1.126 brouard 5243: for(j=1; j<= nlstate; j++){
1.222 brouard 5244: for(i=1; i<=nlstate; i++){
5245: for(h=0; h<=nhstepm-1; h++){
5246: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5247: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5248: }
5249: }
1.126 brouard 5250: }
1.218 brouard 5251:
1.126 brouard 5252: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5253: for(h=0; h<=nhstepm-1; h++){
5254: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5255: }
1.126 brouard 5256: }/* End theta */
5257:
5258:
5259: for(h=0; h<=nhstepm-1; h++)
5260: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5261: for(theta=1; theta <=npar; theta++)
5262: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5263:
1.218 brouard 5264:
1.222 brouard 5265: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5266: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5267: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5268:
1.222 brouard 5269: printf("%d|",(int)age);fflush(stdout);
5270: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5271: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5272: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5273: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5274: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5275: for(ij=1;ij<=nlstate*nlstate;ij++)
5276: for(ji=1;ji<=nlstate*nlstate;ji++)
5277: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5278: }
5279: }
1.218 brouard 5280:
1.126 brouard 5281: /* Computing expectancies */
1.235 brouard 5282: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5283: for(i=1; i<=nlstate;i++)
5284: for(j=1; j<=nlstate;j++)
1.222 brouard 5285: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5286: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5287:
1.222 brouard 5288: /* 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 5289:
1.222 brouard 5290: }
1.218 brouard 5291:
1.126 brouard 5292: fprintf(ficresstdeij,"%3.0f",age );
5293: for(i=1; i<=nlstate;i++){
5294: eip=0.;
5295: vip=0.;
5296: for(j=1; j<=nlstate;j++){
1.222 brouard 5297: eip += eij[i][j][(int)age];
5298: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5299: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5300: 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 5301: }
5302: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5303: }
5304: fprintf(ficresstdeij,"\n");
1.218 brouard 5305:
1.126 brouard 5306: fprintf(ficrescveij,"%3.0f",age );
5307: for(i=1; i<=nlstate;i++)
5308: for(j=1; j<=nlstate;j++){
1.222 brouard 5309: cptj= (j-1)*nlstate+i;
5310: for(i2=1; i2<=nlstate;i2++)
5311: for(j2=1; j2<=nlstate;j2++){
5312: cptj2= (j2-1)*nlstate+i2;
5313: if(cptj2 <= cptj)
5314: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5315: }
1.126 brouard 5316: }
5317: fprintf(ficrescveij,"\n");
1.218 brouard 5318:
1.126 brouard 5319: }
5320: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5321: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5322: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5323: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5324: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5325: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5326: printf("\n");
5327: fprintf(ficlog,"\n");
1.218 brouard 5328:
1.126 brouard 5329: free_vector(xm,1,npar);
5330: free_vector(xp,1,npar);
5331: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5332: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5333: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5334: }
1.218 brouard 5335:
1.126 brouard 5336: /************ Variance ******************/
1.235 brouard 5337: 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 5338: {
5339: /* Variance of health expectancies */
5340: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5341: /* double **newm;*/
5342: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5343:
5344: /* int movingaverage(); */
5345: double **dnewm,**doldm;
5346: double **dnewmp,**doldmp;
5347: int i, j, nhstepm, hstepm, h, nstepm ;
5348: int k;
5349: double *xp;
5350: double **gp, **gm; /* for var eij */
5351: double ***gradg, ***trgradg; /*for var eij */
5352: double **gradgp, **trgradgp; /* for var p point j */
5353: double *gpp, *gmp; /* for var p point j */
5354: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5355: double ***p3mat;
5356: double age,agelim, hf;
5357: /* double ***mobaverage; */
5358: int theta;
5359: char digit[4];
5360: char digitp[25];
5361:
5362: char fileresprobmorprev[FILENAMELENGTH];
5363:
5364: if(popbased==1){
5365: if(mobilav!=0)
5366: strcpy(digitp,"-POPULBASED-MOBILAV_");
5367: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5368: }
5369: else
5370: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5371:
1.218 brouard 5372: /* if (mobilav!=0) { */
5373: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5374: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5375: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5376: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5377: /* } */
5378: /* } */
5379:
5380: strcpy(fileresprobmorprev,"PRMORPREV-");
5381: sprintf(digit,"%-d",ij);
5382: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5383: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5384: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5385: strcat(fileresprobmorprev,fileresu);
5386: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5387: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5388: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5389: }
5390: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5391: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5392: pstamp(ficresprobmorprev);
5393: 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 5394: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5395: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5396: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5397: }
5398: for(j=1;j<=cptcoveff;j++)
5399: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5400: fprintf(ficresprobmorprev,"\n");
5401:
1.218 brouard 5402: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5403: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5404: fprintf(ficresprobmorprev," p.%-d SE",j);
5405: for(i=1; i<=nlstate;i++)
5406: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5407: }
5408: fprintf(ficresprobmorprev,"\n");
5409:
5410: fprintf(ficgp,"\n# Routine varevsij");
5411: fprintf(ficgp,"\nunset title \n");
5412: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5413: 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");
5414: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5415: /* } */
5416: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5417: pstamp(ficresvij);
5418: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5419: if(popbased==1)
5420: 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);
5421: else
5422: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5423: fprintf(ficresvij,"# Age");
5424: for(i=1; i<=nlstate;i++)
5425: for(j=1; j<=nlstate;j++)
5426: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5427: fprintf(ficresvij,"\n");
5428:
5429: xp=vector(1,npar);
5430: dnewm=matrix(1,nlstate,1,npar);
5431: doldm=matrix(1,nlstate,1,nlstate);
5432: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5433: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5434:
5435: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5436: gpp=vector(nlstate+1,nlstate+ndeath);
5437: gmp=vector(nlstate+1,nlstate+ndeath);
5438: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5439:
1.218 brouard 5440: if(estepm < stepm){
5441: printf ("Problem %d lower than %d\n",estepm, stepm);
5442: }
5443: else hstepm=estepm;
5444: /* For example we decided to compute the life expectancy with the smallest unit */
5445: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5446: nhstepm is the number of hstepm from age to agelim
5447: nstepm is the number of stepm from age to agelim.
5448: Look at function hpijx to understand why because of memory size limitations,
5449: we decided (b) to get a life expectancy respecting the most precise curvature of the
5450: survival function given by stepm (the optimization length). Unfortunately it
5451: means that if the survival funtion is printed every two years of age and if
5452: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5453: results. So we changed our mind and took the option of the best precision.
5454: */
5455: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5456: agelim = AGESUP;
5457: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5458: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5459: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5460: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5461: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5462: gp=matrix(0,nhstepm,1,nlstate);
5463: gm=matrix(0,nhstepm,1,nlstate);
5464:
5465:
5466: for(theta=1; theta <=npar; theta++){
5467: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5468: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5469: }
5470:
1.242 brouard 5471: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5472:
5473: if (popbased==1) {
5474: if(mobilav ==0){
5475: for(i=1; i<=nlstate;i++)
5476: prlim[i][i]=probs[(int)age][i][ij];
5477: }else{ /* mobilav */
5478: for(i=1; i<=nlstate;i++)
5479: prlim[i][i]=mobaverage[(int)age][i][ij];
5480: }
5481: }
5482:
1.235 brouard 5483: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=1 to nhstepm */
1.218 brouard 5484: for(j=1; j<= nlstate; j++){
5485: for(h=0; h<=nhstepm; h++){
5486: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5487: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5488: }
5489: }
5490: /* Next for computing probability of death (h=1 means
5491: computed over hstepm matrices product = hstepm*stepm months)
5492: as a weighted average of prlim.
5493: */
5494: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5495: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5496: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5497: }
5498: /* end probability of death */
5499:
5500: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5501: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5502:
1.242 brouard 5503: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5504:
5505: if (popbased==1) {
5506: if(mobilav ==0){
5507: for(i=1; i<=nlstate;i++)
5508: prlim[i][i]=probs[(int)age][i][ij];
5509: }else{ /* mobilav */
5510: for(i=1; i<=nlstate;i++)
5511: prlim[i][i]=mobaverage[(int)age][i][ij];
5512: }
5513: }
5514:
1.235 brouard 5515: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5516:
5517: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5518: for(h=0; h<=nhstepm; h++){
5519: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5520: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5521: }
5522: }
5523: /* This for computing probability of death (h=1 means
5524: computed over hstepm matrices product = hstepm*stepm months)
5525: as a weighted average of prlim.
5526: */
5527: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5528: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5529: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5530: }
5531: /* end probability of death */
5532:
5533: for(j=1; j<= nlstate; j++) /* vareij */
5534: for(h=0; h<=nhstepm; h++){
5535: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5536: }
5537:
5538: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5539: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5540: }
5541:
5542: } /* End theta */
5543:
5544: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5545:
5546: for(h=0; h<=nhstepm; h++) /* veij */
5547: for(j=1; j<=nlstate;j++)
5548: for(theta=1; theta <=npar; theta++)
5549: trgradg[h][j][theta]=gradg[h][theta][j];
5550:
5551: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5552: for(theta=1; theta <=npar; theta++)
5553: trgradgp[j][theta]=gradgp[theta][j];
5554:
5555:
5556: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5557: for(i=1;i<=nlstate;i++)
5558: for(j=1;j<=nlstate;j++)
5559: vareij[i][j][(int)age] =0.;
5560:
5561: for(h=0;h<=nhstepm;h++){
5562: for(k=0;k<=nhstepm;k++){
5563: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5564: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5565: for(i=1;i<=nlstate;i++)
5566: for(j=1;j<=nlstate;j++)
5567: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5568: }
5569: }
5570:
5571: /* pptj */
5572: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5573: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5574: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5575: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5576: varppt[j][i]=doldmp[j][i];
5577: /* end ppptj */
5578: /* x centered again */
5579:
1.242 brouard 5580: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5581:
5582: if (popbased==1) {
5583: if(mobilav ==0){
5584: for(i=1; i<=nlstate;i++)
5585: prlim[i][i]=probs[(int)age][i][ij];
5586: }else{ /* mobilav */
5587: for(i=1; i<=nlstate;i++)
5588: prlim[i][i]=mobaverage[(int)age][i][ij];
5589: }
5590: }
5591:
5592: /* This for computing probability of death (h=1 means
5593: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5594: as a weighted average of prlim.
5595: */
1.235 brouard 5596: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5597: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5598: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5599: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5600: }
5601: /* end probability of death */
5602:
5603: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5604: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5605: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5606: for(i=1; i<=nlstate;i++){
5607: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5608: }
5609: }
5610: fprintf(ficresprobmorprev,"\n");
5611:
5612: fprintf(ficresvij,"%.0f ",age );
5613: for(i=1; i<=nlstate;i++)
5614: for(j=1; j<=nlstate;j++){
5615: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5616: }
5617: fprintf(ficresvij,"\n");
5618: free_matrix(gp,0,nhstepm,1,nlstate);
5619: free_matrix(gm,0,nhstepm,1,nlstate);
5620: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5621: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5622: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5623: } /* End age */
5624: free_vector(gpp,nlstate+1,nlstate+ndeath);
5625: free_vector(gmp,nlstate+1,nlstate+ndeath);
5626: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5627: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5628: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5629: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5630: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5631: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5632: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5633: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5634: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5635: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5636: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5637: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5638: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5639: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5640: 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);
5641: /* 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 5642: */
1.218 brouard 5643: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5644: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5645:
1.218 brouard 5646: free_vector(xp,1,npar);
5647: free_matrix(doldm,1,nlstate,1,nlstate);
5648: free_matrix(dnewm,1,nlstate,1,npar);
5649: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5650: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5651: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5652: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5653: fclose(ficresprobmorprev);
5654: fflush(ficgp);
5655: fflush(fichtm);
5656: } /* end varevsij */
1.126 brouard 5657:
5658: /************ Variance of prevlim ******************/
1.235 brouard 5659: void varprevlim(char fileres[], 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 5660: {
1.205 brouard 5661: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5662: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5663:
1.126 brouard 5664: double **dnewm,**doldm;
5665: int i, j, nhstepm, hstepm;
5666: double *xp;
5667: double *gp, *gm;
5668: double **gradg, **trgradg;
1.208 brouard 5669: double **mgm, **mgp;
1.126 brouard 5670: double age,agelim;
5671: int theta;
5672:
5673: pstamp(ficresvpl);
5674: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5675: fprintf(ficresvpl,"# Age ");
5676: if(nresult >=1)
5677: fprintf(ficresvpl," Result# ");
1.126 brouard 5678: for(i=1; i<=nlstate;i++)
5679: fprintf(ficresvpl," %1d-%1d",i,i);
5680: fprintf(ficresvpl,"\n");
5681:
5682: xp=vector(1,npar);
5683: dnewm=matrix(1,nlstate,1,npar);
5684: doldm=matrix(1,nlstate,1,nlstate);
5685:
5686: hstepm=1*YEARM; /* Every year of age */
5687: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5688: agelim = AGESUP;
5689: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5690: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5691: if (stepm >= YEARM) hstepm=1;
5692: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5693: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5694: mgp=matrix(1,npar,1,nlstate);
5695: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5696: gp=vector(1,nlstate);
5697: gm=vector(1,nlstate);
5698:
5699: for(theta=1; theta <=npar; theta++){
5700: for(i=1; i<=npar; i++){ /* Computes gradient */
5701: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5702: }
1.209 brouard 5703: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5704: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5705: else
1.235 brouard 5706: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5707: for(i=1;i<=nlstate;i++){
1.126 brouard 5708: gp[i] = prlim[i][i];
1.208 brouard 5709: mgp[theta][i] = prlim[i][i];
5710: }
1.126 brouard 5711: for(i=1; i<=npar; i++) /* Computes gradient */
5712: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5713: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5714: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5715: else
1.235 brouard 5716: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5717: for(i=1;i<=nlstate;i++){
1.126 brouard 5718: gm[i] = prlim[i][i];
1.208 brouard 5719: mgm[theta][i] = prlim[i][i];
5720: }
1.126 brouard 5721: for(i=1;i<=nlstate;i++)
5722: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5723: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5724: } /* End theta */
5725:
5726: trgradg =matrix(1,nlstate,1,npar);
5727:
5728: for(j=1; j<=nlstate;j++)
5729: for(theta=1; theta <=npar; theta++)
5730: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5731: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5732: /* printf("\nmgm mgp %d ",(int)age); */
5733: /* for(j=1; j<=nlstate;j++){ */
5734: /* printf(" %d ",j); */
5735: /* for(theta=1; theta <=npar; theta++) */
5736: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5737: /* printf("\n "); */
5738: /* } */
5739: /* } */
5740: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5741: /* printf("\n gradg %d ",(int)age); */
5742: /* for(j=1; j<=nlstate;j++){ */
5743: /* printf("%d ",j); */
5744: /* for(theta=1; theta <=npar; theta++) */
5745: /* printf("%d %lf ",theta,gradg[theta][j]); */
5746: /* printf("\n "); */
5747: /* } */
5748: /* } */
1.126 brouard 5749:
5750: for(i=1;i<=nlstate;i++)
5751: varpl[i][(int)age] =0.;
1.209 brouard 5752: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5753: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5754: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5755: }else{
1.126 brouard 5756: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5757: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5758: }
1.126 brouard 5759: for(i=1;i<=nlstate;i++)
5760: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5761:
5762: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5763: if(nresult >=1)
5764: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5765: for(i=1; i<=nlstate;i++)
5766: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5767: fprintf(ficresvpl,"\n");
5768: free_vector(gp,1,nlstate);
5769: free_vector(gm,1,nlstate);
1.208 brouard 5770: free_matrix(mgm,1,npar,1,nlstate);
5771: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5772: free_matrix(gradg,1,npar,1,nlstate);
5773: free_matrix(trgradg,1,nlstate,1,npar);
5774: } /* End age */
5775:
5776: free_vector(xp,1,npar);
5777: free_matrix(doldm,1,nlstate,1,npar);
5778: free_matrix(dnewm,1,nlstate,1,nlstate);
5779:
5780: }
5781:
5782: /************ Variance of one-step probabilities ******************/
5783: 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 5784: {
5785: int i, j=0, k1, l1, tj;
5786: int k2, l2, j1, z1;
5787: int k=0, l;
5788: int first=1, first1, first2;
5789: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5790: double **dnewm,**doldm;
5791: double *xp;
5792: double *gp, *gm;
5793: double **gradg, **trgradg;
5794: double **mu;
5795: double age, cov[NCOVMAX+1];
5796: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5797: int theta;
5798: char fileresprob[FILENAMELENGTH];
5799: char fileresprobcov[FILENAMELENGTH];
5800: char fileresprobcor[FILENAMELENGTH];
5801: double ***varpij;
5802:
5803: strcpy(fileresprob,"PROB_");
5804: strcat(fileresprob,fileres);
5805: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5806: printf("Problem with resultfile: %s\n", fileresprob);
5807: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5808: }
5809: strcpy(fileresprobcov,"PROBCOV_");
5810: strcat(fileresprobcov,fileresu);
5811: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5812: printf("Problem with resultfile: %s\n", fileresprobcov);
5813: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5814: }
5815: strcpy(fileresprobcor,"PROBCOR_");
5816: strcat(fileresprobcor,fileresu);
5817: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5818: printf("Problem with resultfile: %s\n", fileresprobcor);
5819: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5820: }
5821: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5822: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5823: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5824: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5825: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5826: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5827: pstamp(ficresprob);
5828: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5829: fprintf(ficresprob,"# Age");
5830: pstamp(ficresprobcov);
5831: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5832: fprintf(ficresprobcov,"# Age");
5833: pstamp(ficresprobcor);
5834: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5835: fprintf(ficresprobcor,"# Age");
1.126 brouard 5836:
5837:
1.222 brouard 5838: for(i=1; i<=nlstate;i++)
5839: for(j=1; j<=(nlstate+ndeath);j++){
5840: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5841: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5842: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5843: }
5844: /* fprintf(ficresprob,"\n");
5845: fprintf(ficresprobcov,"\n");
5846: fprintf(ficresprobcor,"\n");
5847: */
5848: xp=vector(1,npar);
5849: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5850: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5851: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5852: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5853: first=1;
5854: fprintf(ficgp,"\n# Routine varprob");
5855: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5856: fprintf(fichtm,"\n");
5857:
5858: 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.</li>\n",optionfilehtmcov);
5859: 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);
5860: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5861: and drawn. It helps understanding how is the covariance between two incidences.\
5862: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5863: 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 5864: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5865: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5866: standard deviations wide on each axis. <br>\
5867: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5868: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5869: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5870:
1.222 brouard 5871: cov[1]=1;
5872: /* tj=cptcoveff; */
1.225 brouard 5873: tj = (int) pow(2,cptcoveff);
1.222 brouard 5874: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5875: j1=0;
1.224 brouard 5876: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5877: if (cptcovn>0) {
5878: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5879: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5880: fprintf(ficresprob, "**********\n#\n");
5881: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5882: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5883: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5884:
1.222 brouard 5885: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5886: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5887: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5888:
5889:
1.222 brouard 5890: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5891: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5892: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5893:
1.222 brouard 5894: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5895: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5896: fprintf(ficresprobcor, "**********\n#");
5897: if(invalidvarcomb[j1]){
5898: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5899: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5900: continue;
5901: }
5902: }
5903: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5904: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5905: gp=vector(1,(nlstate)*(nlstate+ndeath));
5906: gm=vector(1,(nlstate)*(nlstate+ndeath));
5907: for (age=bage; age<=fage; age ++){
5908: cov[2]=age;
5909: if(nagesqr==1)
5910: cov[3]= age*age;
5911: for (k=1; k<=cptcovn;k++) {
5912: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5913: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5914: * 1 1 1 1 1
5915: * 2 2 1 1 1
5916: * 3 1 2 1 1
5917: */
5918: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5919: }
5920: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5921: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5922: for (k=1; k<=cptcovprod;k++)
5923: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5924:
5925:
1.222 brouard 5926: for(theta=1; theta <=npar; theta++){
5927: for(i=1; i<=npar; i++)
5928: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 5929:
1.222 brouard 5930: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 5931:
1.222 brouard 5932: k=0;
5933: for(i=1; i<= (nlstate); i++){
5934: for(j=1; j<=(nlstate+ndeath);j++){
5935: k=k+1;
5936: gp[k]=pmmij[i][j];
5937: }
5938: }
1.220 brouard 5939:
1.222 brouard 5940: for(i=1; i<=npar; i++)
5941: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 5942:
1.222 brouard 5943: pmij(pmmij,cov,ncovmodel,xp,nlstate);
5944: k=0;
5945: for(i=1; i<=(nlstate); i++){
5946: for(j=1; j<=(nlstate+ndeath);j++){
5947: k=k+1;
5948: gm[k]=pmmij[i][j];
5949: }
5950: }
1.220 brouard 5951:
1.222 brouard 5952: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
5953: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
5954: }
1.126 brouard 5955:
1.222 brouard 5956: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
5957: for(theta=1; theta <=npar; theta++)
5958: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 5959:
1.222 brouard 5960: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
5961: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 5962:
1.222 brouard 5963: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 5964:
1.222 brouard 5965: k=0;
5966: for(i=1; i<=(nlstate); i++){
5967: for(j=1; j<=(nlstate+ndeath);j++){
5968: k=k+1;
5969: mu[k][(int) age]=pmmij[i][j];
5970: }
5971: }
5972: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
5973: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
5974: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 5975:
1.222 brouard 5976: /*printf("\n%d ",(int)age);
5977: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5978: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5979: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5980: }*/
1.220 brouard 5981:
1.222 brouard 5982: fprintf(ficresprob,"\n%d ",(int)age);
5983: fprintf(ficresprobcov,"\n%d ",(int)age);
5984: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 5985:
1.222 brouard 5986: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
5987: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
5988: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5989: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
5990: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
5991: }
5992: i=0;
5993: for (k=1; k<=(nlstate);k++){
5994: for (l=1; l<=(nlstate+ndeath);l++){
5995: i++;
5996: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
5997: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
5998: for (j=1; j<=i;j++){
5999: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6000: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6001: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6002: }
6003: }
6004: }/* end of loop for state */
6005: } /* end of loop for age */
6006: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6007: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6008: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6009: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6010:
6011: /* Confidence intervalle of pij */
6012: /*
6013: fprintf(ficgp,"\nunset parametric;unset label");
6014: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6015: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6016: 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);
6017: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6018: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6019: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6020: */
6021:
6022: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6023: first1=1;first2=2;
6024: for (k2=1; k2<=(nlstate);k2++){
6025: for (l2=1; l2<=(nlstate+ndeath);l2++){
6026: if(l2==k2) continue;
6027: j=(k2-1)*(nlstate+ndeath)+l2;
6028: for (k1=1; k1<=(nlstate);k1++){
6029: for (l1=1; l1<=(nlstate+ndeath);l1++){
6030: if(l1==k1) continue;
6031: i=(k1-1)*(nlstate+ndeath)+l1;
6032: if(i<=j) continue;
6033: for (age=bage; age<=fage; age ++){
6034: if ((int)age %5==0){
6035: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6036: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6037: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6038: mu1=mu[i][(int) age]/stepm*YEARM ;
6039: mu2=mu[j][(int) age]/stepm*YEARM;
6040: c12=cv12/sqrt(v1*v2);
6041: /* Computing eigen value of matrix of covariance */
6042: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6043: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6044: if ((lc2 <0) || (lc1 <0) ){
6045: if(first2==1){
6046: first1=0;
6047: 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);
6048: }
6049: 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);
6050: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6051: /* lc2=fabs(lc2); */
6052: }
1.220 brouard 6053:
1.222 brouard 6054: /* Eigen vectors */
6055: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6056: /*v21=sqrt(1.-v11*v11); *//* error */
6057: v21=(lc1-v1)/cv12*v11;
6058: v12=-v21;
6059: v22=v11;
6060: tnalp=v21/v11;
6061: if(first1==1){
6062: first1=0;
6063: 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);
6064: }
6065: 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);
6066: /*printf(fignu*/
6067: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6068: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6069: if(first==1){
6070: first=0;
6071: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6072: fprintf(ficgp,"\nset parametric;unset label");
6073: 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);
6074: fprintf(ficgp,"\nset ter svg size 640, 480");
6075: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6076: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6077: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6078: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6079: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6080: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6081: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6082: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6083: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6084: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6085: 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", \
6086: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6087: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6088: }else{
6089: first=0;
6090: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6091: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6092: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6093: 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", \
6094: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6095: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6096: }/* if first */
6097: } /* age mod 5 */
6098: } /* end loop age */
6099: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6100: first=1;
6101: } /*l12 */
6102: } /* k12 */
6103: } /*l1 */
6104: }/* k1 */
6105: } /* loop on combination of covariates j1 */
6106: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6107: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6108: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6109: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6110: free_vector(xp,1,npar);
6111: fclose(ficresprob);
6112: fclose(ficresprobcov);
6113: fclose(ficresprobcor);
6114: fflush(ficgp);
6115: fflush(fichtmcov);
6116: }
1.126 brouard 6117:
6118:
6119: /******************* Printing html file ***********/
1.201 brouard 6120: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6121: int lastpass, int stepm, int weightopt, char model[],\
6122: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6123: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6124: double jprev1, double mprev1,double anprev1, double dateprev1, \
6125: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6126: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6127:
6128: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6129: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6130: </ul>");
1.237 brouard 6131: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6132: </ul>", model);
1.214 brouard 6133: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6134: 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",
6135: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6136: 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 6137: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6138: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6139: fprintf(fichtm,"\
6140: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6141: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6142: fprintf(fichtm,"\
1.217 brouard 6143: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6144: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6145: fprintf(fichtm,"\
1.126 brouard 6146: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6147: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6148: fprintf(fichtm,"\
1.217 brouard 6149: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6150: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6151: fprintf(fichtm,"\
1.211 brouard 6152: - (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 6153: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6154: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6155: if(prevfcast==1){
6156: fprintf(fichtm,"\
6157: - Prevalence projections by age and states: \
1.201 brouard 6158: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6159: }
1.126 brouard 6160:
1.222 brouard 6161: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6162:
1.225 brouard 6163: m=pow(2,cptcoveff);
1.222 brouard 6164: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6165:
1.222 brouard 6166: jj1=0;
1.237 brouard 6167:
6168: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6169: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6170: if(TKresult[nres]!= k1)
6171: continue;
1.220 brouard 6172:
1.222 brouard 6173: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6174: jj1++;
6175: if (cptcovn > 0) {
6176: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6177: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6178: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6179: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6180: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6181: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6182: }
1.237 brouard 6183: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6184: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6185: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6186: }
6187:
1.230 brouard 6188: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6189: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6190: if(invalidvarcomb[k1]){
6191: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6192: printf("\nCombination (%d) ignored because no cases \n",k1);
6193: continue;
6194: }
6195: }
6196: /* aij, bij */
1.241 brouard 6197: fprintf(fichtm,"<br>- Logit model (yours is: 1+age+%s), for example: logit(pij)=log(pij/pii)= aij+ bij age + V1 age + etc. as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
6198: <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 6199: /* Pij */
1.241 brouard 6200: 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> \
6201: <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 6202: /* Quasi-incidences */
6203: 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 6204: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6205: 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 6206: 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> \
