Annotation of imach/src/imach.c, revision 1.246
1.246 ! brouard 1: /* $Id: imach.c,v 1.245 2016/09/02 07:25:01 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.246 ! brouard 4: Revision 1.245 2016/09/02 07:25:01 brouard
! 5: *** empty log message ***
! 6:
1.245 brouard 7: Revision 1.244 2016/09/02 07:17:34 brouard
8: *** empty log message ***
9:
1.244 brouard 10: Revision 1.243 2016/09/02 06:45:35 brouard
11: *** empty log message ***
12:
1.243 brouard 13: Revision 1.242 2016/08/30 15:01:20 brouard
14: Summary: Fixing a lots
15:
1.242 brouard 16: Revision 1.241 2016/08/29 17:17:25 brouard
17: Summary: gnuplot problem in Back projection to fix
18:
1.241 brouard 19: Revision 1.240 2016/08/29 07:53:18 brouard
20: Summary: Better
21:
1.240 brouard 22: Revision 1.239 2016/08/26 15:51:03 brouard
23: Summary: Improvement in Powell output in order to copy and paste
24:
25: Author:
26:
1.239 brouard 27: Revision 1.238 2016/08/26 14:23:35 brouard
28: Summary: Starting tests of 0.99
29:
1.238 brouard 30: Revision 1.237 2016/08/26 09:20:19 brouard
31: Summary: to valgrind
32:
1.237 brouard 33: Revision 1.236 2016/08/25 10:50:18 brouard
34: *** empty log message ***
35:
1.236 brouard 36: Revision 1.235 2016/08/25 06:59:23 brouard
37: *** empty log message ***
38:
1.235 brouard 39: Revision 1.234 2016/08/23 16:51:20 brouard
40: *** empty log message ***
41:
1.234 brouard 42: Revision 1.233 2016/08/23 07:40:50 brouard
43: Summary: not working
44:
1.233 brouard 45: Revision 1.232 2016/08/22 14:20:21 brouard
46: Summary: not working
47:
1.232 brouard 48: Revision 1.231 2016/08/22 07:17:15 brouard
49: Summary: not working
50:
1.231 brouard 51: Revision 1.230 2016/08/22 06:55:53 brouard
52: Summary: Not working
53:
1.230 brouard 54: Revision 1.229 2016/07/23 09:45:53 brouard
55: Summary: Completing for func too
56:
1.229 brouard 57: Revision 1.228 2016/07/22 17:45:30 brouard
58: Summary: Fixing some arrays, still debugging
59:
1.227 brouard 60: Revision 1.226 2016/07/12 18:42:34 brouard
61: Summary: temp
62:
1.226 brouard 63: Revision 1.225 2016/07/12 08:40:03 brouard
64: Summary: saving but not running
65:
1.225 brouard 66: Revision 1.224 2016/07/01 13:16:01 brouard
67: Summary: Fixes
68:
1.224 brouard 69: Revision 1.223 2016/02/19 09:23:35 brouard
70: Summary: temporary
71:
1.223 brouard 72: Revision 1.222 2016/02/17 08:14:50 brouard
73: Summary: Probably last 0.98 stable version 0.98r6
74:
1.222 brouard 75: Revision 1.221 2016/02/15 23:35:36 brouard
76: Summary: minor bug
77:
1.220 brouard 78: Revision 1.219 2016/02/15 00:48:12 brouard
79: *** empty log message ***
80:
1.219 brouard 81: Revision 1.218 2016/02/12 11:29:23 brouard
82: Summary: 0.99 Back projections
83:
1.218 brouard 84: Revision 1.217 2015/12/23 17:18:31 brouard
85: Summary: Experimental backcast
86:
1.217 brouard 87: Revision 1.216 2015/12/18 17:32:11 brouard
88: Summary: 0.98r4 Warning and status=-2
89:
90: Version 0.98r4 is now:
91: - displaying an error when status is -1, date of interview unknown and date of death known;
92: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
93: Older changes concerning s=-2, dating from 2005 have been supersed.
94:
1.216 brouard 95: Revision 1.215 2015/12/16 08:52:24 brouard
96: Summary: 0.98r4 working
97:
1.215 brouard 98: Revision 1.214 2015/12/16 06:57:54 brouard
99: Summary: temporary not working
100:
1.214 brouard 101: Revision 1.213 2015/12/11 18:22:17 brouard
102: Summary: 0.98r4
103:
1.213 brouard 104: Revision 1.212 2015/11/21 12:47:24 brouard
105: Summary: minor typo
106:
1.212 brouard 107: Revision 1.211 2015/11/21 12:41:11 brouard
108: Summary: 0.98r3 with some graph of projected cross-sectional
109:
110: Author: Nicolas Brouard
111:
1.211 brouard 112: Revision 1.210 2015/11/18 17:41:20 brouard
113: Summary: Start working on projected prevalences
114:
1.210 brouard 115: Revision 1.209 2015/11/17 22:12:03 brouard
116: Summary: Adding ftolpl parameter
117: Author: N Brouard
118:
119: We had difficulties to get smoothed confidence intervals. It was due
120: to the period prevalence which wasn't computed accurately. The inner
121: parameter ftolpl is now an outer parameter of the .imach parameter
122: file after estepm. If ftolpl is small 1.e-4 and estepm too,
123: computation are long.
124:
1.209 brouard 125: Revision 1.208 2015/11/17 14:31:57 brouard
126: Summary: temporary
127:
1.208 brouard 128: Revision 1.207 2015/10/27 17:36:57 brouard
129: *** empty log message ***
130:
1.207 brouard 131: Revision 1.206 2015/10/24 07:14:11 brouard
132: *** empty log message ***
133:
1.206 brouard 134: Revision 1.205 2015/10/23 15:50:53 brouard
135: Summary: 0.98r3 some clarification for graphs on likelihood contributions
136:
1.205 brouard 137: Revision 1.204 2015/10/01 16:20:26 brouard
138: Summary: Some new graphs of contribution to likelihood
139:
1.204 brouard 140: Revision 1.203 2015/09/30 17:45:14 brouard
141: Summary: looking at better estimation of the hessian
142:
143: Also a better criteria for convergence to the period prevalence And
144: therefore adding the number of years needed to converge. (The
145: prevalence in any alive state shold sum to one
146:
1.203 brouard 147: Revision 1.202 2015/09/22 19:45:16 brouard
148: Summary: Adding some overall graph on contribution to likelihood. Might change
149:
1.202 brouard 150: Revision 1.201 2015/09/15 17:34:58 brouard
151: Summary: 0.98r0
152:
153: - Some new graphs like suvival functions
154: - Some bugs fixed like model=1+age+V2.
155:
1.201 brouard 156: Revision 1.200 2015/09/09 16:53:55 brouard
157: Summary: Big bug thanks to Flavia
158:
159: Even model=1+age+V2. did not work anymore
160:
1.200 brouard 161: Revision 1.199 2015/09/07 14:09:23 brouard
162: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
163:
1.199 brouard 164: Revision 1.198 2015/09/03 07:14:39 brouard
165: Summary: 0.98q5 Flavia
166:
1.198 brouard 167: Revision 1.197 2015/09/01 18:24:39 brouard
168: *** empty log message ***
169:
1.197 brouard 170: Revision 1.196 2015/08/18 23:17:52 brouard
171: Summary: 0.98q5
172:
1.196 brouard 173: Revision 1.195 2015/08/18 16:28:39 brouard
174: Summary: Adding a hack for testing purpose
175:
176: After reading the title, ftol and model lines, if the comment line has
177: a q, starting with #q, the answer at the end of the run is quit. It
178: permits to run test files in batch with ctest. The former workaround was
179: $ echo q | imach foo.imach
180:
1.195 brouard 181: Revision 1.194 2015/08/18 13:32:00 brouard
182: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
183:
1.194 brouard 184: Revision 1.193 2015/08/04 07:17:42 brouard
185: Summary: 0.98q4
186:
1.193 brouard 187: Revision 1.192 2015/07/16 16:49:02 brouard
188: Summary: Fixing some outputs
189:
1.192 brouard 190: Revision 1.191 2015/07/14 10:00:33 brouard
191: Summary: Some fixes
192:
1.191 brouard 193: Revision 1.190 2015/05/05 08:51:13 brouard
194: Summary: Adding digits in output parameters (7 digits instead of 6)
195:
196: Fix 1+age+.
197:
1.190 brouard 198: Revision 1.189 2015/04/30 14:45:16 brouard
199: Summary: 0.98q2
200:
1.189 brouard 201: Revision 1.188 2015/04/30 08:27:53 brouard
202: *** empty log message ***
203:
1.188 brouard 204: Revision 1.187 2015/04/29 09:11:15 brouard
205: *** empty log message ***
206:
1.187 brouard 207: Revision 1.186 2015/04/23 12:01:52 brouard
208: Summary: V1*age is working now, version 0.98q1
209:
210: Some codes had been disabled in order to simplify and Vn*age was
211: working in the optimization phase, ie, giving correct MLE parameters,
212: but, as usual, outputs were not correct and program core dumped.
213:
1.186 brouard 214: Revision 1.185 2015/03/11 13:26:42 brouard
215: Summary: Inclusion of compile and links command line for Intel Compiler
216:
1.185 brouard 217: Revision 1.184 2015/03/11 11:52:39 brouard
218: Summary: Back from Windows 8. Intel Compiler
219:
1.184 brouard 220: Revision 1.183 2015/03/10 20:34:32 brouard
221: Summary: 0.98q0, trying with directest, mnbrak fixed
222:
223: We use directest instead of original Powell test; probably no
224: incidence on the results, but better justifications;
225: We fixed Numerical Recipes mnbrak routine which was wrong and gave
226: wrong results.
227:
1.183 brouard 228: Revision 1.182 2015/02/12 08:19:57 brouard
229: Summary: Trying to keep directest which seems simpler and more general
230: Author: Nicolas Brouard
231:
1.182 brouard 232: Revision 1.181 2015/02/11 23:22:24 brouard
233: Summary: Comments on Powell added
234:
235: Author:
236:
1.181 brouard 237: Revision 1.180 2015/02/11 17:33:45 brouard
238: Summary: Finishing move from main to function (hpijx and prevalence_limit)
239:
1.180 brouard 240: Revision 1.179 2015/01/04 09:57:06 brouard
241: Summary: back to OS/X
242:
1.179 brouard 243: Revision 1.178 2015/01/04 09:35:48 brouard
244: *** empty log message ***
245:
1.178 brouard 246: Revision 1.177 2015/01/03 18:40:56 brouard
247: Summary: Still testing ilc32 on OSX
248:
1.177 brouard 249: Revision 1.176 2015/01/03 16:45:04 brouard
250: *** empty log message ***
251:
1.176 brouard 252: Revision 1.175 2015/01/03 16:33:42 brouard
253: *** empty log message ***
254:
1.175 brouard 255: Revision 1.174 2015/01/03 16:15:49 brouard
256: Summary: Still in cross-compilation
257:
1.174 brouard 258: Revision 1.173 2015/01/03 12:06:26 brouard
259: Summary: trying to detect cross-compilation
260:
1.173 brouard 261: Revision 1.172 2014/12/27 12:07:47 brouard
262: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
263:
1.172 brouard 264: Revision 1.171 2014/12/23 13:26:59 brouard
265: Summary: Back from Visual C
266:
267: Still problem with utsname.h on Windows
268:
1.171 brouard 269: Revision 1.170 2014/12/23 11:17:12 brouard
270: Summary: Cleaning some \%% back to %%
271:
272: The escape was mandatory for a specific compiler (which one?), but too many warnings.
273:
1.170 brouard 274: Revision 1.169 2014/12/22 23:08:31 brouard
275: Summary: 0.98p
276:
277: Outputs some informations on compiler used, OS etc. Testing on different platforms.
278:
1.169 brouard 279: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 280: Summary: update
1.169 brouard 281:
1.168 brouard 282: Revision 1.167 2014/12/22 13:50:56 brouard
283: Summary: Testing uname and compiler version and if compiled 32 or 64
284:
285: Testing on Linux 64
286:
1.167 brouard 287: Revision 1.166 2014/12/22 11:40:47 brouard
288: *** empty log message ***
289:
1.166 brouard 290: Revision 1.165 2014/12/16 11:20:36 brouard
291: Summary: After compiling on Visual C
292:
293: * imach.c (Module): Merging 1.61 to 1.162
294:
1.165 brouard 295: Revision 1.164 2014/12/16 10:52:11 brouard
296: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
297:
298: * imach.c (Module): Merging 1.61 to 1.162
299:
1.164 brouard 300: Revision 1.163 2014/12/16 10:30:11 brouard
301: * imach.c (Module): Merging 1.61 to 1.162
302:
1.163 brouard 303: Revision 1.162 2014/09/25 11:43:39 brouard
304: Summary: temporary backup 0.99!
305:
1.162 brouard 306: Revision 1.1 2014/09/16 11:06:58 brouard
307: Summary: With some code (wrong) for nlopt
308:
309: Author:
310:
311: Revision 1.161 2014/09/15 20:41:41 brouard
312: Summary: Problem with macro SQR on Intel compiler
313:
1.161 brouard 314: Revision 1.160 2014/09/02 09:24:05 brouard
315: *** empty log message ***
316:
1.160 brouard 317: Revision 1.159 2014/09/01 10:34:10 brouard
318: Summary: WIN32
319: Author: Brouard
320:
1.159 brouard 321: Revision 1.158 2014/08/27 17:11:51 brouard
322: *** empty log message ***
323:
1.158 brouard 324: Revision 1.157 2014/08/27 16:26:55 brouard
325: Summary: Preparing windows Visual studio version
326: Author: Brouard
327:
328: In order to compile on Visual studio, time.h is now correct and time_t
329: and tm struct should be used. difftime should be used but sometimes I
330: just make the differences in raw time format (time(&now).
331: Trying to suppress #ifdef LINUX
332: Add xdg-open for __linux in order to open default browser.
333:
1.157 brouard 334: Revision 1.156 2014/08/25 20:10:10 brouard
335: *** empty log message ***
336:
1.156 brouard 337: Revision 1.155 2014/08/25 18:32:34 brouard
338: Summary: New compile, minor changes
339: Author: Brouard
340:
1.155 brouard 341: Revision 1.154 2014/06/20 17:32:08 brouard
342: Summary: Outputs now all graphs of convergence to period prevalence
343:
1.154 brouard 344: Revision 1.153 2014/06/20 16:45:46 brouard
345: Summary: If 3 live state, convergence to period prevalence on same graph
346: Author: Brouard
347:
1.153 brouard 348: Revision 1.152 2014/06/18 17:54:09 brouard
349: Summary: open browser, use gnuplot on same dir than imach if not found in the path
350:
1.152 brouard 351: Revision 1.151 2014/06/18 16:43:30 brouard
352: *** empty log message ***
353:
1.151 brouard 354: Revision 1.150 2014/06/18 16:42:35 brouard
355: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
356: Author: brouard
357:
1.150 brouard 358: Revision 1.149 2014/06/18 15:51:14 brouard
359: Summary: Some fixes in parameter files errors
360: Author: Nicolas Brouard
361:
1.149 brouard 362: Revision 1.148 2014/06/17 17:38:48 brouard
363: Summary: Nothing new
364: Author: Brouard
365:
366: Just a new packaging for OS/X version 0.98nS
367:
1.148 brouard 368: Revision 1.147 2014/06/16 10:33:11 brouard
369: *** empty log message ***
370:
1.147 brouard 371: Revision 1.146 2014/06/16 10:20:28 brouard
372: Summary: Merge
373: Author: Brouard
374:
375: Merge, before building revised version.
376:
1.146 brouard 377: Revision 1.145 2014/06/10 21:23:15 brouard
378: Summary: Debugging with valgrind
379: Author: Nicolas Brouard
380:
381: Lot of changes in order to output the results with some covariates
382: After the Edimburgh REVES conference 2014, it seems mandatory to
383: improve the code.
384: No more memory valgrind error but a lot has to be done in order to
385: continue the work of splitting the code into subroutines.
386: Also, decodemodel has been improved. Tricode is still not
387: optimal. nbcode should be improved. Documentation has been added in
388: the source code.
389:
1.144 brouard 390: Revision 1.143 2014/01/26 09:45:38 brouard
391: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
392:
393: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
394: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
395:
1.143 brouard 396: Revision 1.142 2014/01/26 03:57:36 brouard
397: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
398:
399: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
400:
1.142 brouard 401: Revision 1.141 2014/01/26 02:42:01 brouard
402: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
403:
1.141 brouard 404: Revision 1.140 2011/09/02 10:37:54 brouard
405: Summary: times.h is ok with mingw32 now.
406:
1.140 brouard 407: Revision 1.139 2010/06/14 07:50:17 brouard
408: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
409: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
410:
1.139 brouard 411: Revision 1.138 2010/04/30 18:19:40 brouard
412: *** empty log message ***
413:
1.138 brouard 414: Revision 1.137 2010/04/29 18:11:38 brouard
415: (Module): Checking covariates for more complex models
416: than V1+V2. A lot of change to be done. Unstable.
417:
1.137 brouard 418: Revision 1.136 2010/04/26 20:30:53 brouard
419: (Module): merging some libgsl code. Fixing computation
420: of likelione (using inter/intrapolation if mle = 0) in order to
421: get same likelihood as if mle=1.
422: Some cleaning of code and comments added.
423:
1.136 brouard 424: Revision 1.135 2009/10/29 15:33:14 brouard
425: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
426:
1.135 brouard 427: Revision 1.134 2009/10/29 13:18:53 brouard
428: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
429:
1.134 brouard 430: Revision 1.133 2009/07/06 10:21:25 brouard
431: just nforces
432:
1.133 brouard 433: Revision 1.132 2009/07/06 08:22:05 brouard
434: Many tings
435:
1.132 brouard 436: Revision 1.131 2009/06/20 16:22:47 brouard
437: Some dimensions resccaled
438:
1.131 brouard 439: Revision 1.130 2009/05/26 06:44:34 brouard
440: (Module): Max Covariate is now set to 20 instead of 8. A
441: lot of cleaning with variables initialized to 0. Trying to make
442: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
443:
1.130 brouard 444: Revision 1.129 2007/08/31 13:49:27 lievre
445: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
446:
1.129 lievre 447: Revision 1.128 2006/06/30 13:02:05 brouard
448: (Module): Clarifications on computing e.j
449:
1.128 brouard 450: Revision 1.127 2006/04/28 18:11:50 brouard
451: (Module): Yes the sum of survivors was wrong since
452: imach-114 because nhstepm was no more computed in the age
453: loop. Now we define nhstepma in the age loop.
454: (Module): In order to speed up (in case of numerous covariates) we
455: compute health expectancies (without variances) in a first step
456: and then all the health expectancies with variances or standard
457: deviation (needs data from the Hessian matrices) which slows the
458: computation.
459: In the future we should be able to stop the program is only health
460: expectancies and graph are needed without standard deviations.
461:
1.127 brouard 462: Revision 1.126 2006/04/28 17:23:28 brouard
463: (Module): Yes the sum of survivors was wrong since
464: imach-114 because nhstepm was no more computed in the age
465: loop. Now we define nhstepma in the age loop.
466: Version 0.98h
467:
1.126 brouard 468: Revision 1.125 2006/04/04 15:20:31 lievre
469: Errors in calculation of health expectancies. Age was not initialized.
470: Forecasting file added.
471:
472: Revision 1.124 2006/03/22 17:13:53 lievre
473: Parameters are printed with %lf instead of %f (more numbers after the comma).
474: The log-likelihood is printed in the log file
475:
476: Revision 1.123 2006/03/20 10:52:43 brouard
477: * imach.c (Module): <title> changed, corresponds to .htm file
478: name. <head> headers where missing.
479:
480: * imach.c (Module): Weights can have a decimal point as for
481: English (a comma might work with a correct LC_NUMERIC environment,
482: otherwise the weight is truncated).
483: Modification of warning when the covariates values are not 0 or
484: 1.
485: Version 0.98g
486:
487: Revision 1.122 2006/03/20 09:45:41 brouard
488: (Module): Weights can have a decimal point as for
489: English (a comma might work with a correct LC_NUMERIC environment,
490: otherwise the weight is truncated).
491: Modification of warning when the covariates values are not 0 or
492: 1.
493: Version 0.98g
494:
495: Revision 1.121 2006/03/16 17:45:01 lievre
496: * imach.c (Module): Comments concerning covariates added
497:
498: * imach.c (Module): refinements in the computation of lli if
499: status=-2 in order to have more reliable computation if stepm is
500: not 1 month. Version 0.98f
501:
502: Revision 1.120 2006/03/16 15:10:38 lievre
503: (Module): refinements in the computation of lli if
504: status=-2 in order to have more reliable computation if stepm is
505: not 1 month. Version 0.98f
506:
507: Revision 1.119 2006/03/15 17:42:26 brouard
508: (Module): Bug if status = -2, the loglikelihood was
509: computed as likelihood omitting the logarithm. Version O.98e
510:
511: Revision 1.118 2006/03/14 18:20:07 brouard
512: (Module): varevsij Comments added explaining the second
513: table of variances if popbased=1 .
514: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
515: (Module): Function pstamp added
516: (Module): Version 0.98d
517:
518: Revision 1.117 2006/03/14 17:16:22 brouard
519: (Module): varevsij Comments added explaining the second
520: table of variances if popbased=1 .
521: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
522: (Module): Function pstamp added
523: (Module): Version 0.98d
524:
525: Revision 1.116 2006/03/06 10:29:27 brouard
526: (Module): Variance-covariance wrong links and
527: varian-covariance of ej. is needed (Saito).
528:
529: Revision 1.115 2006/02/27 12:17:45 brouard
530: (Module): One freematrix added in mlikeli! 0.98c
531:
532: Revision 1.114 2006/02/26 12:57:58 brouard
533: (Module): Some improvements in processing parameter
534: filename with strsep.
535:
536: Revision 1.113 2006/02/24 14:20:24 brouard
537: (Module): Memory leaks checks with valgrind and:
538: datafile was not closed, some imatrix were not freed and on matrix
539: allocation too.
540:
541: Revision 1.112 2006/01/30 09:55:26 brouard
542: (Module): Back to gnuplot.exe instead of wgnuplot.exe
543:
544: Revision 1.111 2006/01/25 20:38:18 brouard
545: (Module): Lots of cleaning and bugs added (Gompertz)
546: (Module): Comments can be added in data file. Missing date values
547: can be a simple dot '.'.
548:
549: Revision 1.110 2006/01/25 00:51:50 brouard
550: (Module): Lots of cleaning and bugs added (Gompertz)
551:
552: Revision 1.109 2006/01/24 19:37:15 brouard
553: (Module): Comments (lines starting with a #) are allowed in data.
554:
555: Revision 1.108 2006/01/19 18:05:42 lievre
556: Gnuplot problem appeared...
557: To be fixed
558:
559: Revision 1.107 2006/01/19 16:20:37 brouard
560: Test existence of gnuplot in imach path
561:
562: Revision 1.106 2006/01/19 13:24:36 brouard
563: Some cleaning and links added in html output
564:
565: Revision 1.105 2006/01/05 20:23:19 lievre
566: *** empty log message ***
567:
568: Revision 1.104 2005/09/30 16:11:43 lievre
569: (Module): sump fixed, loop imx fixed, and simplifications.
570: (Module): If the status is missing at the last wave but we know
571: that the person is alive, then we can code his/her status as -2
572: (instead of missing=-1 in earlier versions) and his/her
573: contributions to the likelihood is 1 - Prob of dying from last
574: health status (= 1-p13= p11+p12 in the easiest case of somebody in
575: the healthy state at last known wave). Version is 0.98
576:
577: Revision 1.103 2005/09/30 15:54:49 lievre
578: (Module): sump fixed, loop imx fixed, and simplifications.
579:
580: Revision 1.102 2004/09/15 17:31:30 brouard
581: Add the possibility to read data file including tab characters.
582:
583: Revision 1.101 2004/09/15 10:38:38 brouard
584: Fix on curr_time
585:
586: Revision 1.100 2004/07/12 18:29:06 brouard
587: Add version for Mac OS X. Just define UNIX in Makefile
588:
589: Revision 1.99 2004/06/05 08:57:40 brouard
590: *** empty log message ***
591:
592: Revision 1.98 2004/05/16 15:05:56 brouard
593: New version 0.97 . First attempt to estimate force of mortality
594: directly from the data i.e. without the need of knowing the health
595: state at each age, but using a Gompertz model: log u =a + b*age .
596: This is the basic analysis of mortality and should be done before any
597: other analysis, in order to test if the mortality estimated from the
598: cross-longitudinal survey is different from the mortality estimated
599: from other sources like vital statistic data.
600:
601: The same imach parameter file can be used but the option for mle should be -3.
602:
1.133 brouard 603: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 604: former routines in order to include the new code within the former code.
605:
606: The output is very simple: only an estimate of the intercept and of
607: the slope with 95% confident intervals.
608:
609: Current limitations:
610: A) Even if you enter covariates, i.e. with the
611: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
612: B) There is no computation of Life Expectancy nor Life Table.
613:
614: Revision 1.97 2004/02/20 13:25:42 lievre
615: Version 0.96d. Population forecasting command line is (temporarily)
616: suppressed.
617:
618: Revision 1.96 2003/07/15 15:38:55 brouard
619: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
620: rewritten within the same printf. Workaround: many printfs.
621:
622: Revision 1.95 2003/07/08 07:54:34 brouard
623: * imach.c (Repository):
624: (Repository): Using imachwizard code to output a more meaningful covariance
625: matrix (cov(a12,c31) instead of numbers.
626:
627: Revision 1.94 2003/06/27 13:00:02 brouard
628: Just cleaning
629:
630: Revision 1.93 2003/06/25 16:33:55 brouard
631: (Module): On windows (cygwin) function asctime_r doesn't
632: exist so I changed back to asctime which exists.
633: (Module): Version 0.96b
634:
635: Revision 1.92 2003/06/25 16:30:45 brouard
636: (Module): On windows (cygwin) function asctime_r doesn't
637: exist so I changed back to asctime which exists.
638:
639: Revision 1.91 2003/06/25 15:30:29 brouard
640: * imach.c (Repository): Duplicated warning errors corrected.
641: (Repository): Elapsed time after each iteration is now output. It
642: helps to forecast when convergence will be reached. Elapsed time
643: is stamped in powell. We created a new html file for the graphs
644: concerning matrix of covariance. It has extension -cov.htm.
645:
646: Revision 1.90 2003/06/24 12:34:15 brouard
647: (Module): Some bugs corrected for windows. Also, when
648: mle=-1 a template is output in file "or"mypar.txt with the design
649: of the covariance matrix to be input.
650:
651: Revision 1.89 2003/06/24 12:30:52 brouard
652: (Module): Some bugs corrected for windows. Also, when
653: mle=-1 a template is output in file "or"mypar.txt with the design
654: of the covariance matrix to be input.
655:
656: Revision 1.88 2003/06/23 17:54:56 brouard
657: * 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.
658:
659: Revision 1.87 2003/06/18 12:26:01 brouard
660: Version 0.96
661:
662: Revision 1.86 2003/06/17 20:04:08 brouard
663: (Module): Change position of html and gnuplot routines and added
664: routine fileappend.
665:
666: Revision 1.85 2003/06/17 13:12:43 brouard
667: * imach.c (Repository): Check when date of death was earlier that
668: current date of interview. It may happen when the death was just
669: prior to the death. In this case, dh was negative and likelihood
670: was wrong (infinity). We still send an "Error" but patch by
671: assuming that the date of death was just one stepm after the
672: interview.
673: (Repository): Because some people have very long ID (first column)
674: we changed int to long in num[] and we added a new lvector for
675: memory allocation. But we also truncated to 8 characters (left
676: truncation)
677: (Repository): No more line truncation errors.
678:
679: Revision 1.84 2003/06/13 21:44:43 brouard
680: * imach.c (Repository): Replace "freqsummary" at a correct
681: place. It differs from routine "prevalence" which may be called
682: many times. Probs is memory consuming and must be used with
683: parcimony.
684: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
685:
686: Revision 1.83 2003/06/10 13:39:11 lievre
687: *** empty log message ***
688:
689: Revision 1.82 2003/06/05 15:57:20 brouard
690: Add log in imach.c and fullversion number is now printed.
691:
692: */
693: /*
694: Interpolated Markov Chain
695:
696: Short summary of the programme:
697:
1.227 brouard 698: This program computes Healthy Life Expectancies or State-specific
699: (if states aren't health statuses) Expectancies from
700: cross-longitudinal data. Cross-longitudinal data consist in:
701:
702: -1- a first survey ("cross") where individuals from different ages
703: are interviewed on their health status or degree of disability (in
704: the case of a health survey which is our main interest)
705:
706: -2- at least a second wave of interviews ("longitudinal") which
707: measure each change (if any) in individual health status. Health
708: expectancies are computed from the time spent in each health state
709: according to a model. More health states you consider, more time is
710: necessary to reach the Maximum Likelihood of the parameters involved
711: in the model. The simplest model is the multinomial logistic model
712: where pij is the probability to be observed in state j at the second
713: wave conditional to be observed in state i at the first
714: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
715: etc , where 'age' is age and 'sex' is a covariate. If you want to
716: have a more complex model than "constant and age", you should modify
717: the program where the markup *Covariates have to be included here
718: again* invites you to do it. More covariates you add, slower the
1.126 brouard 719: convergence.
720:
721: The advantage of this computer programme, compared to a simple
722: multinomial logistic model, is clear when the delay between waves is not
723: identical for each individual. Also, if a individual missed an
724: intermediate interview, the information is lost, but taken into
725: account using an interpolation or extrapolation.
726:
727: hPijx is the probability to be observed in state i at age x+h
728: conditional to the observed state i at age x. The delay 'h' can be
729: split into an exact number (nh*stepm) of unobserved intermediate
730: states. This elementary transition (by month, quarter,
731: semester or year) is modelled as a multinomial logistic. The hPx
732: matrix is simply the matrix product of nh*stepm elementary matrices
733: and the contribution of each individual to the likelihood is simply
734: hPijx.
735:
736: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 737: of the life expectancies. It also computes the period (stable) prevalence.
738:
739: Back prevalence and projections:
1.227 brouard 740:
741: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
742: double agemaxpar, double ftolpl, int *ncvyearp, double
743: dateprev1,double dateprev2, int firstpass, int lastpass, int
744: mobilavproj)
745:
746: Computes the back prevalence limit for any combination of
747: covariate values k at any age between ageminpar and agemaxpar and
748: returns it in **bprlim. In the loops,
749:
750: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
751: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
752:
753: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 754: Computes for any combination of covariates k and any age between bage and fage
755: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
756: oldm=oldms;savm=savms;
1.227 brouard 757:
758: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 759: Computes the transition matrix starting at age 'age' over
760: 'nhstepm*hstepm*stepm' months (i.e. until
761: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 762: nhstepm*hstepm matrices.
763:
764: Returns p3mat[i][j][h] after calling
765: p3mat[i][j][h]=matprod2(newm,
766: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
767: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
768: oldm);
1.226 brouard 769:
770: Important routines
771:
772: - func (or funcone), computes logit (pij) distinguishing
773: o fixed variables (single or product dummies or quantitative);
774: o varying variables by:
775: (1) wave (single, product dummies, quantitative),
776: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
777: % fixed dummy (treated) or quantitative (not done because time-consuming);
778: % varying dummy (not done) or quantitative (not done);
779: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
780: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
781: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
782: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
783: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 784:
1.226 brouard 785:
786:
1.133 brouard 787: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
788: Institut national d'études démographiques, Paris.
1.126 brouard 789: This software have been partly granted by Euro-REVES, a concerted action
790: from the European Union.
791: It is copyrighted identically to a GNU software product, ie programme and
792: software can be distributed freely for non commercial use. Latest version
793: can be accessed at http://euroreves.ined.fr/imach .
794:
795: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
796: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
797:
798: **********************************************************************/
799: /*
800: main
801: read parameterfile
802: read datafile
803: concatwav
804: freqsummary
805: if (mle >= 1)
806: mlikeli
807: print results files
808: if mle==1
809: computes hessian
810: read end of parameter file: agemin, agemax, bage, fage, estepm
811: begin-prev-date,...
812: open gnuplot file
813: open html file
1.145 brouard 814: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
815: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
816: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
817: freexexit2 possible for memory heap.
818:
819: h Pij x | pij_nom ficrestpij
820: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
821: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
822: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
823:
824: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
825: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
826: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
827: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
828: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
829:
1.126 brouard 830: forecasting if prevfcast==1 prevforecast call prevalence()
831: health expectancies
832: Variance-covariance of DFLE
833: prevalence()
834: movingaverage()
835: varevsij()
836: if popbased==1 varevsij(,popbased)
837: total life expectancies
838: Variance of period (stable) prevalence
839: end
840: */
841:
1.187 brouard 842: /* #define DEBUG */
843: /* #define DEBUGBRENT */
1.203 brouard 844: /* #define DEBUGLINMIN */
845: /* #define DEBUGHESS */
846: #define DEBUGHESSIJ
1.224 brouard 847: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 848: #define POWELL /* Instead of NLOPT */
1.224 brouard 849: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 850: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
851: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 852:
853: #include <math.h>
854: #include <stdio.h>
855: #include <stdlib.h>
856: #include <string.h>
1.226 brouard 857: #include <ctype.h>
1.159 brouard 858:
859: #ifdef _WIN32
860: #include <io.h>
1.172 brouard 861: #include <windows.h>
862: #include <tchar.h>
1.159 brouard 863: #else
1.126 brouard 864: #include <unistd.h>
1.159 brouard 865: #endif
1.126 brouard 866:
867: #include <limits.h>
868: #include <sys/types.h>
1.171 brouard 869:
870: #if defined(__GNUC__)
871: #include <sys/utsname.h> /* Doesn't work on Windows */
872: #endif
873:
1.126 brouard 874: #include <sys/stat.h>
875: #include <errno.h>
1.159 brouard 876: /* extern int errno; */
1.126 brouard 877:
1.157 brouard 878: /* #ifdef LINUX */
879: /* #include <time.h> */
880: /* #include "timeval.h" */
881: /* #else */
882: /* #include <sys/time.h> */
883: /* #endif */
884:
1.126 brouard 885: #include <time.h>
886:
1.136 brouard 887: #ifdef GSL
888: #include <gsl/gsl_errno.h>
889: #include <gsl/gsl_multimin.h>
890: #endif
891:
1.167 brouard 892:
1.162 brouard 893: #ifdef NLOPT
894: #include <nlopt.h>
895: typedef struct {
896: double (* function)(double [] );
897: } myfunc_data ;
898: #endif
899:
1.126 brouard 900: /* #include <libintl.h> */
901: /* #define _(String) gettext (String) */
902:
1.141 brouard 903: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 904:
905: #define GNUPLOTPROGRAM "gnuplot"
906: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
907: #define FILENAMELENGTH 132
908:
909: #define GLOCK_ERROR_NOPATH -1 /* empty path */
910: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
911:
1.144 brouard 912: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
913: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 914:
915: #define NINTERVMAX 8
1.144 brouard 916: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
917: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
918: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 919: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 920: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
921: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 922: #define MAXN 20000
1.144 brouard 923: #define YEARM 12. /**< Number of months per year */
1.218 brouard 924: /* #define AGESUP 130 */
925: #define AGESUP 150
926: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 927: #define AGEBASE 40
1.194 brouard 928: #define AGEOVERFLOW 1.e20
1.164 brouard 929: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 930: #ifdef _WIN32
931: #define DIRSEPARATOR '\\'
932: #define CHARSEPARATOR "\\"
933: #define ODIRSEPARATOR '/'
934: #else
1.126 brouard 935: #define DIRSEPARATOR '/'
936: #define CHARSEPARATOR "/"
937: #define ODIRSEPARATOR '\\'
938: #endif
939:
1.246 ! brouard 940: /* $Id: imach.c,v 1.245 2016/09/02 07:25:01 brouard Exp $ */
1.126 brouard 941: /* $State: Exp $ */
1.196 brouard 942: #include "version.h"
943: char version[]=__IMACH_VERSION__;
1.224 brouard 944: 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";
1.246 ! brouard 945: char fullversion[]="$Revision: 1.245 $ $Date: 2016/09/02 07:25:01 $";
1.126 brouard 946: char strstart[80];
947: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 948: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 949: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 950: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
951: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
952: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 953: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
954: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 955: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
956: int cptcovprodnoage=0; /**< Number of covariate products without age */
957: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 958: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
959: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 960: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 961: int nsd=0; /**< Total number of single dummy variables (output) */
962: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 963: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 964: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 965: int ntveff=0; /**< ntveff number of effective time varying variables */
966: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 967: int cptcov=0; /* Working variable */
1.218 brouard 968: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 969: int npar=NPARMAX;
970: int nlstate=2; /* Number of live states */
971: int ndeath=1; /* Number of dead states */
1.130 brouard 972: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 973: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 974: int popbased=0;
975:
976: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 977: int maxwav=0; /* Maxim number of waves */
978: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
979: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
980: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 981: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 982: int mle=1, weightopt=0;
1.126 brouard 983: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
984: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
985: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
986: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 987: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 988: int selected(int kvar); /* Is covariate kvar selected for printing results */
989:
1.130 brouard 990: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 991: double **matprod2(); /* test */
1.126 brouard 992: double **oldm, **newm, **savm; /* Working pointers to matrices */
993: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 994: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
995:
1.136 brouard 996: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 997: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 998: FILE *ficlog, *ficrespow;
1.130 brouard 999: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1000: double fretone; /* Only one call to likelihood */
1.130 brouard 1001: long ipmx=0; /* Number of contributions */
1.126 brouard 1002: double sw; /* Sum of weights */
1003: char filerespow[FILENAMELENGTH];
1004: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1005: FILE *ficresilk;
1006: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1007: FILE *ficresprobmorprev;
1008: FILE *fichtm, *fichtmcov; /* Html File */
1009: FILE *ficreseij;
1010: char filerese[FILENAMELENGTH];
1011: FILE *ficresstdeij;
1012: char fileresstde[FILENAMELENGTH];
1013: FILE *ficrescveij;
1014: char filerescve[FILENAMELENGTH];
1015: FILE *ficresvij;
1016: char fileresv[FILENAMELENGTH];
1017: FILE *ficresvpl;
1018: char fileresvpl[FILENAMELENGTH];
1019: char title[MAXLINE];
1.234 brouard 1020: char model[MAXLINE]; /**< The model line */
1.217 brouard 1021: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1022: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1023: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1024: char command[FILENAMELENGTH];
1025: int outcmd=0;
1026:
1.217 brouard 1027: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1028: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1029: char filelog[FILENAMELENGTH]; /* Log file */
1030: char filerest[FILENAMELENGTH];
1031: char fileregp[FILENAMELENGTH];
1032: char popfile[FILENAMELENGTH];
1033:
1034: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1035:
1.157 brouard 1036: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1037: /* struct timezone tzp; */
1038: /* extern int gettimeofday(); */
1039: struct tm tml, *gmtime(), *localtime();
1040:
1041: extern time_t time();
1042:
1043: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1044: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1045: struct tm tm;
1046:
1.126 brouard 1047: char strcurr[80], strfor[80];
1048:
1049: char *endptr;
1050: long lval;
1051: double dval;
1052:
1053: #define NR_END 1
1054: #define FREE_ARG char*
1055: #define FTOL 1.0e-10
1056:
1057: #define NRANSI
1.240 brouard 1058: #define ITMAX 200
1059: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1060:
1061: #define TOL 2.0e-4
1062:
1063: #define CGOLD 0.3819660
1064: #define ZEPS 1.0e-10
1065: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1066:
1067: #define GOLD 1.618034
1068: #define GLIMIT 100.0
1069: #define TINY 1.0e-20
1070:
1071: static double maxarg1,maxarg2;
1072: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1073: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1074:
1075: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1076: #define rint(a) floor(a+0.5)
1.166 brouard 1077: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1078: #define mytinydouble 1.0e-16
1.166 brouard 1079: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1080: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1081: /* static double dsqrarg; */
1082: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1083: static double sqrarg;
1084: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1085: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1086: int agegomp= AGEGOMP;
1087:
1088: int imx;
1089: int stepm=1;
1090: /* Stepm, step in month: minimum step interpolation*/
1091:
1092: int estepm;
1093: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1094:
1095: int m,nb;
1096: long *num;
1.197 brouard 1097: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1098: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1099: covariate for which somebody answered excluding
1100: undefined. Usually 2: 0 and 1. */
1101: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1102: covariate for which somebody answered including
1103: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1104: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1105: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1106: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1107: double *ageexmed,*agecens;
1108: double dateintmean=0;
1109:
1110: double *weight;
1111: int **s; /* Status */
1.141 brouard 1112: double *agedc;
1.145 brouard 1113: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1114: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1115: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1116: double **coqvar; /* Fixed quantitative covariate iqv */
1117: double ***cotvar; /* Time varying covariate itv */
1118: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1119: double idx;
1120: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1121: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1122: /*k 1 2 3 4 5 6 7 8 9 */
1123: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1124: /* Tndvar[k] 1 2 3 4 5 */
1125: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1126: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1127: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1128: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1129: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1130: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1131: /* Tprod[i]=k 4 7 */
1132: /* Tage[i]=k 5 8 */
1133: /* */
1134: /* Type */
1135: /* V 1 2 3 4 5 */
1136: /* F F V V V */
1137: /* D Q D D Q */
1138: /* */
1139: int *TvarsD;
1140: int *TvarsDind;
1141: int *TvarsQ;
1142: int *TvarsQind;
1143:
1.235 brouard 1144: #define MAXRESULTLINES 10
1145: int nresult=0;
1146: int TKresult[MAXRESULTLINES];
1.237 brouard 1147: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1148: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1149: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1150: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1151: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1152: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1153:
1.234 brouard 1154: /* 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 1155: 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 */
1156: 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 */
1157: 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 */
1158: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1159: 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 */
1160: 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 1161: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1162: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1163: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1164: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1165: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1166: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1167: 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 */
1168: 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 */
1169:
1.230 brouard 1170: int *Tvarsel; /**< Selected covariates for output */
1171: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1172: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1173: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1174: 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 1175: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1176: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1177: int *Tage;
1.227 brouard 1178: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1179: 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 1180: 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*/
1181: 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 1182: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1183: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1184: int **Tvard;
1185: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1186: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1187: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1188: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1189: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1190: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1191: double *lsurv, *lpop, *tpop;
1192:
1.231 brouard 1193: #define FD 1; /* Fixed dummy covariate */
1194: #define FQ 2; /* Fixed quantitative covariate */
1195: #define FP 3; /* Fixed product covariate */
1196: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1197: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1198: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1199: #define VD 10; /* Varying dummy covariate */
1200: #define VQ 11; /* Varying quantitative covariate */
1201: #define VP 12; /* Varying product covariate */
1202: #define VPDD 13; /* Varying product dummy*dummy covariate */
1203: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1204: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1205: #define APFD 16; /* Age product * fixed dummy covariate */
1206: #define APFQ 17; /* Age product * fixed quantitative covariate */
1207: #define APVD 18; /* Age product * varying dummy covariate */
1208: #define APVQ 19; /* Age product * varying quantitative covariate */
1209:
1210: #define FTYPE 1; /* Fixed covariate */
1211: #define VTYPE 2; /* Varying covariate (loop in wave) */
1212: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1213:
1214: struct kmodel{
1215: int maintype; /* main type */
1216: int subtype; /* subtype */
1217: };
1218: struct kmodel modell[NCOVMAX];
1219:
1.143 brouard 1220: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1221: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1222:
1223: /**************** split *************************/
1224: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1225: {
1226: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1227: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1228: */
1229: char *ss; /* pointer */
1.186 brouard 1230: int l1=0, l2=0; /* length counters */
1.126 brouard 1231:
1232: l1 = strlen(path ); /* length of path */
1233: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1234: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1235: if ( ss == NULL ) { /* no directory, so determine current directory */
1236: strcpy( name, path ); /* we got the fullname name because no directory */
1237: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1238: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1239: /* get current working directory */
1240: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1241: #ifdef WIN32
1242: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1243: #else
1244: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1245: #endif
1.126 brouard 1246: return( GLOCK_ERROR_GETCWD );
1247: }
1248: /* got dirc from getcwd*/
1249: printf(" DIRC = %s \n",dirc);
1.205 brouard 1250: } else { /* strip directory from path */
1.126 brouard 1251: ss++; /* after this, the filename */
1252: l2 = strlen( ss ); /* length of filename */
1253: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1254: strcpy( name, ss ); /* save file name */
1255: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1256: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1257: printf(" DIRC2 = %s \n",dirc);
1258: }
1259: /* We add a separator at the end of dirc if not exists */
1260: l1 = strlen( dirc ); /* length of directory */
1261: if( dirc[l1-1] != DIRSEPARATOR ){
1262: dirc[l1] = DIRSEPARATOR;
1263: dirc[l1+1] = 0;
1264: printf(" DIRC3 = %s \n",dirc);
1265: }
1266: ss = strrchr( name, '.' ); /* find last / */
1267: if (ss >0){
1268: ss++;
1269: strcpy(ext,ss); /* save extension */
1270: l1= strlen( name);
1271: l2= strlen(ss)+1;
1272: strncpy( finame, name, l1-l2);
1273: finame[l1-l2]= 0;
1274: }
1275:
1276: return( 0 ); /* we're done */
1277: }
1278:
1279:
1280: /******************************************/
1281:
1282: void replace_back_to_slash(char *s, char*t)
1283: {
1284: int i;
1285: int lg=0;
1286: i=0;
1287: lg=strlen(t);
1288: for(i=0; i<= lg; i++) {
1289: (s[i] = t[i]);
1290: if (t[i]== '\\') s[i]='/';
1291: }
1292: }
1293:
1.132 brouard 1294: char *trimbb(char *out, char *in)
1.137 brouard 1295: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1296: char *s;
1297: s=out;
1298: while (*in != '\0'){
1.137 brouard 1299: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1300: in++;
1301: }
1302: *out++ = *in++;
1303: }
1304: *out='\0';
1305: return s;
1306: }
1307:
1.187 brouard 1308: /* char *substrchaine(char *out, char *in, char *chain) */
1309: /* { */
1310: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1311: /* char *s, *t; */
1312: /* t=in;s=out; */
1313: /* while ((*in != *chain) && (*in != '\0')){ */
1314: /* *out++ = *in++; */
1315: /* } */
1316:
1317: /* /\* *in matches *chain *\/ */
1318: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1319: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1320: /* } */
1321: /* in--; chain--; */
1322: /* while ( (*in != '\0')){ */
1323: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1324: /* *out++ = *in++; */
1325: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1326: /* } */
1327: /* *out='\0'; */
1328: /* out=s; */
1329: /* return out; */
1330: /* } */
1331: char *substrchaine(char *out, char *in, char *chain)
1332: {
1333: /* Substract chain 'chain' from 'in', return and output 'out' */
1334: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1335:
1336: char *strloc;
1337:
1338: strcpy (out, in);
1339: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1340: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1341: if(strloc != NULL){
1342: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1343: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1344: /* strcpy (strloc, strloc +strlen(chain));*/
1345: }
1346: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1347: return out;
1348: }
1349:
1350:
1.145 brouard 1351: char *cutl(char *blocc, char *alocc, char *in, char occ)
1352: {
1.187 brouard 1353: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1354: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1355: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1356: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1357: */
1.160 brouard 1358: char *s, *t;
1.145 brouard 1359: t=in;s=in;
1360: while ((*in != occ) && (*in != '\0')){
1361: *alocc++ = *in++;
1362: }
1363: if( *in == occ){
1364: *(alocc)='\0';
1365: s=++in;
1366: }
1367:
1368: if (s == t) {/* occ not found */
1369: *(alocc-(in-s))='\0';
1370: in=s;
1371: }
1372: while ( *in != '\0'){
1373: *blocc++ = *in++;
1374: }
1375:
1376: *blocc='\0';
1377: return t;
1378: }
1.137 brouard 1379: char *cutv(char *blocc, char *alocc, char *in, char occ)
1380: {
1.187 brouard 1381: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1382: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1383: gives blocc="abcdef2ghi" and alocc="j".
1384: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1385: */
1386: char *s, *t;
1387: t=in;s=in;
1388: while (*in != '\0'){
1389: while( *in == occ){
1390: *blocc++ = *in++;
1391: s=in;
1392: }
1393: *blocc++ = *in++;
1394: }
1395: if (s == t) /* occ not found */
1396: *(blocc-(in-s))='\0';
1397: else
1398: *(blocc-(in-s)-1)='\0';
1399: in=s;
1400: while ( *in != '\0'){
1401: *alocc++ = *in++;
1402: }
1403:
1404: *alocc='\0';
1405: return s;
1406: }
1407:
1.126 brouard 1408: int nbocc(char *s, char occ)
1409: {
1410: int i,j=0;
1411: int lg=20;
1412: i=0;
1413: lg=strlen(s);
1414: for(i=0; i<= lg; i++) {
1.234 brouard 1415: if (s[i] == occ ) j++;
1.126 brouard 1416: }
1417: return j;
1418: }
1419:
1.137 brouard 1420: /* void cutv(char *u,char *v, char*t, char occ) */
1421: /* { */
1422: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1423: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1424: /* gives u="abcdef2ghi" and v="j" *\/ */
1425: /* int i,lg,j,p=0; */
1426: /* i=0; */
1427: /* lg=strlen(t); */
1428: /* for(j=0; j<=lg-1; j++) { */
1429: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1430: /* } */
1.126 brouard 1431:
1.137 brouard 1432: /* for(j=0; j<p; j++) { */
1433: /* (u[j] = t[j]); */
1434: /* } */
1435: /* u[p]='\0'; */
1.126 brouard 1436:
1.137 brouard 1437: /* for(j=0; j<= lg; j++) { */
1438: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1439: /* } */
1440: /* } */
1.126 brouard 1441:
1.160 brouard 1442: #ifdef _WIN32
1443: char * strsep(char **pp, const char *delim)
1444: {
1445: char *p, *q;
1446:
1447: if ((p = *pp) == NULL)
1448: return 0;
1449: if ((q = strpbrk (p, delim)) != NULL)
1450: {
1451: *pp = q + 1;
1452: *q = '\0';
1453: }
1454: else
1455: *pp = 0;
1456: return p;
1457: }
1458: #endif
1459:
1.126 brouard 1460: /********************** nrerror ********************/
1461:
1462: void nrerror(char error_text[])
1463: {
1464: fprintf(stderr,"ERREUR ...\n");
1465: fprintf(stderr,"%s\n",error_text);
1466: exit(EXIT_FAILURE);
1467: }
1468: /*********************** vector *******************/
1469: double *vector(int nl, int nh)
1470: {
1471: double *v;
1472: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1473: if (!v) nrerror("allocation failure in vector");
1474: return v-nl+NR_END;
1475: }
1476:
1477: /************************ free vector ******************/
1478: void free_vector(double*v, int nl, int nh)
1479: {
1480: free((FREE_ARG)(v+nl-NR_END));
1481: }
1482:
1483: /************************ivector *******************************/
1484: int *ivector(long nl,long nh)
1485: {
1486: int *v;
1487: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1488: if (!v) nrerror("allocation failure in ivector");
1489: return v-nl+NR_END;
1490: }
1491:
1492: /******************free ivector **************************/
1493: void free_ivector(int *v, long nl, long nh)
1494: {
1495: free((FREE_ARG)(v+nl-NR_END));
1496: }
1497:
1498: /************************lvector *******************************/
1499: long *lvector(long nl,long nh)
1500: {
1501: long *v;
1502: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1503: if (!v) nrerror("allocation failure in ivector");
1504: return v-nl+NR_END;
1505: }
1506:
1507: /******************free lvector **************************/
1508: void free_lvector(long *v, long nl, long nh)
1509: {
1510: free((FREE_ARG)(v+nl-NR_END));
1511: }
1512:
1513: /******************* imatrix *******************************/
1514: int **imatrix(long nrl, long nrh, long ncl, long nch)
1515: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1516: {
1517: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1518: int **m;
1519:
1520: /* allocate pointers to rows */
1521: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1522: if (!m) nrerror("allocation failure 1 in matrix()");
1523: m += NR_END;
1524: m -= nrl;
1525:
1526:
1527: /* allocate rows and set pointers to them */
1528: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1529: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1530: m[nrl] += NR_END;
1531: m[nrl] -= ncl;
1532:
1533: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1534:
1535: /* return pointer to array of pointers to rows */
1536: return m;
1537: }
1538:
1539: /****************** free_imatrix *************************/
1540: void free_imatrix(m,nrl,nrh,ncl,nch)
1541: int **m;
1542: long nch,ncl,nrh,nrl;
1543: /* free an int matrix allocated by imatrix() */
1544: {
1545: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1546: free((FREE_ARG) (m+nrl-NR_END));
1547: }
1548:
1549: /******************* matrix *******************************/
1550: double **matrix(long nrl, long nrh, long ncl, long nch)
1551: {
1552: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1553: double **m;
1554:
1555: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1556: if (!m) nrerror("allocation failure 1 in matrix()");
1557: m += NR_END;
1558: m -= nrl;
1559:
1560: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1561: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1562: m[nrl] += NR_END;
1563: m[nrl] -= ncl;
1564:
1565: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1566: return m;
1.145 brouard 1567: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1568: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1569: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1570: */
1571: }
1572:
1573: /*************************free matrix ************************/
1574: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1575: {
1576: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1577: free((FREE_ARG)(m+nrl-NR_END));
1578: }
1579:
1580: /******************* ma3x *******************************/
1581: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1582: {
1583: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1584: double ***m;
1585:
1586: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1587: if (!m) nrerror("allocation failure 1 in matrix()");
1588: m += NR_END;
1589: m -= nrl;
1590:
1591: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1592: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1593: m[nrl] += NR_END;
1594: m[nrl] -= ncl;
1595:
1596: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1597:
1598: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1599: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1600: m[nrl][ncl] += NR_END;
1601: m[nrl][ncl] -= nll;
1602: for (j=ncl+1; j<=nch; j++)
1603: m[nrl][j]=m[nrl][j-1]+nlay;
1604:
1605: for (i=nrl+1; i<=nrh; i++) {
1606: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1607: for (j=ncl+1; j<=nch; j++)
1608: m[i][j]=m[i][j-1]+nlay;
1609: }
1610: return m;
1611: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1612: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1613: */
1614: }
1615:
1616: /*************************free ma3x ************************/
1617: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1618: {
1619: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1620: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1621: free((FREE_ARG)(m+nrl-NR_END));
1622: }
1623:
1624: /*************** function subdirf ***********/
1625: char *subdirf(char fileres[])
1626: {
1627: /* Caution optionfilefiname is hidden */
1628: strcpy(tmpout,optionfilefiname);
1629: strcat(tmpout,"/"); /* Add to the right */
1630: strcat(tmpout,fileres);
1631: return tmpout;
1632: }
1633:
1634: /*************** function subdirf2 ***********/
1635: char *subdirf2(char fileres[], char *preop)
1636: {
1637:
1638: /* Caution optionfilefiname is hidden */
1639: strcpy(tmpout,optionfilefiname);
1640: strcat(tmpout,"/");
1641: strcat(tmpout,preop);
1642: strcat(tmpout,fileres);
1643: return tmpout;
1644: }
1645:
1646: /*************** function subdirf3 ***********/
1647: char *subdirf3(char fileres[], char *preop, char *preop2)
1648: {
1649:
1650: /* Caution optionfilefiname is hidden */
1651: strcpy(tmpout,optionfilefiname);
1652: strcat(tmpout,"/");
1653: strcat(tmpout,preop);
1654: strcat(tmpout,preop2);
1655: strcat(tmpout,fileres);
1656: return tmpout;
1657: }
1.213 brouard 1658:
1659: /*************** function subdirfext ***********/
1660: char *subdirfext(char fileres[], char *preop, char *postop)
1661: {
1662:
1663: strcpy(tmpout,preop);
1664: strcat(tmpout,fileres);
1665: strcat(tmpout,postop);
1666: return tmpout;
1667: }
1.126 brouard 1668:
1.213 brouard 1669: /*************** function subdirfext3 ***********/
1670: char *subdirfext3(char fileres[], char *preop, char *postop)
1671: {
1672:
1673: /* Caution optionfilefiname is hidden */
1674: strcpy(tmpout,optionfilefiname);
1675: strcat(tmpout,"/");
1676: strcat(tmpout,preop);
1677: strcat(tmpout,fileres);
1678: strcat(tmpout,postop);
1679: return tmpout;
1680: }
1681:
1.162 brouard 1682: char *asc_diff_time(long time_sec, char ascdiff[])
1683: {
1684: long sec_left, days, hours, minutes;
1685: days = (time_sec) / (60*60*24);
1686: sec_left = (time_sec) % (60*60*24);
1687: hours = (sec_left) / (60*60) ;
1688: sec_left = (sec_left) %(60*60);
1689: minutes = (sec_left) /60;
1690: sec_left = (sec_left) % (60);
1691: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1692: return ascdiff;
1693: }
1694:
1.126 brouard 1695: /***************** f1dim *************************/
1696: extern int ncom;
1697: extern double *pcom,*xicom;
1698: extern double (*nrfunc)(double []);
1699:
1700: double f1dim(double x)
1701: {
1702: int j;
1703: double f;
1704: double *xt;
1705:
1706: xt=vector(1,ncom);
1707: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1708: f=(*nrfunc)(xt);
1709: free_vector(xt,1,ncom);
1710: return f;
1711: }
1712:
1713: /*****************brent *************************/
1714: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1715: {
1716: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1717: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1718: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1719: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1720: * returned function value.