6207: <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 6208: /* Survival functions (period) in state j */
6209: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6210: fprintf(fichtm,"<br>\n- Survival functions in state %d. Or probability to survive 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> \
6211: <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 6212: }
6213: /* State specific survival functions (period) */
6214: for(cpt=1; cpt<=nlstate;cpt++){
6215: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6216: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6217: <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 6218: }
6219: /* Period (stable) prevalence in each health state */
6220: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6221: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability to be 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> \
6222: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 6223: }
6224: if(backcast==1){
6225: /* Period (stable) back prevalence in each health state */
6226: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6227: fprintf(fichtm,"<br>\n- Convergence to period (stable) back prevalence in state %d. Or probability to be 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> \
6228: <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 6229: }
1.217 brouard 6230: }
1.222 brouard 6231: if(prevfcast==1){
6232: /* Projection of prevalence up to period (stable) prevalence in each health state */
6233: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6234: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f) up to period (stable) prevalence in state %d. Or probability to be 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> \
6235: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6236: }
6237: }
1.220 brouard 6238:
1.222 brouard 6239: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6240: fprintf(fichtm,"\n<br>- Life expectancy by health state (%d) at initial age and its decomposition into health expectancies in each alive state (1 to %d) (or area under each survival functions): <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a> <br> \
6241: <img src=\"%s_%d-%d-%d.svg\">",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.222 brouard 6242: }
6243: /* } /\* end i1 *\/ */
6244: }/* End k1 */
6245: fprintf(fichtm,"</ul>");
1.126 brouard 6246:
1.222 brouard 6247: fprintf(fichtm,"\
1.126 brouard 6248: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6249: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6250: - 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 6251: But because parameters are usually highly correlated (a higher incidence of disability \
6252: and a higher incidence of recovery can give very close observed transition) it might \
6253: be very useful to look not only at linear confidence intervals estimated from the \
6254: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6255: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6256: covariance matrix of the one-step probabilities. \
6257: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6258:
1.222 brouard 6259: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6260: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6261: fprintf(fichtm,"\
1.126 brouard 6262: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6263: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6264:
1.222 brouard 6265: fprintf(fichtm,"\
1.126 brouard 6266: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6267: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6268: fprintf(fichtm,"\
1.126 brouard 6269: - 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): \
6270: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6271: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6272: fprintf(fichtm,"\
1.126 brouard 6273: - (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): \
6274: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6275: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6276: fprintf(fichtm,"\
1.128 brouard 6277: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the 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 6278: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6279: fprintf(fichtm,"\
1.128 brouard 6280: - 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 6281: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6282: fprintf(fichtm,"\
1.126 brouard 6283: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6284: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6285:
6286: /* if(popforecast==1) fprintf(fichtm,"\n */
6287: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6288: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6289: /* <br>",fileres,fileres,fileres,fileres); */
6290: /* else */
6291: /* 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 6292: fflush(fichtm);
6293: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6294:
1.225 brouard 6295: m=pow(2,cptcoveff);
1.222 brouard 6296: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6297:
1.222 brouard 6298: jj1=0;
1.237 brouard 6299:
1.241 brouard 6300: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6301: for(k1=1; k1<=m;k1++){
1.237 brouard 6302: if(TKresult[nres]!= k1)
6303: continue;
1.222 brouard 6304: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6305: jj1++;
1.126 brouard 6306: if (cptcovn > 0) {
6307: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6308: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6309: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6310: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6311: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6312: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6313: }
6314:
1.126 brouard 6315: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6316:
1.222 brouard 6317: if(invalidvarcomb[k1]){
6318: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6319: continue;
6320: }
1.126 brouard 6321: }
6322: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6323: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6324: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6325: <img src=\"%s_%d-%d-%d.svg\">",cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 6326: }
6327: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6328: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6329: true period expectancies (those weighted with period prevalences are also\
6330: drawn in addition to the population based expectancies computed using\
1.241 brouard 6331: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6332: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6333: /* } /\* end i1 *\/ */
6334: }/* End k1 */
1.241 brouard 6335: }/* End nres */
1.222 brouard 6336: fprintf(fichtm,"</ul>");
6337: fflush(fichtm);
1.126 brouard 6338: }
6339:
6340: /******************* Gnuplot file **************/
1.223 brouard 6341: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6342:
6343: char dirfileres[132],optfileres[132];
1.223 brouard 6344: char gplotcondition[132];
1.237 brouard 6345: 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 6346: int lv=0, vlv=0, kl=0;
1.130 brouard 6347: int ng=0;
1.201 brouard 6348: int vpopbased;
1.223 brouard 6349: int ioffset; /* variable offset for columns */
1.235 brouard 6350: int nres=0; /* Index of resultline */
1.219 brouard 6351:
1.126 brouard 6352: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6353: /* printf("Problem with file %s",optionfilegnuplot); */
6354: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6355: /* } */
6356:
6357: /*#ifdef windows */
6358: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6359: /*#endif */
1.225 brouard 6360: m=pow(2,cptcoveff);
1.126 brouard 6361:
1.202 brouard 6362: /* Contribution to likelihood */
6363: /* Plot the probability implied in the likelihood */
1.223 brouard 6364: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6365: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6366: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6367: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6368: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6369: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6370: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6371: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6372: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6373: 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));
6374: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6375: 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));
6376: for (i=1; i<= nlstate ; i ++) {
6377: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6378: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6379: 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);
6380: for (j=2; j<= nlstate+ndeath ; j ++) {
6381: 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);
6382: }
6383: fprintf(ficgp,";\nset out; unset ylabel;\n");
6384: }
6385: /* 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 */
6386: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6387: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6388: fprintf(ficgp,"\nset out;unset log\n");
6389: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6390:
1.126 brouard 6391: strcpy(dirfileres,optionfilefiname);
6392: strcpy(optfileres,"vpl");
1.223 brouard 6393: /* 1eme*/
1.238 brouard 6394: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6395: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6396: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6397: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6398: if(TKresult[nres]!= k1)
6399: continue;
6400: /* We are interested in selected combination by the resultline */
6401: printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6402: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6403: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6404: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6405: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6406: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6407: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6408: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6409: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
6410: printf(" V%d=%d ",Tvaraff[k],vlv);
6411: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6412: }
6413: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6414: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6415: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6416: }
6417: printf("\n#\n");
6418: fprintf(ficgp,"\n#\n");
6419: if(invalidvarcomb[k1]){
6420: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6421: continue;
6422: }
1.235 brouard 6423:
1.241 brouard 6424: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6425: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6426: 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);
1.235 brouard 6427:
1.238 brouard 6428: for (i=1; i<= nlstate ; i ++) {
6429: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6430: else fprintf(ficgp," %%*lf (%%*lf)");
6431: }
1.242 brouard 6432: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6433: for (i=1; i<= nlstate ; i ++) {
6434: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6435: else fprintf(ficgp," %%*lf (%%*lf)");
6436: }
1.242 brouard 6437: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6438: for (i=1; i<= nlstate ; i ++) {
6439: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6440: else fprintf(ficgp," %%*lf (%%*lf)");
6441: }
6442: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6443: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6444: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6445: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6446: if(cptcoveff ==0){
1.245 ! brouard 6447: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6448: }else{
6449: kl=0;
6450: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6451: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6452: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6453: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6454: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6455: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6456: kl++;
1.238 brouard 6457: /* 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 *\/ */
6458: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6459: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6460: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6461: if(k==cptcoveff){
1.245 ! brouard 6462: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6463: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6464: }else{
6465: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6466: kl++;
6467: }
6468: } /* end covariate */
6469: } /* end if no covariate */
6470: } /* end if backcast */
6471: fprintf(ficgp,"\nset out \n");
6472: } /* nres */
1.201 brouard 6473: } /* k1 */
6474: } /* cpt */
1.235 brouard 6475:
6476:
1.126 brouard 6477: /*2 eme*/
1.238 brouard 6478: for (k1=1; k1<= m ; k1 ++){
6479: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6480: if(TKresult[nres]!= k1)
6481: continue;
6482: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6483: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6484: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6485: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6486: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6487: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6488: vlv= nbcode[Tvaraff[k]][lv];
6489: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6490: }
1.237 brouard 6491: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6492: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6493: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6494: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6495: }
1.211 brouard 6496: fprintf(ficgp,"\n#\n");
1.223 brouard 6497: if(invalidvarcomb[k1]){
6498: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6499: continue;
6500: }
1.219 brouard 6501:
1.241 brouard 6502: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6503: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6504: if(vpopbased==0)
6505: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6506: else
6507: fprintf(ficgp,"\nreplot ");
6508: for (i=1; i<= nlstate+1 ; i ++) {
6509: k=2*i;
6510: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1, vpopbased);
6511: for (j=1; j<= nlstate+1 ; j ++) {
6512: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6513: else fprintf(ficgp," %%*lf (%%*lf)");
6514: }
6515: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6516: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6517: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6518: for (j=1; j<= nlstate+1 ; j ++) {
6519: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6520: else fprintf(ficgp," %%*lf (%%*lf)");
6521: }
6522: fprintf(ficgp,"\" t\"\" w l lt 0,");
6523: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6524: for (j=1; j<= nlstate+1 ; j ++) {
6525: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6526: else fprintf(ficgp," %%*lf (%%*lf)");
6527: }
6528: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6529: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6530: } /* state */
6531: } /* vpopbased */
1.244 brouard 6532: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6533: } /* end nres */
6534: } /* k1 end 2 eme*/
6535:
6536:
6537: /*3eme*/
6538: for (k1=1; k1<= m ; k1 ++){
6539: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6540: if(TKresult[nres]!= k1)
1.238 brouard 6541: continue;
6542:
6543: for (cpt=1; cpt<= nlstate ; cpt ++) {
6544: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6545: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6546: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6547: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6548: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6549: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6550: vlv= nbcode[Tvaraff[k]][lv];
6551: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6552: }
6553: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6554: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6555: }
6556: fprintf(ficgp,"\n#\n");
6557: if(invalidvarcomb[k1]){
6558: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6559: continue;
6560: }
6561:
6562: /* k=2+nlstate*(2*cpt-2); */
6563: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6564: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6565: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6566: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),k1-1,k1-1,k,cpt);
1.238 brouard 6567: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6568: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6569: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6570: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6571: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6572: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6573:
1.238 brouard 6574: */
6575: for (i=1; i< nlstate ; i ++) {
6576: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+i,cpt,i+1);
6577: /* 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 6578:
1.238 brouard 6579: }
6580: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6581: }
6582: } /* end nres */
6583: } /* end kl 3eme */
1.126 brouard 6584:
1.223 brouard 6585: /* 4eme */
1.201 brouard 6586: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6587: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6588: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6589: if(TKresult[nres]!= k1)
1.223 brouard 6590: continue;
1.238 brouard 6591: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6592: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6593: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6594: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6595: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6596: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6597: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6598: vlv= nbcode[Tvaraff[k]][lv];
6599: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6600: }
6601: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6602: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6603: }
6604: fprintf(ficgp,"\n#\n");
6605: if(invalidvarcomb[k1]){
6606: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6607: continue;
1.223 brouard 6608: }
1.238 brouard 6609:
1.241 brouard 6610: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6611: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6612: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6613: k=3;
6614: for (i=1; i<= nlstate ; i ++){
6615: if(i==1){
6616: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6617: }else{
6618: fprintf(ficgp,", '' ");
6619: }
6620: l=(nlstate+ndeath)*(i-1)+1;
6621: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6622: for (j=2; j<= nlstate+ndeath ; j ++)
6623: fprintf(ficgp,"+$%d",k+l+j-1);
6624: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6625: } /* nlstate */
6626: fprintf(ficgp,"\nset out\n");
6627: } /* end cpt state*/
6628: } /* end nres */
6629: } /* end covariate k1 */
6630:
1.220 brouard 6631: /* 5eme */
1.201 brouard 6632: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6633: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6634: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6635: if(TKresult[nres]!= k1)
1.227 brouard 6636: continue;
1.238 brouard 6637: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6638: 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);
6639: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6640: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6641: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6642: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6643: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6644: vlv= nbcode[Tvaraff[k]][lv];
6645: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6646: }
6647: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6648: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6649: }
6650: fprintf(ficgp,"\n#\n");
6651: if(invalidvarcomb[k1]){
6652: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6653: continue;
6654: }
1.227 brouard 6655:
1.241 brouard 6656: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6657: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6658: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6659: k=3;
6660: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6661: if(j==1)
6662: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6663: else
6664: fprintf(ficgp,", '' ");
6665: l=(nlstate+ndeath)*(cpt-1) +j;
6666: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6667: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6668: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6669: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6670: } /* nlstate */
6671: fprintf(ficgp,", '' ");
6672: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6673: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6674: l=(nlstate+ndeath)*(cpt-1) +j;
6675: if(j < nlstate)
6676: fprintf(ficgp,"$%d +",k+l);
6677: else
6678: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6679: }
6680: fprintf(ficgp,"\nset out\n");
6681: } /* end cpt state*/
6682: } /* end covariate */
6683: } /* end nres */
1.227 brouard 6684:
1.220 brouard 6685: /* 6eme */
1.202 brouard 6686: /* CV preval stable (period) for each covariate */
1.237 brouard 6687: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6688: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6689: if(TKresult[nres]!= k1)
6690: continue;
1.153 brouard 6691: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6692:
1.211 brouard 6693: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6694: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6695: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6696: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6697: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6698: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6699: vlv= nbcode[Tvaraff[k]][lv];
6700: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6701: }
1.237 brouard 6702: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6703: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6704: }
1.211 brouard 6705: fprintf(ficgp,"\n#\n");
1.223 brouard 6706: if(invalidvarcomb[k1]){
1.227 brouard 6707: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6708: continue;
1.223 brouard 6709: }
1.227 brouard 6710:
1.241 brouard 6711: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6712: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6713: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6714: k=3; /* Offset */
1.153 brouard 6715: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6716: if(i==1)
6717: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6718: else
6719: fprintf(ficgp,", '' ");
6720: l=(nlstate+ndeath)*(i-1)+1;
6721: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6722: for (j=2; j<= nlstate ; j ++)
6723: fprintf(ficgp,"+$%d",k+l+j-1);
6724: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6725: } /* nlstate */
1.201 brouard 6726: fprintf(ficgp,"\nset out\n");
1.153 brouard 6727: } /* end cpt state*/
6728: } /* end covariate */
1.227 brouard 6729:
6730:
1.220 brouard 6731: /* 7eme */
1.218 brouard 6732: if(backcast == 1){
1.217 brouard 6733: /* CV back preval stable (period) for each covariate */
1.237 brouard 6734: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6735: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6736: if(TKresult[nres]!= k1)
6737: continue;
1.218 brouard 6738: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6739: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6740: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6741: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6742: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6743: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6744: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6745: vlv= nbcode[Tvaraff[k]][lv];
6746: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6747: }
1.237 brouard 6748: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6749: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6750: }
1.227 brouard 6751: fprintf(ficgp,"\n#\n");
6752: if(invalidvarcomb[k1]){
6753: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6754: continue;
6755: }
6756:
1.241 brouard 6757: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6758: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6759: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6760: k=3; /* Offset */
6761: for (i=1; i<= nlstate ; i ++){
6762: if(i==1)
6763: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6764: else
6765: fprintf(ficgp,", '' ");
6766: /* l=(nlstate+ndeath)*(i-1)+1; */
6767: l=(nlstate+ndeath)*(cpt-1)+1;
6768: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6769: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6770: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6771: /* for (j=2; j<= nlstate ; j ++) */
6772: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6773: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6774: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6775: } /* nlstate */
6776: fprintf(ficgp,"\nset out\n");
1.218 brouard 6777: } /* end cpt state*/
6778: } /* end covariate */
6779: } /* End if backcast */
6780:
1.223 brouard 6781: /* 8eme */
1.218 brouard 6782: if(prevfcast==1){
6783: /* Projection from cross-sectional to stable (period) for each covariate */
6784:
1.237 brouard 6785: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6786: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6787: if(TKresult[nres]!= k1)
6788: continue;
1.211 brouard 6789: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6790: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6791: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6792: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6793: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6794: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6795: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6796: vlv= nbcode[Tvaraff[k]][lv];
6797: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6798: }
1.237 brouard 6799: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6800: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6801: }
1.227 brouard 6802: fprintf(ficgp,"\n#\n");
6803: if(invalidvarcomb[k1]){
6804: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6805: continue;
6806: }
6807:
6808: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6809: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6810: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6811: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6812: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6813: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6814: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6815: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6816: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6817: if(i==1){
6818: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6819: }else{
6820: fprintf(ficgp,",\\\n '' ");
6821: }
6822: if(cptcoveff ==0){ /* No covariate */
6823: ioffset=2; /* Age is in 2 */
6824: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6825: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6826: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6827: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6828: fprintf(ficgp," u %d:(", ioffset);
6829: if(i==nlstate+1)
6830: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6831: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6832: else
6833: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6834: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6835: }else{ /* more than 2 covariates */
6836: if(cptcoveff ==1){
6837: ioffset=4; /* Age is in 4 */
6838: }else{
6839: ioffset=6; /* Age is in 6 */
6840: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6841: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6842: }
6843: fprintf(ficgp," u %d:(",ioffset);
6844: kl=0;
6845: strcpy(gplotcondition,"(");
6846: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6847: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6848: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6849: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6850: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6851: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6852: kl++;
6853: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6854: kl++;
6855: if(k <cptcoveff && cptcoveff>1)
6856: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6857: }
6858: strcpy(gplotcondition+strlen(gplotcondition),")");
6859: /* 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 *\/ */
6860: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6861: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6862: /* '' 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*/
6863: if(i==nlstate+1){
6864: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6865: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6866: }else{
6867: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6868: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6869: }
6870: } /* end if covariate */
6871: } /* nlstate */
6872: fprintf(ficgp,"\nset out\n");
1.223 brouard 6873: } /* end cpt state*/
6874: } /* end covariate */
6875: } /* End if prevfcast */
1.227 brouard 6876:
6877:
1.238 brouard 6878: /* 9eme writing MLE parameters */
6879: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6880: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6881: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6882: for(k=1; k <=(nlstate+ndeath); k++){
6883: if (k != i) {
1.227 brouard 6884: fprintf(ficgp,"# current state %d\n",k);
6885: for(j=1; j <=ncovmodel; j++){
6886: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6887: jk++;
6888: }
6889: fprintf(ficgp,"\n");
1.126 brouard 6890: }
6891: }
1.223 brouard 6892: }
1.187 brouard 6893: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6894:
1.145 brouard 6895: /*goto avoid;*/
1.238 brouard 6896: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6897: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6898: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6899: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6900: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6901: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6902: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6903: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6904: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6905: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6906: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6907: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6908: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6909: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6910: fprintf(ficgp,"#\n");
1.223 brouard 6911: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6912: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6913: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6914: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6915: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6916: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6917: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6918: if(TKresult[nres]!= jk)
6919: continue;
6920: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6921: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6922: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6923: }
6924: fprintf(ficgp,"\n#\n");
1.241 brouard 6925: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 6926: fprintf(ficgp,"\nset ter svg size 640, 480 ");
6927: if (ng==1){
6928: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
6929: fprintf(ficgp,"\nunset log y");
6930: }else if (ng==2){
6931: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
6932: fprintf(ficgp,"\nset log y");
6933: }else if (ng==3){
6934: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
6935: fprintf(ficgp,"\nset log y");
6936: }else
6937: fprintf(ficgp,"\nunset title ");
6938: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
6939: i=1;
6940: for(k2=1; k2<=nlstate; k2++) {
6941: k3=i;
6942: for(k=1; k<=(nlstate+ndeath); k++) {
6943: if (k != k2){
6944: switch( ng) {
6945: case 1:
6946: if(nagesqr==0)
6947: fprintf(ficgp," p%d+p%d*x",i,i+1);
6948: else /* nagesqr =1 */
6949: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6950: break;
6951: case 2: /* ng=2 */
6952: if(nagesqr==0)
6953: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
6954: else /* nagesqr =1 */
6955: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6956: break;
6957: case 3:
6958: if(nagesqr==0)
6959: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
6960: else /* nagesqr =1 */
6961: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
6962: break;
6963: }
6964: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 6965: ijp=1; /* product no age */
6966: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
6967: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 6968: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 6969: if(j==Tage[ij]) { /* Product by age */
6970: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 6971: if(DummyV[j]==0){
1.237 brouard 6972: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
6973: }else{ /* quantitative */
6974: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
6975: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6976: }
6977: ij++;
6978: }
6979: }else if(j==Tprod[ijp]) { /* */
6980: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
6981: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 6982: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
6983: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 6984: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],nbcode[Tvard[ijp][2]][codtabm(jk,j)]); */
6985: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
6986: }else{ /* Vn is dummy and Vm is quanti */
6987: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
6988: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6989: }
6990: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 6991: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 6992: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
6993: }else{ /* Both quanti */
6994: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6995: }
6996: }
1.238 brouard 6997: ijp++;
1.237 brouard 6998: }
6999: } else{ /* simple covariate */
7000: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7001: if(Dummy[j]==0){
7002: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7003: }else{ /* quantitative */
7004: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7005: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7006: }
1.237 brouard 7007: } /* end simple */
7008: } /* end j */
1.223 brouard 7009: }else{
7010: i=i-ncovmodel;
7011: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7012: fprintf(ficgp," (1.");
7013: }
1.227 brouard 7014:
1.223 brouard 7015: if(ng != 1){
7016: fprintf(ficgp,")/(1");
1.227 brouard 7017:
1.223 brouard 7018: for(k1=1; k1 <=nlstate; k1++){
7019: if(nagesqr==0)
7020: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7021: else /* nagesqr =1 */
7022: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1,k3+(k1-1)*ncovmodel+1+nagesqr);
1.217 brouard 7023:
1.223 brouard 7024: ij=1;
7025: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7026: if((j-2)==Tage[ij]) { /* Bug valgrind */
7027: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7028: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7029: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7030: ij++;
7031: }
7032: }
7033: else
1.225 brouard 7034: fprintf(ficgp,"+p%d*%d",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 7035: }
7036: fprintf(ficgp,")");
7037: }
7038: fprintf(ficgp,")");
7039: if(ng ==2)
7040: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7041: else /* ng= 3 */
7042: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7043: }else{ /* end ng <> 1 */
7044: if( k !=k2) /* logit p11 is hard to draw */
7045: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7046: }
7047: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7048: fprintf(ficgp,",");
7049: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7050: fprintf(ficgp,",");
7051: i=i+ncovmodel;
7052: } /* end k */
7053: } /* end k2 */
7054: fprintf(ficgp,"\n set out\n");
7055: } /* end jk */
7056: } /* end ng */
7057: /* avoid: */
7058: fflush(ficgp);
1.126 brouard 7059: } /* end gnuplot */
7060:
7061:
7062: /*************** Moving average **************/
1.219 brouard 7063: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7064: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7065:
1.222 brouard 7066: int i, cpt, cptcod;
7067: int modcovmax =1;
7068: int mobilavrange, mob;
7069: int iage=0;
7070:
7071: double sum=0.;
7072: double age;
7073: double *sumnewp, *sumnewm;
7074: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7075:
7076:
1.225 brouard 7077: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7078: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7079:
7080: sumnewp = vector(1,ncovcombmax);
7081: sumnewm = vector(1,ncovcombmax);
7082: agemingood = vector(1,ncovcombmax);
7083: agemaxgood = vector(1,ncovcombmax);
7084:
7085: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7086: sumnewm[cptcod]=0.;
7087: sumnewp[cptcod]=0.;
7088: agemingood[cptcod]=0;
7089: agemaxgood[cptcod]=0;
7090: }
7091: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7092:
7093: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7094: if(mobilav==1) mobilavrange=5; /* default */
7095: else mobilavrange=mobilav;
7096: for (age=bage; age<=fage; age++)
7097: for (i=1; i<=nlstate;i++)
7098: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7099: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7100: /* We keep the original values on the extreme ages bage, fage and for
7101: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7102: we use a 5 terms etc. until the borders are no more concerned.