1721: */
1.126 brouard 1722: int iter;
1723: double a,b,d,etemp;
1.159 brouard 1724: double fu=0,fv,fw,fx;
1.164 brouard 1725: double ftemp=0.;
1.126 brouard 1726: double p,q,r,tol1,tol2,u,v,w,x,xm;
1727: double e=0.0;
1728:
1729: a=(ax < cx ? ax : cx);
1730: b=(ax > cx ? ax : cx);
1731: x=w=v=bx;
1732: fw=fv=fx=(*f)(x);
1733: for (iter=1;iter<=ITMAX;iter++) {
1734: xm=0.5*(a+b);
1735: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1736: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1737: printf(".");fflush(stdout);
1738: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1739: #ifdef DEBUGBRENT
1.126 brouard 1740: 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);
1741: 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);
1742: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1743: #endif
1744: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1745: *xmin=x;
1746: return fx;
1747: }
1748: ftemp=fu;
1749: if (fabs(e) > tol1) {
1750: r=(x-w)*(fx-fv);
1751: q=(x-v)*(fx-fw);
1752: p=(x-v)*q-(x-w)*r;
1753: q=2.0*(q-r);
1754: if (q > 0.0) p = -p;
1755: q=fabs(q);
1756: etemp=e;
1757: e=d;
1758: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1759: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1760: else {
1.224 brouard 1761: d=p/q;
1762: u=x+d;
1763: if (u-a < tol2 || b-u < tol2)
1764: d=SIGN(tol1,xm-x);
1.126 brouard 1765: }
1766: } else {
1767: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1768: }
1769: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1770: fu=(*f)(u);
1771: if (fu <= fx) {
1772: if (u >= x) a=x; else b=x;
1773: SHFT(v,w,x,u)
1.183 brouard 1774: SHFT(fv,fw,fx,fu)
1775: } else {
1776: if (u < x) a=u; else b=u;
1777: if (fu <= fw || w == x) {
1.224 brouard 1778: v=w;
1779: w=u;
1780: fv=fw;
1781: fw=fu;
1.183 brouard 1782: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1783: v=u;
1784: fv=fu;
1.183 brouard 1785: }
1786: }
1.126 brouard 1787: }
1788: nrerror("Too many iterations in brent");
1789: *xmin=x;
1790: return fx;
1791: }
1792:
1793: /****************** mnbrak ***********************/
1794:
1795: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1796: double (*func)(double))
1.183 brouard 1797: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1798: the downhill direction (defined by the function as evaluated at the initial points) and returns
1799: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1800: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1801: */
1.126 brouard 1802: double ulim,u,r,q, dum;
1803: double fu;
1.187 brouard 1804:
1805: double scale=10.;
1806: int iterscale=0;
1807:
1808: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1809: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1810:
1811:
1812: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1813: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1814: /* *bx = *ax - (*ax - *bx)/scale; */
1815: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1816: /* } */
1817:
1.126 brouard 1818: if (*fb > *fa) {
1819: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1820: SHFT(dum,*fb,*fa,dum)
1821: }
1.126 brouard 1822: *cx=(*bx)+GOLD*(*bx-*ax);
1823: *fc=(*func)(*cx);
1.183 brouard 1824: #ifdef DEBUG
1.224 brouard 1825: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1826: 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 1827: #endif
1.224 brouard 1828: 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 1829: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1830: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1831: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1832: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1833: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1834: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1835: fu=(*func)(u);
1.163 brouard 1836: #ifdef DEBUG
1837: /* f(x)=A(x-u)**2+f(u) */
1838: double A, fparabu;
1839: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1840: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1841: 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);
1842: 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 1843: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1844: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1845: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1846: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1847: #endif
1.184 brouard 1848: #ifdef MNBRAKORIGINAL
1.183 brouard 1849: #else
1.191 brouard 1850: /* if (fu > *fc) { */
1851: /* #ifdef DEBUG */
1852: /* printf("mnbrak4 fu > fc \n"); */
1853: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1854: /* #endif */
1855: /* /\* 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 *\\/ *\/ */
1856: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1857: /* dum=u; /\* Shifting c and u *\/ */
1858: /* u = *cx; */
1859: /* *cx = dum; */
1860: /* dum = fu; */
1861: /* fu = *fc; */
1862: /* *fc =dum; */
1863: /* } else { /\* end *\/ */
1864: /* #ifdef DEBUG */
1865: /* printf("mnbrak3 fu < fc \n"); */
1866: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1867: /* #endif */
1868: /* dum=u; /\* Shifting c and u *\/ */
1869: /* u = *cx; */
1870: /* *cx = dum; */
1871: /* dum = fu; */
1872: /* fu = *fc; */
1873: /* *fc =dum; */
1874: /* } */
1.224 brouard 1875: #ifdef DEBUGMNBRAK
1876: double A, fparabu;
1877: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1878: fparabu= *fa - A*(*ax-u)*(*ax-u);
1879: 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);
1880: 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 1881: #endif
1.191 brouard 1882: dum=u; /* Shifting c and u */
1883: u = *cx;
1884: *cx = dum;
1885: dum = fu;
1886: fu = *fc;
1887: *fc =dum;
1.183 brouard 1888: #endif
1.162 brouard 1889: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1890: #ifdef DEBUG
1.224 brouard 1891: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1892: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1893: #endif
1.126 brouard 1894: fu=(*func)(u);
1895: if (fu < *fc) {
1.183 brouard 1896: #ifdef DEBUG
1.224 brouard 1897: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1898: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1899: #endif
1900: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1901: SHFT(*fb,*fc,fu,(*func)(u))
1902: #ifdef DEBUG
1903: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1904: #endif
1905: }
1.162 brouard 1906: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1907: #ifdef DEBUG
1.224 brouard 1908: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1909: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1910: #endif
1.126 brouard 1911: u=ulim;
1912: fu=(*func)(u);
1.183 brouard 1913: } else { /* u could be left to b (if r > q parabola has a maximum) */
1914: #ifdef DEBUG
1.224 brouard 1915: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1916: 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 1917: #endif
1.126 brouard 1918: u=(*cx)+GOLD*(*cx-*bx);
1919: fu=(*func)(u);
1.224 brouard 1920: #ifdef DEBUG
1921: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1922: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1923: #endif
1.183 brouard 1924: } /* end tests */
1.126 brouard 1925: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1926: SHFT(*fa,*fb,*fc,fu)
1927: #ifdef DEBUG
1.224 brouard 1928: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1929: 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 1930: #endif
1931: } /* 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 1932: }
1933:
1934: /*************** linmin ************************/
1.162 brouard 1935: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1936: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1937: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1938: the value of func at the returned location p . This is actually all accomplished by calling the
1939: routines mnbrak and brent .*/
1.126 brouard 1940: int ncom;
1941: double *pcom,*xicom;
1942: double (*nrfunc)(double []);
1943:
1.224 brouard 1944: #ifdef LINMINORIGINAL
1.126 brouard 1945: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1946: #else
1947: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1948: #endif
1.126 brouard 1949: {
1950: double brent(double ax, double bx, double cx,
1951: double (*f)(double), double tol, double *xmin);
1952: double f1dim(double x);
1953: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1954: double *fc, double (*func)(double));
1955: int j;
1956: double xx,xmin,bx,ax;
1957: double fx,fb,fa;
1.187 brouard 1958:
1.203 brouard 1959: #ifdef LINMINORIGINAL
1960: #else
1961: double scale=10., axs, xxs; /* Scale added for infinity */
1962: #endif
1963:
1.126 brouard 1964: ncom=n;
1965: pcom=vector(1,n);
1966: xicom=vector(1,n);
1967: nrfunc=func;
1968: for (j=1;j<=n;j++) {
1969: pcom[j]=p[j];
1.202 brouard 1970: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1971: }
1.187 brouard 1972:
1.203 brouard 1973: #ifdef LINMINORIGINAL
1974: xx=1.;
1975: #else
1976: axs=0.0;
1977: xxs=1.;
1978: do{
1979: xx= xxs;
1980: #endif
1.187 brouard 1981: ax=0.;
1982: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1983: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1984: /* 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)) */
1985: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1986: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1987: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
1988: /* 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 1989: #ifdef LINMINORIGINAL
1990: #else
1991: if (fx != fx){
1.224 brouard 1992: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
1993: printf("|");
1994: fprintf(ficlog,"|");
1.203 brouard 1995: #ifdef DEBUGLINMIN
1.224 brouard 1996: 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 1997: #endif
1998: }
1.224 brouard 1999: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2000: #endif
2001:
1.191 brouard 2002: #ifdef DEBUGLINMIN
2003: 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 2004: 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 2005: #endif
1.224 brouard 2006: #ifdef LINMINORIGINAL
2007: #else
2008: if(fb == fx){ /* Flat function in the direction */
2009: xmin=xx;
2010: *flat=1;
2011: }else{
2012: *flat=0;
2013: #endif
2014: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2015: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2016: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2017: /* fmin = f(p[j] + xmin * xi[j]) */
2018: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2019: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2020: #ifdef DEBUG
1.224 brouard 2021: 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);
2022: 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);
2023: #endif
2024: #ifdef LINMINORIGINAL
2025: #else
2026: }
1.126 brouard 2027: #endif
1.191 brouard 2028: #ifdef DEBUGLINMIN
2029: printf("linmin end ");
1.202 brouard 2030: fprintf(ficlog,"linmin end ");
1.191 brouard 2031: #endif
1.126 brouard 2032: for (j=1;j<=n;j++) {
1.203 brouard 2033: #ifdef LINMINORIGINAL
2034: xi[j] *= xmin;
2035: #else
2036: #ifdef DEBUGLINMIN
2037: if(xxs <1.0)
2038: printf(" before xi[%d]=%12.8f", j,xi[j]);
2039: #endif
2040: 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) */
2041: #ifdef DEBUGLINMIN
2042: if(xxs <1.0)
2043: 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 );
2044: #endif
2045: #endif
1.187 brouard 2046: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2047: }
1.191 brouard 2048: #ifdef DEBUGLINMIN
1.203 brouard 2049: printf("\n");
1.191 brouard 2050: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2051: 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 2052: for (j=1;j<=n;j++) {
1.202 brouard 2053: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2054: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2055: if(j % ncovmodel == 0){
1.191 brouard 2056: printf("\n");
1.202 brouard 2057: fprintf(ficlog,"\n");
2058: }
1.191 brouard 2059: }
1.203 brouard 2060: #else
1.191 brouard 2061: #endif
1.126 brouard 2062: free_vector(xicom,1,n);
2063: free_vector(pcom,1,n);
2064: }
2065:
2066:
2067: /*************** powell ************************/
1.162 brouard 2068: /*
2069: Minimization of a function func of n variables. Input consists of an initial starting point
2070: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2071: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2072: such that failure to decrease by more than this amount on one iteration signals doneness. On
2073: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2074: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2075: */
1.224 brouard 2076: #ifdef LINMINORIGINAL
2077: #else
2078: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2079: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2080: #endif
1.126 brouard 2081: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2082: double (*func)(double []))
2083: {
1.224 brouard 2084: #ifdef LINMINORIGINAL
2085: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2086: double (*func)(double []));
1.224 brouard 2087: #else
1.241 brouard 2088: void linmin(double p[], double xi[], int n, double *fret,
2089: double (*func)(double []),int *flat);
1.224 brouard 2090: #endif
1.239 brouard 2091: int i,ibig,j,jk,k;
1.126 brouard 2092: double del,t,*pt,*ptt,*xit;
1.181 brouard 2093: double directest;
1.126 brouard 2094: double fp,fptt;
2095: double *xits;
2096: int niterf, itmp;
1.224 brouard 2097: #ifdef LINMINORIGINAL
2098: #else
2099:
2100: flatdir=ivector(1,n);
2101: for (j=1;j<=n;j++) flatdir[j]=0;
2102: #endif
1.126 brouard 2103:
2104: pt=vector(1,n);
2105: ptt=vector(1,n);
2106: xit=vector(1,n);
2107: xits=vector(1,n);
2108: *fret=(*func)(p);
2109: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2110: rcurr_time = time(NULL);
1.126 brouard 2111: for (*iter=1;;++(*iter)) {
1.187 brouard 2112: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2113: ibig=0;
2114: del=0.0;
1.157 brouard 2115: rlast_time=rcurr_time;
2116: /* (void) gettimeofday(&curr_time,&tzp); */
2117: rcurr_time = time(NULL);
2118: curr_time = *localtime(&rcurr_time);
2119: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2120: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2121: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2122: for (i=1;i<=n;i++) {
1.126 brouard 2123: fprintf(ficrespow," %.12lf", p[i]);
2124: }
1.239 brouard 2125: fprintf(ficrespow,"\n");fflush(ficrespow);
2126: printf("\n#model= 1 + age ");
2127: fprintf(ficlog,"\n#model= 1 + age ");
2128: if(nagesqr==1){
1.241 brouard 2129: printf(" + age*age ");
2130: fprintf(ficlog," + age*age ");
1.239 brouard 2131: }
2132: for(j=1;j <=ncovmodel-2;j++){
2133: if(Typevar[j]==0) {
2134: printf(" + V%d ",Tvar[j]);
2135: fprintf(ficlog," + V%d ",Tvar[j]);
2136: }else if(Typevar[j]==1) {
2137: printf(" + V%d*age ",Tvar[j]);
2138: fprintf(ficlog," + V%d*age ",Tvar[j]);
2139: }else if(Typevar[j]==2) {
2140: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2141: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2142: }
2143: }
1.126 brouard 2144: printf("\n");
1.239 brouard 2145: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2146: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2147: fprintf(ficlog,"\n");
1.239 brouard 2148: for(i=1,jk=1; i <=nlstate; i++){
2149: for(k=1; k <=(nlstate+ndeath); k++){
2150: if (k != i) {
2151: printf("%d%d ",i,k);
2152: fprintf(ficlog,"%d%d ",i,k);
2153: for(j=1; j <=ncovmodel; j++){
2154: printf("%12.7f ",p[jk]);
2155: fprintf(ficlog,"%12.7f ",p[jk]);
2156: jk++;
2157: }
2158: printf("\n");
2159: fprintf(ficlog,"\n");
2160: }
2161: }
2162: }
1.241 brouard 2163: if(*iter <=3 && *iter >1){
1.157 brouard 2164: tml = *localtime(&rcurr_time);
2165: strcpy(strcurr,asctime(&tml));
2166: rforecast_time=rcurr_time;
1.126 brouard 2167: itmp = strlen(strcurr);
2168: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2169: strcurr[itmp-1]='\0';
1.162 brouard 2170: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2171: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2172: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2173: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2174: forecast_time = *localtime(&rforecast_time);
2175: strcpy(strfor,asctime(&forecast_time));
2176: itmp = strlen(strfor);
2177: if(strfor[itmp-1]=='\n')
2178: strfor[itmp-1]='\0';
2179: 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);
2180: 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 2181: }
2182: }
1.187 brouard 2183: for (i=1;i<=n;i++) { /* For each direction i */
2184: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2185: fptt=(*fret);
2186: #ifdef DEBUG
1.203 brouard 2187: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2188: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2189: #endif
1.203 brouard 2190: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2191: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2192: #ifdef LINMINORIGINAL
1.188 brouard 2193: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2194: #else
2195: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2196: flatdir[i]=flat; /* Function is vanishing in that direction i */
2197: #endif
2198: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2199: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2200: /* because that direction will be replaced unless the gain del is small */
2201: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2202: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2203: /* with the new direction. */
2204: del=fabs(fptt-(*fret));
2205: ibig=i;
1.126 brouard 2206: }
2207: #ifdef DEBUG
2208: printf("%d %.12e",i,(*fret));
2209: fprintf(ficlog,"%d %.12e",i,(*fret));
2210: for (j=1;j<=n;j++) {
1.224 brouard 2211: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2212: printf(" x(%d)=%.12e",j,xit[j]);
2213: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2214: }
2215: for(j=1;j<=n;j++) {
1.225 brouard 2216: printf(" p(%d)=%.12e",j,p[j]);
2217: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2218: }
2219: printf("\n");
2220: fprintf(ficlog,"\n");
2221: #endif
1.187 brouard 2222: } /* end loop on each direction i */
2223: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2224: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2225: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2226: for(j=1;j<=n;j++) {
1.225 brouard 2227: if(flatdir[j] >0){
2228: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2229: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2230: }
2231: /* printf("\n"); */
2232: /* fprintf(ficlog,"\n"); */
2233: }
1.243 brouard 2234: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2235: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2236: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2237: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2238: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2239: /* decreased of more than 3.84 */
2240: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2241: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2242: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2243:
1.188 brouard 2244: /* Starting the program with initial values given by a former maximization will simply change */
2245: /* the scales of the directions and the directions, because the are reset to canonical directions */
2246: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2247: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2248: #ifdef DEBUG
2249: int k[2],l;
2250: k[0]=1;
2251: k[1]=-1;
2252: printf("Max: %.12e",(*func)(p));
2253: fprintf(ficlog,"Max: %.12e",(*func)(p));
2254: for (j=1;j<=n;j++) {
2255: printf(" %.12e",p[j]);
2256: fprintf(ficlog," %.12e",p[j]);
2257: }
2258: printf("\n");
2259: fprintf(ficlog,"\n");
2260: for(l=0;l<=1;l++) {
2261: for (j=1;j<=n;j++) {
2262: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2263: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2264: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2265: }
2266: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2267: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2268: }
2269: #endif
2270:
1.224 brouard 2271: #ifdef LINMINORIGINAL
2272: #else
2273: free_ivector(flatdir,1,n);
2274: #endif
1.126 brouard 2275: free_vector(xit,1,n);
2276: free_vector(xits,1,n);
2277: free_vector(ptt,1,n);
2278: free_vector(pt,1,n);
2279: return;
1.192 brouard 2280: } /* enough precision */
1.240 brouard 2281: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2282: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2283: ptt[j]=2.0*p[j]-pt[j];
2284: xit[j]=p[j]-pt[j];
2285: pt[j]=p[j];
2286: }
1.181 brouard 2287: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2288: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2289: if (*iter <=4) {
1.225 brouard 2290: #else
2291: #endif
1.224 brouard 2292: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2293: #else
1.161 brouard 2294: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2295: #endif
1.162 brouard 2296: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2297: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2298: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2299: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2300: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2301: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2302: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2303: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2304: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2305: /* Even if f3 <f1, directest can be negative and t >0 */
2306: /* mu² and del² are equal when f3=f1 */
2307: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2308: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2309: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2310: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2311: #ifdef NRCORIGINAL
2312: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2313: #else
2314: 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 2315: t= t- del*SQR(fp-fptt);
1.183 brouard 2316: #endif
1.202 brouard 2317: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2318: #ifdef DEBUG
1.181 brouard 2319: 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);
2320: 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 2321: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2322: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2323: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2324: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2325: 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);
2326: 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);
2327: #endif
1.183 brouard 2328: #ifdef POWELLORIGINAL
2329: if (t < 0.0) { /* Then we use it for new direction */
2330: #else
1.182 brouard 2331: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2332: 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 2333: 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 2334: 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 2335: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2336: }
1.181 brouard 2337: if (directest < 0.0) { /* Then we use it for new direction */
2338: #endif
1.191 brouard 2339: #ifdef DEBUGLINMIN
1.234 brouard 2340: printf("Before linmin in direction P%d-P0\n",n);
2341: for (j=1;j<=n;j++) {
2342: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2343: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2344: if(j % ncovmodel == 0){
2345: printf("\n");
2346: fprintf(ficlog,"\n");
2347: }
2348: }
1.224 brouard 2349: #endif
2350: #ifdef LINMINORIGINAL
1.234 brouard 2351: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2352: #else
1.234 brouard 2353: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2354: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2355: #endif
1.234 brouard 2356:
1.191 brouard 2357: #ifdef DEBUGLINMIN
1.234 brouard 2358: for (j=1;j<=n;j++) {
2359: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2360: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2361: if(j % ncovmodel == 0){
2362: printf("\n");
2363: fprintf(ficlog,"\n");
2364: }
2365: }
1.224 brouard 2366: #endif
1.234 brouard 2367: for (j=1;j<=n;j++) {
2368: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2369: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2370: }
1.224 brouard 2371: #ifdef LINMINORIGINAL
2372: #else
1.234 brouard 2373: for (j=1, flatd=0;j<=n;j++) {
2374: if(flatdir[j]>0)
2375: flatd++;
2376: }
2377: if(flatd >0){
2378: printf("%d flat directions\n",flatd);
2379: fprintf(ficlog,"%d flat directions\n",flatd);
2380: for (j=1;j<=n;j++) {
2381: if(flatdir[j]>0){
2382: printf("%d ",j);
2383: fprintf(ficlog,"%d ",j);
2384: }
2385: }
2386: printf("\n");
2387: fprintf(ficlog,"\n");
2388: }
1.191 brouard 2389: #endif
1.234 brouard 2390: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2391: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2392:
1.126 brouard 2393: #ifdef DEBUG
1.234 brouard 2394: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2395: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2396: for(j=1;j<=n;j++){
2397: printf(" %lf",xit[j]);
2398: fprintf(ficlog," %lf",xit[j]);
2399: }
2400: printf("\n");
2401: fprintf(ficlog,"\n");
1.126 brouard 2402: #endif
1.192 brouard 2403: } /* end of t or directest negative */
1.224 brouard 2404: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2405: #else
1.234 brouard 2406: } /* end if (fptt < fp) */
1.192 brouard 2407: #endif
1.225 brouard 2408: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2409: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2410: #else
1.224 brouard 2411: #endif
1.234 brouard 2412: } /* loop iteration */
1.126 brouard 2413: }
1.234 brouard 2414:
1.126 brouard 2415: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2416:
1.235 brouard 2417: 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 2418: {
1.235 brouard 2419: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2420: (and selected quantitative values in nres)
2421: by left multiplying the unit
1.234 brouard 2422: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2423: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2424: /* Wx is row vector: population in state 1, population in state 2, population dead */
2425: /* or prevalence in state 1, prevalence in state 2, 0 */
2426: /* newm is the matrix after multiplications, its rows are identical at a factor */
2427: /* Initial matrix pimij */
2428: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2429: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2430: /* 0, 0 , 1} */
2431: /*
2432: * and after some iteration: */
2433: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2434: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2435: /* 0, 0 , 1} */
2436: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2437: /* {0.51571254859325999, 0.4842874514067399, */
2438: /* 0.51326036147820708, 0.48673963852179264} */
2439: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2440:
1.126 brouard 2441: int i, ii,j,k;
1.209 brouard 2442: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2443: /* double **matprod2(); */ /* test */
1.218 brouard 2444: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2445: double **newm;
1.209 brouard 2446: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2447: int ncvloop=0;
1.169 brouard 2448:
1.209 brouard 2449: min=vector(1,nlstate);
2450: max=vector(1,nlstate);
2451: meandiff=vector(1,nlstate);
2452:
1.218 brouard 2453: /* Starting with matrix unity */
1.126 brouard 2454: for (ii=1;ii<=nlstate+ndeath;ii++)
2455: for (j=1;j<=nlstate+ndeath;j++){
2456: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2457: }
1.169 brouard 2458:
2459: cov[1]=1.;
2460:
2461: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2462: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2463: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2464: ncvloop++;
1.126 brouard 2465: newm=savm;
2466: /* Covariates have to be included here again */
1.138 brouard 2467: cov[2]=agefin;
1.187 brouard 2468: if(nagesqr==1)
2469: cov[3]= agefin*agefin;;
1.234 brouard 2470: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2471: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2472: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2473: /* 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 2474: }
2475: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2476: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2477: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2478: /* 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 2479: }
1.237 brouard 2480: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2481: if(Dummy[Tvar[Tage[k]]]){
2482: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2483: } else{
1.235 brouard 2484: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2485: }
1.235 brouard 2486: /* 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 2487: }
1.237 brouard 2488: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2489: /* 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 2490: if(Dummy[Tvard[k][1]==0]){
2491: if(Dummy[Tvard[k][2]==0]){
2492: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2493: }else{
2494: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2495: }
2496: }else{
2497: if(Dummy[Tvard[k][2]==0]){
2498: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2499: }else{
2500: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2501: }
2502: }
1.234 brouard 2503: }
1.138 brouard 2504: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2505: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2506: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2507: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2508: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2509: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2510: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2511:
1.126 brouard 2512: savm=oldm;
2513: oldm=newm;
1.209 brouard 2514:
2515: for(j=1; j<=nlstate; j++){
2516: max[j]=0.;
2517: min[j]=1.;
2518: }
2519: for(i=1;i<=nlstate;i++){
2520: sumnew=0;
2521: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2522: for(j=1; j<=nlstate; j++){
2523: prlim[i][j]= newm[i][j]/(1-sumnew);
2524: max[j]=FMAX(max[j],prlim[i][j]);
2525: min[j]=FMIN(min[j],prlim[i][j]);
2526: }
2527: }
2528:
1.126 brouard 2529: maxmax=0.;
1.209 brouard 2530: for(j=1; j<=nlstate; j++){
2531: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2532: maxmax=FMAX(maxmax,meandiff[j]);
2533: /* 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 2534: } /* j loop */
1.203 brouard 2535: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2536: /* 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 2537: if(maxmax < ftolpl){
1.209 brouard 2538: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2539: free_vector(min,1,nlstate);
2540: free_vector(max,1,nlstate);
2541: free_vector(meandiff,1,nlstate);
1.126 brouard 2542: return prlim;
2543: }
1.169 brouard 2544: } /* age loop */
1.208 brouard 2545: /* After some age loop it doesn't converge */
1.209 brouard 2546: 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 2547: 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 2548: /* 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); */
2549: free_vector(min,1,nlstate);
2550: free_vector(max,1,nlstate);
2551: free_vector(meandiff,1,nlstate);
1.208 brouard 2552:
1.169 brouard 2553: return prlim; /* should not reach here */
1.126 brouard 2554: }
2555:
1.217 brouard 2556:
2557: /**** Back Prevalence limit (stable or period prevalence) ****************/
2558:
1.218 brouard 2559: /* 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) */
2560: /* 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 2561: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2562: {
1.218 brouard 2563: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2564: matrix by transitions matrix until convergence is reached with precision ftolpl */
2565: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2566: /* Wx is row vector: population in state 1, population in state 2, population dead */
2567: /* or prevalence in state 1, prevalence in state 2, 0 */
2568: /* newm is the matrix after multiplications, its rows are identical at a factor */
2569: /* Initial matrix pimij */
2570: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2571: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2572: /* 0, 0 , 1} */
2573: /*
2574: * and after some iteration: */
2575: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2576: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2577: /* 0, 0 , 1} */
2578: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2579: /* {0.51571254859325999, 0.4842874514067399, */
2580: /* 0.51326036147820708, 0.48673963852179264} */
2581: /* If we start from prlim again, prlim tends to a constant matrix */
2582:
2583: int i, ii,j,k;
2584: double *min, *max, *meandiff, maxmax,sumnew=0.;
2585: /* double **matprod2(); */ /* test */
2586: double **out, cov[NCOVMAX+1], **bmij();
2587: double **newm;
1.218 brouard 2588: double **dnewm, **doldm, **dsavm; /* for use */
2589: double **oldm, **savm; /* for use */
2590:
1.217 brouard 2591: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2592: int ncvloop=0;
2593:
2594: min=vector(1,nlstate);
2595: max=vector(1,nlstate);
2596: meandiff=vector(1,nlstate);
2597:
1.218 brouard 2598: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2599: oldm=oldms; savm=savms;
2600:
2601: /* Starting with matrix unity */
2602: for (ii=1;ii<=nlstate+ndeath;ii++)
2603: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2604: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2605: }
2606:
2607: cov[1]=1.;
2608:
2609: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2610: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2611: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2612: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2613: ncvloop++;
1.218 brouard 2614: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2615: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2616: /* Covariates have to be included here again */
2617: cov[2]=agefin;
2618: if(nagesqr==1)
2619: cov[3]= agefin*agefin;;
1.242 brouard 2620: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2621: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2622: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2623: /* 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)); */
2624: }
2625: /* for (k=1; k<=cptcovn;k++) { */
2626: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2627: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2628: /* /\* 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])]); *\/ */
2629: /* } */
2630: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2631: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2632: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2633: /* 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]); */
2634: }
2635: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2636: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2637: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2638: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2639: for (k=1; k<=cptcovage;k++){ /* For product with age */
2640: if(Dummy[Tvar[Tage[k]]]){
2641: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2642: } else{
2643: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2644: }
2645: /* 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]); */
2646: }
2647: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2648: /* 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]); */
2649: if(Dummy[Tvard[k][1]==0]){
2650: if(Dummy[Tvard[k][2]==0]){
2651: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2652: }else{
2653: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2654: }
2655: }else{
2656: if(Dummy[Tvard[k][2]==0]){
2657: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2658: }else{
2659: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2660: }
2661: }
1.217 brouard 2662: }
2663:
2664: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2665: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2666: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2667: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2668: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2669: /* ij should be linked to the correct index of cov */
2670: /* age and covariate values ij are in 'cov', but we need to pass
2671: * ij for the observed prevalence at age and status and covariate
2672: * number: prevacurrent[(int)agefin][ii][ij]
2673: */
2674: /* 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 *\/ */
2675: /* 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 *\/ */
2676: 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 2677: savm=oldm;
2678: oldm=newm;
2679: for(j=1; j<=nlstate; j++){
2680: max[j]=0.;
2681: min[j]=1.;
2682: }
2683: for(j=1; j<=nlstate; j++){
2684: for(i=1;i<=nlstate;i++){
1.234 brouard 2685: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2686: bprlim[i][j]= newm[i][j];
2687: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2688: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2689: }
2690: }
1.218 brouard 2691:
1.217 brouard 2692: maxmax=0.;
2693: for(i=1; i<=nlstate; i++){
2694: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2695: maxmax=FMAX(maxmax,meandiff[i]);
2696: /* 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); */
2697: } /* j loop */
2698: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2699: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2700: if(maxmax < ftolpl){
1.220 brouard 2701: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2702: free_vector(min,1,nlstate);
2703: free_vector(max,1,nlstate);
2704: free_vector(meandiff,1,nlstate);
2705: return bprlim;
2706: }
2707: } /* age loop */
2708: /* After some age loop it doesn't converge */
2709: 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\
2710: 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);
2711: /* 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); */
2712: free_vector(min,1,nlstate);
2713: free_vector(max,1,nlstate);
2714: free_vector(meandiff,1,nlstate);
2715:
2716: return bprlim; /* should not reach here */
2717: }
2718:
1.126 brouard 2719: /*************** transition probabilities ***************/
2720:
2721: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2722: {
1.138 brouard 2723: /* According to parameters values stored in x and the covariate's values stored in cov,
2724: computes the probability to be observed in state j being in state i by appying the
2725: model to the ncovmodel covariates (including constant and age).