7103: */
7104: for (mob=3;mob <=mobilavrange;mob=mob+2){
7105: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7106: for (i=1; i<=nlstate;i++){
7107: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7108: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7109: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7110: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7111: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7112: }
7113: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7114: }
7115: }
7116: }/* end age */
7117: }/* end mob */
7118: }else
7119: return -1;
7120: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7121: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7122: if(invalidvarcomb[cptcod]){
7123: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7124: continue;
7125: }
1.219 brouard 7126:
1.222 brouard 7127: agemingood[cptcod]=fage-(mob-1)/2;
7128: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7129: sumnewm[cptcod]=0.;
7130: for (i=1; i<=nlstate;i++){
7131: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7132: }
7133: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7134: agemingood[cptcod]=age;
7135: }else{ /* bad */
7136: for (i=1; i<=nlstate;i++){
7137: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7138: } /* i */
7139: } /* end bad */
7140: }/* age */
7141: sum=0.;
7142: for (i=1; i<=nlstate;i++){
7143: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7144: }
7145: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7146: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any descending age!\n",cptcod);
7147: /* for (i=1; i<=nlstate;i++){ */
7148: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7149: /* } /\* i *\/ */
7150: } /* end bad */
7151: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7152: /* From youngest, finding the oldest wrong */
7153: agemaxgood[cptcod]=bage+(mob-1)/2;
7154: for (age=bage+(mob-1)/2; age<=fage; age++){
7155: sumnewm[cptcod]=0.;
7156: for (i=1; i<=nlstate;i++){
7157: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7158: }
7159: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7160: agemaxgood[cptcod]=age;
7161: }else{ /* bad */
7162: for (i=1; i<=nlstate;i++){
7163: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7164: } /* i */
7165: } /* end bad */
7166: }/* age */
7167: sum=0.;
7168: for (i=1; i<=nlstate;i++){
7169: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7170: }
7171: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7172: printf("For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one at any ascending age!\n",cptcod);
7173: /* for (i=1; i<=nlstate;i++){ */
7174: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7175: /* } /\* i *\/ */
7176: } /* end bad */
7177:
7178: for (age=bage; age<=fage; age++){
1.235 brouard 7179: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7180: sumnewp[cptcod]=0.;
7181: sumnewm[cptcod]=0.;
7182: for (i=1; i<=nlstate;i++){
7183: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7184: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7185: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7186: }
7187: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7188: }
7189: /* printf("\n"); */
7190: /* } */
7191: /* brutal averaging */
7192: for (i=1; i<=nlstate;i++){
7193: for (age=1; age<=bage; age++){
7194: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7195: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7196: }
7197: for (age=fage; age<=AGESUP; age++){
7198: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7199: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7200: }
7201: } /* end i status */
7202: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7203: for (age=1; age<=AGESUP; age++){
7204: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7205: mobaverage[(int)age][i][cptcod]=0.;
7206: }
7207: }
7208: }/* end cptcod */
7209: free_vector(sumnewm,1, ncovcombmax);
7210: free_vector(sumnewp,1, ncovcombmax);
7211: free_vector(agemaxgood,1, ncovcombmax);
7212: free_vector(agemingood,1, ncovcombmax);
7213: return 0;
7214: }/* End movingaverage */
1.218 brouard 7215:
1.126 brouard 7216:
7217: /************** Forecasting ******************/
1.235 brouard 7218: void prevforecast(char fileres[], double anproj1, double mproj1, double jproj1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anproj2, double p[], int cptcoveff){
1.126 brouard 7219: /* proj1, year, month, day of starting projection
7220: agemin, agemax range of age
7221: dateprev1 dateprev2 range of dates during which prevalence is computed
7222: anproj2 year of en of projection (same day and month as proj1).
7223: */
1.235 brouard 7224: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7225: double agec; /* generic age */
7226: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7227: double *popeffectif,*popcount;
7228: double ***p3mat;
1.218 brouard 7229: /* double ***mobaverage; */
1.126 brouard 7230: char fileresf[FILENAMELENGTH];
7231:
7232: agelim=AGESUP;
1.211 brouard 7233: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7234: in each health status at the date of interview (if between dateprev1 and dateprev2).
7235: We still use firstpass and lastpass as another selection.
7236: */
1.214 brouard 7237: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7238: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7239:
1.201 brouard 7240: strcpy(fileresf,"F_");
7241: strcat(fileresf,fileresu);
1.126 brouard 7242: if((ficresf=fopen(fileresf,"w"))==NULL) {
7243: printf("Problem with forecast resultfile: %s\n", fileresf);
7244: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7245: }
1.235 brouard 7246: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7247: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7248:
1.225 brouard 7249: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7250:
7251:
7252: stepsize=(int) (stepm+YEARM-1)/YEARM;
7253: if (stepm<=12) stepsize=1;
7254: if(estepm < stepm){
7255: printf ("Problem %d lower than %d\n",estepm, stepm);
7256: }
7257: else hstepm=estepm;
7258:
7259: hstepm=hstepm/stepm;
7260: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7261: fractional in yp1 */
7262: anprojmean=yp;
7263: yp2=modf((yp1*12),&yp);
7264: mprojmean=yp;
7265: yp1=modf((yp2*30.5),&yp);
7266: jprojmean=yp;
7267: if(jprojmean==0) jprojmean=1;
7268: if(mprojmean==0) jprojmean=1;
7269:
1.227 brouard 7270: i1=pow(2,cptcoveff);
1.126 brouard 7271: if (cptcovn < 1){i1=1;}
7272:
7273: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7274:
7275: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7276:
1.126 brouard 7277: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7278: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7279: for(k=1; k<=i1;k++){
7280: if(TKresult[nres]!= k)
7281: continue;
1.227 brouard 7282: if(invalidvarcomb[k]){
7283: printf("\nCombination (%d) projection ignored because no cases \n",k);
7284: continue;
7285: }
7286: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7287: for(j=1;j<=cptcoveff;j++) {
7288: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7289: }
1.235 brouard 7290: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7291: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7292: }
1.227 brouard 7293: fprintf(ficresf," yearproj age");
7294: for(j=1; j<=nlstate+ndeath;j++){
7295: for(i=1; i<=nlstate;i++)
7296: fprintf(ficresf," p%d%d",i,j);
7297: fprintf(ficresf," wp.%d",j);
7298: }
7299: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7300: fprintf(ficresf,"\n");
7301: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7302: for (agec=fage; agec>=(ageminpar-1); agec--){
7303: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7304: nhstepm = nhstepm/hstepm;
7305: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7306: oldm=oldms;savm=savms;
1.235 brouard 7307: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7308:
7309: for (h=0; h<=nhstepm; h++){
7310: if (h*hstepm/YEARM*stepm ==yearp) {
7311: fprintf(ficresf,"\n");
7312: for(j=1;j<=cptcoveff;j++)
7313: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7314: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7315: }
7316: for(j=1; j<=nlstate+ndeath;j++) {
7317: ppij=0.;
7318: for(i=1; i<=nlstate;i++) {
7319: if (mobilav==1)
7320: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7321: else {
7322: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7323: }
7324: if (h*hstepm/YEARM*stepm== yearp) {
7325: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7326: }
7327: } /* end i */
7328: if (h*hstepm/YEARM*stepm==yearp) {
7329: fprintf(ficresf," %.3f", ppij);
7330: }
7331: }/* end j */
7332: } /* end h */
7333: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7334: } /* end agec */
7335: } /* end yearp */
7336: } /* end k */
1.219 brouard 7337:
1.126 brouard 7338: fclose(ficresf);
1.215 brouard 7339: printf("End of Computing forecasting \n");
7340: fprintf(ficlog,"End of Computing forecasting\n");
7341:
1.126 brouard 7342: }
7343:
1.218 brouard 7344: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7345: /* void prevbackforecast(char fileres[], 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){ */
1.218 brouard 7346: /* /\* back1, year, month, day of starting backection */
7347: /* agemin, agemax range of age */
7348: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7349: /* anback2 year of en of backection (same day and month as back1). */
7350: /* *\/ */
7351: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7352: /* double agec; /\* generic age *\/ */
7353: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7354: /* double *popeffectif,*popcount; */
7355: /* double ***p3mat; */
7356: /* /\* double ***mobaverage; *\/ */
7357: /* char fileresfb[FILENAMELENGTH]; */
7358:
7359: /* agelim=AGESUP; */
7360: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7361: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7362: /* We still use firstpass and lastpass as another selection. */
7363: /* *\/ */
7364: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7365: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7366: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7367:
7368: /* strcpy(fileresfb,"FB_"); */
7369: /* strcat(fileresfb,fileresu); */
7370: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7371: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7372: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7373: /* } */
7374: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7375: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7376:
1.225 brouard 7377: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7378:
7379: /* /\* if (mobilav!=0) { *\/ */
7380: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7381: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7382: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7383: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7384: /* /\* } *\/ */
7385: /* /\* } *\/ */
7386:
7387: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7388: /* if (stepm<=12) stepsize=1; */
7389: /* if(estepm < stepm){ */
7390: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7391: /* } */
7392: /* else hstepm=estepm; */
7393:
7394: /* hstepm=hstepm/stepm; */
7395: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7396: /* fractional in yp1 *\/ */
7397: /* anprojmean=yp; */
7398: /* yp2=modf((yp1*12),&yp); */
7399: /* mprojmean=yp; */
7400: /* yp1=modf((yp2*30.5),&yp); */
7401: /* jprojmean=yp; */
7402: /* if(jprojmean==0) jprojmean=1; */
7403: /* if(mprojmean==0) jprojmean=1; */
7404:
1.225 brouard 7405: /* i1=cptcoveff; */
1.218 brouard 7406: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7407:
1.218 brouard 7408: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7409:
1.218 brouard 7410: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7411:
7412: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7413: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7414: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7415: /* k=k+1; */
7416: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7417: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7418: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7419: /* } */
7420: /* fprintf(ficresfb," yearbproj age"); */
7421: /* for(j=1; j<=nlstate+ndeath;j++){ */
7422: /* for(i=1; i<=nlstate;i++) */
7423: /* fprintf(ficresfb," p%d%d",i,j); */
7424: /* fprintf(ficresfb," p.%d",j); */
7425: /* } */
7426: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7427: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7428: /* fprintf(ficresfb,"\n"); */
7429: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7430: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7431: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7432: /* nhstepm = nhstepm/hstepm; */
7433: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7434: /* oldm=oldms;savm=savms; */
7435: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7436: /* for (h=0; h<=nhstepm; h++){ */
7437: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7438: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7439: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7440: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7441: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7442: /* } */
7443: /* for(j=1; j<=nlstate+ndeath;j++) { */
7444: /* ppij=0.; */
7445: /* for(i=1; i<=nlstate;i++) { */
7446: /* if (mobilav==1) */
7447: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7448: /* else { */
7449: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7450: /* } */
7451: /* if (h*hstepm/YEARM*stepm== yearp) { */
7452: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7453: /* } */
7454: /* } /\* end i *\/ */
7455: /* if (h*hstepm/YEARM*stepm==yearp) { */
7456: /* fprintf(ficresfb," %.3f", ppij); */
7457: /* } */
7458: /* }/\* end j *\/ */
7459: /* } /\* end h *\/ */
7460: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7461: /* } /\* end agec *\/ */
7462: /* } /\* end yearp *\/ */
7463: /* } /\* end cptcod *\/ */
7464: /* } /\* end cptcov *\/ */
7465:
7466: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7467:
7468: /* fclose(ficresfb); */
7469: /* printf("End of Computing Back forecasting \n"); */
7470: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7471:
1.218 brouard 7472: /* } */
1.217 brouard 7473:
1.126 brouard 7474: /************** Forecasting *****not tested NB*************/
1.227 brouard 7475: /* 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 7476:
1.227 brouard 7477: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7478: /* int *popage; */
7479: /* double calagedatem, agelim, kk1, kk2; */
7480: /* double *popeffectif,*popcount; */
7481: /* double ***p3mat,***tabpop,***tabpopprev; */
7482: /* /\* double ***mobaverage; *\/ */
7483: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7484:
1.227 brouard 7485: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7486: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7487: /* agelim=AGESUP; */
7488: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7489:
1.227 brouard 7490: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7491:
7492:
1.227 brouard 7493: /* strcpy(filerespop,"POP_"); */
7494: /* strcat(filerespop,fileresu); */
7495: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7496: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7497: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7498: /* } */
7499: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7500: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7501:
1.227 brouard 7502: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7503:
1.227 brouard 7504: /* /\* if (mobilav!=0) { *\/ */
7505: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7506: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7507: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7508: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7509: /* /\* } *\/ */
7510: /* /\* } *\/ */
1.126 brouard 7511:
1.227 brouard 7512: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7513: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7514:
1.227 brouard 7515: /* agelim=AGESUP; */
1.126 brouard 7516:
1.227 brouard 7517: /* hstepm=1; */
7518: /* hstepm=hstepm/stepm; */
1.218 brouard 7519:
1.227 brouard 7520: /* if (popforecast==1) { */
7521: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7522: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7523: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7524: /* } */
7525: /* popage=ivector(0,AGESUP); */
7526: /* popeffectif=vector(0,AGESUP); */
7527: /* popcount=vector(0,AGESUP); */
1.126 brouard 7528:
1.227 brouard 7529: /* i=1; */
7530: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7531:
1.227 brouard 7532: /* imx=i; */
7533: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7534: /* } */
1.218 brouard 7535:
1.227 brouard 7536: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7537: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7538: /* k=k+1; */
7539: /* fprintf(ficrespop,"\n#******"); */
7540: /* for(j=1;j<=cptcoveff;j++) { */
7541: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7542: /* } */
7543: /* fprintf(ficrespop,"******\n"); */
7544: /* fprintf(ficrespop,"# Age"); */
7545: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7546: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7547:
1.227 brouard 7548: /* for (cpt=0; cpt<=0;cpt++) { */
7549: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7550:
1.227 brouard 7551: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7552: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7553: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7554:
1.227 brouard 7555: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7556: /* oldm=oldms;savm=savms; */
7557: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7558:
1.227 brouard 7559: /* for (h=0; h<=nhstepm; h++){ */
7560: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7561: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7562: /* } */
7563: /* for(j=1; j<=nlstate+ndeath;j++) { */
7564: /* kk1=0.;kk2=0; */
7565: /* for(i=1; i<=nlstate;i++) { */
7566: /* if (mobilav==1) */
7567: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7568: /* else { */
7569: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7570: /* } */
7571: /* } */
7572: /* if (h==(int)(calagedatem+12*cpt)){ */
7573: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7574: /* /\*fprintf(ficrespop," %.3f", kk1); */
7575: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7576: /* } */
7577: /* } */
7578: /* for(i=1; i<=nlstate;i++){ */
7579: /* kk1=0.; */
7580: /* for(j=1; j<=nlstate;j++){ */
7581: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7582: /* } */
7583: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7584: /* } */
1.218 brouard 7585:
1.227 brouard 7586: /* if (h==(int)(calagedatem+12*cpt)) */
7587: /* for(j=1; j<=nlstate;j++) */
7588: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7589: /* } */
7590: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7591: /* } */
7592: /* } */
1.218 brouard 7593:
1.227 brouard 7594: /* /\******\/ */
1.218 brouard 7595:
1.227 brouard 7596: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7597: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7598: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7599: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7600: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7601:
1.227 brouard 7602: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7603: /* oldm=oldms;savm=savms; */
7604: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7605: /* for (h=0; h<=nhstepm; h++){ */
7606: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7607: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7608: /* } */
7609: /* for(j=1; j<=nlstate+ndeath;j++) { */
7610: /* kk1=0.;kk2=0; */
7611: /* for(i=1; i<=nlstate;i++) { */
7612: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7613: /* } */
7614: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7615: /* } */
7616: /* } */
7617: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7618: /* } */
7619: /* } */
7620: /* } */
7621: /* } */
1.218 brouard 7622:
1.227 brouard 7623: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7624:
1.227 brouard 7625: /* if (popforecast==1) { */
7626: /* free_ivector(popage,0,AGESUP); */
7627: /* free_vector(popeffectif,0,AGESUP); */
7628: /* free_vector(popcount,0,AGESUP); */
7629: /* } */
7630: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7631: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7632: /* fclose(ficrespop); */
7633: /* } /\* End of popforecast *\/ */
1.218 brouard 7634:
1.126 brouard 7635: int fileappend(FILE *fichier, char *optionfich)
7636: {
7637: if((fichier=fopen(optionfich,"a"))==NULL) {
7638: printf("Problem with file: %s\n", optionfich);
7639: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7640: return (0);
7641: }
7642: fflush(fichier);
7643: return (1);
7644: }
7645:
7646:
7647: /**************** function prwizard **********************/
7648: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7649: {
7650:
7651: /* Wizard to print covariance matrix template */
7652:
1.164 brouard 7653: char ca[32], cb[32];
7654: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7655: int numlinepar;
7656:
7657: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7658: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7659: for(i=1; i <=nlstate; i++){
7660: jj=0;
7661: for(j=1; j <=nlstate+ndeath; j++){
7662: if(j==i) continue;
7663: jj++;
7664: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7665: printf("%1d%1d",i,j);
7666: fprintf(ficparo,"%1d%1d",i,j);
7667: for(k=1; k<=ncovmodel;k++){
7668: /* printf(" %lf",param[i][j][k]); */
7669: /* fprintf(ficparo," %lf",param[i][j][k]); */
7670: printf(" 0.");
7671: fprintf(ficparo," 0.");
7672: }
7673: printf("\n");
7674: fprintf(ficparo,"\n");
7675: }
7676: }
7677: printf("# Scales (for hessian or gradient estimation)\n");
7678: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7679: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7680: for(i=1; i <=nlstate; i++){
7681: jj=0;
7682: for(j=1; j <=nlstate+ndeath; j++){
7683: if(j==i) continue;
7684: jj++;
7685: fprintf(ficparo,"%1d%1d",i,j);
7686: printf("%1d%1d",i,j);
7687: fflush(stdout);
7688: for(k=1; k<=ncovmodel;k++){
7689: /* printf(" %le",delti3[i][j][k]); */
7690: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7691: printf(" 0.");
7692: fprintf(ficparo," 0.");
7693: }
7694: numlinepar++;
7695: printf("\n");
7696: fprintf(ficparo,"\n");
7697: }
7698: }
7699: printf("# Covariance matrix\n");
7700: /* # 121 Var(a12)\n\ */
7701: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7702: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7703: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7704: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7705: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7706: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7707: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7708: fflush(stdout);
7709: fprintf(ficparo,"# Covariance matrix\n");
7710: /* # 121 Var(a12)\n\ */
7711: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7712: /* # ...\n\ */
7713: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7714:
7715: for(itimes=1;itimes<=2;itimes++){
7716: jj=0;
7717: for(i=1; i <=nlstate; i++){
7718: for(j=1; j <=nlstate+ndeath; j++){
7719: if(j==i) continue;
7720: for(k=1; k<=ncovmodel;k++){
7721: jj++;
7722: ca[0]= k+'a'-1;ca[1]='\0';
7723: if(itimes==1){
7724: printf("#%1d%1d%d",i,j,k);
7725: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7726: }else{
7727: printf("%1d%1d%d",i,j,k);
7728: fprintf(ficparo,"%1d%1d%d",i,j,k);
7729: /* printf(" %.5le",matcov[i][j]); */
7730: }
7731: ll=0;
7732: for(li=1;li <=nlstate; li++){
7733: for(lj=1;lj <=nlstate+ndeath; lj++){
7734: if(lj==li) continue;
7735: for(lk=1;lk<=ncovmodel;lk++){
7736: ll++;
7737: if(ll<=jj){
7738: cb[0]= lk +'a'-1;cb[1]='\0';
7739: if(ll<jj){
7740: if(itimes==1){
7741: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7742: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7743: }else{
7744: printf(" 0.");
7745: fprintf(ficparo," 0.");
7746: }
7747: }else{
7748: if(itimes==1){
7749: printf(" Var(%s%1d%1d)",ca,i,j);
7750: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7751: }else{
7752: printf(" 0.");
7753: fprintf(ficparo," 0.");
7754: }
7755: }
7756: }
7757: } /* end lk */
7758: } /* end lj */
7759: } /* end li */
7760: printf("\n");
7761: fprintf(ficparo,"\n");
7762: numlinepar++;
7763: } /* end k*/
7764: } /*end j */
7765: } /* end i */
7766: } /* end itimes */
7767:
7768: } /* end of prwizard */
7769: /******************* Gompertz Likelihood ******************************/
7770: double gompertz(double x[])
7771: {
7772: double A,B,L=0.0,sump=0.,num=0.;
7773: int i,n=0; /* n is the size of the sample */