2726: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2727: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2728: ncth covariate in the global vector x is given by the formula:
2729: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2730: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2731: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2732: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2733: Outputs ps[i][j] the probability to be observed in j being in j according to
2734: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2735: */
2736: double s1, lnpijopii;
1.126 brouard 2737: /*double t34;*/
1.164 brouard 2738: int i,j, nc, ii, jj;
1.126 brouard 2739:
1.223 brouard 2740: for(i=1; i<= nlstate; i++){
2741: for(j=1; j<i;j++){
2742: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2743: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2744: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2745: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2746: }
2747: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2748: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2749: }
2750: for(j=i+1; j<=nlstate+ndeath;j++){
2751: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2752: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2753: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2754: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2755: }
2756: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2757: }
2758: }
1.218 brouard 2759:
1.223 brouard 2760: for(i=1; i<= nlstate; i++){
2761: s1=0;
2762: for(j=1; j<i; j++){
2763: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2764: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2765: }
2766: for(j=i+1; j<=nlstate+ndeath; j++){
2767: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2768: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2769: }
2770: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2771: ps[i][i]=1./(s1+1.);
2772: /* Computing other pijs */
2773: for(j=1; j<i; j++)
2774: ps[i][j]= exp(ps[i][j])*ps[i][i];
2775: for(j=i+1; j<=nlstate+ndeath; j++)
2776: ps[i][j]= exp(ps[i][j])*ps[i][i];
2777: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2778: } /* end i */
1.218 brouard 2779:
1.223 brouard 2780: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2781: for(jj=1; jj<= nlstate+ndeath; jj++){
2782: ps[ii][jj]=0;
2783: ps[ii][ii]=1;
2784: }
2785: }
1.218 brouard 2786:
2787:
1.223 brouard 2788: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2789: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2790: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2791: /* } */
2792: /* printf("\n "); */
2793: /* } */
2794: /* printf("\n ");printf("%lf ",cov[2]);*/
2795: /*
2796: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2797: goto end;*/
1.223 brouard 2798: return ps;
1.126 brouard 2799: }
2800:
1.218 brouard 2801: /*************** backward transition probabilities ***************/
2802:
2803: /* 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 ) */
2804: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2805: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2806: {
1.222 brouard 2807: /* Computes the backward probability at age agefin and covariate ij
2808: * and returns in **ps as well as **bmij.
2809: */
1.218 brouard 2810: int i, ii, j,k;
1.222 brouard 2811:
2812: double **out, **pmij();
2813: double sumnew=0.;
1.218 brouard 2814: double agefin;
1.222 brouard 2815:
2816: double **dnewm, **dsavm, **doldm;
2817: double **bbmij;
2818:
1.218 brouard 2819: doldm=ddoldms; /* global pointers */
1.222 brouard 2820: dnewm=ddnewms;
2821: dsavm=ddsavms;
2822:
2823: agefin=cov[2];
2824: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2825: the observed prevalence (with this covariate ij) */
2826: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2827: /* We do have the matrix Px in savm and we need pij */
2828: for (j=1;j<=nlstate+ndeath;j++){
2829: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2830: for (ii=1;ii<=nlstate;ii++){
2831: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2832: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2833: for (ii=1;ii<=nlstate+ndeath;ii++){
2834: if(sumnew >= 1.e-10){
2835: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2836: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2837: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2838: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2839: /* }else */
2840: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2841: }else{
1.242 brouard 2842: ;
2843: /* 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 2844: }
2845: } /*End ii */
2846: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2847: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2848: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2849: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2850: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2851: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2852: /* left Product of this matrix by diag matrix of prevalences (savm) */
2853: for (j=1;j<=nlstate+ndeath;j++){
2854: for (ii=1;ii<=nlstate+ndeath;ii++){
2855: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2856: }
2857: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2858: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2859: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2860: /* end bmij */
2861: return ps;
1.218 brouard 2862: }
1.217 brouard 2863: /*************** transition probabilities ***************/
2864:
1.218 brouard 2865: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2866: {
2867: /* According to parameters values stored in x and the covariate's values stored in cov,
2868: computes the probability to be observed in state j being in state i by appying the
2869: model to the ncovmodel covariates (including constant and age).
2870: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2871: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2872: ncth covariate in the global vector x is given by the formula:
2873: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2874: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2875: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2876: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2877: Outputs ps[i][j] the probability to be observed in j being in j according to
2878: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2879: */
2880: double s1, lnpijopii;
2881: /*double t34;*/
2882: int i,j, nc, ii, jj;
2883:
1.234 brouard 2884: for(i=1; i<= nlstate; i++){
2885: for(j=1; j<i;j++){
2886: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2887: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2888: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2889: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2890: }
2891: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2892: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2893: }
2894: for(j=i+1; j<=nlstate+ndeath;j++){
2895: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2896: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2897: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2898: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2899: }
2900: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2901: }
2902: }
2903:
2904: for(i=1; i<= nlstate; i++){
2905: s1=0;
2906: for(j=1; j<i; j++){
2907: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2908: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2909: }
2910: for(j=i+1; j<=nlstate+ndeath; j++){
2911: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2912: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2913: }
2914: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2915: ps[i][i]=1./(s1+1.);
2916: /* Computing other pijs */
2917: for(j=1; j<i; j++)
2918: ps[i][j]= exp(ps[i][j])*ps[i][i];
2919: for(j=i+1; j<=nlstate+ndeath; j++)
2920: ps[i][j]= exp(ps[i][j])*ps[i][i];
2921: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2922: } /* end i */
2923:
2924: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2925: for(jj=1; jj<= nlstate+ndeath; jj++){
2926: ps[ii][jj]=0;
2927: ps[ii][ii]=1;
2928: }
2929: }
2930: /* Added for backcast */ /* Transposed matrix too */
2931: for(jj=1; jj<= nlstate+ndeath; jj++){
2932: s1=0.;
2933: for(ii=1; ii<= nlstate+ndeath; ii++){
2934: s1+=ps[ii][jj];
2935: }
2936: for(ii=1; ii<= nlstate; ii++){
2937: ps[ii][jj]=ps[ii][jj]/s1;
2938: }
2939: }
2940: /* Transposition */
2941: for(jj=1; jj<= nlstate+ndeath; jj++){
2942: for(ii=jj; ii<= nlstate+ndeath; ii++){
2943: s1=ps[ii][jj];
2944: ps[ii][jj]=ps[jj][ii];
2945: ps[jj][ii]=s1;
2946: }
2947: }
2948: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2949: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2950: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2951: /* } */
2952: /* printf("\n "); */
2953: /* } */
2954: /* printf("\n ");printf("%lf ",cov[2]);*/
2955: /*
2956: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2957: goto end;*/
2958: return ps;
1.217 brouard 2959: }
2960:
2961:
1.126 brouard 2962: /**************** Product of 2 matrices ******************/
2963:
1.145 brouard 2964: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2965: {
2966: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2967: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2968: /* in, b, out are matrice of pointers which should have been initialized
2969: before: only the contents of out is modified. The function returns
2970: a pointer to pointers identical to out */
1.145 brouard 2971: int i, j, k;
1.126 brouard 2972: for(i=nrl; i<= nrh; i++)
1.145 brouard 2973: for(k=ncolol; k<=ncoloh; k++){
2974: out[i][k]=0.;
2975: for(j=ncl; j<=nch; j++)
2976: out[i][k] +=in[i][j]*b[j][k];
2977: }
1.126 brouard 2978: return out;
2979: }
2980:
2981:
2982: /************* Higher Matrix Product ***************/
2983:
1.235 brouard 2984: 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 2985: {
1.218 brouard 2986: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 2987: 'nhstepm*hstepm*stepm' months (i.e. until
2988: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
2989: nhstepm*hstepm matrices.
2990: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
2991: (typically every 2 years instead of every month which is too big
2992: for the memory).
2993: Model is determined by parameters x and covariates have to be
2994: included manually here.
2995:
2996: */
2997:
2998: int i, j, d, h, k;
1.131 brouard 2999: double **out, cov[NCOVMAX+1];
1.126 brouard 3000: double **newm;
1.187 brouard 3001: double agexact;
1.214 brouard 3002: double agebegin, ageend;
1.126 brouard 3003:
3004: /* Hstepm could be zero and should return the unit matrix */
3005: for (i=1;i<=nlstate+ndeath;i++)
3006: for (j=1;j<=nlstate+ndeath;j++){
3007: oldm[i][j]=(i==j ? 1.0 : 0.0);
3008: po[i][j][0]=(i==j ? 1.0 : 0.0);
3009: }
3010: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3011: for(h=1; h <=nhstepm; h++){
3012: for(d=1; d <=hstepm; d++){
3013: newm=savm;
3014: /* Covariates have to be included here again */
3015: cov[1]=1.;
1.214 brouard 3016: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3017: cov[2]=agexact;
3018: if(nagesqr==1)
1.227 brouard 3019: cov[3]= agexact*agexact;
1.235 brouard 3020: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3021: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3022: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3023: /* 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)); */
3024: }
3025: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3026: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3027: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3028: /* 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]); */
3029: }
3030: for (k=1; k<=cptcovage;k++){
3031: if(Dummy[Tvar[Tage[k]]]){
3032: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3033: } else{
3034: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3035: }
3036: /* 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]); */
3037: }
3038: for (k=1; k<=cptcovprod;k++){ /* */
3039: /* 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]); */
3040: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3041: }
3042: /* for (k=1; k<=cptcovn;k++) */
3043: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3044: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3045: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3046: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3047: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3048:
3049:
1.126 brouard 3050: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3051: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3052: /* right multiplication of oldm by the current matrix */
1.126 brouard 3053: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3054: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3055: /* if((int)age == 70){ */
3056: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3057: /* for(i=1; i<=nlstate+ndeath; i++) { */
3058: /* printf("%d pmmij ",i); */
3059: /* for(j=1;j<=nlstate+ndeath;j++) { */
3060: /* printf("%f ",pmmij[i][j]); */
3061: /* } */
3062: /* printf(" oldm "); */
3063: /* for(j=1;j<=nlstate+ndeath;j++) { */
3064: /* printf("%f ",oldm[i][j]); */
3065: /* } */
3066: /* printf("\n"); */
3067: /* } */
3068: /* } */
1.126 brouard 3069: savm=oldm;
3070: oldm=newm;
3071: }
3072: for(i=1; i<=nlstate+ndeath; i++)
3073: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3074: po[i][j][h]=newm[i][j];
3075: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3076: }
1.128 brouard 3077: /*printf("h=%d ",h);*/
1.126 brouard 3078: } /* end h */
1.218 brouard 3079: /* printf("\n H=%d \n",h); */
1.126 brouard 3080: return po;
3081: }
3082:
1.217 brouard 3083: /************* Higher Back Matrix Product ***************/
1.218 brouard 3084: /* 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 3085: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3086: {
1.218 brouard 3087: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3088: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3089: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3090: nhstepm*hstepm matrices.
3091: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3092: (typically every 2 years instead of every month which is too big
1.217 brouard 3093: for the memory).
1.218 brouard 3094: Model is determined by parameters x and covariates have to be
3095: included manually here.
1.217 brouard 3096:
1.222 brouard 3097: */
1.217 brouard 3098:
3099: int i, j, d, h, k;
3100: double **out, cov[NCOVMAX+1];
3101: double **newm;
3102: double agexact;
3103: double agebegin, ageend;
1.222 brouard 3104: double **oldm, **savm;
1.217 brouard 3105:
1.222 brouard 3106: oldm=oldms;savm=savms;
1.217 brouard 3107: /* Hstepm could be zero and should return the unit matrix */
3108: for (i=1;i<=nlstate+ndeath;i++)
3109: for (j=1;j<=nlstate+ndeath;j++){
3110: oldm[i][j]=(i==j ? 1.0 : 0.0);
3111: po[i][j][0]=(i==j ? 1.0 : 0.0);
3112: }
3113: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3114: for(h=1; h <=nhstepm; h++){
3115: for(d=1; d <=hstepm; d++){
3116: newm=savm;
3117: /* Covariates have to be included here again */
3118: cov[1]=1.;
3119: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3120: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3121: cov[2]=agexact;
3122: if(nagesqr==1)
1.222 brouard 3123: cov[3]= agexact*agexact;
1.218 brouard 3124: for (k=1; k<=cptcovn;k++)
1.222 brouard 3125: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3126: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3127: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3128: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3129: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3130: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3131: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3132: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3133: /* 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 3134:
3135:
1.217 brouard 3136: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3137: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3138: /* Careful transposed matrix */
1.222 brouard 3139: /* age is in cov[2] */
1.218 brouard 3140: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3141: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3142: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3143: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3144: /* if((int)age == 70){ */
3145: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3146: /* for(i=1; i<=nlstate+ndeath; i++) { */
3147: /* printf("%d pmmij ",i); */
3148: /* for(j=1;j<=nlstate+ndeath;j++) { */
3149: /* printf("%f ",pmmij[i][j]); */
3150: /* } */
3151: /* printf(" oldm "); */
3152: /* for(j=1;j<=nlstate+ndeath;j++) { */
3153: /* printf("%f ",oldm[i][j]); */
3154: /* } */
3155: /* printf("\n"); */
3156: /* } */
3157: /* } */
3158: savm=oldm;
3159: oldm=newm;
3160: }
3161: for(i=1; i<=nlstate+ndeath; i++)
3162: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3163: po[i][j][h]=newm[i][j];
3164: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3165: }
3166: /*printf("h=%d ",h);*/
3167: } /* end h */
1.222 brouard 3168: /* printf("\n H=%d \n",h); */
1.217 brouard 3169: return po;
3170: }
3171:
3172:
1.162 brouard 3173: #ifdef NLOPT
3174: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3175: double fret;
3176: double *xt;
3177: int j;
3178: myfunc_data *d2 = (myfunc_data *) pd;
3179: /* xt = (p1-1); */
3180: xt=vector(1,n);
3181: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3182:
3183: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3184: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3185: printf("Function = %.12lf ",fret);
3186: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3187: printf("\n");
3188: free_vector(xt,1,n);
3189: return fret;
3190: }
3191: #endif
1.126 brouard 3192:
3193: /*************** log-likelihood *************/
3194: double func( double *x)
3195: {
1.226 brouard 3196: int i, ii, j, k, mi, d, kk;
3197: int ioffset=0;
3198: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3199: double **out;
3200: double lli; /* Individual log likelihood */
3201: int s1, s2;
1.228 brouard 3202: 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 3203: double bbh, survp;
3204: long ipmx;
3205: double agexact;
3206: /*extern weight */
3207: /* We are differentiating ll according to initial status */
3208: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3209: /*for(i=1;i<imx;i++)
3210: printf(" %d\n",s[4][i]);
3211: */
1.162 brouard 3212:
1.226 brouard 3213: ++countcallfunc;
1.162 brouard 3214:
1.226 brouard 3215: cov[1]=1.;
1.126 brouard 3216:
1.226 brouard 3217: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3218: ioffset=0;
1.226 brouard 3219: if(mle==1){
3220: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3221: /* Computes the values of the ncovmodel covariates of the model
3222: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3223: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3224: to be observed in j being in i according to the model.
3225: */
1.243 brouard 3226: ioffset=2+nagesqr ;
1.233 brouard 3227: /* Fixed */
1.234 brouard 3228: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3229: 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)*/
3230: }
1.226 brouard 3231: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3232: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3233: has been calculated etc */
3234: /* For an individual i, wav[i] gives the number of effective waves */
3235: /* We compute the contribution to Likelihood of each effective transition
3236: mw[mi][i] is real wave of the mi th effectve wave */
3237: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3238: s2=s[mw[mi+1][i]][i];
3239: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3240: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3241: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3242: */
3243: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3244: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3245: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3246: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3247: }
3248: for (ii=1;ii<=nlstate+ndeath;ii++)
3249: for (j=1;j<=nlstate+ndeath;j++){
3250: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3251: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3252: }
3253: for(d=0; d<dh[mi][i]; d++){
3254: newm=savm;
3255: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3256: cov[2]=agexact;
3257: if(nagesqr==1)
3258: cov[3]= agexact*agexact; /* Should be changed here */
3259: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3260: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3261: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3262: else
3263: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3264: }
3265: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3266: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3267: savm=oldm;
3268: oldm=newm;
3269: } /* end mult */
3270:
3271: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3272: /* But now since version 0.9 we anticipate for bias at large stepm.
3273: * If stepm is larger than one month (smallest stepm) and if the exact delay
3274: * (in months) between two waves is not a multiple of stepm, we rounded to
3275: * the nearest (and in case of equal distance, to the lowest) interval but now
3276: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3277: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3278: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3279: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3280: * -stepm/2 to stepm/2 .
3281: * For stepm=1 the results are the same as for previous versions of Imach.
3282: * For stepm > 1 the results are less biased than in previous versions.
3283: */
1.234 brouard 3284: s1=s[mw[mi][i]][i];
3285: s2=s[mw[mi+1][i]][i];
3286: bbh=(double)bh[mi][i]/(double)stepm;
3287: /* bias bh is positive if real duration
3288: * is higher than the multiple of stepm and negative otherwise.
3289: */
3290: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3291: if( s2 > nlstate){
3292: /* i.e. if s2 is a death state and if the date of death is known
3293: then the contribution to the likelihood is the probability to
3294: die between last step unit time and current step unit time,
3295: which is also equal to probability to die before dh
3296: minus probability to die before dh-stepm .
3297: In version up to 0.92 likelihood was computed
3298: as if date of death was unknown. Death was treated as any other
3299: health state: the date of the interview describes the actual state
3300: and not the date of a change in health state. The former idea was
3301: to consider that at each interview the state was recorded
3302: (healthy, disable or death) and IMaCh was corrected; but when we
3303: introduced the exact date of death then we should have modified
3304: the contribution of an exact death to the likelihood. This new
3305: contribution is smaller and very dependent of the step unit
3306: stepm. It is no more the probability to die between last interview
3307: and month of death but the probability to survive from last
3308: interview up to one month before death multiplied by the
3309: probability to die within a month. Thanks to Chris
3310: Jackson for correcting this bug. Former versions increased
3311: mortality artificially. The bad side is that we add another loop
3312: which slows down the processing. The difference can be up to 10%
3313: lower mortality.
3314: */
3315: /* If, at the beginning of the maximization mostly, the
3316: cumulative probability or probability to be dead is
3317: constant (ie = 1) over time d, the difference is equal to
3318: 0. out[s1][3] = savm[s1][3]: probability, being at state
3319: s1 at precedent wave, to be dead a month before current
3320: wave is equal to probability, being at state s1 at
3321: precedent wave, to be dead at mont of the current
3322: wave. Then the observed probability (that this person died)
3323: is null according to current estimated parameter. In fact,
3324: it should be very low but not zero otherwise the log go to
3325: infinity.
3326: */
1.183 brouard 3327: /* #ifdef INFINITYORIGINAL */
3328: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3329: /* #else */
3330: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3331: /* lli=log(mytinydouble); */
3332: /* else */
3333: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3334: /* #endif */
1.226 brouard 3335: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3336:
1.226 brouard 3337: } else if ( s2==-1 ) { /* alive */
3338: for (j=1,survp=0. ; j<=nlstate; j++)
3339: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3340: /*survp += out[s1][j]; */
3341: lli= log(survp);
3342: }
3343: else if (s2==-4) {
3344: for (j=3,survp=0. ; j<=nlstate; j++)
3345: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3346: lli= log(survp);
3347: }
3348: else if (s2==-5) {
3349: for (j=1,survp=0. ; j<=2; j++)
3350: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3351: lli= log(survp);
3352: }
3353: else{
3354: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3355: /* 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 */
3356: }
3357: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3358: /*if(lli ==000.0)*/
3359: /*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); */
3360: ipmx +=1;
3361: sw += weight[i];
3362: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3363: /* if (lli < log(mytinydouble)){ */
3364: /* 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); */
3365: /* 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]); */
3366: /* } */
3367: } /* end of wave */
3368: } /* end of individual */
3369: } else if(mle==2){
3370: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3371: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3372: for(mi=1; mi<= wav[i]-1; mi++){
3373: for (ii=1;ii<=nlstate+ndeath;ii++)
3374: for (j=1;j<=nlstate+ndeath;j++){
3375: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3376: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3377: }
3378: for(d=0; d<=dh[mi][i]; d++){
3379: newm=savm;
3380: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3381: cov[2]=agexact;
3382: if(nagesqr==1)
3383: cov[3]= agexact*agexact;
3384: for (kk=1; kk<=cptcovage;kk++) {
3385: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3386: }
3387: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3388: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3389: savm=oldm;
3390: oldm=newm;
3391: } /* end mult */
3392:
3393: s1=s[mw[mi][i]][i];
3394: s2=s[mw[mi+1][i]][i];
3395: bbh=(double)bh[mi][i]/(double)stepm;
3396: 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 */
3397: ipmx +=1;
3398: sw += weight[i];
3399: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3400: } /* end of wave */
3401: } /* end of individual */
3402: } else if(mle==3){ /* exponential inter-extrapolation */
3403: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3404: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3405: for(mi=1; mi<= wav[i]-1; mi++){
3406: for (ii=1;ii<=nlstate+ndeath;ii++)
3407: for (j=1;j<=nlstate+ndeath;j++){
3408: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3409: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3410: }
3411: for(d=0; d<dh[mi][i]; d++){
3412: newm=savm;
3413: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3414: cov[2]=agexact;
3415: if(nagesqr==1)
3416: cov[3]= agexact*agexact;
3417: for (kk=1; kk<=cptcovage;kk++) {
3418: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3419: }
3420: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3421: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3422: savm=oldm;
3423: oldm=newm;
3424: } /* end mult */
3425:
3426: s1=s[mw[mi][i]][i];
3427: s2=s[mw[mi+1][i]][i];
3428: bbh=(double)bh[mi][i]/(double)stepm;
3429: 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 */
3430: ipmx +=1;
3431: sw += weight[i];
3432: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3433: } /* end of wave */
3434: } /* end of individual */
3435: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3436: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3437: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3438: for(mi=1; mi<= wav[i]-1; mi++){
3439: for (ii=1;ii<=nlstate+ndeath;ii++)
3440: for (j=1;j<=nlstate+ndeath;j++){
3441: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3442: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3443: }
3444: for(d=0; d<dh[mi][i]; d++){
3445: newm=savm;
3446: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3447: cov[2]=agexact;
3448: if(nagesqr==1)
3449: cov[3]= agexact*agexact;
3450: for (kk=1; kk<=cptcovage;kk++) {
3451: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3452: }
1.126 brouard 3453:
1.226 brouard 3454: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3455: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3456: savm=oldm;
3457: oldm=newm;
3458: } /* end mult */
3459:
3460: s1=s[mw[mi][i]][i];
3461: s2=s[mw[mi+1][i]][i];
3462: if( s2 > nlstate){
3463: lli=log(out[s1][s2] - savm[s1][s2]);
3464: } else if ( s2==-1 ) { /* alive */
3465: for (j=1,survp=0. ; j<=nlstate; j++)
3466: survp += out[s1][j];
3467: lli= log(survp);
3468: }else{
3469: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3470: }
3471: ipmx +=1;
3472: sw += weight[i];
3473: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3474: /* 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 3475: } /* end of wave */
3476: } /* end of individual */
3477: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3478: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3479: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3480: for(mi=1; mi<= wav[i]-1; mi++){
3481: for (ii=1;ii<=nlstate+ndeath;ii++)
3482: for (j=1;j<=nlstate+ndeath;j++){
3483: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3484: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3485: }
3486: for(d=0; d<dh[mi][i]; d++){
3487: newm=savm;
3488: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3489: cov[2]=agexact;
3490: if(nagesqr==1)
3491: cov[3]= agexact*agexact;
3492: for (kk=1; kk<=cptcovage;kk++) {
3493: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3494: }
1.126 brouard 3495:
1.226 brouard 3496: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3497: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3498: savm=oldm;
3499: oldm=newm;
3500: } /* end mult */
3501:
3502: s1=s[mw[mi][i]][i];
3503: s2=s[mw[mi+1][i]][i];
3504: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3505: ipmx +=1;
3506: sw += weight[i];
3507: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3508: /*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]);*/
3509: } /* end of wave */
3510: } /* end of individual */
3511: } /* End of if */
3512: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3513: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3514: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3515: return -l;
1.126 brouard 3516: }
3517:
3518: /*************** log-likelihood *************/
3519: double funcone( double *x)
3520: {
1.228 brouard 3521: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3522: int i, ii, j, k, mi, d, kk;
1.228 brouard 3523: int ioffset=0;
1.131 brouard 3524: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3525: double **out;
3526: double lli; /* Individual log likelihood */
3527: double llt;
3528: int s1, s2;
1.228 brouard 3529: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3530:
1.126 brouard 3531: double bbh, survp;
1.187 brouard 3532: double agexact;
1.214 brouard 3533: double agebegin, ageend;
1.126 brouard 3534: /*extern weight */
3535: /* We are differentiating ll according to initial status */
3536: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3537: /*for(i=1;i<imx;i++)
3538: printf(" %d\n",s[4][i]);
3539: */
3540: cov[1]=1.;
3541:
3542: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3543: ioffset=0;
3544: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3545: /* ioffset=2+nagesqr+cptcovage; */
3546: ioffset=2+nagesqr;
1.232 brouard 3547: /* Fixed */
1.224 brouard 3548: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3549: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3550: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3551: 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)*/
3552: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3553: /* cov[2+6]=covar[Tvar[6]][i]; */
3554: /* cov[2+6]=covar[2][i]; V2 */
3555: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3556: /* cov[2+7]=covar[Tvar[7]][i]; */
3557: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3558: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3559: /* cov[2+9]=covar[Tvar[9]][i]; */
3560: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3561: }
1.232 brouard 3562: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3563: /* 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?)*\/ */
3564: /* } */
1.231 brouard 3565: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3566: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3567: /* } */
1.225 brouard 3568:
1.233 brouard 3569:
3570: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3571: /* Wave varying (but not age varying) */
3572: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3573: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3574: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3575: }
1.232 brouard 3576: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3577: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3578: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3579: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3580: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3581: /* 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 3582: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3583: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3584: /* /\* 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]); *\/ */
3585: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3586: /* } */
1.126 brouard 3587: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3588: for (j=1;j<=nlstate+ndeath;j++){
3589: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3590: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3591: }
1.214 brouard 3592:
3593: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3594: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3595: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.242 brouard 3596: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3597: and mw[mi+1][i]. dh depends on stepm.*/
3598: newm=savm;
3599: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3600: cov[2]=agexact;
3601: if(nagesqr==1)
3602: cov[3]= agexact*agexact;
3603: for (kk=1; kk<=cptcovage;kk++) {
3604: if(!FixedV[Tvar[Tage[kk]]])
3605: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3606: else
3607: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3608: }
3609: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3610: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3611: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3612: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3613: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3614: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3615: savm=oldm;
3616: oldm=newm;
1.126 brouard 3617: } /* end mult */
3618:
3619: s1=s[mw[mi][i]][i];
3620: s2=s[mw[mi+1][i]][i];
1.217 brouard 3621: /* if(s2==-1){ */
3622: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3623: /* /\* exit(1); *\/ */
3624: /* } */
1.126 brouard 3625: bbh=(double)bh[mi][i]/(double)stepm;
3626: /* bias is positive if real duration
3627: * is higher than the multiple of stepm and negative otherwise.
3628: */
3629: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3630: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3631: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3632: for (j=1,survp=0. ; j<=nlstate; j++)
3633: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3634: lli= log(survp);
1.126 brouard 3635: }else if (mle==1){
1.242 brouard 3636: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3637: } else if(mle==2){
1.242 brouard 3638: 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 3639: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3640: 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 3641: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3642: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3643: } else{ /* mle=0 back to 1 */
1.242 brouard 3644: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3645: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3646: } /* End of if */
3647: ipmx +=1;
3648: sw += weight[i];
3649: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3650: /*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 3651: if(globpr){
1.246 ! brouard 3652: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3653: %11.6f %11.6f %11.6f ", \
1.242 brouard 3654: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3655: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3656: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3657: llt +=ll[k]*gipmx/gsw;
3658: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3659: }
3660: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3661: }
1.232 brouard 3662: } /* end of wave */
3663: } /* end of individual */
3664: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3665: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3666: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3667: if(globpr==0){ /* First time we count the contributions and weights */
3668: gipmx=ipmx;
3669: gsw=sw;
3670: }
3671: return -l;
1.126 brouard 3672: }
3673:
3674:
3675: /*************** function likelione ***********/
3676: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3677: {
3678: /* This routine should help understanding what is done with
3679: the selection of individuals/waves and
3680: to check the exact contribution to the likelihood.
3681: Plotting could be done.
3682: */
3683: int k;
3684:
3685: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3686: strcpy(fileresilk,"ILK_");
1.202 brouard 3687: strcat(fileresilk,fileresu);
1.126 brouard 3688: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3689: printf("Problem with resultfile: %s\n", fileresilk);
3690: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3691: }
1.214 brouard 3692: 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");
3693: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3694: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3695: for(k=1; k<=nlstate; k++)
3696: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3697: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3698: }
3699:
3700: *fretone=(*funcone)(p);
3701: if(*globpri !=0){
3702: fclose(ficresilk);
1.205 brouard 3703: if (mle ==0)
3704: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3705: else if(mle >=1)
3706: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3707: 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 3708:
1.208 brouard 3709:
3710: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3711: 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 3712: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3713: }
1.207 brouard 3714: 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 3715: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3716: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3717: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3718: fflush(fichtm);
1.205 brouard 3719: }
1.126 brouard 3720: return;
3721: }
3722:
3723:
3724: /*********** Maximum Likelihood Estimation ***************/
3725:
3726: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3727: {
1.165 brouard 3728: int i,j, iter=0;
1.126 brouard 3729: double **xi;
3730: double fret;
3731: double fretone; /* Only one call to likelihood */
3732: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3733:
3734: #ifdef NLOPT
3735: int creturn;
3736: nlopt_opt opt;
3737: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3738: double *lb;
3739: double minf; /* the minimum objective value, upon return */
3740: double * p1; /* Shifted parameters from 0 instead of 1 */
3741: myfunc_data dinst, *d = &dinst;
3742: #endif
3743:
3744:
1.126 brouard 3745: xi=matrix(1,npar,1,npar);
3746: for (i=1;i<=npar;i++)
3747: for (j=1;j<=npar;j++)
3748: xi[i][j]=(i==j ? 1.0 : 0.0);
3749: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3750: strcpy(filerespow,"POW_");
1.126 brouard 3751: strcat(filerespow,fileres);
3752: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3753: printf("Problem with resultfile: %s\n", filerespow);
3754: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3755: }
3756: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3757: for (i=1;i<=nlstate;i++)
3758: for(j=1;j<=nlstate+ndeath;j++)
3759: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3760: fprintf(ficrespow,"\n");
1.162 brouard 3761: #ifdef POWELL
1.126 brouard 3762: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3763: #endif
1.126 brouard 3764:
1.162 brouard 3765: #ifdef NLOPT
3766: #ifdef NEWUOA
3767: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3768: #else
3769: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3770: #endif
3771: lb=vector(0,npar-1);
3772: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3773: nlopt_set_lower_bounds(opt, lb);
3774: nlopt_set_initial_step1(opt, 0.1);
3775:
3776: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3777: d->function = func;
3778: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3779: nlopt_set_min_objective(opt, myfunc, d);
3780: nlopt_set_xtol_rel(opt, ftol);
3781: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3782: printf("nlopt failed! %d\n",creturn);
3783: }
3784: else {
3785: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3786: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3787: iter=1; /* not equal */
3788: }
3789: nlopt_destroy(opt);
3790: #endif
1.126 brouard 3791: free_matrix(xi,1,npar,1,npar);
3792: fclose(ficrespow);
1.203 brouard 3793: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3794: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3795: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3796:
3797: }
3798:
3799: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3800: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3801: {
3802: double **a,**y,*x,pd;
1.203 brouard 3803: /* double **hess; */
1.164 brouard 3804: int i, j;
1.126 brouard 3805: int *indx;
3806:
3807: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3808: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3809: void lubksb(double **a, int npar, int *indx, double b[]) ;
3810: void ludcmp(double **a, int npar, int *indx, double *d) ;
3811: double gompertz(double p[]);
1.203 brouard 3812: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3813:
3814: printf("\nCalculation of the hessian matrix. Wait...\n");
3815: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3816: for (i=1;i<=npar;i++){
1.203 brouard 3817: printf("%d-",i);fflush(stdout);
3818: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3819:
3820: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3821:
3822: /* printf(" %f ",p[i]);
3823: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3824: }
3825:
3826: for (i=1;i<=npar;i++) {
3827: for (j=1;j<=npar;j++) {
3828: if (j>i) {
1.203 brouard 3829: printf(".%d-%d",i,j);fflush(stdout);
3830: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3831: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3832:
3833: hess[j][i]=hess[i][j];
3834: /*printf(" %lf ",hess[i][j]);*/
3835: }
3836: }
3837: }
3838: printf("\n");
3839: fprintf(ficlog,"\n");
3840:
3841: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3842: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3843:
3844: a=matrix(1,npar,1,npar);
3845: y=matrix(1,npar,1,npar);
3846: x=vector(1,npar);
3847: indx=ivector(1,npar);
3848: for (i=1;i<=npar;i++)
3849: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3850: ludcmp(a,npar,indx,&pd);
3851:
3852: for (j=1;j<=npar;j++) {
3853: for (i=1;i<=npar;i++) x[i]=0;
3854: x[j]=1;
3855: lubksb(a,npar,indx,x);
3856: for (i=1;i<=npar;i++){
3857: matcov[i][j]=x[i];
3858: }
3859: }
3860:
3861: printf("\n#Hessian matrix#\n");
3862: fprintf(ficlog,"\n#Hessian matrix#\n");
3863: for (i=1;i<=npar;i++) {
3864: for (j=1;j<=npar;j++) {
1.203 brouard 3865: printf("%.6e ",hess[i][j]);
3866: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3867: }
3868: printf("\n");
3869: fprintf(ficlog,"\n");
3870: }
3871:
1.203 brouard 3872: /* printf("\n#Covariance matrix#\n"); */
3873: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3874: /* for (i=1;i<=npar;i++) { */
3875: /* for (j=1;j<=npar;j++) { */
3876: /* printf("%.6e ",matcov[i][j]); */
3877: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3878: /* } */
3879: /* printf("\n"); */
3880: /* fprintf(ficlog,"\n"); */
3881: /* } */
3882:
1.126 brouard 3883: /* Recompute Inverse */
1.203 brouard 3884: /* for (i=1;i<=npar;i++) */
3885: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3886: /* ludcmp(a,npar,indx,&pd); */
3887:
3888: /* printf("\n#Hessian matrix recomputed#\n"); */
3889:
3890: /* for (j=1;j<=npar;j++) { */
3891: /* for (i=1;i<=npar;i++) x[i]=0; */
3892: /* x[j]=1; */
3893: /* lubksb(a,npar,indx,x); */
3894: /* for (i=1;i<=npar;i++){ */
3895: /* y[i][j]=x[i]; */
3896: /* printf("%.3e ",y[i][j]); */
3897: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3898: /* } */
3899: /* printf("\n"); */
3900: /* fprintf(ficlog,"\n"); */
3901: /* } */
3902:
3903: /* Verifying the inverse matrix */
3904: #ifdef DEBUGHESS
3905: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3906:
1.203 brouard 3907: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3908: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3909:
3910: for (j=1;j<=npar;j++) {
3911: for (i=1;i<=npar;i++){
1.203 brouard 3912: printf("%.2f ",y[i][j]);
3913: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3914: }
3915: printf("\n");
3916: fprintf(ficlog,"\n");
3917: }
1.203 brouard 3918: #endif
1.126 brouard 3919:
3920: free_matrix(a,1,npar,1,npar);
3921: free_matrix(y,1,npar,1,npar);
3922: free_vector(x,1,npar);
3923: free_ivector(indx,1,npar);
1.203 brouard 3924: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3925:
3926:
3927: }
3928:
3929: /*************** hessian matrix ****************/
3930: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3931: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3932: int i;
3933: int l=1, lmax=20;
1.203 brouard 3934: double k1,k2, res, fx;
1.132 brouard 3935: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3936: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3937: int k=0,kmax=10;
3938: double l1;
3939:
3940: fx=func(x);
3941: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3942: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3943: l1=pow(10,l);
3944: delts=delt;
3945: for(k=1 ; k <kmax; k=k+1){
3946: delt = delta*(l1*k);
3947: p2[theta]=x[theta] +delt;
1.145 brouard 3948: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3949: p2[theta]=x[theta]-delt;
3950: k2=func(p2)-fx;
3951: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3952: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3953:
1.203 brouard 3954: #ifdef DEBUGHESSII
1.126 brouard 3955: 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);
3956: 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);
3957: #endif
3958: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3959: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3960: k=kmax;
3961: }
3962: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3963: k=kmax; l=lmax*10;
1.126 brouard 3964: }
3965: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3966: delts=delt;
3967: }
1.203 brouard 3968: } /* End loop k */
1.126 brouard 3969: }
3970: delti[theta]=delts;
3971: return res;
3972:
3973: }
3974:
1.203 brouard 3975: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3976: {
3977: int i;
1.164 brouard 3978: int l=1, lmax=20;
1.126 brouard 3979: double k1,k2,k3,k4,res,fx;
1.132 brouard 3980: double p2[MAXPARM+1];
1.203 brouard 3981: int k, kmax=1;
3982: double v1, v2, cv12, lc1, lc2;
1.208 brouard 3983:
3984: int firstime=0;
1.203 brouard 3985:
1.126 brouard 3986: fx=func(x);
1.203 brouard 3987: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 3988: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 3989: p2[thetai]=x[thetai]+delti[thetai]*k;
3990: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3991: k1=func(p2)-fx;
3992:
1.203 brouard 3993: p2[thetai]=x[thetai]+delti[thetai]*k;
3994: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 3995: k2=func(p2)-fx;
3996:
1.203 brouard 3997: p2[thetai]=x[thetai]-delti[thetai]*k;
3998: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 3999: k3=func(p2)-fx;
4000:
1.203 brouard 4001: p2[thetai]=x[thetai]-delti[thetai]*k;
4002: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4003: k4=func(p2)-fx;
1.203 brouard 4004: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4005: if(k1*k2*k3*k4 <0.){
1.208 brouard 4006: firstime=1;
1.203 brouard 4007: kmax=kmax+10;
1.208 brouard 4008: }
4009: if(kmax >=10 || firstime ==1){
1.246 ! brouard 4010: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
! 4011: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 4012: 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);
4013: 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);
4014: }
4015: #ifdef DEBUGHESSIJ
4016: v1=hess[thetai][thetai];
4017: v2=hess[thetaj][thetaj];
4018: cv12=res;
4019: /* Computing eigen value of Hessian matrix */
4020: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4021: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4022: if ((lc2 <0) || (lc1 <0) ){
4023: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4024: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4025: 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);
4026: 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);
4027: }
1.126 brouard 4028: #endif
4029: }
4030: return res;
4031: }
4032:
1.203 brouard 4033: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4034: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4035: /* { */
4036: /* int i; */
4037: /* int l=1, lmax=20; */
4038: /* double k1,k2,k3,k4,res,fx; */
4039: /* double p2[MAXPARM+1]; */
4040: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4041: /* int k=0,kmax=10; */
4042: /* double l1; */
4043:
4044: /* fx=func(x); */
4045: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4046: /* l1=pow(10,l); */
4047: /* delts=delt; */
4048: /* for(k=1 ; k <kmax; k=k+1){ */
4049: /* delt = delti*(l1*k); */
4050: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4051: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4052: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4053: /* k1=func(p2)-fx; */
4054:
4055: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4056: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4057: /* k2=func(p2)-fx; */
4058:
4059: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4060: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4061: /* k3=func(p2)-fx; */
4062:
4063: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4064: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4065: /* k4=func(p2)-fx; */
4066: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4067: /* #ifdef DEBUGHESSIJ */
4068: /* 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); */
4069: /* 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); */
4070: /* #endif */
4071: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4072: /* k=kmax; */
4073: /* } */
4074: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4075: /* k=kmax; l=lmax*10; */
4076: /* } */
4077: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4078: /* delts=delt; */
4079: /* } */
4080: /* } /\* End loop k *\/ */
4081: /* } */
4082: /* delti[theta]=delts; */
4083: /* return res; */
4084: /* } */
4085:
4086:
1.126 brouard 4087: /************** Inverse of matrix **************/
4088: void ludcmp(double **a, int n, int *indx, double *d)
4089: {
4090: int i,imax,j,k;
4091: double big,dum,sum,temp;
4092: double *vv;
4093:
4094: vv=vector(1,n);
4095: *d=1.0;
4096: for (i=1;i<=n;i++) {
4097: big=0.0;
4098: for (j=1;j<=n;j++)
4099: if ((temp=fabs(a[i][j])) > big) big=temp;
4100: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4101: vv[i]=1.0/big;
4102: }
4103: for (j=1;j<=n;j++) {
4104: for (i=1;i<j;i++) {
4105: sum=a[i][j];
4106: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4107: a[i][j]=sum;
4108: }
4109: big=0.0;
4110: for (i=j;i<=n;i++) {
4111: sum=a[i][j];
4112: for (k=1;k<j;k++)
4113: sum -= a[i][k]*a[k][j];
4114: a[i][j]=sum;
4115: if ( (dum=vv[i]*fabs(sum)) >= big) {
4116: big=dum;
4117: imax=i;
4118: }
4119: }
4120: if (j != imax) {
4121: for (k=1;k<=n;k++) {
4122: dum=a[imax][k];
4123: a[imax][k]=a[j][k];
4124: a[j][k]=dum;
4125: }
4126: *d = -(*d);
4127: vv[imax]=vv[j];
4128: }
4129: indx[j]=imax;
4130: if (a[j][j] == 0.0) a[j][j]=TINY;
4131: if (j != n) {
4132: dum=1.0/(a[j][j]);
4133: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4134: }
4135: }
4136: free_vector(vv,1,n); /* Doesn't work */
4137: ;
4138: }
4139:
4140: void lubksb(double **a, int n, int *indx, double b[])
4141: {
4142: int i,ii=0,ip,j;
4143: double sum;
4144:
4145: for (i=1;i<=n;i++) {
4146: ip=indx[i];
4147: sum=b[ip];
4148: b[ip]=b[i];
4149: if (ii)
4150: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4151: else if (sum) ii=i;
4152: b[i]=sum;
4153: }
4154: for (i=n;i>=1;i--) {
4155: sum=b[i];
4156: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4157: b[i]=sum/a[i][i];
4158: }
4159: }
4160:
4161: void pstamp(FILE *fichier)
4162: {
1.196 brouard 4163: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4164: }
4165:
4166: /************ Frequencies ********************/
1.226 brouard 4167: void freqsummary(char fileres[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
4168: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4169: int firstpass, int lastpass, int stepm, int weightopt, char model[])
4170: { /* Some frequencies */
4171:
1.227 brouard 4172: int i, m, jk, j1, bool, z1,j, k, iv;
1.226 brouard 4173: int iind=0, iage=0;
4174: int mi; /* Effective wave */
4175: int first;
4176: double ***freq; /* Frequencies */
4177: double *meanq;
4178: double **meanqt;
4179: double *pp, **prop, *posprop, *pospropt;
4180: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4181: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4182: double agebegin, ageend;
4183:
4184: pp=vector(1,nlstate);
4185: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4186: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4187: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4188: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4189: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4190: meanqt=matrix(1,lastpass,1,nqtveff);
4191: strcpy(fileresp,"P_");
4192: strcat(fileresp,fileresu);
4193: /*strcat(fileresphtm,fileresu);*/
4194: if((ficresp=fopen(fileresp,"w"))==NULL) {
4195: printf("Problem with prevalence resultfile: %s\n", fileresp);
4196: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4197: exit(0);
4198: }
1.240 brouard 4199:
1.226 brouard 4200: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4201: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4202: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4203: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4204: fflush(ficlog);
4205: exit(70);
4206: }
4207: else{
4208: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4209: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4210: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4211: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4212: }
1.237 brouard 4213: 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 4214:
1.226 brouard 4215: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4216: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4217: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4218: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4219: fflush(ficlog);
4220: exit(70);
1.240 brouard 4221: } else{
1.226 brouard 4222: 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 4223: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4224: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4225: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4226: }
1.240 brouard 4227: 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);
4228:
1.226 brouard 4229: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4230: j1=0;
1.126 brouard 4231:
1.227 brouard 4232: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4233: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4234: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4235:
1.226 brouard 4236: first=1;
1.240 brouard 4237:
1.226 brouard 4238: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4239: reference=low_education V1=0,V2=0
4240: med_educ V1=1 V2=0,
4241: high_educ V1=0 V2=1
4242: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4243: */
1.240 brouard 4244:
1.227 brouard 4245: 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 4246: posproptt=0.;
4247: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4248: scanf("%d", i);*/
4249: for (i=-5; i<=nlstate+ndeath; i++)
4250: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4251: for(m=iagemin; m <= iagemax+3; m++)
4252: freq[i][jk][m]=0;
4253:
1.226 brouard 4254: for (i=1; i<=nlstate; i++) {
4255: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4256: prop[i][m]=0;
1.226 brouard 4257: posprop[i]=0;
4258: pospropt[i]=0;
4259: }
1.227 brouard 4260: /* for (z1=1; z1<= nqfveff; z1++) { */
4261: /* meanq[z1]+=0.; */
4262: /* for(m=1;m<=lastpass;m++){ */
4263: /* meanqt[m][z1]=0.; */
4264: /* } */
4265: /* } */
1.240 brouard 4266:
1.226 brouard 4267: dateintsum=0;
4268: k2cpt=0;
1.227 brouard 4269: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4270: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4271: bool=1;
1.227 brouard 4272: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4273: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4274: /* for (z1=1; z1<= nqfveff; z1++) { */
4275: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4276: /* } */
1.234 brouard 4277: for (z1=1; z1<=cptcoveff; z1++) {
4278: /* if(Tvaraff[z1] ==-20){ */
4279: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4280: /* }else if(Tvaraff[z1] ==-10){ */
4281: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4282: /* }else */
4283: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){
4284: /* Tests if this individual iind responded to j1 (V4=1 V3=0) */
4285: bool=0;
4286: /* 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",
4287: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4288: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4289: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4290: } /* Onlyf fixed */
4291: } /* end z1 */
4292: } /* cptcovn > 0 */
1.227 brouard 4293: } /* end any */
4294: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4295: /* for(m=firstpass; m<=lastpass; m++){ */
4296: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4297: m=mw[mi][iind];
4298: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4299: for (z1=1; z1<=cptcoveff; z1++) {
4300: if( Fixed[Tmodelind[z1]]==1){
4301: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4302: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4303: bool=0;
4304: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4305: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4306: bool=0;
4307: }
4308: }
4309: }
4310: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4311: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4312: if(bool==1){
4313: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4314: and mw[mi+1][iind]. dh depends on stepm. */
4315: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4316: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4317: if(m >=firstpass && m <=lastpass){
4318: k2=anint[m][iind]+(mint[m][iind]/12.);
4319: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4320: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4321: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4322: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4323: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4324: if (m<lastpass) {
4325: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4326: /* 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]); */
4327: if(s[m][iind]==-1)
4328: 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.));
4329: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4330: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4331: 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 */
4332: }
4333: } /* end if between passes */
4334: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99)) {
4335: dateintsum=dateintsum+k2;
4336: k2cpt++;
4337: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4338: }
4339: } /* end bool 2 */
4340: } /* end m */
1.226 brouard 4341: } /* end bool */
4342: } /* end iind = 1 to imx */
4343: /* prop[s][age] is feeded for any initial and valid live state as well as
4344: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4345:
4346:
1.226 brouard 4347: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4348: pstamp(ficresp);
1.240 brouard 4349: if (cptcoveff>0){
1.226 brouard 4350: fprintf(ficresp, "\n#********** Variable ");
4351: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4352: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4353: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4354: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4355: if(DummyV[z1]){
4356: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4357: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4358: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4359: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4360: }else{
4361: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4362: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4363: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4364: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4365: }
1.226 brouard 4366: }
4367: fprintf(ficresp, "**********\n#");
4368: fprintf(ficresphtm, "**********</h3>\n");
4369: fprintf(ficresphtmfr, "**********</h3>\n");
4370: fprintf(ficlog, "**********\n");
4371: }
4372: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4373: for(i=1; i<=nlstate;i++) {
1.240 brouard 4374: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4375: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4376: }
4377: fprintf(ficresp, "\n");
4378: fprintf(ficresphtm, "\n");
1.240 brouard 4379:
1.226 brouard 4380: /* Header of frequency table by age */
4381: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4382: fprintf(ficresphtmfr,"<th>Age</th> ");
4383: for(jk=-1; jk <=nlstate+ndeath; jk++){
4384: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4385: if(jk!=0 && m!=0)
4386: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4387: }
4388: }
4389: fprintf(ficresphtmfr, "\n");
1.240 brouard 4390:
1.226 brouard 4391: /* For each age */
4392: for(iage=iagemin; iage <= iagemax+3; iage++){
4393: fprintf(ficresphtm,"<tr>");
4394: if(iage==iagemax+1){
1.240 brouard 4395: fprintf(ficlog,"1");
4396: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4397: }else if(iage==iagemax+2){
1.240 brouard 4398: fprintf(ficlog,"0");
4399: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4400: }else if(iage==iagemax+3){
1.240 brouard 4401: fprintf(ficlog,"Total");
4402: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4403: }else{
1.240 brouard 4404: if(first==1){
4405: first=0;
4406: printf("See log file for details...\n");
4407: }
4408: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4409: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4410: }
4411: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4412: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4413: pp[jk] += freq[jk][m][iage];
1.226 brouard 4414: }
4415: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4416: for(m=-1, pos=0; m <=0 ; m++)
4417: pos += freq[jk][m][iage];
4418: if(pp[jk]>=1.e-10){
4419: if(first==1){
4420: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4421: }
4422: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4423: }else{
4424: if(first==1)
4425: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4426: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4427: }
1.226 brouard 4428: }
1.240 brouard 4429:
1.226 brouard 4430: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4431: /* posprop[jk]=0; */
4432: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4433: pp[jk] += freq[jk][m][iage];
1.226 brouard 4434: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4435:
1.226 brouard 4436: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4437: pos += pp[jk]; /* pos is the total number of transitions until this age */
4438: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4439: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4440: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4441: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4442: }
4443: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4444: if(pos>=1.e-5){
4445: if(first==1)
4446: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4447: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4448: }else{
4449: if(first==1)
4450: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4451: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4452: }
4453: if( iage <= iagemax){
4454: if(pos>=1.e-5){
4455: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4456: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4457: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4458: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4459: }
4460: else{
4461: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4462: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4463: }
4464: }
4465: pospropt[jk] +=posprop[jk];
1.226 brouard 4466: } /* end loop jk */
4467: /* pospropt=0.; */
4468: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4469: for(m=-1; m <=nlstate+ndeath; m++){
4470: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4471: if(first==1){
4472: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4473: }
4474: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4475: }
4476: if(jk!=0 && m!=0)
4477: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4478: }
1.226 brouard 4479: } /* end loop jk */
4480: posproptt=0.;
4481: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4482: posproptt += pospropt[jk];
1.226 brouard 4483: }
4484: fprintf(ficresphtmfr,"</tr>\n ");
4485: if(iage <= iagemax){
1.240 brouard 4486: fprintf(ficresp,"\n");
4487: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4488: }
4489: if(first==1)
1.240 brouard 4490: printf("Others in log...\n");
1.226 brouard 4491: fprintf(ficlog,"\n");
4492: } /* end loop age iage */
4493: fprintf(ficresphtm,"<tr><th>Tot</th>");
4494: for(jk=1; jk <=nlstate ; jk++){
4495: if(posproptt < 1.e-5){
1.240 brouard 4496: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4497: }else{
1.240 brouard 4498: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4499: }
4500: }
4501: fprintf(ficresphtm,"</tr>\n");
4502: fprintf(ficresphtm,"</table>\n");
4503: fprintf(ficresphtmfr,"</table>\n");
4504: if(posproptt < 1.e-5){
4505: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4506: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4507: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4508: invalidvarcomb[j1]=1;
4509: }else{
4510: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4511: invalidvarcomb[j1]=0;
4512: }
4513: fprintf(ficresphtmfr,"</table>\n");
4514: } /* end selected combination of covariate j1 */
4515: dateintmean=dateintsum/k2cpt;
1.240 brouard 4516:
1.226 brouard 4517: fclose(ficresp);
4518: fclose(ficresphtm);
4519: fclose(ficresphtmfr);
4520: free_vector(meanq,1,nqfveff);
4521: free_matrix(meanqt,1,lastpass,1,nqtveff);
4522: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4523: free_vector(pospropt,1,nlstate);
4524: free_vector(posprop,1,nlstate);
4525: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4526: free_vector(pp,1,nlstate);
4527: /* End of freqsummary */
4528: }
1.126 brouard 4529:
4530: /************ Prevalence ********************/
1.227 brouard 4531: 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)
4532: {
4533: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4534: in each health status at the date of interview (if between dateprev1 and dateprev2).