7774:
1.220 brouard 7775: for (i=1;i<=imx ; i++) {
1.126 brouard 7776: sump=sump+weight[i];
7777: /* sump=sump+1;*/
7778: num=num+1;
7779: }
7780:
7781:
7782: /* for (i=0; i<=imx; i++)
7783: 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]);*/
7784:
7785: for (i=1;i<=imx ; i++)
7786: {
7787: if (cens[i] == 1 && wav[i]>1)
7788: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7789:
7790: if (cens[i] == 0 && wav[i]>1)
7791: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7792: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7793:
7794: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7795: if (wav[i] > 1 ) { /* ??? */
7796: L=L+A*weight[i];
7797: /* 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]);*/
7798: }
7799: }
7800:
7801: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7802:
7803: return -2*L*num/sump;
7804: }
7805:
1.136 brouard 7806: #ifdef GSL
7807: /******************* Gompertz_f Likelihood ******************************/
7808: double gompertz_f(const gsl_vector *v, void *params)
7809: {
7810: double A,B,LL=0.0,sump=0.,num=0.;
7811: double *x= (double *) v->data;
7812: int i,n=0; /* n is the size of the sample */
7813:
7814: for (i=0;i<=imx-1 ; i++) {
7815: sump=sump+weight[i];
7816: /* sump=sump+1;*/
7817: num=num+1;
7818: }
7819:
7820:
7821: /* for (i=0; i<=imx; i++)
7822: 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]);*/
7823: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7824: for (i=1;i<=imx ; i++)
7825: {
7826: if (cens[i] == 1 && wav[i]>1)
7827: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7828:
7829: if (cens[i] == 0 && wav[i]>1)
7830: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7831: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7832:
7833: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7834: if (wav[i] > 1 ) { /* ??? */
7835: LL=LL+A*weight[i];
7836: /* 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]);*/
7837: }
7838: }
7839:
7840: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7841: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7842:
7843: return -2*LL*num/sump;
7844: }
7845: #endif
7846:
1.126 brouard 7847: /******************* Printing html file ***********/
1.201 brouard 7848: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7849: int lastpass, int stepm, int weightopt, char model[],\
7850: int imx, double p[],double **matcov,double agemortsup){
7851: int i,k;
7852:
7853: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7854: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7855: for (i=1;i<=2;i++)
7856: 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 7857: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7858: fprintf(fichtm,"</ul>");
7859:
7860: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7861:
7862: 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>");
7863:
7864: for (k=agegomp;k<(agemortsup-2);k++)
7865: 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]);
7866:
7867:
7868: fflush(fichtm);
7869: }
7870:
7871: /******************* Gnuplot file **************/
1.201 brouard 7872: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7873:
7874: char dirfileres[132],optfileres[132];
1.164 brouard 7875:
1.126 brouard 7876: int ng;
7877:
7878:
7879: /*#ifdef windows */
7880: fprintf(ficgp,"cd \"%s\" \n",pathc);
7881: /*#endif */
7882:
7883:
7884: strcpy(dirfileres,optionfilefiname);
7885: strcpy(optfileres,"vpl");
1.199 brouard 7886: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7887: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7888: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7889: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7890: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7891:
7892: }
7893:
1.136 brouard 7894: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7895: {
1.126 brouard 7896:
1.136 brouard 7897: /*-------- data file ----------*/
7898: FILE *fic;
7899: char dummy[]=" ";
1.240 brouard 7900: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7901: int lstra;
1.136 brouard 7902: int linei, month, year,iout;
7903: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7904: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7905: char *stratrunc;
1.223 brouard 7906:
1.240 brouard 7907: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7908: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7909:
1.240 brouard 7910: for(v=1; v <=ncovcol;v++){
7911: DummyV[v]=0;
7912: FixedV[v]=0;
7913: }
7914: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7915: DummyV[v]=1;
7916: FixedV[v]=0;
7917: }
7918: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7919: DummyV[v]=0;
7920: FixedV[v]=1;
7921: }
7922: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7923: DummyV[v]=1;
7924: FixedV[v]=1;
7925: }
7926: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
7927: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
7928: 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]);
7929: }
1.126 brouard 7930:
1.136 brouard 7931: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 7932: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
7933: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 7934: }
1.126 brouard 7935:
1.136 brouard 7936: i=1;
7937: linei=0;
7938: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
7939: linei=linei+1;
7940: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
7941: if(line[j] == '\t')
7942: line[j] = ' ';
7943: }
7944: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
7945: ;
7946: };
7947: line[j+1]=0; /* Trims blanks at end of line */
7948: if(line[0]=='#'){
7949: fprintf(ficlog,"Comment line\n%s\n",line);
7950: printf("Comment line\n%s\n",line);
7951: continue;
7952: }
7953: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 7954: strcpy(line, linetmp);
1.223 brouard 7955:
7956: /* Loops on waves */
7957: for (j=maxwav;j>=1;j--){
7958: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 7959: cutv(stra, strb, line, ' ');
7960: if(strb[0]=='.') { /* Missing value */
7961: lval=-1;
7962: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
7963: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
7964: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
7965: 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);
7966: 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);
7967: return 1;
7968: }
7969: }else{
7970: errno=0;
7971: /* what_kind_of_number(strb); */
7972: dval=strtod(strb,&endptr);
7973: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
7974: /* if(strb != endptr && *endptr == '\0') */
7975: /* dval=dlval; */
7976: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
7977: if( strb[0]=='\0' || (*endptr != '\0')){
7978: 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);
7979: 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);
7980: return 1;
7981: }
7982: cotqvar[j][iv][i]=dval;
7983: cotvar[j][ntv+iv][i]=dval;
7984: }
7985: strcpy(line,stra);
1.223 brouard 7986: }/* end loop ntqv */
1.225 brouard 7987:
1.223 brouard 7988: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 7989: cutv(stra, strb, line, ' ');
7990: if(strb[0]=='.') { /* Missing value */
7991: lval=-1;
7992: }else{
7993: errno=0;
7994: lval=strtol(strb,&endptr,10);
7995: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
7996: if( strb[0]=='\0' || (*endptr != '\0')){
7997: 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);
7998: 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);
7999: return 1;
8000: }
8001: }
8002: if(lval <-1 || lval >1){
8003: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8004: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8005: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8006: For example, for multinomial values like 1, 2 and 3,\n \
8007: build V1=0 V2=0 for the reference value (1),\n \
8008: V1=1 V2=0 for (2) \n \
1.223 brouard 8009: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8010: output of IMaCh is often meaningless.\n \
1.223 brouard 8011: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8012: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8013: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8014: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8015: For example, for multinomial values like 1, 2 and 3,\n \
8016: build V1=0 V2=0 for the reference value (1),\n \
8017: V1=1 V2=0 for (2) \n \
1.223 brouard 8018: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8019: output of IMaCh is often meaningless.\n \
1.223 brouard 8020: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8021: return 1;
8022: }
8023: cotvar[j][iv][i]=(double)(lval);
8024: strcpy(line,stra);
1.223 brouard 8025: }/* end loop ntv */
1.225 brouard 8026:
1.223 brouard 8027: /* Statuses at wave */
1.137 brouard 8028: cutv(stra, strb, line, ' ');
1.223 brouard 8029: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8030: lval=-1;
1.136 brouard 8031: }else{
1.238 brouard 8032: errno=0;
8033: lval=strtol(strb,&endptr,10);
8034: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8035: if( strb[0]=='\0' || (*endptr != '\0')){
8036: 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);
8037: 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);
8038: return 1;
8039: }
1.136 brouard 8040: }
1.225 brouard 8041:
1.136 brouard 8042: s[j][i]=lval;
1.225 brouard 8043:
1.223 brouard 8044: /* Date of Interview */
1.136 brouard 8045: strcpy(line,stra);
8046: cutv(stra, strb,line,' ');
1.169 brouard 8047: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8048: }
1.169 brouard 8049: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8050: month=99;
8051: year=9999;
1.136 brouard 8052: }else{
1.225 brouard 8053: 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);
8054: 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);
8055: return 1;
1.136 brouard 8056: }
8057: anint[j][i]= (double) year;
8058: mint[j][i]= (double)month;
8059: strcpy(line,stra);
1.223 brouard 8060: } /* End loop on waves */
1.225 brouard 8061:
1.223 brouard 8062: /* Date of death */
1.136 brouard 8063: cutv(stra, strb,line,' ');
1.169 brouard 8064: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8065: }
1.169 brouard 8066: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8067: month=99;
8068: year=9999;
8069: }else{
1.141 brouard 8070: 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 8071: 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);
8072: return 1;
1.136 brouard 8073: }
8074: andc[i]=(double) year;
8075: moisdc[i]=(double) month;
8076: strcpy(line,stra);
8077:
1.223 brouard 8078: /* Date of birth */
1.136 brouard 8079: cutv(stra, strb,line,' ');
1.169 brouard 8080: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8081: }
1.169 brouard 8082: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8083: month=99;
8084: year=9999;
8085: }else{
1.141 brouard 8086: 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);
8087: 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 8088: return 1;
1.136 brouard 8089: }
8090: if (year==9999) {
1.141 brouard 8091: 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);
8092: 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 8093: return 1;
8094:
1.136 brouard 8095: }
8096: annais[i]=(double)(year);
8097: moisnais[i]=(double)(month);
8098: strcpy(line,stra);
1.225 brouard 8099:
1.223 brouard 8100: /* Sample weight */
1.136 brouard 8101: cutv(stra, strb,line,' ');
8102: errno=0;
8103: dval=strtod(strb,&endptr);
8104: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8105: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8106: 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 8107: fflush(ficlog);
8108: return 1;
8109: }
8110: weight[i]=dval;
8111: strcpy(line,stra);
1.225 brouard 8112:
1.223 brouard 8113: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8114: cutv(stra, strb, line, ' ');
8115: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8116: lval=-1;
1.223 brouard 8117: }else{
1.225 brouard 8118: errno=0;
8119: /* what_kind_of_number(strb); */
8120: dval=strtod(strb,&endptr);
8121: /* if(strb != endptr && *endptr == '\0') */
8122: /* dval=dlval; */
8123: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8124: if( strb[0]=='\0' || (*endptr != '\0')){
8125: 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);
8126: 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);
8127: return 1;
8128: }
8129: coqvar[iv][i]=dval;
1.226 brouard 8130: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8131: }
8132: strcpy(line,stra);
8133: }/* end loop nqv */
1.136 brouard 8134:
1.223 brouard 8135: /* Covariate values */
1.136 brouard 8136: for (j=ncovcol;j>=1;j--){
8137: cutv(stra, strb,line,' ');
1.223 brouard 8138: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8139: lval=-1;
1.136 brouard 8140: }else{
1.225 brouard 8141: errno=0;
8142: lval=strtol(strb,&endptr,10);
8143: if( strb[0]=='\0' || (*endptr != '\0')){
8144: 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);
8145: 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);
8146: return 1;
8147: }
1.136 brouard 8148: }
8149: if(lval <-1 || lval >1){
1.225 brouard 8150: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8151: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8152: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8153: For example, for multinomial values like 1, 2 and 3,\n \
8154: build V1=0 V2=0 for the reference value (1),\n \
8155: V1=1 V2=0 for (2) \n \
1.136 brouard 8156: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8157: output of IMaCh is often meaningless.\n \
1.136 brouard 8158: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8159: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8160: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8161: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8162: For example, for multinomial values like 1, 2 and 3,\n \
8163: build V1=0 V2=0 for the reference value (1),\n \
8164: V1=1 V2=0 for (2) \n \
1.136 brouard 8165: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8166: output of IMaCh is often meaningless.\n \
1.136 brouard 8167: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8168: return 1;
1.136 brouard 8169: }
8170: covar[j][i]=(double)(lval);
8171: strcpy(line,stra);
8172: }
8173: lstra=strlen(stra);
1.225 brouard 8174:
1.136 brouard 8175: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8176: stratrunc = &(stra[lstra-9]);
8177: num[i]=atol(stratrunc);
8178: }
8179: else
8180: num[i]=atol(stra);
8181: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8182: 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;}*/
8183:
8184: i=i+1;
8185: } /* End loop reading data */
1.225 brouard 8186:
1.136 brouard 8187: *imax=i-1; /* Number of individuals */
8188: fclose(fic);
1.225 brouard 8189:
1.136 brouard 8190: return (0);
1.164 brouard 8191: /* endread: */
1.225 brouard 8192: printf("Exiting readdata: ");
8193: fclose(fic);
8194: return (1);
1.223 brouard 8195: }
1.126 brouard 8196:
1.234 brouard 8197: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8198: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8199: while (*p2 == ' ')
1.234 brouard 8200: p2++;
8201: /* while ((*p1++ = *p2++) !=0) */
8202: /* ; */
8203: /* do */
8204: /* while (*p2 == ' ') */
8205: /* p2++; */
8206: /* while (*p1++ == *p2++); */
8207: *stri=p2;
1.145 brouard 8208: }
8209:
1.235 brouard 8210: int decoderesult ( char resultline[], int nres)
1.230 brouard 8211: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8212: {
1.235 brouard 8213: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8214: char resultsav[MAXLINE];
1.234 brouard 8215: int resultmodel[MAXLINE];
8216: int modelresult[MAXLINE];
1.230 brouard 8217: char stra[80], strb[80], strc[80], strd[80],stre[80];
8218:
1.234 brouard 8219: removefirstspace(&resultline);
1.233 brouard 8220: printf("decoderesult:%s\n",resultline);
1.230 brouard 8221:
8222: if (strstr(resultline,"v") !=0){
8223: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8224: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8225: return 1;
8226: }
8227: trimbb(resultsav, resultline);
8228: if (strlen(resultsav) >1){
8229: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8230: }
1.234 brouard 8231: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8232: printf("ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
8233: fprintf(ficlog,"ERROR: the number of variable in the resultline, %d, differs from the number of variable used in the model line, %d.\n",j, cptcovs);
8234: }
8235: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8236: if(nbocc(resultsav,'=') >1){
8237: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8238: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8239: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8240: }else
8241: cutl(strc,strd,resultsav,'=');
1.230 brouard 8242: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8243:
1.230 brouard 8244: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8245: Tvarsel[k]=atoi(strc);
8246: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8247: /* cptcovsel++; */
8248: if (nbocc(stra,'=') >0)
8249: strcpy(resultsav,stra); /* and analyzes it */
8250: }
1.235 brouard 8251: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8252: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8253: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8254: match=0;
1.236 brouard 8255: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8256: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8257: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8258: match=1;
8259: break;
8260: }
8261: }
8262: if(match == 0){
8263: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8264: }
8265: }
8266: }
1.235 brouard 8267: /* Checking for missing or useless values in comparison of current model needs */
8268: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8269: match=0;
1.235 brouard 8270: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8271: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8272: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8273: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8274: ++match;
8275: }
8276: }
8277: }
8278: if(match == 0){
8279: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8280: }else if(match > 1){
8281: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8282: }
8283: }
1.235 brouard 8284:
1.234 brouard 8285: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8286: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8287: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8288: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8289: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8290: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8291: /* 1 0 0 0 */
8292: /* 2 1 0 0 */
8293: /* 3 0 1 0 */
8294: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8295: /* 5 0 0 1 */
8296: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8297: /* 7 0 1 1 */
8298: /* 8 1 1 1 */
1.237 brouard 8299: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8300: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8301: /* V5*age V5 known which value for nres? */
8302: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8303: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8304: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8305: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8306: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8307: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8308: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8309: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8310: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8311: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8312: k4++;;
8313: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8314: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8315: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8316: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8317: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8318: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8319: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8320: k4q++;;
8321: }
8322: }
1.234 brouard 8323:
1.235 brouard 8324: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8325: return (0);
8326: }
1.235 brouard 8327:
1.230 brouard 8328: int decodemodel( char model[], int lastobs)
8329: /**< This routine decodes the model and returns:
1.224 brouard 8330: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8331: * - nagesqr = 1 if age*age in the model, otherwise 0.
8332: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8333: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8334: * - cptcovage number of covariates with age*products =2
8335: * - cptcovs number of simple covariates
8336: * - 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
8337: * which is a new column after the 9 (ncovcol) variables.
8338: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8339: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8340: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8341: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8342: */
1.136 brouard 8343: {
1.238 brouard 8344: int i, j, k, ks, v;
1.227 brouard 8345: int j1, k1, k2, k3, k4;
1.136 brouard 8346: char modelsav[80];
1.145 brouard 8347: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8348: char *strpt;
1.136 brouard 8349:
1.145 brouard 8350: /*removespace(model);*/
1.136 brouard 8351: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8352: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8353: if (strstr(model,"AGE") !=0){
1.192 brouard 8354: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8355: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8356: return 1;
8357: }
1.141 brouard 8358: if (strstr(model,"v") !=0){
8359: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8360: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8361: return 1;
8362: }
1.187 brouard 8363: strcpy(modelsav,model);
8364: if ((strpt=strstr(model,"age*age")) !=0){
8365: printf(" strpt=%s, model=%s\n",strpt, model);
8366: if(strpt != model){
1.234 brouard 8367: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8368: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8369: corresponding column of parameters.\n",model);
1.234 brouard 8370: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8371: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8372: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8373: return 1;
1.225 brouard 8374: }
1.187 brouard 8375: nagesqr=1;
8376: if (strstr(model,"+age*age") !=0)
1.234 brouard 8377: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8378: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8379: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8380: else
1.234 brouard 8381: substrchaine(modelsav, model, "age*age");
1.187 brouard 8382: }else
8383: nagesqr=0;
8384: if (strlen(modelsav) >1){
8385: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8386: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8387: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8388: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8389: * cst, age and age*age
8390: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8391: /* including age products which are counted in cptcovage.
8392: * but the covariates which are products must be treated
8393: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8394: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8395: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8396:
8397:
1.187 brouard 8398: /* Design
8399: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8400: * < ncovcol=8 >
8401: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8402: * k= 1 2 3 4 5 6 7 8
8403: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8404: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8405: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8406: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8407: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8408: * Tage[++cptcovage]=k
8409: * if products, new covar are created after ncovcol with k1
8410: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8411: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8412: * 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
8413: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8414: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8415: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8416: * < ncovcol=8 >
8417: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8418: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8419: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8420: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8421: * p Tprod[1]@2={ 6, 5}
8422: *p Tvard[1][1]@4= {7, 8, 5, 6}
8423: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8424: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8425: *How to reorganize?