4535: We still use firstpass and lastpass as another selection.
4536: */
1.126 brouard 4537:
1.227 brouard 4538: int i, m, jk, j1, bool, z1,j, iv;
4539: int mi; /* Effective wave */
4540: int iage;
4541: double agebegin, ageend;
4542:
4543: double **prop;
4544: double posprop;
4545: double y2; /* in fractional years */
4546: int iagemin, iagemax;
4547: int first; /** to stop verbosity which is redirected to log file */
4548:
4549: iagemin= (int) agemin;
4550: iagemax= (int) agemax;
4551: /*pp=vector(1,nlstate);*/
4552: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4553: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4554: j1=0;
1.222 brouard 4555:
1.227 brouard 4556: /*j=cptcoveff;*/
4557: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4558:
1.227 brouard 4559: first=1;
4560: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4561: for (i=1; i<=nlstate; i++)
4562: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4563: prop[i][iage]=0.0;
4564: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4565: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4566: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4567:
4568: for (i=1; i<=imx; i++) { /* Each individual */
4569: bool=1;
4570: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4571: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4572: m=mw[mi][i];
4573: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4574: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4575: for (z1=1; z1<=cptcoveff; z1++){
4576: if( Fixed[Tmodelind[z1]]==1){
4577: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4578: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4579: bool=0;
4580: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4581: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4582: bool=0;
4583: }
4584: }
4585: if(bool==1){ /* Otherwise we skip that wave/person */
4586: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4587: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4588: if(m >=firstpass && m <=lastpass){
4589: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4590: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4591: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4592: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4593: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4594: 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);
4595: exit(1);
4596: }
4597: if (s[m][i]>0 && s[m][i]<=nlstate) {
4598: /*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]]);*/
4599: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4600: prop[s[m][i]][iagemax+3] += weight[i];
4601: } /* end valid statuses */
4602: } /* end selection of dates */
4603: } /* end selection of waves */
4604: } /* end bool */
4605: } /* end wave */
4606: } /* end individual */
4607: for(i=iagemin; i <= iagemax+3; i++){
4608: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4609: posprop += prop[jk][i];
4610: }
4611:
4612: for(jk=1; jk <=nlstate ; jk++){
4613: if( i <= iagemax){
4614: if(posprop>=1.e-5){
4615: probs[i][jk][j1]= prop[jk][i]/posprop;
4616: } else{
4617: if(first==1){
4618: first=0;
4619: 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]);
4620: }
4621: }
4622: }
4623: }/* end jk */
4624: }/* end i */
1.222 brouard 4625: /*} *//* end i1 */
1.227 brouard 4626: } /* end j1 */
1.222 brouard 4627:
1.227 brouard 4628: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4629: /*free_vector(pp,1,nlstate);*/
4630: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4631: } /* End of prevalence */
1.126 brouard 4632:
4633: /************* Waves Concatenation ***************/
4634:
4635: 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)
4636: {
4637: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4638: Death is a valid wave (if date is known).
4639: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4640: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4641: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4642: */
1.126 brouard 4643:
1.224 brouard 4644: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4645: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4646: double sum=0., jmean=0.;*/
1.224 brouard 4647: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4648: int j, k=0,jk, ju, jl;
4649: double sum=0.;
4650: first=0;
1.214 brouard 4651: firstwo=0;
1.217 brouard 4652: firsthree=0;
1.218 brouard 4653: firstfour=0;
1.164 brouard 4654: jmin=100000;
1.126 brouard 4655: jmax=-1;
4656: jmean=0.;
1.224 brouard 4657:
4658: /* Treating live states */
1.214 brouard 4659: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4660: mi=0; /* First valid wave */
1.227 brouard 4661: mli=0; /* Last valid wave */
1.126 brouard 4662: m=firstpass;
1.214 brouard 4663: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4664: 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 */
4665: mli=m-1;/* mw[++mi][i]=m-1; */
4666: }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 */
4667: mw[++mi][i]=m;
4668: mli=m;
1.224 brouard 4669: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4670: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4671: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4672: }
1.227 brouard 4673: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4674: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4675: break;
1.224 brouard 4676: #else
1.227 brouard 4677: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4678: if(firsthree == 0){
4679: 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);
4680: firsthree=1;
4681: }
4682: 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);
4683: mw[++mi][i]=m;
4684: mli=m;
4685: }
4686: if(s[m][i]==-2){ /* Vital status is really unknown */
4687: nbwarn++;
4688: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4689: 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);
4690: 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);
4691: }
4692: break;
4693: }
4694: break;
1.224 brouard 4695: #endif
1.227 brouard 4696: }/* End m >= lastpass */
1.126 brouard 4697: }/* end while */
1.224 brouard 4698:
1.227 brouard 4699: /* 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 4700: /* After last pass */
1.224 brouard 4701: /* Treating death states */
1.214 brouard 4702: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4703: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4704: /* } */
1.126 brouard 4705: mi++; /* Death is another wave */
4706: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4707: /* Only death is a correct wave */
1.126 brouard 4708: mw[mi][i]=m;
1.224 brouard 4709: }
4710: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4711: 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 4712: /* m++; */
4713: /* mi++; */
4714: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4715: /* mw[mi][i]=m; */
1.218 brouard 4716: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4717: 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 */
4718: nbwarn++;
4719: if(firstfiv==0){
4720: 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 );
4721: firstfiv=1;
4722: }else{
4723: 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 );
4724: }
4725: }else{ /* Death occured afer last wave potential bias */
4726: nberr++;
4727: if(firstwo==0){
4728: 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 );
4729: firstwo=1;
4730: }
4731: 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 );
4732: }
1.218 brouard 4733: }else{ /* end date of interview is known */
1.227 brouard 4734: /* death is known but not confirmed by death status at any wave */
4735: if(firstfour==0){
4736: 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 );
4737: firstfour=1;
4738: }
4739: 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 4740: }
1.224 brouard 4741: } /* end if date of death is known */
4742: #endif
4743: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4744: /* wav[i]=mw[mi][i]; */
1.126 brouard 4745: if(mi==0){
4746: nbwarn++;
4747: if(first==0){
1.227 brouard 4748: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4749: first=1;
1.126 brouard 4750: }
4751: if(first==1){
1.227 brouard 4752: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4753: }
4754: } /* end mi==0 */
4755: } /* End individuals */
1.214 brouard 4756: /* wav and mw are no more changed */
1.223 brouard 4757:
1.214 brouard 4758:
1.126 brouard 4759: for(i=1; i<=imx; i++){
4760: for(mi=1; mi<wav[i];mi++){
4761: if (stepm <=0)
1.227 brouard 4762: dh[mi][i]=1;
1.126 brouard 4763: else{
1.227 brouard 4764: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4765: if (agedc[i] < 2*AGESUP) {
4766: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4767: if(j==0) j=1; /* Survives at least one month after exam */
4768: else if(j<0){
4769: nberr++;
4770: 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]);
4771: j=1; /* Temporary Dangerous patch */
4772: 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);
4773: 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]);
4774: 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);
4775: }
4776: k=k+1;
4777: if (j >= jmax){
4778: jmax=j;
4779: ijmax=i;
4780: }
4781: if (j <= jmin){
4782: jmin=j;
4783: ijmin=i;
4784: }
4785: sum=sum+j;
4786: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4787: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4788: }
4789: }
4790: else{
4791: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4792: /* 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 4793:
1.227 brouard 4794: k=k+1;
4795: if (j >= jmax) {
4796: jmax=j;
4797: ijmax=i;
4798: }
4799: else if (j <= jmin){
4800: jmin=j;
4801: ijmin=i;
4802: }
4803: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4804: /*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]);*/
4805: if(j<0){
4806: nberr++;
4807: 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]);
4808: 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]);
4809: }
4810: sum=sum+j;
4811: }
4812: jk= j/stepm;
4813: jl= j -jk*stepm;
4814: ju= j -(jk+1)*stepm;
4815: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4816: if(jl==0){
4817: dh[mi][i]=jk;
4818: bh[mi][i]=0;
4819: }else{ /* We want a negative bias in order to only have interpolation ie
4820: * to avoid the price of an extra matrix product in likelihood */
4821: dh[mi][i]=jk+1;
4822: bh[mi][i]=ju;
4823: }
4824: }else{
4825: if(jl <= -ju){
4826: dh[mi][i]=jk;
4827: bh[mi][i]=jl; /* bias is positive if real duration
4828: * is higher than the multiple of stepm and negative otherwise.
4829: */
4830: }
4831: else{
4832: dh[mi][i]=jk+1;
4833: bh[mi][i]=ju;
4834: }
4835: if(dh[mi][i]==0){
4836: dh[mi][i]=1; /* At least one step */
4837: bh[mi][i]=ju; /* At least one step */
4838: /* 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);*/
4839: }
4840: } /* end if mle */
1.126 brouard 4841: }
4842: } /* end wave */
4843: }
4844: jmean=sum/k;
4845: 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 4846: 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 4847: }
1.126 brouard 4848:
4849: /*********** Tricode ****************************/
1.220 brouard 4850: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4851: {
4852: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4853: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4854: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4855: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4856: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4857: */
1.130 brouard 4858:
1.242 brouard 4859: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
4860: int modmaxcovj=0; /* Modality max of covariates j */
4861: int cptcode=0; /* Modality max of covariates j */
4862: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4863:
4864:
1.242 brouard 4865: /* cptcoveff=0; */
4866: /* *cptcov=0; */
1.126 brouard 4867:
1.242 brouard 4868: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4869:
1.242 brouard 4870: /* Loop on covariates without age and products and no quantitative variable */
4871: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
4872: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4873: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4874: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4875: switch(Fixed[k]) {
4876: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
4877: 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*/
4878: ij=(int)(covar[Tvar[k]][i]);
4879: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4880: * If product of Vn*Vm, still boolean *:
4881: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4882: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4883: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4884: modality of the nth covariate of individual i. */
4885: if (ij > modmaxcovj)
4886: modmaxcovj=ij;
4887: else if (ij < modmincovj)
4888: modmincovj=ij;
4889: if ((ij < -1) && (ij > NCOVMAX)){
4890: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
4891: exit(1);
4892: }else
4893: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
4894: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
4895: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
4896: /* getting the maximum value of the modality of the covariate
4897: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
4898: female ies 1, then modmaxcovj=1.
4899: */
4900: } /* end for loop on individuals i */
4901: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4902: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4903: cptcode=modmaxcovj;
4904: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
4905: /*for (i=0; i<=cptcode; i++) {*/
4906: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
4907: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4908: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4909: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
4910: if( j != -1){
4911: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
4912: covariate for which somebody answered excluding
4913: undefined. Usually 2: 0 and 1. */
4914: }
4915: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
4916: covariate for which somebody answered including
4917: undefined. Usually 3: -1, 0 and 1. */
4918: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
4919: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
4920: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 4921:
1.242 brouard 4922: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
4923: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
4924: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
4925: /* modmincovj=3; modmaxcovj = 7; */
4926: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
4927: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
4928: /* defining two dummy variables: variables V1_1 and V1_2.*/
4929: /* nbcode[Tvar[j]][ij]=k; */
4930: /* nbcode[Tvar[j]][1]=0; */
4931: /* nbcode[Tvar[j]][2]=1; */
4932: /* nbcode[Tvar[j]][3]=2; */
4933: /* To be continued (not working yet). */
4934: ij=0; /* ij is similar to i but can jump over null modalities */
4935: 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*/
4936: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
4937: break;
4938: }
4939: ij++;
4940: 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*/
4941: cptcode = ij; /* New max modality for covar j */
4942: } /* end of loop on modality i=-1 to 1 or more */
4943: break;
4944: case 1: /* Testing on varying covariate, could be simple and
4945: * should look at waves or product of fixed *
4946: * varying. No time to test -1, assuming 0 and 1 only */
4947: ij=0;
4948: for(i=0; i<=1;i++){
4949: nbcode[Tvar[k]][++ij]=i;
4950: }
4951: break;
4952: default:
4953: break;
4954: } /* end switch */
4955: } /* end dummy test */
4956:
4957: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
4958: /* /\*recode from 0 *\/ */
4959: /* k is a modality. If we have model=V1+V1*sex */
4960: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
4961: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
4962: /* } */
4963: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
4964: /* if (ij > ncodemax[j]) { */
4965: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4966: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4967: /* break; */
4968: /* } */
4969: /* } /\* end of loop on modality k *\/ */
4970: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
4971:
4972: for (k=-1; k< maxncov; k++) Ndum[k]=0;
4973: /* Look at fixed dummy (single or product) covariates to check empty modalities */
4974: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
4975: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
4976: 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 */
4977: 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 */
4978: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
4979: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
4980:
4981: ij=0;
4982: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
4983: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
4984: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
4985: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
4986: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
4987: /* If product not in single variable we don't print results */
4988: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
4989: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
4990: 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*/
4991: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
4992: 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 */
4993: if(Fixed[k]!=0)
4994: anyvaryingduminmodel=1;
4995: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
4996: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
4997: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
4998: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
4999: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5000: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5001: }
5002: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5003: /* ij--; */
5004: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5005: *cptcov=ij; /*Number of total real effective covariates: effective
5006: * because they can be excluded from the model and real
5007: * if in the model but excluded because missing values, but how to get k from ij?*/
5008: for(j=ij+1; j<= cptcovt; j++){
5009: Tvaraff[j]=0;
5010: Tmodelind[j]=0;
5011: }
5012: for(j=ntveff+1; j<= cptcovt; j++){
5013: TmodelInvind[j]=0;
5014: }
5015: /* To be sorted */
5016: ;
5017: }
1.126 brouard 5018:
1.145 brouard 5019:
1.126 brouard 5020: /*********** Health Expectancies ****************/
5021:
1.235 brouard 5022: 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 5023:
5024: {
5025: /* Health expectancies, no variances */
1.164 brouard 5026: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5027: int nhstepma, nstepma; /* Decreasing with age */
5028: double age, agelim, hf;
5029: double ***p3mat;
5030: double eip;
5031:
1.238 brouard 5032: /* pstamp(ficreseij); */
1.126 brouard 5033: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5034: fprintf(ficreseij,"# Age");
5035: for(i=1; i<=nlstate;i++){
5036: for(j=1; j<=nlstate;j++){
5037: fprintf(ficreseij," e%1d%1d ",i,j);
5038: }
5039: fprintf(ficreseij," e%1d. ",i);
5040: }
5041: fprintf(ficreseij,"\n");
5042:
5043:
5044: if(estepm < stepm){
5045: printf ("Problem %d lower than %d\n",estepm, stepm);
5046: }
5047: else hstepm=estepm;
5048: /* We compute the life expectancy from trapezoids spaced every estepm months
5049: * This is mainly to measure the difference between two models: for example
5050: * if stepm=24 months pijx are given only every 2 years and by summing them
5051: * we are calculating an estimate of the Life Expectancy assuming a linear
5052: * progression in between and thus overestimating or underestimating according
5053: * to the curvature of the survival function. If, for the same date, we
5054: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5055: * to compare the new estimate of Life expectancy with the same linear
5056: * hypothesis. A more precise result, taking into account a more precise
5057: * curvature will be obtained if estepm is as small as stepm. */
5058:
5059: /* For example we decided to compute the life expectancy with the smallest unit */
5060: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5061: nhstepm is the number of hstepm from age to agelim
5062: nstepm is the number of stepm from age to agelin.
5063: Look at hpijx to understand the reason of that which relies in memory size
5064: and note for a fixed period like estepm months */
5065: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5066: survival function given by stepm (the optimization length). Unfortunately it
5067: means that if the survival funtion is printed only each two years of age and if
5068: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5069: results. So we changed our mind and took the option of the best precision.
5070: */
5071: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5072:
5073: agelim=AGESUP;
5074: /* If stepm=6 months */
5075: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5076: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5077:
5078: /* nhstepm age range expressed in number of stepm */
5079: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5080: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5081: /* if (stepm >= YEARM) hstepm=1;*/
5082: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5083: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5084:
5085: for (age=bage; age<=fage; age ++){
5086: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5087: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5088: /* if (stepm >= YEARM) hstepm=1;*/
5089: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5090:
5091: /* If stepm=6 months */
5092: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5093: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5094:
1.235 brouard 5095: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5096:
5097: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5098:
5099: printf("%d|",(int)age);fflush(stdout);
5100: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5101:
5102: /* Computing expectancies */
5103: for(i=1; i<=nlstate;i++)
5104: for(j=1; j<=nlstate;j++)
5105: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5106: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5107:
5108: /* 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]);*/
5109:
5110: }
5111:
5112: fprintf(ficreseij,"%3.0f",age );
5113: for(i=1; i<=nlstate;i++){
5114: eip=0;
5115: for(j=1; j<=nlstate;j++){
5116: eip +=eij[i][j][(int)age];
5117: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5118: }
5119: fprintf(ficreseij,"%9.4f", eip );
5120: }
5121: fprintf(ficreseij,"\n");
5122:
5123: }
5124: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5125: printf("\n");
5126: fprintf(ficlog,"\n");
5127:
5128: }
5129:
1.235 brouard 5130: 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 5131:
5132: {
5133: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5134: to initial status i, ei. .
1.126 brouard 5135: */
5136: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5137: int nhstepma, nstepma; /* Decreasing with age */
5138: double age, agelim, hf;
5139: double ***p3matp, ***p3matm, ***varhe;
5140: double **dnewm,**doldm;
5141: double *xp, *xm;
5142: double **gp, **gm;
5143: double ***gradg, ***trgradg;
5144: int theta;
5145:
5146: double eip, vip;
5147:
5148: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5149: xp=vector(1,npar);
5150: xm=vector(1,npar);
5151: dnewm=matrix(1,nlstate*nlstate,1,npar);
5152: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5153:
5154: pstamp(ficresstdeij);
5155: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5156: fprintf(ficresstdeij,"# Age");
5157: for(i=1; i<=nlstate;i++){
5158: for(j=1; j<=nlstate;j++)
5159: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5160: fprintf(ficresstdeij," e%1d. ",i);
5161: }
5162: fprintf(ficresstdeij,"\n");
5163:
5164: pstamp(ficrescveij);
5165: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5166: fprintf(ficrescveij,"# Age");
5167: for(i=1; i<=nlstate;i++)
5168: for(j=1; j<=nlstate;j++){
5169: cptj= (j-1)*nlstate+i;
5170: for(i2=1; i2<=nlstate;i2++)
5171: for(j2=1; j2<=nlstate;j2++){
5172: cptj2= (j2-1)*nlstate+i2;
5173: if(cptj2 <= cptj)
5174: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5175: }
5176: }
5177: fprintf(ficrescveij,"\n");
5178:
5179: if(estepm < stepm){
5180: printf ("Problem %d lower than %d\n",estepm, stepm);
5181: }
5182: else hstepm=estepm;
5183: /* We compute the life expectancy from trapezoids spaced every estepm months
5184: * This is mainly to measure the difference between two models: for example
5185: * if stepm=24 months pijx are given only every 2 years and by summing them
5186: * we are calculating an estimate of the Life Expectancy assuming a linear
5187: * progression in between and thus overestimating or underestimating according
5188: * to the curvature of the survival function. If, for the same date, we
5189: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5190: * to compare the new estimate of Life expectancy with the same linear
5191: * hypothesis. A more precise result, taking into account a more precise
5192: * curvature will be obtained if estepm is as small as stepm. */
5193:
5194: /* For example we decided to compute the life expectancy with the smallest unit */
5195: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5196: nhstepm is the number of hstepm from age to agelim
5197: nstepm is the number of stepm from age to agelin.
5198: Look at hpijx to understand the reason of that which relies in memory size
5199: and note for a fixed period like estepm months */
5200: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5201: survival function given by stepm (the optimization length). Unfortunately it
5202: means that if the survival funtion is printed only each two years of age and if
5203: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5204: results. So we changed our mind and took the option of the best precision.
5205: */
5206: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5207:
5208: /* If stepm=6 months */
5209: /* nhstepm age range expressed in number of stepm */
5210: agelim=AGESUP;
5211: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5212: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5213: /* if (stepm >= YEARM) hstepm=1;*/
5214: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5215:
5216: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5217: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5218: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5219: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5220: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5221: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5222:
5223: for (age=bage; age<=fage; age ++){
5224: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5225: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5226: /* if (stepm >= YEARM) hstepm=1;*/
5227: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5228:
1.126 brouard 5229: /* If stepm=6 months */
5230: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5231: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5232:
5233: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5234:
1.126 brouard 5235: /* Computing Variances of health expectancies */
5236: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5237: decrease memory allocation */
5238: for(theta=1; theta <=npar; theta++){
5239: for(i=1; i<=npar; i++){
1.222 brouard 5240: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5241: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5242: }
1.235 brouard 5243: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5244: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5245:
1.126 brouard 5246: for(j=1; j<= nlstate; j++){
1.222 brouard 5247: for(i=1; i<=nlstate; i++){
5248: for(h=0; h<=nhstepm-1; h++){
5249: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5250: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5251: }
5252: }
1.126 brouard 5253: }
1.218 brouard 5254:
1.126 brouard 5255: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5256: for(h=0; h<=nhstepm-1; h++){
5257: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5258: }
1.126 brouard 5259: }/* End theta */
5260:
5261:
5262: for(h=0; h<=nhstepm-1; h++)
5263: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5264: for(theta=1; theta <=npar; theta++)
5265: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5266:
1.218 brouard 5267:
1.222 brouard 5268: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5269: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5270: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5271:
1.222 brouard 5272: printf("%d|",(int)age);fflush(stdout);
5273: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5274: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5275: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5276: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5277: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5278: for(ij=1;ij<=nlstate*nlstate;ij++)
5279: for(ji=1;ji<=nlstate*nlstate;ji++)
5280: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5281: }
5282: }
1.218 brouard 5283:
1.126 brouard 5284: /* Computing expectancies */
1.235 brouard 5285: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5286: for(i=1; i<=nlstate;i++)
5287: for(j=1; j<=nlstate;j++)
1.222 brouard 5288: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5289: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5290:
1.222 brouard 5291: /* 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 5292:
1.222 brouard 5293: }
1.218 brouard 5294:
1.126 brouard 5295: fprintf(ficresstdeij,"%3.0f",age );
5296: for(i=1; i<=nlstate;i++){
5297: eip=0.;
5298: vip=0.;
5299: for(j=1; j<=nlstate;j++){
1.222 brouard 5300: eip += eij[i][j][(int)age];
5301: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5302: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5303: 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 5304: }
5305: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5306: }
5307: fprintf(ficresstdeij,"\n");
1.218 brouard 5308:
1.126 brouard 5309: fprintf(ficrescveij,"%3.0f",age );
5310: for(i=1; i<=nlstate;i++)
5311: for(j=1; j<=nlstate;j++){
1.222 brouard 5312: cptj= (j-1)*nlstate+i;
5313: for(i2=1; i2<=nlstate;i2++)
5314: for(j2=1; j2<=nlstate;j2++){
5315: cptj2= (j2-1)*nlstate+i2;
5316: if(cptj2 <= cptj)
5317: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5318: }
1.126 brouard 5319: }
5320: fprintf(ficrescveij,"\n");
1.218 brouard 5321:
1.126 brouard 5322: }
5323: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5324: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5325: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5326: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5327: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5328: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5329: printf("\n");
5330: fprintf(ficlog,"\n");
1.218 brouard 5331:
1.126 brouard 5332: free_vector(xm,1,npar);
5333: free_vector(xp,1,npar);
5334: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5335: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5336: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5337: }
1.218 brouard 5338:
1.126 brouard 5339: /************ Variance ******************/
1.235 brouard 5340: 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 5341: {
5342: /* Variance of health expectancies */
5343: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5344: /* double **newm;*/
5345: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5346:
5347: /* int movingaverage(); */
5348: double **dnewm,**doldm;
5349: double **dnewmp,**doldmp;
5350: int i, j, nhstepm, hstepm, h, nstepm ;
5351: int k;
5352: double *xp;
5353: double **gp, **gm; /* for var eij */
5354: double ***gradg, ***trgradg; /*for var eij */
5355: double **gradgp, **trgradgp; /* for var p point j */
5356: double *gpp, *gmp; /* for var p point j */
5357: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5358: double ***p3mat;
5359: double age,agelim, hf;
5360: /* double ***mobaverage; */
5361: int theta;
5362: char digit[4];
5363: char digitp[25];
5364:
5365: char fileresprobmorprev[FILENAMELENGTH];
5366:
5367: if(popbased==1){
5368: if(mobilav!=0)
5369: strcpy(digitp,"-POPULBASED-MOBILAV_");
5370: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5371: }
5372: else
5373: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5374:
1.218 brouard 5375: /* if (mobilav!=0) { */
5376: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5377: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5378: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5379: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5380: /* } */
5381: /* } */
5382:
5383: strcpy(fileresprobmorprev,"PRMORPREV-");
5384: sprintf(digit,"%-d",ij);
5385: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5386: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5387: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5388: strcat(fileresprobmorprev,fileresu);
5389: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5390: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5391: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5392: }
5393: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5394: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5395: pstamp(ficresprobmorprev);
5396: 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 5397: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5398: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5399: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5400: }
5401: for(j=1;j<=cptcoveff;j++)
5402: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5403: fprintf(ficresprobmorprev,"\n");
5404:
1.218 brouard 5405: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5406: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5407: fprintf(ficresprobmorprev," p.%-d SE",j);
5408: for(i=1; i<=nlstate;i++)
5409: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5410: }
5411: fprintf(ficresprobmorprev,"\n");
5412:
5413: fprintf(ficgp,"\n# Routine varevsij");
5414: fprintf(ficgp,"\nunset title \n");
5415: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5416: 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");
5417: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5418: /* } */
5419: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5420: pstamp(ficresvij);
5421: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5422: if(popbased==1)
5423: 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);
5424: else
5425: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5426: fprintf(ficresvij,"# Age");
5427: for(i=1; i<=nlstate;i++)
5428: for(j=1; j<=nlstate;j++)
5429: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5430: fprintf(ficresvij,"\n");
5431:
5432: xp=vector(1,npar);
5433: dnewm=matrix(1,nlstate,1,npar);
5434: doldm=matrix(1,nlstate,1,nlstate);
5435: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5436: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5437:
5438: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5439: gpp=vector(nlstate+1,nlstate+ndeath);
5440: gmp=vector(nlstate+1,nlstate+ndeath);
5441: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5442:
1.218 brouard 5443: if(estepm < stepm){
5444: printf ("Problem %d lower than %d\n",estepm, stepm);
5445: }
5446: else hstepm=estepm;
5447: /* For example we decided to compute the life expectancy with the smallest unit */
5448: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5449: nhstepm is the number of hstepm from age to agelim
5450: nstepm is the number of stepm from age to agelim.
5451: Look at function hpijx to understand why because of memory size limitations,
5452: we decided (b) to get a life expectancy respecting the most precise curvature of the
5453: survival function given by stepm (the optimization length). Unfortunately it
5454: means that if the survival funtion is printed every two years of age and if
5455: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5456: results. So we changed our mind and took the option of the best precision.
5457: */
5458: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5459: agelim = AGESUP;
5460: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5461: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5462: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5463: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5464: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5465: gp=matrix(0,nhstepm,1,nlstate);
5466: gm=matrix(0,nhstepm,1,nlstate);
5467:
5468:
5469: for(theta=1; theta <=npar; theta++){
5470: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5471: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5472: }
5473:
1.242 brouard 5474: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5475:
5476: if (popbased==1) {
5477: if(mobilav ==0){
5478: for(i=1; i<=nlstate;i++)
5479: prlim[i][i]=probs[(int)age][i][ij];
5480: }else{ /* mobilav */
5481: for(i=1; i<=nlstate;i++)
5482: prlim[i][i]=mobaverage[(int)age][i][ij];
5483: }
5484: }
5485:
1.235 brouard 5486: 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 5487: for(j=1; j<= nlstate; j++){
5488: for(h=0; h<=nhstepm; h++){
5489: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5490: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5491: }
5492: }
5493: /* Next for computing probability of death (h=1 means
5494: computed over hstepm matrices product = hstepm*stepm months)
5495: as a weighted average of prlim.
5496: */
5497: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5498: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5499: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5500: }
5501: /* end probability of death */
5502:
5503: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5504: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5505:
1.242 brouard 5506: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5507:
5508: if (popbased==1) {
5509: if(mobilav ==0){
5510: for(i=1; i<=nlstate;i++)
5511: prlim[i][i]=probs[(int)age][i][ij];
5512: }else{ /* mobilav */
5513: for(i=1; i<=nlstate;i++)
5514: prlim[i][i]=mobaverage[(int)age][i][ij];
5515: }
5516: }
5517:
1.235 brouard 5518: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5519:
5520: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5521: for(h=0; h<=nhstepm; h++){
5522: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5523: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5524: }
5525: }
5526: /* This for computing probability of death (h=1 means
5527: computed over hstepm matrices product = hstepm*stepm months)
5528: as a weighted average of prlim.
5529: */
5530: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5531: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5532: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5533: }
5534: /* end probability of death */
5535:
5536: for(j=1; j<= nlstate; j++) /* vareij */
5537: for(h=0; h<=nhstepm; h++){
5538: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5539: }
5540:
5541: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5542: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5543: }
5544:
5545: } /* End theta */
5546:
5547: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5548:
5549: for(h=0; h<=nhstepm; h++) /* veij */
5550: for(j=1; j<=nlstate;j++)
5551: for(theta=1; theta <=npar; theta++)
5552: trgradg[h][j][theta]=gradg[h][theta][j];
5553:
5554: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5555: for(theta=1; theta <=npar; theta++)
5556: trgradgp[j][theta]=gradgp[theta][j];
5557:
5558:
5559: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5560: for(i=1;i<=nlstate;i++)
5561: for(j=1;j<=nlstate;j++)
5562: vareij[i][j][(int)age] =0.;
5563:
5564: for(h=0;h<=nhstepm;h++){
5565: for(k=0;k<=nhstepm;k++){
5566: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5567: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5568: for(i=1;i<=nlstate;i++)
5569: for(j=1;j<=nlstate;j++)
5570: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5571: }
5572: }
5573:
5574: /* pptj */
5575: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5576: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5577: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5578: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5579: varppt[j][i]=doldmp[j][i];
5580: /* end ppptj */
5581: /* x centered again */
5582:
1.242 brouard 5583: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5584:
5585: if (popbased==1) {
5586: if(mobilav ==0){
5587: for(i=1; i<=nlstate;i++)
5588: prlim[i][i]=probs[(int)age][i][ij];
5589: }else{ /* mobilav */
5590: for(i=1; i<=nlstate;i++)
5591: prlim[i][i]=mobaverage[(int)age][i][ij];
5592: }
5593: }
5594:
5595: /* This for computing probability of death (h=1 means
5596: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5597: as a weighted average of prlim.
5598: */
1.235 brouard 5599: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5600: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5601: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5602: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5603: }
5604: /* end probability of death */
5605:
5606: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5607: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5608: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5609: for(i=1; i<=nlstate;i++){
5610: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5611: }
5612: }
5613: fprintf(ficresprobmorprev,"\n");
5614:
5615: fprintf(ficresvij,"%.0f ",age );
5616: for(i=1; i<=nlstate;i++)
5617: for(j=1; j<=nlstate;j++){
5618: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5619: }
5620: fprintf(ficresvij,"\n");
5621: free_matrix(gp,0,nhstepm,1,nlstate);
5622: free_matrix(gm,0,nhstepm,1,nlstate);
5623: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5624: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5625: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5626: } /* End age */
5627: free_vector(gpp,nlstate+1,nlstate+ndeath);
5628: free_vector(gmp,nlstate+1,nlstate+ndeath);
5629: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5630: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5631: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5632: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5633: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5634: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5635: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5636: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5637: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5638: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5639: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5640: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5641: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5642: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5643: 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);
5644: /* 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 5645: */
1.218 brouard 5646: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5647: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5648:
1.218 brouard 5649: free_vector(xp,1,npar);
5650: free_matrix(doldm,1,nlstate,1,nlstate);
5651: free_matrix(dnewm,1,nlstate,1,npar);
5652: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5653: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5654: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5655: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5656: fclose(ficresprobmorprev);
5657: fflush(ficgp);
5658: fflush(fichtm);
5659: } /* end varevsij */
1.126 brouard 5660:
5661: /************ Variance of prevlim ******************/
1.235 brouard 5662: 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 5663: {
1.205 brouard 5664: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5665: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5666:
1.126 brouard 5667: double **dnewm,**doldm;
5668: int i, j, nhstepm, hstepm;
5669: double *xp;
5670: double *gp, *gm;
5671: double **gradg, **trgradg;
1.208 brouard 5672: double **mgm, **mgp;
1.126 brouard 5673: double age,agelim;
5674: int theta;
5675:
5676: pstamp(ficresvpl);
5677: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5678: fprintf(ficresvpl,"# Age ");
5679: if(nresult >=1)
5680: fprintf(ficresvpl," Result# ");
1.126 brouard 5681: for(i=1; i<=nlstate;i++)
5682: fprintf(ficresvpl," %1d-%1d",i,i);
5683: fprintf(ficresvpl,"\n");
5684:
5685: xp=vector(1,npar);
5686: dnewm=matrix(1,nlstate,1,npar);
5687: doldm=matrix(1,nlstate,1,nlstate);
5688:
5689: hstepm=1*YEARM; /* Every year of age */
5690: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5691: agelim = AGESUP;
5692: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5693: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5694: if (stepm >= YEARM) hstepm=1;
5695: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5696: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5697: mgp=matrix(1,npar,1,nlstate);
5698: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5699: gp=vector(1,nlstate);
5700: gm=vector(1,nlstate);
5701:
5702: for(theta=1; theta <=npar; theta++){
5703: for(i=1; i<=npar; i++){ /* Computes gradient */
5704: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5705: }
1.209 brouard 5706: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5707: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5708: else
1.235 brouard 5709: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5710: for(i=1;i<=nlstate;i++){
1.126 brouard 5711: gp[i] = prlim[i][i];
1.208 brouard 5712: mgp[theta][i] = prlim[i][i];
5713: }
1.126 brouard 5714: for(i=1; i<=npar; i++) /* Computes gradient */
5715: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5716: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5717: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5718: else
1.235 brouard 5719: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5720: for(i=1;i<=nlstate;i++){
1.126 brouard 5721: gm[i] = prlim[i][i];
1.208 brouard 5722: mgm[theta][i] = prlim[i][i];
5723: }
1.126 brouard 5724: for(i=1;i<=nlstate;i++)
5725: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5726: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5727: } /* End theta */
5728:
5729: trgradg =matrix(1,nlstate,1,npar);
5730:
5731: for(j=1; j<=nlstate;j++)
5732: for(theta=1; theta <=npar; theta++)
5733: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5734: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5735: /* printf("\nmgm mgp %d ",(int)age); */
5736: /* for(j=1; j<=nlstate;j++){ */
5737: /* printf(" %d ",j); */
5738: /* for(theta=1; theta <=npar; theta++) */
5739: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5740: /* printf("\n "); */
5741: /* } */
5742: /* } */
5743: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5744: /* printf("\n gradg %d ",(int)age); */
5745: /* for(j=1; j<=nlstate;j++){ */
5746: /* printf("%d ",j); */
5747: /* for(theta=1; theta <=npar; theta++) */
5748: /* printf("%d %lf ",theta,gradg[theta][j]); */
5749: /* printf("\n "); */
5750: /* } */
5751: /* } */
1.126 brouard 5752:
5753: for(i=1;i<=nlstate;i++)
5754: varpl[i][(int)age] =0.;
1.209 brouard 5755: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5756: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5757: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5758: }else{
1.126 brouard 5759: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5760: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5761: }
1.126 brouard 5762: for(i=1;i<=nlstate;i++)
5763: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5764:
5765: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5766: if(nresult >=1)
5767: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5768: for(i=1; i<=nlstate;i++)
5769: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5770: fprintf(ficresvpl,"\n");
5771: free_vector(gp,1,nlstate);
5772: free_vector(gm,1,nlstate);
1.208 brouard 5773: free_matrix(mgm,1,npar,1,nlstate);
5774: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5775: free_matrix(gradg,1,npar,1,nlstate);
5776: free_matrix(trgradg,1,nlstate,1,npar);
5777: } /* End age */
5778:
5779: free_vector(xp,1,npar);
5780: free_matrix(doldm,1,nlstate,1,npar);
5781: free_matrix(dnewm,1,nlstate,1,nlstate);
5782:
5783: }
5784:
5785: /************ Variance of one-step probabilities ******************/
5786: 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 5787: {
5788: int i, j=0, k1, l1, tj;
5789: int k2, l2, j1, z1;
5790: int k=0, l;
5791: int first=1, first1, first2;
5792: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5793: double **dnewm,**doldm;
5794: double *xp;
5795: double *gp, *gm;
5796: double **gradg, **trgradg;
5797: double **mu;
5798: double age, cov[NCOVMAX+1];
5799: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5800: int theta;
5801: char fileresprob[FILENAMELENGTH];
5802: char fileresprobcov[FILENAMELENGTH];
5803: char fileresprobcor[FILENAMELENGTH];
5804: double ***varpij;
5805:
5806: strcpy(fileresprob,"PROB_");
5807: strcat(fileresprob,fileres);
5808: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5809: printf("Problem with resultfile: %s\n", fileresprob);
5810: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5811: }
5812: strcpy(fileresprobcov,"PROBCOV_");
5813: strcat(fileresprobcov,fileresu);
5814: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5815: printf("Problem with resultfile: %s\n", fileresprobcov);
5816: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5817: }
5818: strcpy(fileresprobcor,"PROBCOR_");
5819: strcat(fileresprobcor,fileresu);
5820: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5821: printf("Problem with resultfile: %s\n", fileresprobcor);
5822: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5823: }
5824: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5825: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5826: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5827: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5828: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5829: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5830: pstamp(ficresprob);
5831: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5832: fprintf(ficresprob,"# Age");
5833: pstamp(ficresprobcov);
5834: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5835: fprintf(ficresprobcov,"# Age");
5836: pstamp(ficresprobcor);
5837: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5838: fprintf(ficresprobcor,"# Age");
1.126 brouard 5839:
5840:
1.222 brouard 5841: for(i=1; i<=nlstate;i++)
5842: for(j=1; j<=(nlstate+ndeath);j++){
5843: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5844: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5845: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5846: }
5847: /* fprintf(ficresprob,"\n");
5848: fprintf(ficresprobcov,"\n");
5849: fprintf(ficresprobcor,"\n");
5850: */
5851: xp=vector(1,npar);
5852: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5853: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5854: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5855: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5856: first=1;
5857: fprintf(ficgp,"\n# Routine varprob");
5858: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5859: fprintf(fichtm,"\n");
5860:
5861: 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);
5862: 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);
5863: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5864: and drawn. It helps understanding how is the covariance between two incidences.\
5865: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5866: 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 5867: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5868: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5869: standard deviations wide on each axis. <br>\
5870: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5871: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5872: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5873:
1.222 brouard 5874: cov[1]=1;
5875: /* tj=cptcoveff; */
1.225 brouard 5876: tj = (int) pow(2,cptcoveff);
1.222 brouard 5877: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5878: j1=0;
1.224 brouard 5879: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5880: if (cptcovn>0) {
5881: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5882: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5883: fprintf(ficresprob, "**********\n#\n");
5884: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5885: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5886: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5887:
1.222 brouard 5888: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5889: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5890: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5891:
5892:
1.222 brouard 5893: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5894: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5895: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5896:
1.222 brouard 5897: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5898: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5899: fprintf(ficresprobcor, "**********\n#");
5900: if(invalidvarcomb[j1]){
5901: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5902: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5903: continue;
5904: }
5905: }
5906: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5907: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5908: gp=vector(1,(nlstate)*(nlstate+ndeath));
5909: gm=vector(1,(nlstate)*(nlstate+ndeath));
5910: for (age=bage; age<=fage; age ++){
5911: cov[2]=age;
5912: if(nagesqr==1)
5913: cov[3]= age*age;
5914: for (k=1; k<=cptcovn;k++) {
5915: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5916: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5917: * 1 1 1 1 1
5918: * 2 2 1 1 1
5919: * 3 1 2 1 1
5920: */
5921: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5922: }
5923: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5924: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5925: for (k=1; k<=cptcovprod;k++)
5926: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5927:
5928:
1.222 brouard 5929: for(theta=1; theta <=npar; theta++){
5930: for(i=1; i<=npar; i++)
5931: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 5932:
1.222 brouard 5933: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 5934:
1.222 brouard 5935: k=0;
5936: for(i=1; i<= (nlstate); i++){
5937: for(j=1; j<=(nlstate+ndeath);j++){
5938: k=k+1;
5939: gp[k]=pmmij[i][j];
5940: }
5941: }
1.220 brouard 5942:
1.222 brouard 5943: for(i=1; i<=npar; i++)
5944: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 5945:
1.222 brouard 5946: pmij(pmmij,cov,ncovmodel,xp,nlstate);
5947: k=0;
5948: for(i=1; i<=(nlstate); i++){
5949: for(j=1; j<=(nlstate+ndeath);j++){
5950: k=k+1;
5951: gm[k]=pmmij[i][j];
5952: }
5953: }
1.220 brouard 5954:
1.222 brouard 5955: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
5956: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
5957: }
1.126 brouard 5958:
1.222 brouard 5959: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
5960: for(theta=1; theta <=npar; theta++)
5961: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 5962:
1.222 brouard 5963: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
5964: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 5965:
1.222 brouard 5966: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 5967:
1.222 brouard 5968: k=0;
5969: for(i=1; i<=(nlstate); i++){
5970: for(j=1; j<=(nlstate+ndeath);j++){
5971: k=k+1;
5972: mu[k][(int) age]=pmmij[i][j];
5973: }
5974: }
5975: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
5976: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
5977: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 5978:
1.222 brouard 5979: /*printf("\n%d ",(int)age);
5980: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5981: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5982: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5983: }*/
1.220 brouard 5984:
1.222 brouard 5985: fprintf(ficresprob,"\n%d ",(int)age);
5986: fprintf(ficresprobcov,"\n%d ",(int)age);
5987: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 5988:
1.222 brouard 5989: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
5990: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
5991: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5992: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
5993: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
5994: }
5995: i=0;
5996: for (k=1; k<=(nlstate);k++){
5997: for (l=1; l<=(nlstate+ndeath);l++){
5998: i++;
5999: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6000: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6001: for (j=1; j<=i;j++){
6002: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6003: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6004: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6005: }
6006: }
6007: }/* end of loop for state */
6008: } /* end of loop for age */
6009: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6010: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6011: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6012: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6013:
6014: /* Confidence intervalle of pij */
6015: /*
6016: fprintf(ficgp,"\nunset parametric;unset label");
6017: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6018: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6019: 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);
6020: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6021: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6022: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6023: */
6024:
6025: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6026: first1=1;first2=2;
6027: for (k2=1; k2<=(nlstate);k2++){
6028: for (l2=1; l2<=(nlstate+ndeath);l2++){
6029: if(l2==k2) continue;
6030: j=(k2-1)*(nlstate+ndeath)+l2;
6031: for (k1=1; k1<=(nlstate);k1++){
6032: for (l1=1; l1<=(nlstate+ndeath);l1++){
6033: if(l1==k1) continue;
6034: i=(k1-1)*(nlstate+ndeath)+l1;
6035: if(i<=j) continue;
6036: for (age=bage; age<=fage; age ++){
6037: if ((int)age %5==0){
6038: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6039: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6040: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6041: mu1=mu[i][(int) age]/stepm*YEARM ;
6042: mu2=mu[j][(int) age]/stepm*YEARM;
6043: c12=cv12/sqrt(v1*v2);
6044: /* Computing eigen value of matrix of covariance */
6045: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6046: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6047: if ((lc2 <0) || (lc1 <0) ){
6048: if(first2==1){
6049: first1=0;
6050: 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);
6051: }
6052: 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);
6053: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6054: /* lc2=fabs(lc2); */
6055: }
1.220 brouard 6056:
1.222 brouard 6057: /* Eigen vectors */
6058: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6059: /*v21=sqrt(1.-v11*v11); *//* error */
6060: v21=(lc1-v1)/cv12*v11;
6061: v12=-v21;
6062: v22=v11;
6063: tnalp=v21/v11;
6064: if(first1==1){
6065: first1=0;
6066: 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);
6067: }
6068: 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);
6069: /*printf(fignu*/
6070: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6071: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6072: if(first==1){
6073: first=0;
6074: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6075: fprintf(ficgp,"\nset parametric;unset label");
6076: 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);
6077: fprintf(ficgp,"\nset ter svg size 640, 480");
6078: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6079: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6080: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6081: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6082: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6083: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6084: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6085: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6086: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6087: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6088: 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", \
6089: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6090: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6091: }else{
6092: first=0;
6093: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6094: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6095: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6096: 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", \
6097: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6098: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6099: }/* if first */
6100: } /* age mod 5 */
6101: } /* end loop age */
6102: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6103: first=1;
6104: } /*l12 */
6105: } /* k12 */
6106: } /*l1 */
6107: }/* k1 */
6108: } /* loop on combination of covariates j1 */
6109: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6110: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6111: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6112: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6113: free_vector(xp,1,npar);
6114: fclose(ficresprob);
6115: fclose(ficresprobcov);
6116: fclose(ficresprobcor);
6117: fflush(ficgp);
6118: fflush(fichtmcov);
6119: }
1.126 brouard 6120:
6121:
6122: /******************* Printing html file ***********/
1.201 brouard 6123: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6124: int lastpass, int stepm, int weightopt, char model[],\
6125: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6126: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6127: double jprev1, double mprev1,double anprev1, double dateprev1, \
6128: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6129: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6130:
6131: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6132: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6133: </ul>");
1.237 brouard 6134: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6135: </ul>", model);
1.214 brouard 6136: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6137: 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",
6138: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6139: 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 6140: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6141: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6142: fprintf(fichtm,"\
6143: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6144: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6145: fprintf(fichtm,"\
1.217 brouard 6146: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6147: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6148: fprintf(fichtm,"\
1.126 brouard 6149: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6150: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6151: fprintf(fichtm,"\
1.217 brouard 6152: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6153: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6154: fprintf(fichtm,"\
1.211 brouard 6155: - (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 6156: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6157: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6158: if(prevfcast==1){
6159: fprintf(fichtm,"\
6160: - Prevalence projections by age and states: \
1.201 brouard 6161: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6162: }
1.126 brouard 6163:
1.222 brouard 6164: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6165:
1.225 brouard 6166: m=pow(2,cptcoveff);
1.222 brouard 6167: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6168:
1.222 brouard 6169: jj1=0;
1.237 brouard 6170:
6171: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6172: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6173: if(TKresult[nres]!= k1)
6174: continue;
1.220 brouard 6175:
1.222 brouard 6176: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6177: jj1++;
6178: if (cptcovn > 0) {
6179: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6180: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6181: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6182: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6183: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6184: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6185: }
1.237 brouard 6186: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6187: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6188: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6189: }
6190:
1.230 brouard 6191: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6192: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6193: if(invalidvarcomb[k1]){
6194: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6195: printf("\nCombination (%d) ignored because no cases \n",k1);
6196: continue;
6197: }
6198: }
6199: /* aij, bij */
1.241 brouard 6200: 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> \
6201: <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 6202: /* Pij */
1.241 brouard 6203: 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> \
6204: <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 6205: /* Quasi-incidences */
6206: 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 6207: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6208: 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 6209: 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> \
6210: <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 6211: /* Survival functions (period) in state j */
6212: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6213: 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> \
6214: <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 6215: }
6216: /* State specific survival functions (period) */
6217: for(cpt=1; cpt<=nlstate;cpt++){
6218: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6219: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6220: <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 6221: }
6222: /* Period (stable) prevalence in each health state */
6223: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6224: 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> \
6225: <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 6226: }
6227: if(backcast==1){
6228: /* Period (stable) back prevalence in each health state */
6229: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6230: 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> \
6231: <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 6232: }
1.217 brouard 6233: }
1.222 brouard 6234: if(prevfcast==1){
6235: /* Projection of prevalence up to period (stable) prevalence in each health state */
6236: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6237: 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> \
6238: <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 6239: }
6240: }
1.220 brouard 6241:
1.222 brouard 6242: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6243: 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> \
6244: <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 6245: }
6246: /* } /\* end i1 *\/ */
6247: }/* End k1 */
6248: fprintf(fichtm,"</ul>");
1.126 brouard 6249:
1.222 brouard 6250: fprintf(fichtm,"\
1.126 brouard 6251: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6252: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6253: - 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 6254: But because parameters are usually highly correlated (a higher incidence of disability \
6255: and a higher incidence of recovery can give very close observed transition) it might \
6256: be very useful to look not only at linear confidence intervals estimated from the \
6257: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6258: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6259: covariance matrix of the one-step probabilities. \
6260: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6261:
1.222 brouard 6262: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6263: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6264: fprintf(fichtm,"\
1.126 brouard 6265: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6266: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6267:
1.222 brouard 6268: fprintf(fichtm,"\
1.126 brouard 6269: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6270: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6271: fprintf(fichtm,"\
1.126 brouard 6272: - 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): \
6273: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6274: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6275: fprintf(fichtm,"\
1.126 brouard 6276: - (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): \
6277: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6278: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6279: fprintf(fichtm,"\
1.128 brouard 6280: - 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 6281: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6282: fprintf(fichtm,"\
1.128 brouard 6283: - 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 6284: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6285: fprintf(fichtm,"\
1.126 brouard 6286: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6287: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6288:
6289: /* if(popforecast==1) fprintf(fichtm,"\n */
6290: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6291: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6292: /* <br>",fileres,fileres,fileres,fileres); */
6293: /* else */
6294: /* 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 6295: fflush(fichtm);
6296: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6297:
1.225 brouard 6298: m=pow(2,cptcoveff);
1.222 brouard 6299: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6300:
1.222 brouard 6301: jj1=0;
1.237 brouard 6302:
1.241 brouard 6303: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6304: for(k1=1; k1<=m;k1++){
1.237 brouard 6305: if(TKresult[nres]!= k1)
6306: continue;
1.222 brouard 6307: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6308: jj1++;
1.126 brouard 6309: if (cptcovn > 0) {
6310: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6311: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6312: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6313: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6314: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6315: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6316: }
6317:
1.126 brouard 6318: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6319:
1.222 brouard 6320: if(invalidvarcomb[k1]){
6321: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6322: continue;
6323: }
1.126 brouard 6324: }
6325: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6326: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6327: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6328: <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 6329: }
6330: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6331: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6332: true period expectancies (those weighted with period prevalences are also\
6333: drawn in addition to the population based expectancies computed using\
1.241 brouard 6334: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6335: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6336: /* } /\* end i1 *\/ */
6337: }/* End k1 */
1.241 brouard 6338: }/* End nres */
1.222 brouard 6339: fprintf(fichtm,"</ul>");
6340: fflush(fichtm);
1.126 brouard 6341: }
6342:
6343: /******************* Gnuplot file **************/
1.223 brouard 6344: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6345:
6346: char dirfileres[132],optfileres[132];
1.223 brouard 6347: char gplotcondition[132];
1.237 brouard 6348: 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 6349: int lv=0, vlv=0, kl=0;
1.130 brouard 6350: int ng=0;
1.201 brouard 6351: int vpopbased;
1.223 brouard 6352: int ioffset; /* variable offset for columns */
1.235 brouard 6353: int nres=0; /* Index of resultline */
1.219 brouard 6354:
1.126 brouard 6355: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6356: /* printf("Problem with file %s",optionfilegnuplot); */
6357: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6358: /* } */
6359:
6360: /*#ifdef windows */
6361: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6362: /*#endif */
1.225 brouard 6363: m=pow(2,cptcoveff);
1.126 brouard 6364:
1.202 brouard 6365: /* Contribution to likelihood */
6366: /* Plot the probability implied in the likelihood */
1.223 brouard 6367: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6368: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6369: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6370: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6371: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6372: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6373: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6374: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6375: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6376: 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));
6377: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6378: 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));
6379: for (i=1; i<= nlstate ; i ++) {
6380: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6381: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6382: 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);
6383: for (j=2; j<= nlstate+ndeath ; j ++) {
6384: 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);
6385: }
6386: fprintf(ficgp,";\nset out; unset ylabel;\n");
6387: }
6388: /* 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 */
6389: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6390: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6391: fprintf(ficgp,"\nset out;unset log\n");
6392: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6393:
1.126 brouard 6394: strcpy(dirfileres,optionfilefiname);
6395: strcpy(optfileres,"vpl");
1.223 brouard 6396: /* 1eme*/
1.238 brouard 6397: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6398: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6399: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6400: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6401: if(TKresult[nres]!= k1)
6402: continue;
6403: /* We are interested in selected combination by the resultline */
1.246 ! brouard 6404: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6405: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6406: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6407: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6408: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6409: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6410: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6411: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6412: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 ! brouard 6413: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6414: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6415: }
6416: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 ! brouard 6417: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6418: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6419: }
1.246 ! brouard 6420: /* printf("\n#\n"); */
1.238 brouard 6421: fprintf(ficgp,"\n#\n");
6422: if(invalidvarcomb[k1]){
6423: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6424: continue;
6425: }
1.235 brouard 6426:
1.241 brouard 6427: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6428: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6429: 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 6430:
1.238 brouard 6431: for (i=1; i<= nlstate ; i ++) {
6432: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6433: else fprintf(ficgp," %%*lf (%%*lf)");
6434: }
1.242 brouard 6435: 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 6436: for (i=1; i<= nlstate ; i ++) {
6437: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6438: else fprintf(ficgp," %%*lf (%%*lf)");
6439: }
1.242 brouard 6440: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6441: for (i=1; i<= nlstate ; i ++) {
6442: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6443: else fprintf(ficgp," %%*lf (%%*lf)");
6444: }
6445: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6446: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6447: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6448: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6449: if(cptcoveff ==0){
1.245 brouard 6450: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6451: }else{
6452: kl=0;
6453: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6454: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6455: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6456: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6457: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6458: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6459: kl++;
1.238 brouard 6460: /* 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 *\/ */
6461: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6462: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6463: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6464: if(k==cptcoveff){