8426: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8427: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8428: * {2, 1, 4, 8, 5, 6, 3, 7}
8429: * Struct []
8430: */
1.225 brouard 8431:
1.187 brouard 8432: /* This loop fills the array Tvar from the string 'model'.*/
8433: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8434: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8435: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8436: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8437: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8438: /* k=1 Tvar[1]=2 (from V2) */
8439: /* k=5 Tvar[5] */
8440: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8441: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8442: /* } */
1.198 brouard 8443: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8444: /*
8445: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8446: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8447: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8448: }
1.187 brouard 8449: cptcovage=0;
8450: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8451: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8452: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8453: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8454: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8455: /*scanf("%d",i);*/
8456: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8457: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8458: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8459: /* covar is not filled and then is empty */
8460: cptcovprod--;
8461: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8462: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8463: Typevar[k]=1; /* 1 for age product */
8464: cptcovage++; /* Sums the number of covariates which include age as a product */
8465: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8466: /*printf("stre=%s ", stre);*/
8467: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8468: cptcovprod--;
8469: cutl(stre,strb,strc,'V');
8470: Tvar[k]=atoi(stre);
8471: Typevar[k]=1; /* 1 for age product */
8472: cptcovage++;
8473: Tage[cptcovage]=k;
8474: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8475: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8476: cptcovn++;
8477: cptcovprodnoage++;k1++;
8478: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8479: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8480: because this model-covariate is a construction we invent a new column
8481: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8482: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8483: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8484: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8485: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8486: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8487: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8488: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8489: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8490: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8491: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8492: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8493: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8494: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8495: for (i=1; i<=lastobs;i++){
8496: /* Computes the new covariate which is a product of
8497: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8498: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8499: }
8500: } /* End age is not in the model */
8501: } /* End if model includes a product */
8502: else { /* no more sum */
8503: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8504: /* scanf("%d",i);*/
8505: cutl(strd,strc,strb,'V');
8506: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8507: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8508: Tvar[k]=atoi(strd);
8509: Typevar[k]=0; /* 0 for simple covariates */
8510: }
8511: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8512: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8513: scanf("%d",i);*/
1.187 brouard 8514: } /* end of loop + on total covariates */
8515: } /* end if strlen(modelsave == 0) age*age might exist */
8516: } /* end if strlen(model == 0) */
1.136 brouard 8517:
8518: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8519: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8520:
1.136 brouard 8521: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8522: printf("cptcovprod=%d ", cptcovprod);
8523: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8524: scanf("%d ",i);*/
8525:
8526:
1.230 brouard 8527: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8528: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8529: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8530: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8531: k = 1 2 3 4 5 6 7 8 9
8532: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8533: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8534: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8535: Dummy[k] 1 0 0 0 3 1 1 2 3
8536: Tmodelind[combination of covar]=k;
1.225 brouard 8537: */
8538: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8539: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8540: /* 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 8541: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8542: printf("Model=%s\n\
8543: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8544: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8545: 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);
8546: fprintf(ficlog,"Model=%s\n\
8547: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8548: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8549: 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);
1.240 brouard 8550: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8551: 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 */
8552: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8553: Fixed[k]= 0;
8554: Dummy[k]= 0;
1.225 brouard 8555: ncoveff++;
1.232 brouard 8556: ncovf++;
1.234 brouard 8557: nsd++;
8558: modell[k].maintype= FTYPE;
8559: TvarsD[nsd]=Tvar[k];
8560: TvarsDind[nsd]=k;
8561: TvarF[ncovf]=Tvar[k];
8562: TvarFind[ncovf]=k;
8563: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8564: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8565: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8566: Fixed[k]= 0;
8567: Dummy[k]= 0;
8568: ncoveff++;
8569: ncovf++;
8570: modell[k].maintype= FTYPE;
8571: TvarF[ncovf]=Tvar[k];
8572: TvarFind[ncovf]=k;
1.230 brouard 8573: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8574: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8575: }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 8576: Fixed[k]= 0;
8577: Dummy[k]= 1;
1.230 brouard 8578: nqfveff++;
1.234 brouard 8579: modell[k].maintype= FTYPE;
8580: modell[k].subtype= FQ;
8581: nsq++;
8582: TvarsQ[nsq]=Tvar[k];
8583: TvarsQind[nsq]=k;
1.232 brouard 8584: ncovf++;
1.234 brouard 8585: TvarF[ncovf]=Tvar[k];
8586: TvarFind[ncovf]=k;
1.231 brouard 8587: 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 8588: 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 8589: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8590: Fixed[k]= 1;
8591: Dummy[k]= 0;
1.225 brouard 8592: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8593: modell[k].maintype= VTYPE;
8594: modell[k].subtype= VD;
8595: nsd++;
8596: TvarsD[nsd]=Tvar[k];
8597: TvarsDind[nsd]=k;
8598: ncovv++; /* Only simple time varying variables */
8599: TvarV[ncovv]=Tvar[k];
1.242 brouard 8600: 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 8601: 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 */
8602: 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 8603: 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);
8604: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8605: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8606: Fixed[k]= 1;
8607: Dummy[k]= 1;
8608: nqtveff++;
8609: modell[k].maintype= VTYPE;
8610: modell[k].subtype= VQ;
8611: ncovv++; /* Only simple time varying variables */
8612: nsq++;
8613: TvarsQ[nsq]=Tvar[k];
8614: TvarsQind[nsq]=k;
8615: TvarV[ncovv]=Tvar[k];
1.242 brouard 8616: 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 8617: 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 */
8618: 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 8619: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8620: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8621: 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 8622: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8623: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8624: ncova++;
8625: TvarA[ncova]=Tvar[k];
8626: TvarAind[ncova]=k;
1.231 brouard 8627: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8628: Fixed[k]= 2;
8629: Dummy[k]= 2;
8630: modell[k].maintype= ATYPE;
8631: modell[k].subtype= APFD;
8632: /* ncoveff++; */
1.227 brouard 8633: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8634: Fixed[k]= 2;
8635: Dummy[k]= 3;
8636: modell[k].maintype= ATYPE;
8637: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8638: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8639: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8640: Fixed[k]= 3;
8641: Dummy[k]= 2;
8642: modell[k].maintype= ATYPE;
8643: modell[k].subtype= APVD; /* Product age * varying dummy */
8644: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8645: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8646: Fixed[k]= 3;
8647: Dummy[k]= 3;
8648: modell[k].maintype= ATYPE;
8649: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8650: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8651: }
8652: }else if (Typevar[k] == 2) { /* product without age */
8653: k1=Tposprod[k];
8654: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8655: if(Tvard[k1][2] <=ncovcol){
8656: Fixed[k]= 1;
8657: Dummy[k]= 0;
8658: modell[k].maintype= FTYPE;
8659: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8660: ncovf++; /* Fixed variables without age */
8661: TvarF[ncovf]=Tvar[k];
8662: TvarFind[ncovf]=k;
8663: }else if(Tvard[k1][2] <=ncovcol+nqv){
8664: Fixed[k]= 0; /* or 2 ?*/
8665: Dummy[k]= 1;
8666: modell[k].maintype= FTYPE;
8667: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8668: ncovf++; /* Varying variables without age */
8669: TvarF[ncovf]=Tvar[k];
8670: TvarFind[ncovf]=k;
8671: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8672: Fixed[k]= 1;
8673: Dummy[k]= 0;
8674: modell[k].maintype= VTYPE;
8675: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8676: ncovv++; /* Varying variables without age */
8677: TvarV[ncovv]=Tvar[k];
8678: TvarVind[ncovv]=k;
8679: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8680: Fixed[k]= 1;
8681: Dummy[k]= 1;
8682: modell[k].maintype= VTYPE;
8683: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8684: ncovv++; /* Varying variables without age */
8685: TvarV[ncovv]=Tvar[k];
8686: TvarVind[ncovv]=k;
8687: }
1.227 brouard 8688: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8689: if(Tvard[k1][2] <=ncovcol){
8690: Fixed[k]= 0; /* or 2 ?*/
8691: Dummy[k]= 1;
8692: modell[k].maintype= FTYPE;
8693: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8694: ncovf++; /* Fixed variables without age */
8695: TvarF[ncovf]=Tvar[k];
8696: TvarFind[ncovf]=k;
8697: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8698: Fixed[k]= 1;
8699: Dummy[k]= 1;
8700: modell[k].maintype= VTYPE;
8701: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8702: ncovv++; /* Varying variables without age */
8703: TvarV[ncovv]=Tvar[k];
8704: TvarVind[ncovv]=k;
8705: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8706: Fixed[k]= 1;
8707: Dummy[k]= 1;
8708: modell[k].maintype= VTYPE;
8709: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8710: ncovv++; /* Varying variables without age */
8711: TvarV[ncovv]=Tvar[k];
8712: TvarVind[ncovv]=k;
8713: ncovv++; /* Varying variables without age */
8714: TvarV[ncovv]=Tvar[k];
8715: TvarVind[ncovv]=k;
8716: }
1.227 brouard 8717: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8718: if(Tvard[k1][2] <=ncovcol){
8719: Fixed[k]= 1;
8720: Dummy[k]= 1;
8721: modell[k].maintype= VTYPE;
8722: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8723: ncovv++; /* Varying variables without age */
8724: TvarV[ncovv]=Tvar[k];
8725: TvarVind[ncovv]=k;
8726: }else if(Tvard[k1][2] <=ncovcol+nqv){
8727: Fixed[k]= 1;
8728: Dummy[k]= 1;
8729: modell[k].maintype= VTYPE;
8730: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8731: ncovv++; /* Varying variables without age */
8732: TvarV[ncovv]=Tvar[k];
8733: TvarVind[ncovv]=k;
8734: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8735: Fixed[k]= 1;
8736: Dummy[k]= 0;
8737: modell[k].maintype= VTYPE;
8738: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8739: ncovv++; /* Varying variables without age */
8740: TvarV[ncovv]=Tvar[k];
8741: TvarVind[ncovv]=k;
8742: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8743: Fixed[k]= 1;
8744: Dummy[k]= 1;
8745: modell[k].maintype= VTYPE;
8746: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8747: ncovv++; /* Varying variables without age */
8748: TvarV[ncovv]=Tvar[k];
8749: TvarVind[ncovv]=k;
8750: }
1.227 brouard 8751: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8752: if(Tvard[k1][2] <=ncovcol){
8753: Fixed[k]= 1;
8754: Dummy[k]= 1;
8755: modell[k].maintype= VTYPE;
8756: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8757: ncovv++; /* Varying variables without age */
8758: TvarV[ncovv]=Tvar[k];
8759: TvarVind[ncovv]=k;
8760: }else if(Tvard[k1][2] <=ncovcol+nqv){
8761: Fixed[k]= 1;
8762: Dummy[k]= 1;
8763: modell[k].maintype= VTYPE;
8764: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8765: ncovv++; /* Varying variables without age */
8766: TvarV[ncovv]=Tvar[k];
8767: TvarVind[ncovv]=k;
8768: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8769: Fixed[k]= 1;
8770: Dummy[k]= 1;
8771: modell[k].maintype= VTYPE;
8772: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8773: ncovv++; /* Varying variables without age */
8774: TvarV[ncovv]=Tvar[k];
8775: TvarVind[ncovv]=k;
8776: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8777: Fixed[k]= 1;
8778: Dummy[k]= 1;
8779: modell[k].maintype= VTYPE;
8780: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8781: ncovv++; /* Varying variables without age */
8782: TvarV[ncovv]=Tvar[k];
8783: TvarVind[ncovv]=k;
8784: }
1.227 brouard 8785: }else{
1.240 brouard 8786: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8787: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8788: } /*end k1*/
1.225 brouard 8789: }else{
1.226 brouard 8790: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8791: 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 8792: }
1.227 brouard 8793: 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 8794: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8795: 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]);
8796: }
8797: /* Searching for doublons in the model */
8798: for(k1=1; k1<= cptcovt;k1++){
8799: for(k2=1; k2 <k1;k2++){
8800: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8801: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8802: if(Tvar[k1]==Tvar[k2]){
8803: 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[Tvar[k1]],Dummy[Tvar[k1]]);
8804: 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[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
8805: return(1);
8806: }
8807: }else if (Typevar[k1] ==2){
8808: k3=Tposprod[k1];
8809: k4=Tposprod[k2];
8810: 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])) ){
8811: 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]]);
8812: 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);
8813: return(1);
8814: }
8815: }
1.227 brouard 8816: }
8817: }
1.225 brouard 8818: }
8819: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8820: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8821: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8822: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8823: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8824: /*endread:*/
1.225 brouard 8825: printf("Exiting decodemodel: ");
8826: return (1);
1.136 brouard 8827: }
8828:
1.169 brouard 8829: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.136 brouard 8830: {
8831: int i, m;
1.218 brouard 8832: int firstone=0;
8833:
1.136 brouard 8834: for (i=1; i<=imx; i++) {
8835: for(m=2; (m<= maxwav); m++) {
8836: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8837: anint[m][i]=9999;
1.216 brouard 8838: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8839: s[m][i]=-1;
1.136 brouard 8840: }
8841: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8842: *nberr = *nberr + 1;
1.218 brouard 8843: if(firstone == 0){
8844: firstone=1;
8845: printf("Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\nOther similar cases in log file\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
8846: }
8847: fprintf(ficlog,"Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
1.136 brouard 8848: s[m][i]=-1;
8849: }
8850: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8851: (*nberr)++;
1.136 brouard 8852: printf("Error! Month of death of individual %ld on line %d was unknown %2d, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,(int)moisdc[i]);
8853: fprintf(ficlog,"Error! Month of death of individual %ld on line %d was unknown %f, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,moisdc[i]);
8854: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8855: }
8856: }
8857: }
8858:
8859: for (i=1; i<=imx; i++) {
8860: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8861: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8862: 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 8863: if (s[m][i] >= nlstate+1) {
1.169 brouard 8864: if(agedc[i]>0){
8865: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8866: agev[m][i]=agedc[i];
1.214 brouard 8867: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8868: }else {
1.136 brouard 8869: if ((int)andc[i]!=9999){
8870: nbwarn++;
8871: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8872: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8873: agev[m][i]=-1;
8874: }
8875: }
1.169 brouard 8876: } /* agedc > 0 */
1.214 brouard 8877: } /* end if */
1.136 brouard 8878: else if(s[m][i] !=9){ /* Standard case, age in fractional
8879: years but with the precision of a month */
8880: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8881: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8882: agev[m][i]=1;
8883: else if(agev[m][i] < *agemin){
8884: *agemin=agev[m][i];
8885: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8886: }
8887: else if(agev[m][i] >*agemax){
8888: *agemax=agev[m][i];
1.156 brouard 8889: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8890: }
8891: /*agev[m][i]=anint[m][i]-annais[i];*/
8892: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8893: } /* en if 9*/
1.136 brouard 8894: else { /* =9 */
1.214 brouard 8895: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8896: agev[m][i]=1;
8897: s[m][i]=-1;
8898: }
8899: }
1.214 brouard 8900: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8901: agev[m][i]=1;
1.214 brouard 8902: else{
8903: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8904: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8905: agev[m][i]=0;
8906: }
8907: } /* End for lastpass */
8908: }
1.136 brouard 8909:
8910: for (i=1; i<=imx; i++) {
8911: for(m=firstpass; (m<=lastpass); m++){
8912: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8913: (*nberr)++;
1.136 brouard 8914: 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);
8915: 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);
8916: return 1;
8917: }
8918: }
8919: }
8920:
8921: /*for (i=1; i<=imx; i++){
8922: for (m=firstpass; (m<lastpass); m++){
8923: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8924: }
8925:
8926: }*/
8927:
8928:
1.139 brouard 8929: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
8930: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 8931:
8932: return (0);
1.164 brouard 8933: /* endread:*/
1.136 brouard 8934: printf("Exiting calandcheckages: ");
8935: return (1);
8936: }
8937:
1.172 brouard 8938: #if defined(_MSC_VER)
8939: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8940: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8941: //#include "stdafx.h"
8942: //#include <stdio.h>
8943: //#include <tchar.h>
8944: //#include <windows.h>
8945: //#include <iostream>
8946: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
8947:
8948: LPFN_ISWOW64PROCESS fnIsWow64Process;
8949:
8950: BOOL IsWow64()
8951: {
8952: BOOL bIsWow64 = FALSE;
8953:
8954: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
8955: // (HANDLE, PBOOL);
8956:
8957: //LPFN_ISWOW64PROCESS fnIsWow64Process;
8958:
8959: HMODULE module = GetModuleHandle(_T("kernel32"));
8960: const char funcName[] = "IsWow64Process";
8961: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
8962: GetProcAddress(module, funcName);
8963:
8964: if (NULL != fnIsWow64Process)
8965: {
8966: if (!fnIsWow64Process(GetCurrentProcess(),
8967: &bIsWow64))
8968: //throw std::exception("Unknown error");
8969: printf("Unknown error\n");
8970: }
8971: return bIsWow64 != FALSE;
8972: }
8973: #endif
1.177 brouard 8974:
1.191 brouard 8975: void syscompilerinfo(int logged)
1.167 brouard 8976: {
8977: /* #include "syscompilerinfo.h"*/
1.185 brouard 8978: /* command line Intel compiler 32bit windows, XP compatible:*/
8979: /* /GS /W3 /Gy
8980: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
8981: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
8982: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 8983: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
8984: */
8985: /* 64 bits */
1.185 brouard 8986: /*
8987: /GS /W3 /Gy
8988: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
8989: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
8990: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
8991: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
8992: /* Optimization are useless and O3 is slower than O2 */
8993: /*
8994: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
8995: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
8996: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
8997: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
8998: */
1.186 brouard 8999: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9000: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9001: /PDB:"visual studio
9002: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9003: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9004: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9005: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9006: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9007: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9008: uiAccess='false'"
9009: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9010: /NOLOGO /TLBID:1
9011: */
1.177 brouard 9012: #if defined __INTEL_COMPILER
1.178 brouard 9013: #if defined(__GNUC__)
9014: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9015: #endif
1.177 brouard 9016: #elif defined(__GNUC__)
1.179 brouard 9017: #ifndef __APPLE__
1.174 brouard 9018: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9019: #endif
1.177 brouard 9020: struct utsname sysInfo;
1.178 brouard 9021: int cross = CROSS;
9022: if (cross){
9023: printf("Cross-");
1.191 brouard 9024: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9025: }
1.174 brouard 9026: #endif
9027:
1.171 brouard 9028: #include <stdint.h>
1.178 brouard 9029:
1.191 brouard 9030: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9031: #if defined(__clang__)
1.191 brouard 9032: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9033: #endif
9034: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9035: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9036: #endif
9037: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9038: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9039: #endif
9040: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9041: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9042: #endif
9043: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9044: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9045: #endif
9046: #if defined(_MSC_VER)
1.191 brouard 9047: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9048: #endif
9049: #if defined(__PGI)
1.191 brouard 9050: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9051: #endif
9052: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9053: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9054: #endif
1.191 brouard 9055: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9056:
1.167 brouard 9057: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9058: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9059: // Windows (x64 and x86)
1.191 brouard 9060: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9061: #elif __unix__ // all unices, not all compilers
9062: // Unix
1.191 brouard 9063: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9064: #elif __linux__
9065: // linux
1.191 brouard 9066: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9067: #elif __APPLE__
1.174 brouard 9068: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9069: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9070: #endif
9071:
9072: /* __MINGW32__ */
9073: /* __CYGWIN__ */
9074: /* __MINGW64__ */
9075: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9076: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9077: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9078: /* _WIN64 // Defined for applications for Win64. */
9079: /* _M_X64 // Defined for compilations that target x64 processors. */
9080: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9081:
1.167 brouard 9082: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9083: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9084: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9085: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9086: #else
1.191 brouard 9087: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9088: #endif
9089:
1.169 brouard 9090: #if defined(__GNUC__)
9091: # if defined(__GNUC_PATCHLEVEL__)
9092: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9093: + __GNUC_MINOR__ * 100 \
9094: + __GNUC_PATCHLEVEL__)
9095: # else
9096: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9097: + __GNUC_MINOR__ * 100)
9098: # endif
1.174 brouard 9099: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9100: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9101:
9102: if (uname(&sysInfo) != -1) {
9103: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9104: 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 9105: }
9106: else
9107: perror("uname() error");
1.179 brouard 9108: //#ifndef __INTEL_COMPILER
9109: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9110: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9111: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9112: #endif
1.169 brouard 9113: #endif
1.172 brouard 9114:
9115: // void main()
9116: // {
1.169 brouard 9117: #if defined(_MSC_VER)
1.174 brouard 9118: if (IsWow64()){
1.191 brouard 9119: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9120: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9121: }
9122: else{
1.191 brouard 9123: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9124: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9125: }
1.172 brouard 9126: // printf("\nPress Enter to continue...");
9127: // getchar();
9128: // }
9129:
1.169 brouard 9130: #endif
9131:
1.167 brouard 9132:
1.219 brouard 9133: }
1.136 brouard 9134:
1.219 brouard 9135: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9136: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9137: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9138: /* double ftolpl = 1.e-10; */
1.180 brouard 9139: double age, agebase, agelim;
1.203 brouard 9140: double tot;
1.180 brouard 9141:
1.202 brouard 9142: strcpy(filerespl,"PL_");
9143: strcat(filerespl,fileresu);
9144: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9145: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9146: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9147: }
1.227 brouard 9148: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9149: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9150: pstamp(ficrespl);
1.203 brouard 9151: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9152: fprintf(ficrespl,"#Age ");
9153: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9154: fprintf(ficrespl,"\n");
1.180 brouard 9155:
1.219 brouard 9156: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9157:
1.219 brouard 9158: agebase=ageminpar;
9159: agelim=agemaxpar;
1.180 brouard 9160:
1.227 brouard 9161: /* i1=pow(2,ncoveff); */
1.234 brouard 9162: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9163: if (cptcovn < 1){i1=1;}
1.180 brouard 9164:
1.238 brouard 9165: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9166: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9167: if(TKresult[nres]!= k)
9168: continue;
1.235 brouard 9169:
1.238 brouard 9170: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9171: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9172: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9173: /* k=k+1; */
9174: /* to clean */
9175: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9176: fprintf(ficrespl,"#******");
9177: printf("#******");
9178: fprintf(ficlog,"#******");
9179: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9180: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9181: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9182: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9183: }
9184: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9185: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9186: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9187: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9188: }
9189: fprintf(ficrespl,"******\n");
9190: printf("******\n");
9191: fprintf(ficlog,"******\n");
9192: if(invalidvarcomb[k]){
9193: printf("\nCombination (%d) ignored because no case \n",k);
9194: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9195: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9196: continue;
9197: }
1.219 brouard 9198:
1.238 brouard 9199: fprintf(ficrespl,"#Age ");
9200: for(j=1;j<=cptcoveff;j++) {
9201: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9202: }
9203: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9204: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9205:
1.238 brouard 9206: for (age=agebase; age<=agelim; age++){
9207: /* for (age=agebase; age<=agebase; age++){ */
9208: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9209: fprintf(ficrespl,"%.0f ",age );
9210: for(j=1;j<=cptcoveff;j++)
9211: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9212: tot=0.;
9213: for(i=1; i<=nlstate;i++){
9214: tot += prlim[i][i];
9215: fprintf(ficrespl," %.5f", prlim[i][i]);
9216: }
9217: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9218: } /* Age */
9219: /* was end of cptcod */
9220: } /* cptcov */
9221: } /* nres */
1.219 brouard 9222: return 0;
1.180 brouard 9223: }
9224:
1.218 brouard 9225: 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){
9226: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9227:
9228: /* Computes the back prevalence limit for any combination of covariate values
9229: * at any age between ageminpar and agemaxpar
9230: */
1.235 brouard 9231: int i, j, k, i1, nres=0 ;
1.217 brouard 9232: /* double ftolpl = 1.e-10; */
9233: double age, agebase, agelim;
9234: double tot;
1.218 brouard 9235: /* double ***mobaverage; */
9236: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9237:
9238: strcpy(fileresplb,"PLB_");
9239: strcat(fileresplb,fileresu);
9240: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9241: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9242: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9243: }
9244: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9245: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9246: pstamp(ficresplb);
9247: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9248: fprintf(ficresplb,"#Age ");
9249: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9250: fprintf(ficresplb,"\n");
9251:
1.218 brouard 9252:
9253: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9254:
9255: agebase=ageminpar;
9256: agelim=agemaxpar;
9257:
9258:
1.227 brouard 9259: i1=pow(2,cptcoveff);
1.218 brouard 9260: if (cptcovn < 1){i1=1;}
1.227 brouard 9261:
1.238 brouard 9262: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9263: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9264: if(TKresult[nres]!= k)
9265: continue;
9266: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9267: fprintf(ficresplb,"#******");
9268: printf("#******");
9269: fprintf(ficlog,"#******");
9270: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9271: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9272: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9273: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9274: }
9275: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9276: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9277: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9278: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9279: }
9280: fprintf(ficresplb,"******\n");
9281: printf("******\n");
9282: fprintf(ficlog,"******\n");
9283: if(invalidvarcomb[k]){
9284: printf("\nCombination (%d) ignored because no cases \n",k);
9285: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9286: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9287: continue;
9288: }
1.218 brouard 9289:
1.238 brouard 9290: fprintf(ficresplb,"#Age ");
9291: for(j=1;j<=cptcoveff;j++) {
9292: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9293: }
9294: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9295: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9296:
9297:
1.238 brouard 9298: for (age=agebase; age<=agelim; age++){
9299: /* for (age=agebase; age<=agebase; age++){ */
9300: if(mobilavproj > 0){
9301: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9302: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9303: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9304: }else if (mobilavproj == 0){
9305: 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);
9306: 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);
9307: exit(1);
9308: }else{
9309: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9310: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9311: }
9312: fprintf(ficresplb,"%.0f ",age );
9313: for(j=1;j<=cptcoveff;j++)
9314: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9315: tot=0.;
9316: for(i=1; i<=nlstate;i++){
9317: tot += bprlim[i][i];
9318: fprintf(ficresplb," %.5f", bprlim[i][i]);
9319: }
9320: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9321: } /* Age */
9322: /* was end of cptcod */
9323: } /* end of any combination */
9324: } /* end of nres */
1.218 brouard 9325: /* hBijx(p, bage, fage); */
9326: /* fclose(ficrespijb); */
9327:
9328: return 0;
1.217 brouard 9329: }
1.218 brouard 9330:
1.180 brouard 9331: int hPijx(double *p, int bage, int fage){
9332: /*------------- h Pij x at various ages ------------*/
9333:
9334: int stepsize;
9335: int agelim;
9336: int hstepm;
9337: int nhstepm;
1.235 brouard 9338: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9339:
9340: double agedeb;
9341: double ***p3mat;
9342:
1.201 brouard 9343: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9344: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9345: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9346: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9347: }
9348: printf("Computing pij: result on file '%s' \n", filerespij);
9349: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9350:
9351: stepsize=(int) (stepm+YEARM-1)/YEARM;
9352: /*if (stepm<=24) stepsize=2;*/
9353:
9354: agelim=AGESUP;
9355: hstepm=stepsize*YEARM; /* Every year of age */
9356: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9357:
1.180 brouard 9358: /* hstepm=1; aff par mois*/
9359: pstamp(ficrespij);
9360: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9361: i1= pow(2,cptcoveff);
1.218 brouard 9362: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9363: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9364: /* k=k+1; */
1.235 brouard 9365: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9366: for(k=1; k<=i1;k++){
9367: if(TKresult[nres]!= k)
9368: continue;
1.183 brouard 9369: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9370: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9371: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9372: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9373: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9374: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9375: }
1.183 brouard 9376: fprintf(ficrespij,"******\n");
9377:
9378: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9379: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9380: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9381:
9382: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9383:
1.183 brouard 9384: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9385: oldm=oldms;savm=savms;
1.235 brouard 9386: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9387: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9388: for(i=1; i<=nlstate;i++)
9389: for(j=1; j<=nlstate+ndeath;j++)
9390: fprintf(ficrespij," %1d-%1d",i,j);
9391: fprintf(ficrespij,"\n");
9392: for (h=0; h<=nhstepm; h++){
9393: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9394: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9395: for(i=1; i<=nlstate;i++)
9396: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9397: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9398: fprintf(ficrespij,"\n");
9399: }
1.183 brouard 9400: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9401: fprintf(ficrespij,"\n");
9402: }
1.180 brouard 9403: /*}*/
9404: }
1.218 brouard 9405: return 0;
1.180 brouard 9406: }
1.218 brouard 9407:
9408: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9409: /*------------- h Bij x at various ages ------------*/
9410:
9411: int stepsize;
1.218 brouard 9412: /* int agelim; */
9413: int ageminl;
1.217 brouard 9414: int hstepm;
9415: int nhstepm;
1.238 brouard 9416: int h, i, i1, j, k, nres;
1.218 brouard 9417:
1.217 brouard 9418: double agedeb;
9419: double ***p3mat;
1.218 brouard 9420:
9421: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9422: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9423: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9424: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9425: }
9426: printf("Computing pij back: result on file '%s' \n", filerespijb);
9427: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9428:
9429: stepsize=(int) (stepm+YEARM-1)/YEARM;
9430: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9431:
1.218 brouard 9432: /* agelim=AGESUP; */
9433: ageminl=30;
9434: hstepm=stepsize*YEARM; /* Every year of age */
9435: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9436:
9437: /* hstepm=1; aff par mois*/
9438: pstamp(ficrespijb);
9439: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9440: i1= pow(2,cptcoveff);
1.218 brouard 9441: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9442: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9443: /* k=k+1; */
1.238 brouard 9444: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9445: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9446: if(TKresult[nres]!= k)
9447: continue;
9448: fprintf(ficrespijb,"\n#****** ");
9449: for(j=1;j<=cptcoveff;j++)
9450: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9451: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9452: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9453: }
9454: fprintf(ficrespijb,"******\n");
9455: if(invalidvarcomb[k]){
9456: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9457: continue;
9458: }
9459:
9460: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9461: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9462: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9463: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9464: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9465:
9466: /* nhstepm=nhstepm*YEARM; aff par mois*/
9467:
9468: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9469: /* oldm=oldms;savm=savms; */
9470: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9471: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9472: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9473: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9474: for(i=1; i<=nlstate;i++)
9475: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9476: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9477: fprintf(ficrespijb,"\n");
1.238 brouard 9478: for (h=0; h<=nhstepm; h++){
9479: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9480: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9481: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9482: for(i=1; i<=nlstate;i++)
9483: for(j=1; j<=nlstate+ndeath;j++)
9484: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9485: fprintf(ficrespijb,"\n");
9486: }
9487: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9488: fprintf(ficrespijb,"\n");
9489: } /* end age deb */
9490: } /* end combination */
9491: } /* end nres */
1.218 brouard 9492: return 0;
9493: } /* hBijx */
1.217 brouard 9494:
1.180 brouard 9495:
1.136 brouard 9496: /***********************************************/
9497: /**************** Main Program *****************/
9498: /***********************************************/
9499:
9500: int main(int argc, char *argv[])
9501: {
9502: #ifdef GSL
9503: const gsl_multimin_fminimizer_type *T;
9504: size_t iteri = 0, it;
9505: int rval = GSL_CONTINUE;
9506: int status = GSL_SUCCESS;
9507: double ssval;
9508: #endif
9509: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9510: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9511: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9512: int jj, ll, li, lj, lk;
1.136 brouard 9513: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9514: int num_filled;
1.136 brouard 9515: int itimes;
9516: int NDIM=2;
9517: int vpopbased=0;
1.235 brouard 9518: int nres=0;
1.136 brouard 9519:
1.164 brouard 9520: char ca[32], cb[32];
1.136 brouard 9521: /* FILE *fichtm; *//* Html File */
9522: /* FILE *ficgp;*/ /*Gnuplot File */
9523: struct stat info;
1.191 brouard 9524: double agedeb=0.;
1.194 brouard 9525:
9526: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9527: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9528:
1.165 brouard 9529: double fret;
1.191 brouard 9530: double dum=0.; /* Dummy variable */
1.136 brouard 9531: double ***p3mat;
1.218 brouard 9532: /* double ***mobaverage; */
1.164 brouard 9533:
9534: char line[MAXLINE];
1.197 brouard 9535: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9536:
1.234 brouard 9537: char modeltemp[MAXLINE];
1.230 brouard 9538: char resultline[MAXLINE];
9539:
1.136 brouard 9540: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9541: char *tok, *val; /* pathtot */
1.136 brouard 9542: int firstobs=1, lastobs=10;
1.195 brouard 9543: int c, h , cpt, c2;
1.191 brouard 9544: int jl=0;
9545: int i1, j1, jk, stepsize=0;
1.194 brouard 9546: int count=0;
9547:
1.164 brouard 9548: int *tab;
1.136 brouard 9549: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9550: int backcast=0;
1.136 brouard 9551: int mobilav=0,popforecast=0;
1.191 brouard 9552: int hstepm=0, nhstepm=0;
1.136 brouard 9553: int agemortsup;
9554: float sumlpop=0.;
9555: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9556: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9557:
1.191 brouard 9558: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9559: double ftolpl=FTOL;
9560: double **prlim;
1.217 brouard 9561: double **bprlim;
1.136 brouard 9562: double ***param; /* Matrix of parameters */
9563: double *p;
9564: double **matcov; /* Matrix of covariance */
1.203 brouard 9565: double **hess; /* Hessian matrix */
1.136 brouard 9566: double ***delti3; /* Scale */
9567: double *delti; /* Scale */
9568: double ***eij, ***vareij;
9569: double **varpl; /* Variances of prevalence limits by age */
9570: double *epj, vepp;
1.164 brouard 9571:
1.136 brouard 9572: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9573: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9574:
1.136 brouard 9575: double **ximort;
1.145 brouard 9576: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9577: int *dcwave;
9578:
1.164 brouard 9579: char z[1]="c";
1.136 brouard 9580:
9581: /*char *strt;*/
9582: char strtend[80];
1.126 brouard 9583:
1.164 brouard 9584:
1.126 brouard 9585: /* setlocale (LC_ALL, ""); */
9586: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9587: /* textdomain (PACKAGE); */
9588: /* setlocale (LC_CTYPE, ""); */
9589: /* setlocale (LC_MESSAGES, ""); */
9590:
9591: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9592: rstart_time = time(NULL);
9593: /* (void) gettimeofday(&start_time,&tzp);*/
9594: start_time = *localtime(&rstart_time);
1.126 brouard 9595: curr_time=start_time;
1.157 brouard 9596: /*tml = *localtime(&start_time.tm_sec);*/
9597: /* strcpy(strstart,asctime(&tml)); */
9598: strcpy(strstart,asctime(&start_time));
1.126 brouard 9599:
9600: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9601: /* tp.tm_sec = tp.tm_sec +86400; */
9602: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9603: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9604: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9605: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9606: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9607: /* strt=asctime(&tmg); */
9608: /* printf("Time(after) =%s",strstart); */
9609: /* (void) time (&time_value);
9610: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9611: * tm = *localtime(&time_value);
9612: * strstart=asctime(&tm);
9613: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9614: */
9615:
9616: nberr=0; /* Number of errors and warnings */
9617: nbwarn=0;
1.184 brouard 9618: #ifdef WIN32
9619: _getcwd(pathcd, size);
9620: #else
1.126 brouard 9621: getcwd(pathcd, size);
1.184 brouard 9622: #endif
1.191 brouard 9623: syscompilerinfo(0);
1.196 brouard 9624: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9625: if(argc <=1){
9626: printf("\nEnter the parameter file name: ");
1.205 brouard 9627: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9628: printf("ERROR Empty parameter file name\n");
9629: goto end;
9630: }
1.126 brouard 9631: i=strlen(pathr);
9632: if(pathr[i-1]=='\n')
9633: pathr[i-1]='\0';
1.156 brouard 9634: i=strlen(pathr);
1.205 brouard 9635: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9636: pathr[i-1]='\0';
1.205 brouard 9637: }
9638: i=strlen(pathr);
9639: if( i==0 ){
9640: printf("ERROR Empty parameter file name\n");
9641: goto end;
9642: }
9643: for (tok = pathr; tok != NULL; ){
1.126 brouard 9644: printf("Pathr |%s|\n",pathr);
9645: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9646: printf("val= |%s| pathr=%s\n",val,pathr);
9647: strcpy (pathtot, val);
9648: if(pathr[0] == '\0') break; /* Dirty */
9649: }
9650: }
9651: else{
9652: strcpy(pathtot,argv[1]);
9653: }
9654: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9655: /*cygwin_split_path(pathtot,path,optionfile);
9656: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9657: /* cutv(path,optionfile,pathtot,'\\');*/
9658:
9659: /* Split argv[0], imach program to get pathimach */
9660: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9661: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9662: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9663: /* strcpy(pathimach,argv[0]); */
9664: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9665: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9666: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9667: #ifdef WIN32
9668: _chdir(path); /* Can be a relative path */
9669: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9670: #else
1.126 brouard 9671: chdir(path); /* Can be a relative path */
1.184 brouard 9672: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9673: #endif
9674: printf("Current directory %s!\n",pathcd);
1.126 brouard 9675: strcpy(command,"mkdir ");
9676: strcat(command,optionfilefiname);
9677: if((outcmd=system(command)) != 0){
1.169 brouard 9678: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9679: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9680: /* fclose(ficlog); */
9681: /* exit(1); */
9682: }
9683: /* if((imk=mkdir(optionfilefiname))<0){ */
9684: /* perror("mkdir"); */
9685: /* } */
9686:
9687: /*-------- arguments in the command line --------*/
9688:
1.186 brouard 9689: /* Main Log file */
1.126 brouard 9690: strcat(filelog, optionfilefiname);
9691: strcat(filelog,".log"); /* */
9692: if((ficlog=fopen(filelog,"w"))==NULL) {
9693: printf("Problem with logfile %s\n",filelog);
9694: goto end;
9695: }
9696: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9697: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9698: fprintf(ficlog,"\nEnter the parameter file name: \n");
9699: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9700: path=%s \n\
9701: optionfile=%s\n\
9702: optionfilext=%s\n\
1.156 brouard 9703: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9704:
1.197 brouard 9705: syscompilerinfo(1);
1.167 brouard 9706:
1.126 brouard 9707: printf("Local time (at start):%s",strstart);
9708: fprintf(ficlog,"Local time (at start): %s",strstart);
9709: fflush(ficlog);
9710: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9711: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9712:
9713: /* */
9714: strcpy(fileres,"r");
9715: strcat(fileres, optionfilefiname);
1.201 brouard 9716: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9717: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9718: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9719:
1.186 brouard 9720: /* Main ---------arguments file --------*/
1.126 brouard 9721:
9722: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9723: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9724: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9725: fflush(ficlog);
1.149 brouard 9726: /* goto end; */
9727: exit(70);
1.126 brouard 9728: }
9729:
9730:
9731:
9732: strcpy(filereso,"o");
1.201 brouard 9733: strcat(filereso,fileresu);
1.126 brouard 9734: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9735: printf("Problem with Output resultfile: %s\n", filereso);
9736: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9737: fflush(ficlog);
9738: goto end;
9739: }
9740:
9741: /* Reads comments: lines beginning with '#' */
9742: numlinepar=0;
1.197 brouard 9743:
9744: /* First parameter line */
9745: while(fgets(line, MAXLINE, ficpar)) {
9746: /* If line starts with a # it is a comment */
9747: if (line[0] == '#') {
9748: numlinepar++;
9749: fputs(line,stdout);
9750: fputs(line,ficparo);
9751: fputs(line,ficlog);
9752: continue;
9753: }else
9754: break;
9755: }
9756: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9757: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9758: if (num_filled != 5) {
9759: printf("Should be 5 parameters\n");
9760: }
1.126 brouard 9761: numlinepar++;
1.197 brouard 9762: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9763: }
9764: /* Second parameter line */
9765: while(fgets(line, MAXLINE, ficpar)) {
9766: /* If line starts with a # it is a comment */
9767: if (line[0] == '#') {
9768: numlinepar++;
9769: fputs(line,stdout);
9770: fputs(line,ficparo);
9771: fputs(line,ficlog);
9772: continue;
9773: }else
9774: break;
9775: }
1.223 brouard 9776: 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", \
9777: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9778: if (num_filled != 11) {
9779: 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 9780: printf("but line=%s\n",line);
1.197 brouard 9781: }
1.223 brouard 9782: 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);
1.126 brouard 9783: }
1.203 brouard 9784: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9785: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9786: /* Third parameter line */
9787: while(fgets(line, MAXLINE, ficpar)) {
9788: /* If line starts with a # it is a comment */
9789: if (line[0] == '#') {
9790: numlinepar++;
9791: fputs(line,stdout);
9792: fputs(line,ficparo);
9793: fputs(line,ficlog);
9794: continue;
9795: }else
9796: break;
9797: }
1.201 brouard 9798: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9799: if (num_filled == 0)
9800: model[0]='\0';
9801: else if (num_filled != 1){
1.197 brouard 9802: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9803: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9804: model[0]='\0';
9805: goto end;
9806: }
9807: else{
9808: if (model[0]=='+'){
9809: for(i=1; i<=strlen(model);i++)
9810: modeltemp[i-1]=model[i];
1.201 brouard 9811: strcpy(model,modeltemp);
1.197 brouard 9812: }
9813: }
1.199 brouard 9814: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9815: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9816: }
9817: /* 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); */
9818: /* numlinepar=numlinepar+3; /\* In general *\/ */
9819: /* 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); */
1.223 brouard 9820: 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);
9821: 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 9822: fflush(ficlog);
1.190 brouard 9823: /* if(model[0]=='#'|| model[0]== '\0'){ */
9824: if(model[0]=='#'){
1.187 brouard 9825: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9826: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9827: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9828: if(mle != -1){
9829: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9830: exit(1);
9831: }
9832: }
1.126 brouard 9833: while((c=getc(ficpar))=='#' && c!= EOF){
9834: ungetc(c,ficpar);
9835: fgets(line, MAXLINE, ficpar);
9836: numlinepar++;
1.195 brouard 9837: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9838: z[0]=line[1];
9839: }
9840: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9841: fputs(line, stdout);
9842: //puts(line);
1.126 brouard 9843: fputs(line,ficparo);
9844: fputs(line,ficlog);
9845: }
9846: ungetc(c,ficpar);
9847:
9848:
1.145 brouard 9849: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9850: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9851: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9852: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9853: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9854: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9855: v1+v2*age+v2*v3 makes cptcovn = 3
9856: */
9857: if (strlen(model)>1)
1.187 brouard 9858: 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 9859: else
1.187 brouard 9860: ncovmodel=2; /* Constant and age */
1.133 brouard 9861: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9862: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9863: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9864: 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);
9865: 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);
9866: fflush(stdout);
9867: fclose (ficlog);
9868: goto end;
9869: }
1.126 brouard 9870: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9871: delti=delti3[1][1];
9872: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9873: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
9874: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9875: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9876: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9877: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9878: fclose (ficparo);
9879: fclose (ficlog);
9880: goto end;
9881: exit(0);
1.220 brouard 9882: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9883: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9884: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9885: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9886: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9887: matcov=matrix(1,npar,1,npar);
1.203 brouard 9888: hess=matrix(1,npar,1,npar);
1.220 brouard 9889: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9890: /* Read guessed parameters */
1.126 brouard 9891: /* Reads comments: lines beginning with '#' */
9892: while((c=getc(ficpar))=='#' && c!= EOF){
9893: ungetc(c,ficpar);
9894: fgets(line, MAXLINE, ficpar);
9895: numlinepar++;
1.141 brouard 9896: fputs(line,stdout);
1.126 brouard 9897: fputs(line,ficparo);
9898: fputs(line,ficlog);
9899: }
9900: ungetc(c,ficpar);
9901:
9902: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9903: for(i=1; i <=nlstate; i++){
1.234 brouard 9904: j=0;
1.126 brouard 9905: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9906: if(jj==i) continue;
9907: j++;
9908: fscanf(ficpar,"%1d%1d",&i1,&j1);
9909: if ((i1 != i) || (j1 != jj)){
9910: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9911: It might be a problem of design; if ncovcol and the model are correct\n \
9912: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9913: exit(1);
9914: }
9915: fprintf(ficparo,"%1d%1d",i1,j1);
9916: if(mle==1)
9917: printf("%1d%1d",i,jj);
9918: fprintf(ficlog,"%1d%1d",i,jj);
9919: for(k=1; k<=ncovmodel;k++){
9920: fscanf(ficpar," %lf",¶m[i][j][k]);
9921: if(mle==1){
9922: printf(" %lf",param[i][j][k]);
9923: fprintf(ficlog," %lf",param[i][j][k]);
9924: }
9925: else
9926: fprintf(ficlog," %lf",param[i][j][k]);
9927: fprintf(ficparo," %lf",param[i][j][k]);
9928: }
9929: fscanf(ficpar,"\n");
9930: numlinepar++;
9931: if(mle==1)
9932: printf("\n");
9933: fprintf(ficlog,"\n");
9934: fprintf(ficparo,"\n");
1.126 brouard 9935: }
9936: }
9937: fflush(ficlog);
1.234 brouard 9938:
1.145 brouard 9939: /* Reads scales values */
1.126 brouard 9940: p=param[1][1];
9941:
9942: /* Reads comments: lines beginning with '#' */
9943: while((c=getc(ficpar))=='#' && c!= EOF){
9944: ungetc(c,ficpar);
9945: fgets(line, MAXLINE, ficpar);
9946: numlinepar++;
1.141 brouard 9947: fputs(line,stdout);
1.126 brouard 9948: fputs(line,ficparo);
9949: fputs(line,ficlog);
9950: }
9951: ungetc(c,ficpar);
9952:
9953: for(i=1; i <=nlstate; i++){
9954: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 9955: fscanf(ficpar,"%1d%1d",&i1,&j1);
9956: if ( (i1-i) * (j1-j) != 0){
9957: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
9958: exit(1);
9959: }
9960: printf("%1d%1d",i,j);
9961: fprintf(ficparo,"%1d%1d",i1,j1);
9962: fprintf(ficlog,"%1d%1d",i1,j1);
9963: for(k=1; k<=ncovmodel;k++){
9964: fscanf(ficpar,"%le",&delti3[i][j][k]);
9965: printf(" %le",delti3[i][j][k]);
9966: fprintf(ficparo," %le",delti3[i][j][k]);
9967: fprintf(ficlog," %le",delti3[i][j][k]);
9968: }
9969: fscanf(ficpar,"\n");
9970: numlinepar++;
9971: printf("\n");
9972: fprintf(ficparo,"\n");
9973: fprintf(ficlog,"\n");
1.126 brouard 9974: }
9975: }
9976: fflush(ficlog);
1.234 brouard 9977:
1.145 brouard 9978: /* Reads covariance matrix */
1.126 brouard 9979: delti=delti3[1][1];
1.220 brouard 9980:
9981:
1.126 brouard 9982: /* 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 9983:
1.126 brouard 9984: /* Reads comments: lines beginning with '#' */
9985: while((c=getc(ficpar))=='#' && c!= EOF){
9986: ungetc(c,ficpar);
9987: fgets(line, MAXLINE, ficpar);
9988: numlinepar++;
1.141 brouard 9989: fputs(line,stdout);
1.126 brouard 9990: fputs(line,ficparo);
9991: fputs(line,ficlog);
9992: }
9993: ungetc(c,ficpar);
1.220 brouard 9994:
1.126 brouard 9995: matcov=matrix(1,npar,1,npar);
1.203 brouard 9996: hess=matrix(1,npar,1,npar);
1.131 brouard 9997: for(i=1; i <=npar; i++)
9998: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 9999:
1.194 brouard 10000: /* Scans npar lines */
1.126 brouard 10001: for(i=1; i <=npar; i++){
1.226 brouard 10002: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10003: if(count != 3){
1.226 brouard 10004: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10005: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10006: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10007: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10008: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10009: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10010: exit(1);
1.220 brouard 10011: }else{
1.226 brouard 10012: if(mle==1)
10013: printf("%1d%1d%d",i1,j1,jk);
10014: }
10015: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10016: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10017: for(j=1; j <=i; j++){
1.226 brouard 10018: fscanf(ficpar," %le",&matcov[i][j]);
10019: if(mle==1){
10020: printf(" %.5le",matcov[i][j]);
10021: }
10022: fprintf(ficlog," %.5le",matcov[i][j]);
10023: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10024: }
10025: fscanf(ficpar,"\n");
10026: numlinepar++;
10027: if(mle==1)
1.220 brouard 10028: printf("\n");
1.126 brouard 10029: fprintf(ficlog,"\n");
10030: fprintf(ficparo,"\n");
10031: }
1.194 brouard 10032: /* End of read covariance matrix npar lines */
1.126 brouard 10033: for(i=1; i <=npar; i++)
10034: for(j=i+1;j<=npar;j++)
1.226 brouard 10035: matcov[i][j]=matcov[j][i];
1.126 brouard 10036:
10037: if(mle==1)
10038: printf("\n");
10039: fprintf(ficlog,"\n");
10040:
10041: fflush(ficlog);
10042:
10043: /*-------- Rewriting parameter file ----------*/
10044: strcpy(rfileres,"r"); /* "Rparameterfile */
10045: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10046: strcat(rfileres,"."); /* */
10047: strcat(rfileres,optionfilext); /* Other files have txt extension */
10048: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10049: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10050: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10051: }
10052: fprintf(ficres,"#%s\n",version);
10053: } /* End of mle != -3 */
1.218 brouard 10054:
1.186 brouard 10055: /* Main data
10056: */
1.126 brouard 10057: n= lastobs;
10058: num=lvector(1,n);
10059: moisnais=vector(1,n);
10060: annais=vector(1,n);
10061: moisdc=vector(1,n);
10062: andc=vector(1,n);
1.220 brouard 10063: weight=vector(1,n);
1.126 brouard 10064: agedc=vector(1,n);
10065: cod=ivector(1,n);
1.220 brouard 10066: for(i=1;i<=n;i++){
1.234 brouard 10067: num[i]=0;
10068: moisnais[i]=0;
10069: annais[i]=0;
10070: moisdc[i]=0;
10071: andc[i]=0;
10072: agedc[i]=0;
10073: cod[i]=0;
10074: weight[i]=1.0; /* Equal weights, 1 by default */
10075: }
1.126 brouard 10076: mint=matrix(1,maxwav,1,n);
10077: anint=matrix(1,maxwav,1,n);
1.131 brouard 10078: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10079: tab=ivector(1,NCOVMAX);
1.144 brouard 10080: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10081: 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 10082:
1.136 brouard 10083: /* Reads data from file datafile */
10084: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10085: goto end;
10086:
10087: /* Calculation of the number of parameters from char model */
1.234 brouard 10088: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10089: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10090: k=3 V4 Tvar[k=3]= 4 (from V4)
10091: k=2 V1 Tvar[k=2]= 1 (from V1)
10092: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10093: */
10094:
10095: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10096: TvarsDind=ivector(1,NCOVMAX); /* */
10097: TvarsD=ivector(1,NCOVMAX); /* */
10098: TvarsQind=ivector(1,NCOVMAX); /* */
10099: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10100: TvarF=ivector(1,NCOVMAX); /* */
10101: TvarFind=ivector(1,NCOVMAX); /* */
10102: TvarV=ivector(1,NCOVMAX); /* */
10103: TvarVind=ivector(1,NCOVMAX); /* */
10104: TvarA=ivector(1,NCOVMAX); /* */
10105: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10106: TvarFD=ivector(1,NCOVMAX); /* */
10107: TvarFDind=ivector(1,NCOVMAX); /* */
10108: TvarFQ=ivector(1,NCOVMAX); /* */
10109: TvarFQind=ivector(1,NCOVMAX); /* */
10110: TvarVD=ivector(1,NCOVMAX); /* */
10111: TvarVDind=ivector(1,NCOVMAX); /* */
10112: TvarVQ=ivector(1,NCOVMAX); /* */
10113: TvarVQind=ivector(1,NCOVMAX); /* */
10114:
1.230 brouard 10115: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10116: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10117: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10118: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10119: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10120: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10121: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10122: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10123: */
10124: /* For model-covariate k tells which data-covariate to use but
10125: because this model-covariate is a construction we invent a new column
10126: ncovcol + k1
10127: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10128: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10129: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10130: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10131: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10132: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10133: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10134: */
1.145 brouard 10135: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10136: 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 10137: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10138: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10139: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10140: 4 covariates (3 plus signs)
10141: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10142: */
1.230 brouard 10143: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10144: * individual dummy, fixed or varying:
10145: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10146: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10147: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10148: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10149: * Tmodelind[1]@9={9,0,3,2,}*/
10150: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10151: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10152: * individual quantitative, fixed or varying:
10153: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10154: * 3, 1, 0, 0, 0, 0, 0, 0},
10155: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10156: /* Main decodemodel */
10157:
1.187 brouard 10158:
1.223 brouard 10159: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10160: goto end;
10161:
1.137 brouard 10162: if((double)(lastobs-imx)/(double)imx > 1.10){
10163: nbwarn++;
10164: 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);
10165: 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);
10166: }
1.136 brouard 10167: /* if(mle==1){*/
1.137 brouard 10168: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10169: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10170: }
10171:
10172: /*-calculation of age at interview from date of interview and age at death -*/
10173: agev=matrix(1,maxwav,1,imx);
10174:
10175: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10176: goto end;
10177:
1.126 brouard 10178:
1.136 brouard 10179: agegomp=(int)agemin;
10180: free_vector(moisnais,1,n);
10181: free_vector(annais,1,n);
1.126 brouard 10182: /* free_matrix(mint,1,maxwav,1,n);
10183: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10184: /* free_vector(moisdc,1,n); */
10185: /* free_vector(andc,1,n); */
1.145 brouard 10186: /* */
10187:
1.126 brouard 10188: wav=ivector(1,imx);
1.214 brouard 10189: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10190: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10191: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10192: 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.*/
10193: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10194: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10195:
10196: /* Concatenates waves */
1.214 brouard 10197: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10198: Death is a valid wave (if date is known).
10199: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10200: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10201: and mw[mi+1][i]. dh depends on stepm.
10202: */
10203:
1.126 brouard 10204: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.145 brouard 10205: /* */
10206:
1.215 brouard 10207: free_vector(moisdc,1,n);
10208: free_vector(andc,1,n);
10209:
1.126 brouard 10210: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10211: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10212: ncodemax[1]=1;
1.145 brouard 10213: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10214: cptcoveff=0;
1.220 brouard 10215: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10216: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10217: }
10218:
10219: ncovcombmax=pow(2,cptcoveff);
10220: invalidvarcomb=ivector(1, ncovcombmax);
10221: for(i=1;i<ncovcombmax;i++)
10222: invalidvarcomb[i]=0;
10223:
1.211 brouard 10224: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10225: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10226: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10227:
1.200 brouard 10228: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10229: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10230: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10231: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10232: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10233: * (currently 0 or 1) in the data.
10234: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10235: * corresponding modality (h,j).
10236: */
10237:
1.145 brouard 10238: h=0;
10239: /*if (cptcovn > 0) */
1.126 brouard 10240: m=pow(2,cptcoveff);
10241:
1.144 brouard 10242: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10243: * For k=4 covariates, h goes from 1 to m=2**k
10244: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10245: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10246: * h\k 1 2 3 4
1.143 brouard 10247: *______________________________
10248: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10249: * 2 2 1 1 1
10250: * 3 i=2 1 2 1 1
10251: * 4 2 2 1 1
10252: * 5 i=3 1 i=2 1 2 1
10253: * 6 2 1 2 1
10254: * 7 i=4 1 2 2 1
10255: * 8 2 2 2 1
1.197 brouard 10256: * 9 i=5 1 i=3 1 i=2 1 2
10257: * 10 2 1 1 2
10258: * 11 i=6 1 2 1 2
10259: * 12 2 2 1 2
10260: * 13 i=7 1 i=4 1 2 2
10261: * 14 2 1 2 2
10262: * 15 i=8 1 2 2 2
10263: * 16 2 2 2 2
1.143 brouard 10264: */
1.212 brouard 10265: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10266: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10267: * and the value of each covariate?
10268: * V1=1, V2=1, V3=2, V4=1 ?
10269: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10270: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10271: * In order to get the real value in the data, we use nbcode
10272: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10273: * We are keeping this crazy system in order to be able (in the future?)
10274: * to have more than 2 values (0 or 1) for a covariate.
10275: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10276: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10277: * bbbbbbbb
10278: * 76543210
10279: * h-1 00000101 (6-1=5)
1.219 brouard 10280: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10281: * &
10282: * 1 00000001 (1)
1.219 brouard 10283: * 00000000 = 1 & ((h-1) >> (k-1))
10284: * +1= 00000001 =1
1.211 brouard 10285: *
10286: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10287: * h' 1101 =2^3+2^2+0x2^1+2^0
10288: * >>k' 11
10289: * & 00000001
10290: * = 00000001
10291: * +1 = 00000010=2 = codtabm(14,3)
10292: * Reverse h=6 and m=16?
10293: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10294: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10295: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10296: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10297: * V3=decodtabm(14,3,2**4)=2
10298: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10299: *(h-1) >> (j-1) 0011 =13 >> 2
10300: * &1 000000001
10301: * = 000000001
10302: * +1= 000000010 =2
10303: * 2211
10304: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10305: * V3=2
1.220 brouard 10306: * codtabm and decodtabm are identical
1.211 brouard 10307: */
10308:
1.145 brouard 10309:
10310: free_ivector(Ndum,-1,NCOVMAX);
10311:
10312:
1.126 brouard 10313:
1.186 brouard 10314: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10315: strcpy(optionfilegnuplot,optionfilefiname);
10316: if(mle==-3)
1.201 brouard 10317: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10318: strcat(optionfilegnuplot,".gp");
10319:
10320: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10321: printf("Problem with file %s",optionfilegnuplot);
10322: }
10323: else{
1.204 brouard 10324: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10325: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10326: //fprintf(ficgp,"set missing 'NaNq'\n");
10327: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10328: }
10329: /* fclose(ficgp);*/
1.186 brouard 10330:
10331:
10332: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10333:
10334: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10335: if(mle==-3)
1.201 brouard 10336: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10337: strcat(optionfilehtm,".htm");
10338: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10339: printf("Problem with %s \n",optionfilehtm);
10340: exit(0);
1.126 brouard 10341: }
10342:
10343: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10344: strcat(optionfilehtmcov,"-cov.htm");
10345: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10346: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10347: }
10348: else{
10349: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10350: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10351: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10352: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10353: }
10354:
1.213 brouard 10355: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本学術振興会 (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \
1.204 brouard 10356: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10357: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10358: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10359: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10360: \n\
10361: <hr size=\"2\" color=\"#EC5E5E\">\
10362: <ul><li><h4>Parameter files</h4>\n\
10363: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10364: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10365: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10366: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10367: - Date and time at start: %s</ul>\n",\
10368: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10369: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10370: fileres,fileres,\
10371: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10372: fflush(fichtm);
10373:
10374: strcpy(pathr,path);
10375: strcat(pathr,optionfilefiname);
1.184 brouard 10376: #ifdef WIN32
10377: _chdir(optionfilefiname); /* Move to directory named optionfile */
10378: #else
1.126 brouard 10379: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10380: #endif
10381:
1.126 brouard 10382:
1.220 brouard 10383: /* Calculates basic frequencies. Computes observed prevalence at single age
10384: and for any valid combination of covariates
1.126 brouard 10385: and prints on file fileres'p'. */
1.227 brouard 10386: freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
10387: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10388:
10389: fprintf(fichtm,"\n");
10390: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10391: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10392: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10393: imx,agemin,agemax,jmin,jmax,jmean);
10394: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10395: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10396: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10397: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10398: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10399:
1.126 brouard 10400: /* For Powell, parameters are in a vector p[] starting at p[1]
10401: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10402: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10403:
10404: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10405: /* For mortality only */
1.126 brouard 10406: if (mle==-3){
1.136 brouard 10407: ximort=matrix(1,NDIM,1,NDIM);
1.220 brouard 10408: for(i=1;i<=NDIM;i++)
10409: for(j=1;j<=NDIM;j++)
10410: ximort[i][j]=0.;
1.186 brouard 10411: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10412: cens=ivector(1,n);
10413: ageexmed=vector(1,n);
10414: agecens=vector(1,n);
10415: dcwave=ivector(1,n);
1.223 brouard 10416:
1.126 brouard 10417: for (i=1; i<=imx; i++){
10418: dcwave[i]=-1;
10419: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10420: if (s[m][i]>nlstate) {
10421: dcwave[i]=m;
10422: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10423: break;
10424: }
1.126 brouard 10425: }
1.226 brouard 10426:
1.126 brouard 10427: for (i=1; i<=imx; i++) {
10428: if (wav[i]>0){
1.226 brouard 10429: ageexmed[i]=agev[mw[1][i]][i];
10430: j=wav[i];
10431: agecens[i]=1.;
10432:
10433: if (ageexmed[i]> 1 && wav[i] > 0){
10434: agecens[i]=agev[mw[j][i]][i];
10435: cens[i]= 1;
10436: }else if (ageexmed[i]< 1)
10437: cens[i]= -1;
10438: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10439: cens[i]=0 ;
1.126 brouard 10440: }
10441: else cens[i]=-1;
10442: }
10443:
10444: for (i=1;i<=NDIM;i++) {
10445: for (j=1;j<=NDIM;j++)
1.226 brouard 10446: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10447: }
10448:
1.145 brouard 10449: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10450: /*printf("%lf %lf", p[1], p[2]);*/
10451:
10452:
1.136 brouard 10453: #ifdef GSL
10454: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10455: #else
1.126 brouard 10456: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10457: #endif
1.201 brouard 10458: strcpy(filerespow,"POW-MORT_");
10459: strcat(filerespow,fileresu);
1.126 brouard 10460: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10461: printf("Problem with resultfile: %s\n", filerespow);
10462: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10463: }
1.136 brouard 10464: #ifdef GSL
10465: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10466: #else
1.126 brouard 10467: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10468: #endif
1.126 brouard 10469: /* for (i=1;i<=nlstate;i++)
10470: for(j=1;j<=nlstate+ndeath;j++)
10471: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10472: */
10473: fprintf(ficrespow,"\n");
1.136 brouard 10474: #ifdef GSL
10475: /* gsl starts here */
10476: T = gsl_multimin_fminimizer_nmsimplex;
10477: gsl_multimin_fminimizer *sfm = NULL;
10478: gsl_vector *ss, *x;
10479: gsl_multimin_function minex_func;
10480:
10481: /* Initial vertex size vector */
10482: ss = gsl_vector_alloc (NDIM);
10483:
10484: if (ss == NULL){
10485: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10486: }
10487: /* Set all step sizes to 1 */
10488: gsl_vector_set_all (ss, 0.001);
10489:
10490: /* Starting point */
1.126 brouard 10491:
1.136 brouard 10492: x = gsl_vector_alloc (NDIM);
10493:
10494: if (x == NULL){
10495: gsl_vector_free(ss);
10496: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10497: }
10498:
10499: /* Initialize method and iterate */
10500: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10501: /* gsl_vector_set(x, 0, 0.0268); */
10502: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10503: gsl_vector_set(x, 0, p[1]);
10504: gsl_vector_set(x, 1, p[2]);
10505:
10506: minex_func.f = &gompertz_f;
10507: minex_func.n = NDIM;
10508: minex_func.params = (void *)&p; /* ??? */
10509:
10510: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10511: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10512:
10513: printf("Iterations beginning .....\n\n");
10514: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10515:
10516: iteri=0;
10517: while (rval == GSL_CONTINUE){
10518: iteri++;
10519: status = gsl_multimin_fminimizer_iterate(sfm);
10520:
10521: if (status) printf("error: %s\n", gsl_strerror (status));
10522: fflush(0);
10523:
10524: if (status)
10525: break;
10526:
10527: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10528: ssval = gsl_multimin_fminimizer_size (sfm);
10529:
10530: if (rval == GSL_SUCCESS)
10531: printf ("converged to a local maximum at\n");
10532:
10533: printf("%5d ", iteri);
10534: for (it = 0; it < NDIM; it++){
10535: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10536: }
10537: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10538: }
10539:
10540: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10541:
10542: gsl_vector_free(x); /* initial values */
10543: gsl_vector_free(ss); /* inital step size */
10544: for (it=0; it<NDIM; it++){
10545: p[it+1]=gsl_vector_get(sfm->x,it);
10546: fprintf(ficrespow," %.12lf", p[it]);
10547: }
10548: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10549: #endif
10550: #ifdef POWELL
10551: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10552: #endif
1.126 brouard 10553: fclose(ficrespow);
10554:
1.203 brouard 10555: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10556:
10557: for(i=1; i <=NDIM; i++)
10558: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10559: matcov[i][j]=matcov[j][i];
1.126 brouard 10560:
10561: printf("\nCovariance matrix\n ");
1.203 brouard 10562: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10563: for(i=1; i <=NDIM; i++) {
10564: for(j=1;j<=NDIM;j++){
1.220 brouard 10565: printf("%f ",matcov[i][j]);
10566: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10567: }
1.203 brouard 10568: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10569: }
10570:
10571: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10572: for (i=1;i<=NDIM;i++) {
1.126 brouard 10573: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10574: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10575: }
1.126 brouard 10576: lsurv=vector(1,AGESUP);
10577: lpop=vector(1,AGESUP);
10578: tpop=vector(1,AGESUP);
10579: lsurv[agegomp]=100000;
10580:
10581: for (k=agegomp;k<=AGESUP;k++) {
10582: agemortsup=k;
10583: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10584: }
10585:
10586: for (k=agegomp;k<agemortsup;k++)
10587: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10588:
10589: for (k=agegomp;k<agemortsup;k++){
10590: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10591: sumlpop=sumlpop+lpop[k];
10592: }
10593:
10594: tpop[agegomp]=sumlpop;
10595: for (k=agegomp;k<(agemortsup-3);k++){
10596: /* tpop[k+1]=2;*/
10597: tpop[k+1]=tpop[k]-lpop[k];
10598: }
10599:
10600:
10601: printf("\nAge lx qx dx Lx Tx e(x)\n");
10602: for (k=agegomp;k<(agemortsup-2);k++)
10603: 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]);
10604:
10605:
10606: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10607: ageminpar=50;
10608: agemaxpar=100;
1.194 brouard 10609: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10610: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10611: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10612: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10613: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10614: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10615: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10616: }else{
10617: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10618: 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 10619: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10620: }
1.201 brouard 10621: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10622: stepm, weightopt,\
10623: model,imx,p,matcov,agemortsup);
10624:
10625: free_vector(lsurv,1,AGESUP);
10626: free_vector(lpop,1,AGESUP);
10627: free_vector(tpop,1,AGESUP);
1.220 brouard 10628: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10629: free_ivector(cens,1,n);
10630: free_vector(agecens,1,n);
10631: free_ivector(dcwave,1,n);
1.220 brouard 10632: #ifdef GSL
1.136 brouard 10633: #endif
1.186 brouard 10634: } /* Endof if mle==-3 mortality only */
1.205 brouard 10635: /* Standard */
10636: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10637: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10638: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10639: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10640: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10641: for (k=1; k<=npar;k++)
10642: printf(" %d %8.5f",k,p[k]);
10643: printf("\n");
1.205 brouard 10644: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10645: /* mlikeli uses func not funcone */
10646: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10647: }
10648: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10649: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10650: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10651: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10652: }
10653: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10654: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10655: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10656: for (k=1; k<=npar;k++)
10657: printf(" %d %8.5f",k,p[k]);
10658: printf("\n");
10659:
10660: /*--------- results files --------------*/
1.224 brouard 10661: 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 10662:
10663:
10664: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10665: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10666: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10667: for(i=1,jk=1; i <=nlstate; i++){
10668: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10669: if (k != i) {
10670: printf("%d%d ",i,k);
10671: fprintf(ficlog,"%d%d ",i,k);
10672: fprintf(ficres,"%1d%1d ",i,k);
10673: for(j=1; j <=ncovmodel; j++){
10674: printf("%12.7f ",p[jk]);
10675: fprintf(ficlog,"%12.7f ",p[jk]);
10676: fprintf(ficres,"%12.7f ",p[jk]);
10677: jk++;
10678: }
10679: printf("\n");
10680: fprintf(ficlog,"\n");
10681: fprintf(ficres,"\n");
10682: }
1.126 brouard 10683: }
10684: }
1.203 brouard 10685: if(mle != 0){
10686: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10687: ftolhess=ftol; /* Usually correct */
1.203 brouard 10688: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10689: 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");
10690: 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");
10691: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10692: for(k=1; k <=(nlstate+ndeath); k++){
10693: if (k != i) {
10694: printf("%d%d ",i,k);
10695: fprintf(ficlog,"%d%d ",i,k);
10696: for(j=1; j <=ncovmodel; j++){
10697: 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]));
10698: 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]));
10699: jk++;
10700: }
10701: printf("\n");
10702: fprintf(ficlog,"\n");
10703: }
10704: }
1.193 brouard 10705: }
1.203 brouard 10706: } /* end of hesscov and Wald tests */
1.225 brouard 10707:
1.203 brouard 10708: /* */
1.126 brouard 10709: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10710: printf("# Scales (for hessian or gradient estimation)\n");
10711: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10712: for(i=1,jk=1; i <=nlstate; i++){
10713: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10714: if (j!=i) {
10715: fprintf(ficres,"%1d%1d",i,j);
10716: printf("%1d%1d",i,j);
10717: fprintf(ficlog,"%1d%1d",i,j);
10718: for(k=1; k<=ncovmodel;k++){
10719: printf(" %.5e",delti[jk]);
10720: fprintf(ficlog," %.5e",delti[jk]);
10721: fprintf(ficres," %.5e",delti[jk]);
10722: jk++;
10723: }
10724: printf("\n");
10725: fprintf(ficlog,"\n");
10726: fprintf(ficres,"\n");
10727: }
1.126 brouard 10728: }
10729: }
10730:
10731: 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 10732: if(mle >= 1) /* To big for the screen */
1.126 brouard 10733: 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");
10734: 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");
10735: /* # 121 Var(a12)\n\ */
10736: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10737: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10738: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10739: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10740: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10741: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10742: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10743:
10744:
10745: /* Just to have a covariance matrix which will be more understandable
10746: even is we still don't want to manage dictionary of variables
10747: */
10748: for(itimes=1;itimes<=2;itimes++){
10749: jj=0;
10750: for(i=1; i <=nlstate; i++){
1.225 brouard 10751: for(j=1; j <=nlstate+ndeath; j++){
10752: if(j==i) continue;
10753: for(k=1; k<=ncovmodel;k++){
10754: jj++;
10755: ca[0]= k+'a'-1;ca[1]='\0';
10756: if(itimes==1){
10757: if(mle>=1)
10758: printf("#%1d%1d%d",i,j,k);
10759: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10760: fprintf(ficres,"#%1d%1d%d",i,j,k);
10761: }else{
10762: if(mle>=1)
10763: printf("%1d%1d%d",i,j,k);
10764: fprintf(ficlog,"%1d%1d%d",i,j,k);
10765: fprintf(ficres,"%1d%1d%d",i,j,k);
10766: }
10767: ll=0;
10768: for(li=1;li <=nlstate; li++){
10769: for(lj=1;lj <=nlstate+ndeath; lj++){
10770: if(lj==li) continue;
10771: for(lk=1;lk<=ncovmodel;lk++){
10772: ll++;
10773: if(ll<=jj){
10774: cb[0]= lk +'a'-1;cb[1]='\0';
10775: if(ll<jj){
10776: if(itimes==1){
10777: if(mle>=1)
10778: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10779: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10780: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10781: }else{
10782: if(mle>=1)
10783: printf(" %.5e",matcov[jj][ll]);
10784: fprintf(ficlog," %.5e",matcov[jj][ll]);
10785: fprintf(ficres," %.5e",matcov[jj][ll]);
10786: }
10787: }else{
10788: if(itimes==1){
10789: if(mle>=1)
10790: printf(" Var(%s%1d%1d)",ca,i,j);
10791: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10792: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10793: }else{
10794: if(mle>=1)
10795: printf(" %.7e",matcov[jj][ll]);
10796: fprintf(ficlog," %.7e",matcov[jj][ll]);
10797: fprintf(ficres," %.7e",matcov[jj][ll]);
10798: }
10799: }
10800: }
10801: } /* end lk */
10802: } /* end lj */
10803: } /* end li */
10804: if(mle>=1)
10805: printf("\n");
10806: fprintf(ficlog,"\n");
10807: fprintf(ficres,"\n");
10808: numlinepar++;
10809: } /* end k*/
10810: } /*end j */
1.126 brouard 10811: } /* end i */
10812: } /* end itimes */
10813:
10814: fflush(ficlog);
10815: fflush(ficres);
1.225 brouard 10816: while(fgets(line, MAXLINE, ficpar)) {
10817: /* If line starts with a # it is a comment */
10818: if (line[0] == '#') {
10819: numlinepar++;
10820: fputs(line,stdout);
10821: fputs(line,ficparo);
10822: fputs(line,ficlog);
10823: continue;
10824: }else
10825: break;
10826: }
10827:
1.209 brouard 10828: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10829: /* ungetc(c,ficpar); */
10830: /* fgets(line, MAXLINE, ficpar); */
10831: /* fputs(line,stdout); */
10832: /* fputs(line,ficparo); */
10833: /* } */
10834: /* ungetc(c,ficpar); */
1.126 brouard 10835:
10836: estepm=0;
1.209 brouard 10837: 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 10838:
10839: if (num_filled != 6) {
10840: 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);
10841: 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);
10842: goto end;
10843: }
10844: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10845: }
10846: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10847: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10848:
1.209 brouard 10849: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10850: if (estepm==0 || estepm < stepm) estepm=stepm;
10851: if (fage <= 2) {
10852: bage = ageminpar;
10853: fage = agemaxpar;
10854: }
10855:
10856: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10857: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10858: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10859:
1.186 brouard 10860: /* Other stuffs, more or less useful */
1.126 brouard 10861: while((c=getc(ficpar))=='#' && c!= EOF){
10862: ungetc(c,ficpar);
10863: fgets(line, MAXLINE, ficpar);
1.141 brouard 10864: fputs(line,stdout);
1.126 brouard 10865: fputs(line,ficparo);
10866: }
10867: ungetc(c,ficpar);
10868:
10869: fscanf(ficpar,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav);
10870: 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);
10871: 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);
10872: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10873: 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);
10874:
10875: while((c=getc(ficpar))=='#' && c!= EOF){
10876: ungetc(c,ficpar);
10877: fgets(line, MAXLINE, ficpar);
1.141 brouard 10878: fputs(line,stdout);
1.126 brouard 10879: fputs(line,ficparo);
10880: }
10881: ungetc(c,ficpar);
10882:
10883:
10884: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10885: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10886:
10887: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10888: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10889: fprintf(ficparo,"pop_based=%d\n",popbased);
10890: fprintf(ficres,"pop_based=%d\n",popbased);
10891:
10892: while((c=getc(ficpar))=='#' && c!= EOF){
10893: ungetc(c,ficpar);
10894: fgets(line, MAXLINE, ficpar);
1.141 brouard 10895: fputs(line,stdout);
1.238 brouard 10896: fputs(line,ficres);
1.126 brouard 10897: fputs(line,ficparo);
10898: }
10899: ungetc(c,ficpar);
10900:
10901: fscanf(ficpar,"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);
10902: 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);
10903: 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);
10904: 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);
10905: 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);
10906: /* day and month of proj2 are not used but only year anproj2.*/
10907:
1.217 brouard 10908: while((c=getc(ficpar))=='#' && c!= EOF){
10909: ungetc(c,ficpar);
10910: fgets(line, MAXLINE, ficpar);
10911: fputs(line,stdout);
10912: fputs(line,ficparo);
1.238 brouard 10913: fputs(line,ficres);
1.217 brouard 10914: }
10915: ungetc(c,ficpar);
10916:
10917: fscanf(ficpar,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj);
1.223 brouard 10918: fprintf(ficparo,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
10919: fprintf(ficlog,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
10920: fprintf(ficres,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
1.217 brouard 10921: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 10922:
1.230 brouard 10923: /* Results */
1.235 brouard 10924: nresult=0;
1.230 brouard 10925: while(fgets(line, MAXLINE, ficpar)) {
10926: /* If line starts with a # it is a comment */
10927: if (line[0] == '#') {
10928: numlinepar++;
10929: fputs(line,stdout);
10930: fputs(line,ficparo);
10931: fputs(line,ficlog);
1.238 brouard 10932: fputs(line,ficres);
1.230 brouard 10933: continue;
10934: }else
10935: break;
10936: }
1.240 brouard 10937: if (!feof(ficpar))
1.230 brouard 10938: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 10939: if (num_filled == 0){
1.230 brouard 10940: resultline[0]='\0';
1.240 brouard 10941: break;
10942: } else if (num_filled != 1){
1.230 brouard 10943: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
10944: }
1.235 brouard 10945: nresult++; /* Sum of resultlines */
10946: printf("Result %d: result=%s\n",nresult, resultline);
10947: if(nresult > MAXRESULTLINES){
10948: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10949: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10950: goto end;
10951: }
10952: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 10953: fprintf(ficparo,"result: %s\n",resultline);
10954: fprintf(ficres,"result: %s\n",resultline);
10955: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 10956: while(fgets(line, MAXLINE, ficpar)) {
10957: /* If line starts with a # it is a comment */
10958: if (line[0] == '#') {
10959: numlinepar++;
10960: fputs(line,stdout);
10961: fputs(line,ficparo);
1.238 brouard 10962: fputs(line,ficres);
1.230 brouard 10963: fputs(line,ficlog);
10964: continue;
10965: }else
10966: break;
10967: }
10968: if (feof(ficpar))
10969: break;
10970: else{ /* Processess output results for this combination of covariate values */
10971: }
1.240 brouard 10972: } /* end while */
1.230 brouard 10973:
10974:
1.126 brouard 10975:
1.230 brouard 10976: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 10977: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 10978:
10979: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 10980: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 10981: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10982: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10983: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 10984: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10985: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10986: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10987: }else{
1.218 brouard 10988: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 10989: }
10990: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 10991: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
10992: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 10993:
1.225 brouard 10994: /*------------ free_vector -------------*/
10995: /* chdir(path); */
1.220 brouard 10996:
1.215 brouard 10997: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
10998: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
10999: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11000: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11001: free_lvector(num,1,n);
11002: free_vector(agedc,1,n);
11003: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11004: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11005: fclose(ficparo);
11006: fclose(ficres);
1.220 brouard 11007:
11008:
1.186 brouard 11009: /* Other results (useful)*/
1.220 brouard 11010:
11011:
1.126 brouard 11012: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11013: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11014: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11015: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11016: fclose(ficrespl);
11017:
11018: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11019: /*#include "hpijx.h"*/
11020: hPijx(p, bage, fage);
1.145 brouard 11021: fclose(ficrespij);
1.227 brouard 11022:
1.220 brouard 11023: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11024: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11025: k=1;
1.126 brouard 11026: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11027:
1.219 brouard 11028: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11029: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11030: for(i=1;i<=AGESUP;i++)
1.219 brouard 11031: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11032: for(k=1;k<=ncovcombmax;k++)
11033: probs[i][j][k]=0.;
1.219 brouard 11034: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11035: if (mobilav!=0 ||mobilavproj !=0 ) {
11036: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11037: for(i=1;i<=AGESUP;i++)
11038: for(j=1;j<=nlstate;j++)
11039: for(k=1;k<=ncovcombmax;k++)
11040: mobaverages[i][j][k]=0.;
1.219 brouard 11041: mobaverage=mobaverages;
11042: if (mobilav!=0) {
1.235 brouard 11043: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11044: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11045: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11046: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11047: }
1.219 brouard 11048: }
11049: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11050: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11051: else if (mobilavproj !=0) {
1.235 brouard 11052: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11053: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11054: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11055: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11056: }
1.219 brouard 11057: }
11058: }/* end if moving average */
1.227 brouard 11059:
1.126 brouard 11060: /*---------- Forecasting ------------------*/
11061: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11062: if(prevfcast==1){
11063: /* if(stepm ==1){*/
1.225 brouard 11064: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11065: }
1.217 brouard 11066: if(backcast==1){
1.219 brouard 11067: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11068: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11069: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11070:
11071: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11072:
11073: bprlim=matrix(1,nlstate,1,nlstate);
11074: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11075: fclose(ficresplb);
11076:
1.222 brouard 11077: hBijx(p, bage, fage, mobaverage);
11078: fclose(ficrespijb);
1.219 brouard 11079: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11080:
11081: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11082: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11083: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11084: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11085: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11086: }
1.217 brouard 11087:
1.186 brouard 11088:
11089: /* ------ Other prevalence ratios------------ */
1.126 brouard 11090:
1.215 brouard 11091: free_ivector(wav,1,imx);
11092: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11093: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11094: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11095:
11096:
1.127 brouard 11097: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11098:
1.201 brouard 11099: strcpy(filerese,"E_");
11100: strcat(filerese,fileresu);
1.126 brouard 11101: if((ficreseij=fopen(filerese,"w"))==NULL) {
11102: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11103: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11104: }
1.208 brouard 11105: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11106: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11107:
11108: pstamp(ficreseij);
1.219 brouard 11109:
1.235 brouard 11110: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11111: if (cptcovn < 1){i1=1;}
11112:
11113: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11114: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11115: if(TKresult[nres]!= k)
11116: continue;
1.219 brouard 11117: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11118: printf("\n#****** ");
1.225 brouard 11119: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11120: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11121: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11122: }
11123: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11124: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11125: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11126: }
11127: fprintf(ficreseij,"******\n");
1.235 brouard 11128: printf("******\n");
1.219 brouard 11129:
11130: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11131: oldm=oldms;savm=savms;
1.235 brouard 11132: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11133:
1.219 brouard 11134: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11135: }
11136: fclose(ficreseij);
1.208 brouard 11137: printf("done evsij\n");fflush(stdout);
11138: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11139:
1.227 brouard 11140: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11141:
11142:
1.201 brouard 11143: strcpy(filerest,"T_");
11144: strcat(filerest,fileresu);
1.127 brouard 11145: if((ficrest=fopen(filerest,"w"))==NULL) {
11146: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11147: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11148: }
1.208 brouard 11149: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11150: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11151:
1.126 brouard 11152:
1.201 brouard 11153: strcpy(fileresstde,"STDE_");
11154: strcat(fileresstde,fileresu);
1.126 brouard 11155: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11156: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11157: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11158: }
1.227 brouard 11159: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11160: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11161:
1.201 brouard 11162: strcpy(filerescve,"CVE_");
11163: strcat(filerescve,fileresu);
1.126 brouard 11164: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11165: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11166: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11167: }
1.227 brouard 11168: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11169: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11170:
1.201 brouard 11171: strcpy(fileresv,"V_");
11172: strcat(fileresv,fileresu);
1.126 brouard 11173: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11174: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11175: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11176: }
1.227 brouard 11177: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11178: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11179:
1.145 brouard 11180: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11181: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11182:
1.235 brouard 11183: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11184: if (cptcovn < 1){i1=1;}
11185:
11186: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11187: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11188: if(TKresult[nres]!= k)
11189: continue;
1.242 brouard 11190: printf("\n#****** Result for:");
11191: fprintf(ficrest,"\n#****** Result for:");
11192: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11193: for(j=1;j<=cptcoveff;j++){
11194: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11195: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11196: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11197: }
1.235 brouard 11198: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11199: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11200: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11201: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11202: }
1.208 brouard 11203: fprintf(ficrest,"******\n");
1.227 brouard 11204: fprintf(ficlog,"******\n");
11205: printf("******\n");
1.208 brouard 11206:
11207: fprintf(ficresstdeij,"\n#****** ");
11208: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11209: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11210: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11211: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11212: }
1.235 brouard 11213: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11214: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11215: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11216: }
1.208 brouard 11217: fprintf(ficresstdeij,"******\n");
11218: fprintf(ficrescveij,"******\n");
11219:
11220: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11221: /* pstamp(ficresvij); */
1.225 brouard 11222: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11223: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11224: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11225: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11226: }
1.208 brouard 11227: fprintf(ficresvij,"******\n");
11228:
11229: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11230: oldm=oldms;savm=savms;
1.235 brouard 11231: printf(" cvevsij ");
11232: fprintf(ficlog, " cvevsij ");
11233: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11234: printf(" end cvevsij \n ");
11235: fprintf(ficlog, " end cvevsij \n ");
11236:
11237: /*
11238: */
11239: /* goto endfree; */
11240:
11241: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11242: pstamp(ficrest);
11243:
11244:
11245: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11246: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11247: cptcod= 0; /* To be deleted */
11248: printf("varevsij vpopbased=%d \n",vpopbased);
11249: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11250: 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 11251: 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 ");
11252: if(vpopbased==1)
11253: 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);
11254: else
11255: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11256: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11257: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11258: fprintf(ficrest,"\n");
11259: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11260: epj=vector(1,nlstate+1);
11261: printf("Computing age specific period (stable) prevalences in each health state \n");
11262: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11263: for(age=bage; age <=fage ;age++){
1.235 brouard 11264: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11265: if (vpopbased==1) {
11266: if(mobilav ==0){
11267: for(i=1; i<=nlstate;i++)
11268: prlim[i][i]=probs[(int)age][i][k];
11269: }else{ /* mobilav */
11270: for(i=1; i<=nlstate;i++)
11271: prlim[i][i]=mobaverage[(int)age][i][k];
11272: }
11273: }
1.219 brouard 11274:
1.227 brouard 11275: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11276: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11277: /* printf(" age %4.0f ",age); */
11278: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11279: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11280: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11281: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11282: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11283: }
11284: epj[nlstate+1] +=epj[j];
11285: }
11286: /* printf(" age %4.0f \n",age); */
1.219 brouard 11287:
1.227 brouard 11288: for(i=1, vepp=0.;i <=nlstate;i++)
11289: for(j=1;j <=nlstate;j++)
11290: vepp += vareij[i][j][(int)age];
11291: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11292: for(j=1;j <=nlstate;j++){
11293: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11294: }
11295: fprintf(ficrest,"\n");
11296: }
1.208 brouard 11297: } /* End vpopbased */
11298: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11299: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11300: free_vector(epj,1,nlstate+1);
1.235 brouard 11301: printf("done selection\n");fflush(stdout);
11302: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11303:
1.145 brouard 11304: /*}*/
1.235 brouard 11305: } /* End k selection */
1.227 brouard 11306:
11307: printf("done State-specific expectancies\n");fflush(stdout);
11308: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11309:
1.126 brouard 11310: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11311:
1.201 brouard 11312: strcpy(fileresvpl,"VPL_");
11313: strcat(fileresvpl,fileresu);
1.126 brouard 11314: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11315: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11316: exit(0);
11317: }
1.208 brouard 11318: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11319: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11320:
1.145 brouard 11321: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11322: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11323:
1.235 brouard 11324: i1=pow(2,cptcoveff);
11325: if (cptcovn < 1){i1=1;}
11326:
11327: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11328: for(k=1; k<=i1;k++){
11329: if(TKresult[nres]!= k)
11330: continue;
1.227 brouard 11331: fprintf(ficresvpl,"\n#****** ");
11332: printf("\n#****** ");
11333: fprintf(ficlog,"\n#****** ");
11334: for(j=1;j<=cptcoveff;j++) {
11335: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11336: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11337: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11338: }
1.235 brouard 11339: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11340: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11341: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11342: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11343: }
1.227 brouard 11344: fprintf(ficresvpl,"******\n");
11345: printf("******\n");
11346: fprintf(ficlog,"******\n");
11347:
11348: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11349: oldm=oldms;savm=savms;
1.235 brouard 11350: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11351: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11352: /*}*/
1.126 brouard 11353: }
1.227 brouard 11354:
1.126 brouard 11355: fclose(ficresvpl);
1.208 brouard 11356: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11357: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11358:
11359: free_vector(weight,1,n);
11360: free_imatrix(Tvard,1,NCOVMAX,1,2);
11361: free_imatrix(s,1,maxwav+1,1,n);
11362: free_matrix(anint,1,maxwav,1,n);
11363: free_matrix(mint,1,maxwav,1,n);
11364: free_ivector(cod,1,n);
11365: free_ivector(tab,1,NCOVMAX);
11366: fclose(ficresstdeij);
11367: fclose(ficrescveij);
11368: fclose(ficresvij);
11369: fclose(ficrest);
11370: fclose(ficpar);
11371:
11372:
1.126 brouard 11373: /*---------- End : free ----------------*/
1.219 brouard 11374: if (mobilav!=0 ||mobilavproj !=0)
11375: free_ma3x(mobaverages,1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
1.218 brouard 11376: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11377: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11378: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11379: } /* mle==-3 arrives here for freeing */
1.227 brouard 11380: /* endfree:*/
11381: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11382: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11383: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11384: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11385: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11386: free_matrix(coqvar,1,maxwav,1,n);
11387: free_matrix(covar,0,NCOVMAX,1,n);
11388: free_matrix(matcov,1,npar,1,npar);
11389: free_matrix(hess,1,npar,1,npar);
11390: /*free_vector(delti,1,npar);*/
11391: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11392: free_matrix(agev,1,maxwav,1,imx);
11393: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11394:
11395: free_ivector(ncodemax,1,NCOVMAX);
11396: free_ivector(ncodemaxwundef,1,NCOVMAX);
11397: free_ivector(Dummy,-1,NCOVMAX);
11398: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11399: free_ivector(DummyV,1,NCOVMAX);
11400: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11401: free_ivector(Typevar,-1,NCOVMAX);
11402: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11403: free_ivector(TvarsQ,1,NCOVMAX);
11404: free_ivector(TvarsQind,1,NCOVMAX);
11405: free_ivector(TvarsD,1,NCOVMAX);
11406: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11407: free_ivector(TvarFD,1,NCOVMAX);
11408: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11409: free_ivector(TvarF,1,NCOVMAX);
11410: free_ivector(TvarFind,1,NCOVMAX);
11411: free_ivector(TvarV,1,NCOVMAX);
11412: free_ivector(TvarVind,1,NCOVMAX);
11413: free_ivector(TvarA,1,NCOVMAX);
11414: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11415: free_ivector(TvarFQ,1,NCOVMAX);
11416: free_ivector(TvarFQind,1,NCOVMAX);
11417: free_ivector(TvarVD,1,NCOVMAX);
11418: free_ivector(TvarVDind,1,NCOVMAX);
11419: free_ivector(TvarVQ,1,NCOVMAX);
11420: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11421: free_ivector(Tvarsel,1,NCOVMAX);
11422: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11423: free_ivector(Tposprod,1,NCOVMAX);
11424: free_ivector(Tprod,1,NCOVMAX);
11425: free_ivector(Tvaraff,1,NCOVMAX);
11426: free_ivector(invalidvarcomb,1,ncovcombmax);
11427: free_ivector(Tage,1,NCOVMAX);
11428: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11429: free_ivector(TmodelInvind,1,NCOVMAX);
11430: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11431:
11432: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11433: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11434: fflush(fichtm);
11435: fflush(ficgp);
11436:
1.227 brouard 11437:
1.126 brouard 11438: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11439: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11440: 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 11441: }else{
11442: printf("End of Imach\n");
11443: fprintf(ficlog,"End of Imach\n");
11444: }
11445: printf("See log file on %s\n",filelog);
11446: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11447: /*(void) gettimeofday(&end_time,&tzp);*/
11448: rend_time = time(NULL);
11449: end_time = *localtime(&rend_time);
11450: /* tml = *localtime(&end_time.tm_sec); */
11451: strcpy(strtend,asctime(&end_time));
1.126 brouard 11452: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11453: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11454: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11455:
1.157 brouard 11456: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11457: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11458: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11459: /* printf("Total time was %d uSec.\n", total_usecs);*/
11460: /* if(fileappend(fichtm,optionfilehtm)){ */
11461: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11462: fclose(fichtm);
11463: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11464: fclose(fichtmcov);
11465: fclose(ficgp);
11466: fclose(ficlog);
11467: /*------ End -----------*/
1.227 brouard 11468:
11469:
11470: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11471: #ifdef WIN32
1.227 brouard 11472: if (_chdir(pathcd) != 0)
11473: printf("Can't move to directory %s!\n",path);
11474: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11475: #else
1.227 brouard 11476: if(chdir(pathcd) != 0)
11477: printf("Can't move to directory %s!\n", path);
11478: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11479: #endif
1.126 brouard 11480: printf("Current directory %s!\n",pathcd);
11481: /*strcat(plotcmd,CHARSEPARATOR);*/
11482: sprintf(plotcmd,"gnuplot");
1.157 brouard 11483: #ifdef _WIN32
1.126 brouard 11484: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11485: #endif
11486: if(!stat(plotcmd,&info)){
1.158 brouard 11487: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11488: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11489: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11490: }else
11491: strcpy(pplotcmd,plotcmd);
1.157 brouard 11492: #ifdef __unix
1.126 brouard 11493: strcpy(plotcmd,GNUPLOTPROGRAM);
11494: if(!stat(plotcmd,&info)){
1.158 brouard 11495: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11496: }else
11497: strcpy(pplotcmd,plotcmd);
11498: #endif
11499: }else
11500: strcpy(pplotcmd,plotcmd);
11501:
11502: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11503: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11504:
1.126 brouard 11505: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11506: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11507: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11508: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11509: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11510: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11511: }
1.158 brouard 11512: printf(" Successful, please wait...");
1.126 brouard 11513: while (z[0] != 'q') {
11514: /* chdir(path); */
1.154 brouard 11515: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11516: scanf("%s",z);
11517: /* if (z[0] == 'c') system("./imach"); */
11518: if (z[0] == 'e') {
1.158 brouard 11519: #ifdef __APPLE__
1.152 brouard 11520: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11521: #elif __linux
11522: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11523: #else
1.152 brouard 11524: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11525: #endif
11526: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11527: system(pplotcmd);
1.126 brouard 11528: }
11529: else if (z[0] == 'g') system(plotcmd);
11530: else if (z[0] == 'q') exit(0);
11531: }
1.227 brouard 11532: end:
1.126 brouard 11533: while (z[0] != 'q') {
1.195 brouard 11534: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11535: scanf("%s",z);
11536: }
11537: }
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