1.245 brouard 6465: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6466: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6467: }else{
6468: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6469: kl++;
6470: }
6471: } /* end covariate */
6472: } /* end if no covariate */
6473: } /* end if backcast */
6474: fprintf(ficgp,"\nset out \n");
6475: } /* nres */
1.201 brouard 6476: } /* k1 */
6477: } /* cpt */
1.235 brouard 6478:
6479:
1.126 brouard 6480: /*2 eme*/
1.238 brouard 6481: for (k1=1; k1<= m ; k1 ++){
6482: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6483: if(TKresult[nres]!= k1)
6484: continue;
6485: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6486: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6487: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6488: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6489: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6490: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6491: vlv= nbcode[Tvaraff[k]][lv];
6492: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6493: }
1.237 brouard 6494: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6495: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6496: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6497: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6498: }
1.211 brouard 6499: fprintf(ficgp,"\n#\n");
1.223 brouard 6500: if(invalidvarcomb[k1]){
6501: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6502: continue;
6503: }
1.219 brouard 6504:
1.241 brouard 6505: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6506: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6507: if(vpopbased==0)
6508: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6509: else
6510: fprintf(ficgp,"\nreplot ");
6511: for (i=1; i<= nlstate+1 ; i ++) {
6512: k=2*i;
6513: 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);
6514: for (j=1; j<= nlstate+1 ; j ++) {
6515: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6516: else fprintf(ficgp," %%*lf (%%*lf)");
6517: }
6518: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6519: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6520: 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);
6521: for (j=1; j<= nlstate+1 ; j ++) {
6522: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6523: else fprintf(ficgp," %%*lf (%%*lf)");
6524: }
6525: fprintf(ficgp,"\" t\"\" w l lt 0,");
6526: 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);
6527: for (j=1; j<= nlstate+1 ; j ++) {
6528: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6529: else fprintf(ficgp," %%*lf (%%*lf)");
6530: }
6531: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6532: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6533: } /* state */
6534: } /* vpopbased */
1.244 brouard 6535: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6536: } /* end nres */
6537: } /* k1 end 2 eme*/
6538:
6539:
6540: /*3eme*/
6541: for (k1=1; k1<= m ; k1 ++){
6542: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6543: if(TKresult[nres]!= k1)
1.238 brouard 6544: continue;
6545:
6546: for (cpt=1; cpt<= nlstate ; cpt ++) {
6547: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6548: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6549: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6550: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6551: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6552: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6553: vlv= nbcode[Tvaraff[k]][lv];
6554: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6555: }
6556: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6557: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6558: }
6559: fprintf(ficgp,"\n#\n");
6560: if(invalidvarcomb[k1]){
6561: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6562: continue;
6563: }
6564:
6565: /* k=2+nlstate*(2*cpt-2); */
6566: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6567: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6568: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6569: 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 6570: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6571: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6572: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6573: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6574: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6575: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6576:
1.238 brouard 6577: */
6578: for (i=1; i< nlstate ; i ++) {
6579: 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);
6580: /* 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 6581:
1.238 brouard 6582: }
6583: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6584: }
6585: } /* end nres */
6586: } /* end kl 3eme */
1.126 brouard 6587:
1.223 brouard 6588: /* 4eme */
1.201 brouard 6589: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6590: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6591: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6592: if(TKresult[nres]!= k1)
1.223 brouard 6593: continue;
1.238 brouard 6594: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6595: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6596: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6597: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6598: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6599: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6600: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6601: vlv= nbcode[Tvaraff[k]][lv];
6602: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6603: }
6604: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6605: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6606: }
6607: fprintf(ficgp,"\n#\n");
6608: if(invalidvarcomb[k1]){
6609: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6610: continue;
1.223 brouard 6611: }
1.238 brouard 6612:
1.241 brouard 6613: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6614: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6615: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6616: k=3;
6617: for (i=1; i<= nlstate ; i ++){
6618: if(i==1){
6619: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6620: }else{
6621: fprintf(ficgp,", '' ");
6622: }
6623: l=(nlstate+ndeath)*(i-1)+1;
6624: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6625: for (j=2; j<= nlstate+ndeath ; j ++)
6626: fprintf(ficgp,"+$%d",k+l+j-1);
6627: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6628: } /* nlstate */
6629: fprintf(ficgp,"\nset out\n");
6630: } /* end cpt state*/
6631: } /* end nres */
6632: } /* end covariate k1 */
6633:
1.220 brouard 6634: /* 5eme */
1.201 brouard 6635: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6636: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6637: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6638: if(TKresult[nres]!= k1)
1.227 brouard 6639: continue;
1.238 brouard 6640: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6641: 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);
6642: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6643: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6644: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6645: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6646: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6647: vlv= nbcode[Tvaraff[k]][lv];
6648: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6649: }
6650: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6651: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6652: }
6653: fprintf(ficgp,"\n#\n");
6654: if(invalidvarcomb[k1]){
6655: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6656: continue;
6657: }
1.227 brouard 6658:
1.241 brouard 6659: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6660: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6661: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6662: k=3;
6663: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6664: if(j==1)
6665: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6666: else
6667: fprintf(ficgp,", '' ");
6668: l=(nlstate+ndeath)*(cpt-1) +j;
6669: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6670: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6671: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6672: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6673: } /* nlstate */
6674: fprintf(ficgp,", '' ");
6675: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6676: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6677: l=(nlstate+ndeath)*(cpt-1) +j;
6678: if(j < nlstate)
6679: fprintf(ficgp,"$%d +",k+l);
6680: else
6681: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6682: }
6683: fprintf(ficgp,"\nset out\n");
6684: } /* end cpt state*/
6685: } /* end covariate */
6686: } /* end nres */
1.227 brouard 6687:
1.220 brouard 6688: /* 6eme */
1.202 brouard 6689: /* CV preval stable (period) for each covariate */
1.237 brouard 6690: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6691: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6692: if(TKresult[nres]!= k1)
6693: continue;
1.153 brouard 6694: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6695:
1.211 brouard 6696: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6697: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6698: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6699: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6700: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6701: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6702: vlv= nbcode[Tvaraff[k]][lv];
6703: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6704: }
1.237 brouard 6705: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6706: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6707: }
1.211 brouard 6708: fprintf(ficgp,"\n#\n");
1.223 brouard 6709: if(invalidvarcomb[k1]){
1.227 brouard 6710: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6711: continue;
1.223 brouard 6712: }
1.227 brouard 6713:
1.241 brouard 6714: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6715: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6716: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6717: k=3; /* Offset */
1.153 brouard 6718: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6719: if(i==1)
6720: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6721: else
6722: fprintf(ficgp,", '' ");
6723: l=(nlstate+ndeath)*(i-1)+1;
6724: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6725: for (j=2; j<= nlstate ; j ++)
6726: fprintf(ficgp,"+$%d",k+l+j-1);
6727: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6728: } /* nlstate */
1.201 brouard 6729: fprintf(ficgp,"\nset out\n");
1.153 brouard 6730: } /* end cpt state*/
6731: } /* end covariate */
1.227 brouard 6732:
6733:
1.220 brouard 6734: /* 7eme */
1.218 brouard 6735: if(backcast == 1){
1.217 brouard 6736: /* CV back preval stable (period) for each covariate */
1.237 brouard 6737: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6738: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6739: if(TKresult[nres]!= k1)
6740: continue;
1.218 brouard 6741: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6742: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6743: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6744: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6745: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6746: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6747: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6748: vlv= nbcode[Tvaraff[k]][lv];
6749: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6750: }
1.237 brouard 6751: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6752: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6753: }
1.227 brouard 6754: fprintf(ficgp,"\n#\n");
6755: if(invalidvarcomb[k1]){
6756: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6757: continue;
6758: }
6759:
1.241 brouard 6760: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6761: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6762: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6763: k=3; /* Offset */
6764: for (i=1; i<= nlstate ; i ++){
6765: if(i==1)
6766: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6767: else
6768: fprintf(ficgp,", '' ");
6769: /* l=(nlstate+ndeath)*(i-1)+1; */
6770: l=(nlstate+ndeath)*(cpt-1)+1;
6771: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6772: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6773: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6774: /* for (j=2; j<= nlstate ; j ++) */
6775: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6776: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6777: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6778: } /* nlstate */
6779: fprintf(ficgp,"\nset out\n");
1.218 brouard 6780: } /* end cpt state*/
6781: } /* end covariate */
6782: } /* End if backcast */
6783:
1.223 brouard 6784: /* 8eme */
1.218 brouard 6785: if(prevfcast==1){
6786: /* Projection from cross-sectional to stable (period) for each covariate */
6787:
1.237 brouard 6788: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6789: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6790: if(TKresult[nres]!= k1)
6791: continue;
1.211 brouard 6792: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6793: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6794: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6795: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6796: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6797: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6798: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6799: vlv= nbcode[Tvaraff[k]][lv];
6800: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6801: }
1.237 brouard 6802: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6803: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6804: }
1.227 brouard 6805: fprintf(ficgp,"\n#\n");
6806: if(invalidvarcomb[k1]){
6807: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6808: continue;
6809: }
6810:
6811: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6812: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6813: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6814: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6815: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6816: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6817: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6818: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6819: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6820: if(i==1){
6821: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6822: }else{
6823: fprintf(ficgp,",\\\n '' ");
6824: }
6825: if(cptcoveff ==0){ /* No covariate */
6826: ioffset=2; /* Age is in 2 */
6827: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6828: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6829: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6830: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6831: fprintf(ficgp," u %d:(", ioffset);
6832: if(i==nlstate+1)
6833: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6834: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6835: else
6836: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6837: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6838: }else{ /* more than 2 covariates */
6839: if(cptcoveff ==1){
6840: ioffset=4; /* Age is in 4 */
6841: }else{
6842: ioffset=6; /* Age is in 6 */
6843: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6844: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6845: }
6846: fprintf(ficgp," u %d:(",ioffset);
6847: kl=0;
6848: strcpy(gplotcondition,"(");
6849: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6850: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6851: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6852: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6853: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6854: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6855: kl++;
6856: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6857: kl++;
6858: if(k <cptcoveff && cptcoveff>1)
6859: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6860: }
6861: strcpy(gplotcondition+strlen(gplotcondition),")");
6862: /* 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 *\/ */
6863: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6864: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6865: /* '' 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*/
6866: if(i==nlstate+1){
6867: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6868: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6869: }else{
6870: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6871: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6872: }
6873: } /* end if covariate */
6874: } /* nlstate */
6875: fprintf(ficgp,"\nset out\n");
1.223 brouard 6876: } /* end cpt state*/
6877: } /* end covariate */
6878: } /* End if prevfcast */
1.227 brouard 6879:
6880:
1.238 brouard 6881: /* 9eme writing MLE parameters */
6882: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6883: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6884: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6885: for(k=1; k <=(nlstate+ndeath); k++){
6886: if (k != i) {
1.227 brouard 6887: fprintf(ficgp,"# current state %d\n",k);
6888: for(j=1; j <=ncovmodel; j++){
6889: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6890: jk++;
6891: }
6892: fprintf(ficgp,"\n");
1.126 brouard 6893: }
6894: }
1.223 brouard 6895: }
1.187 brouard 6896: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6897:
1.145 brouard 6898: /*goto avoid;*/
1.238 brouard 6899: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6900: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6901: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6902: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6903: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6904: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6905: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6906: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6907: fprintf(ficgp,"# p11=1/(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,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6910: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6911: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6912: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6913: fprintf(ficgp,"#\n");
1.223 brouard 6914: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6915: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6916: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6917: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6918: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6919: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6920: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6921: if(TKresult[nres]!= jk)
6922: continue;
6923: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6924: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6925: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6926: }
6927: fprintf(ficgp,"\n#\n");
1.241 brouard 6928: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 6929: fprintf(ficgp,"\nset ter svg size 640, 480 ");
6930: if (ng==1){
6931: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
6932: fprintf(ficgp,"\nunset log y");
6933: }else if (ng==2){
6934: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
6935: fprintf(ficgp,"\nset log y");
6936: }else if (ng==3){
6937: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
6938: fprintf(ficgp,"\nset log y");
6939: }else
6940: fprintf(ficgp,"\nunset title ");
6941: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
6942: i=1;
6943: for(k2=1; k2<=nlstate; k2++) {
6944: k3=i;
6945: for(k=1; k<=(nlstate+ndeath); k++) {
6946: if (k != k2){
6947: switch( ng) {
6948: case 1:
6949: if(nagesqr==0)
6950: fprintf(ficgp," p%d+p%d*x",i,i+1);
6951: else /* nagesqr =1 */
6952: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6953: break;
6954: case 2: /* ng=2 */
6955: if(nagesqr==0)
6956: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
6957: else /* nagesqr =1 */
6958: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6959: break;
6960: case 3:
6961: if(nagesqr==0)
6962: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
6963: else /* nagesqr =1 */
6964: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
6965: break;
6966: }
6967: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 6968: ijp=1; /* product no age */
6969: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
6970: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 6971: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 6972: if(j==Tage[ij]) { /* Product by age */
6973: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 6974: if(DummyV[j]==0){
1.237 brouard 6975: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
6976: }else{ /* quantitative */
6977: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
6978: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6979: }
6980: ij++;
6981: }
6982: }else if(j==Tprod[ijp]) { /* */
6983: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
6984: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 6985: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
6986: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 6987: /* 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)]); */
6988: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
6989: }else{ /* Vn is dummy and Vm is quanti */
6990: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
6991: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6992: }
6993: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 6994: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 6995: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
6996: }else{ /* Both quanti */
6997: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
6998: }
6999: }
1.238 brouard 7000: ijp++;
1.237 brouard 7001: }
7002: } else{ /* simple covariate */
7003: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7004: if(Dummy[j]==0){
7005: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7006: }else{ /* quantitative */
7007: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7008: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7009: }
1.237 brouard 7010: } /* end simple */
7011: } /* end j */
1.223 brouard 7012: }else{
7013: i=i-ncovmodel;
7014: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7015: fprintf(ficgp," (1.");
7016: }
1.227 brouard 7017:
1.223 brouard 7018: if(ng != 1){
7019: fprintf(ficgp,")/(1");
1.227 brouard 7020:
1.223 brouard 7021: for(k1=1; k1 <=nlstate; k1++){
7022: if(nagesqr==0)
7023: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7024: else /* nagesqr =1 */
7025: 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 7026:
1.223 brouard 7027: ij=1;
7028: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7029: if((j-2)==Tage[ij]) { /* Bug valgrind */
7030: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7031: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7032: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7033: ij++;
7034: }
7035: }
7036: else
1.225 brouard 7037: 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 7038: }
7039: fprintf(ficgp,")");
7040: }
7041: fprintf(ficgp,")");
7042: if(ng ==2)
7043: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7044: else /* ng= 3 */
7045: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7046: }else{ /* end ng <> 1 */
7047: if( k !=k2) /* logit p11 is hard to draw */
7048: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7049: }
7050: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7051: fprintf(ficgp,",");
7052: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7053: fprintf(ficgp,",");
7054: i=i+ncovmodel;
7055: } /* end k */
7056: } /* end k2 */
7057: fprintf(ficgp,"\n set out\n");
7058: } /* end jk */
7059: } /* end ng */
7060: /* avoid: */
7061: fflush(ficgp);
1.126 brouard 7062: } /* end gnuplot */
7063:
7064:
7065: /*************** Moving average **************/
1.219 brouard 7066: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7067: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7068:
1.222 brouard 7069: int i, cpt, cptcod;
7070: int modcovmax =1;
7071: int mobilavrange, mob;
7072: int iage=0;
7073:
7074: double sum=0.;
7075: double age;
7076: double *sumnewp, *sumnewm;
7077: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7078:
7079:
1.225 brouard 7080: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7081: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7082:
7083: sumnewp = vector(1,ncovcombmax);
7084: sumnewm = vector(1,ncovcombmax);
7085: agemingood = vector(1,ncovcombmax);
7086: agemaxgood = vector(1,ncovcombmax);
7087:
7088: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7089: sumnewm[cptcod]=0.;
7090: sumnewp[cptcod]=0.;
7091: agemingood[cptcod]=0;
7092: agemaxgood[cptcod]=0;
7093: }
7094: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7095:
7096: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7097: if(mobilav==1) mobilavrange=5; /* default */
7098: else mobilavrange=mobilav;
7099: for (age=bage; age<=fage; age++)
7100: for (i=1; i<=nlstate;i++)
7101: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7102: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7103: /* We keep the original values on the extreme ages bage, fage and for
7104: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7105: we use a 5 terms etc. until the borders are no more concerned.
7106: */
7107: for (mob=3;mob <=mobilavrange;mob=mob+2){
7108: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7109: for (i=1; i<=nlstate;i++){
7110: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7111: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7112: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7113: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7114: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7115: }
7116: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7117: }
7118: }
7119: }/* end age */
7120: }/* end mob */
7121: }else
7122: return -1;
7123: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7124: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7125: if(invalidvarcomb[cptcod]){
7126: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7127: continue;
7128: }
1.219 brouard 7129:
1.222 brouard 7130: agemingood[cptcod]=fage-(mob-1)/2;
7131: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7132: sumnewm[cptcod]=0.;
7133: for (i=1; i<=nlstate;i++){
7134: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7135: }
7136: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7137: agemingood[cptcod]=age;
7138: }else{ /* bad */
7139: for (i=1; i<=nlstate;i++){
7140: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7141: } /* i */
7142: } /* end bad */
7143: }/* age */
7144: sum=0.;
7145: for (i=1; i<=nlstate;i++){
7146: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7147: }
7148: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7149: 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);
7150: /* for (i=1; i<=nlstate;i++){ */
7151: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7152: /* } /\* i *\/ */
7153: } /* end bad */
7154: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7155: /* From youngest, finding the oldest wrong */
7156: agemaxgood[cptcod]=bage+(mob-1)/2;
7157: for (age=bage+(mob-1)/2; age<=fage; age++){
7158: sumnewm[cptcod]=0.;
7159: for (i=1; i<=nlstate;i++){
7160: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7161: }
7162: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7163: agemaxgood[cptcod]=age;
7164: }else{ /* bad */
7165: for (i=1; i<=nlstate;i++){
7166: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7167: } /* i */
7168: } /* end bad */
7169: }/* age */
7170: sum=0.;
7171: for (i=1; i<=nlstate;i++){
7172: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7173: }
7174: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7175: 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);
7176: /* for (i=1; i<=nlstate;i++){ */
7177: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7178: /* } /\* i *\/ */
7179: } /* end bad */
7180:
7181: for (age=bage; age<=fage; age++){
1.235 brouard 7182: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7183: sumnewp[cptcod]=0.;
7184: sumnewm[cptcod]=0.;
7185: for (i=1; i<=nlstate;i++){
7186: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7187: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7188: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7189: }
7190: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7191: }
7192: /* printf("\n"); */
7193: /* } */
7194: /* brutal averaging */
7195: for (i=1; i<=nlstate;i++){
7196: for (age=1; age<=bage; age++){
7197: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7198: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7199: }
7200: for (age=fage; age<=AGESUP; age++){
7201: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7202: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7203: }
7204: } /* end i status */
7205: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7206: for (age=1; age<=AGESUP; age++){
7207: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7208: mobaverage[(int)age][i][cptcod]=0.;
7209: }
7210: }
7211: }/* end cptcod */
7212: free_vector(sumnewm,1, ncovcombmax);
7213: free_vector(sumnewp,1, ncovcombmax);
7214: free_vector(agemaxgood,1, ncovcombmax);
7215: free_vector(agemingood,1, ncovcombmax);
7216: return 0;
7217: }/* End movingaverage */
1.218 brouard 7218:
1.126 brouard 7219:
7220: /************** Forecasting ******************/
1.235 brouard 7221: 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 7222: /* proj1, year, month, day of starting projection
7223: agemin, agemax range of age
7224: dateprev1 dateprev2 range of dates during which prevalence is computed
7225: anproj2 year of en of projection (same day and month as proj1).
7226: */
1.235 brouard 7227: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7228: double agec; /* generic age */
7229: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7230: double *popeffectif,*popcount;
7231: double ***p3mat;
1.218 brouard 7232: /* double ***mobaverage; */
1.126 brouard 7233: char fileresf[FILENAMELENGTH];
7234:
7235: agelim=AGESUP;
1.211 brouard 7236: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7237: in each health status at the date of interview (if between dateprev1 and dateprev2).
7238: We still use firstpass and lastpass as another selection.
7239: */
1.214 brouard 7240: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7241: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7242:
1.201 brouard 7243: strcpy(fileresf,"F_");
7244: strcat(fileresf,fileresu);
1.126 brouard 7245: if((ficresf=fopen(fileresf,"w"))==NULL) {
7246: printf("Problem with forecast resultfile: %s\n", fileresf);
7247: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7248: }
1.235 brouard 7249: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7250: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7251:
1.225 brouard 7252: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7253:
7254:
7255: stepsize=(int) (stepm+YEARM-1)/YEARM;
7256: if (stepm<=12) stepsize=1;
7257: if(estepm < stepm){
7258: printf ("Problem %d lower than %d\n",estepm, stepm);
7259: }
7260: else hstepm=estepm;
7261:
7262: hstepm=hstepm/stepm;
7263: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7264: fractional in yp1 */
7265: anprojmean=yp;
7266: yp2=modf((yp1*12),&yp);
7267: mprojmean=yp;
7268: yp1=modf((yp2*30.5),&yp);
7269: jprojmean=yp;
7270: if(jprojmean==0) jprojmean=1;
7271: if(mprojmean==0) jprojmean=1;
7272:
1.227 brouard 7273: i1=pow(2,cptcoveff);
1.126 brouard 7274: if (cptcovn < 1){i1=1;}
7275:
7276: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7277:
7278: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7279:
1.126 brouard 7280: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7281: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7282: for(k=1; k<=i1;k++){
7283: if(TKresult[nres]!= k)
7284: continue;
1.227 brouard 7285: if(invalidvarcomb[k]){
7286: printf("\nCombination (%d) projection ignored because no cases \n",k);
7287: continue;
7288: }
7289: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7290: for(j=1;j<=cptcoveff;j++) {
7291: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7292: }
1.235 brouard 7293: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7294: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7295: }
1.227 brouard 7296: fprintf(ficresf," yearproj age");
7297: for(j=1; j<=nlstate+ndeath;j++){
7298: for(i=1; i<=nlstate;i++)
7299: fprintf(ficresf," p%d%d",i,j);
7300: fprintf(ficresf," wp.%d",j);
7301: }
7302: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7303: fprintf(ficresf,"\n");
7304: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7305: for (agec=fage; agec>=(ageminpar-1); agec--){
7306: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7307: nhstepm = nhstepm/hstepm;
7308: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7309: oldm=oldms;savm=savms;
1.235 brouard 7310: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7311:
7312: for (h=0; h<=nhstepm; h++){
7313: if (h*hstepm/YEARM*stepm ==yearp) {
7314: fprintf(ficresf,"\n");
7315: for(j=1;j<=cptcoveff;j++)
7316: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7317: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7318: }
7319: for(j=1; j<=nlstate+ndeath;j++) {
7320: ppij=0.;
7321: for(i=1; i<=nlstate;i++) {
7322: if (mobilav==1)
7323: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7324: else {
7325: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7326: }
7327: if (h*hstepm/YEARM*stepm== yearp) {
7328: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7329: }
7330: } /* end i */
7331: if (h*hstepm/YEARM*stepm==yearp) {
7332: fprintf(ficresf," %.3f", ppij);
7333: }
7334: }/* end j */
7335: } /* end h */
7336: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7337: } /* end agec */
7338: } /* end yearp */
7339: } /* end k */
1.219 brouard 7340:
1.126 brouard 7341: fclose(ficresf);
1.215 brouard 7342: printf("End of Computing forecasting \n");
7343: fprintf(ficlog,"End of Computing forecasting\n");
7344:
1.126 brouard 7345: }
7346:
1.218 brouard 7347: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7348: /* 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 7349: /* /\* back1, year, month, day of starting backection */
7350: /* agemin, agemax range of age */
7351: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7352: /* anback2 year of en of backection (same day and month as back1). */
7353: /* *\/ */
7354: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7355: /* double agec; /\* generic age *\/ */
7356: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7357: /* double *popeffectif,*popcount; */
7358: /* double ***p3mat; */
7359: /* /\* double ***mobaverage; *\/ */
7360: /* char fileresfb[FILENAMELENGTH]; */
7361:
7362: /* agelim=AGESUP; */
7363: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7364: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7365: /* We still use firstpass and lastpass as another selection. */
7366: /* *\/ */
7367: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7368: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7369: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7370:
7371: /* strcpy(fileresfb,"FB_"); */
7372: /* strcat(fileresfb,fileresu); */
7373: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7374: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7375: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7376: /* } */
7377: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7378: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7379:
1.225 brouard 7380: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7381:
7382: /* /\* if (mobilav!=0) { *\/ */
7383: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7384: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7385: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7386: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7387: /* /\* } *\/ */
7388: /* /\* } *\/ */
7389:
7390: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7391: /* if (stepm<=12) stepsize=1; */
7392: /* if(estepm < stepm){ */
7393: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7394: /* } */
7395: /* else hstepm=estepm; */
7396:
7397: /* hstepm=hstepm/stepm; */
7398: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7399: /* fractional in yp1 *\/ */
7400: /* anprojmean=yp; */
7401: /* yp2=modf((yp1*12),&yp); */
7402: /* mprojmean=yp; */
7403: /* yp1=modf((yp2*30.5),&yp); */
7404: /* jprojmean=yp; */
7405: /* if(jprojmean==0) jprojmean=1; */
7406: /* if(mprojmean==0) jprojmean=1; */
7407:
1.225 brouard 7408: /* i1=cptcoveff; */
1.218 brouard 7409: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7410:
1.218 brouard 7411: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7412:
1.218 brouard 7413: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7414:
7415: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7416: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7417: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7418: /* k=k+1; */
7419: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7420: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7421: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7422: /* } */
7423: /* fprintf(ficresfb," yearbproj age"); */
7424: /* for(j=1; j<=nlstate+ndeath;j++){ */
7425: /* for(i=1; i<=nlstate;i++) */
7426: /* fprintf(ficresfb," p%d%d",i,j); */
7427: /* fprintf(ficresfb," p.%d",j); */
7428: /* } */
7429: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7430: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7431: /* fprintf(ficresfb,"\n"); */
7432: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7433: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7434: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7435: /* nhstepm = nhstepm/hstepm; */
7436: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7437: /* oldm=oldms;savm=savms; */
7438: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7439: /* for (h=0; h<=nhstepm; h++){ */
7440: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7441: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7442: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7443: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7444: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7445: /* } */
7446: /* for(j=1; j<=nlstate+ndeath;j++) { */
7447: /* ppij=0.; */
7448: /* for(i=1; i<=nlstate;i++) { */
7449: /* if (mobilav==1) */
7450: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7451: /* else { */
7452: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7453: /* } */
7454: /* if (h*hstepm/YEARM*stepm== yearp) { */
7455: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7456: /* } */
7457: /* } /\* end i *\/ */
7458: /* if (h*hstepm/YEARM*stepm==yearp) { */
7459: /* fprintf(ficresfb," %.3f", ppij); */
7460: /* } */
7461: /* }/\* end j *\/ */
7462: /* } /\* end h *\/ */
7463: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7464: /* } /\* end agec *\/ */
7465: /* } /\* end yearp *\/ */
7466: /* } /\* end cptcod *\/ */
7467: /* } /\* end cptcov *\/ */
7468:
7469: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7470:
7471: /* fclose(ficresfb); */
7472: /* printf("End of Computing Back forecasting \n"); */
7473: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7474:
1.218 brouard 7475: /* } */
1.217 brouard 7476:
1.126 brouard 7477: /************** Forecasting *****not tested NB*************/
1.227 brouard 7478: /* 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 7479:
1.227 brouard 7480: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7481: /* int *popage; */
7482: /* double calagedatem, agelim, kk1, kk2; */
7483: /* double *popeffectif,*popcount; */
7484: /* double ***p3mat,***tabpop,***tabpopprev; */
7485: /* /\* double ***mobaverage; *\/ */
7486: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7487:
1.227 brouard 7488: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7489: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7490: /* agelim=AGESUP; */
7491: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7492:
1.227 brouard 7493: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7494:
7495:
1.227 brouard 7496: /* strcpy(filerespop,"POP_"); */
7497: /* strcat(filerespop,fileresu); */
7498: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7499: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7500: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7501: /* } */
7502: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7503: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7504:
1.227 brouard 7505: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7506:
1.227 brouard 7507: /* /\* if (mobilav!=0) { *\/ */
7508: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7509: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7510: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7511: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7512: /* /\* } *\/ */
7513: /* /\* } *\/ */
1.126 brouard 7514:
1.227 brouard 7515: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7516: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7517:
1.227 brouard 7518: /* agelim=AGESUP; */
1.126 brouard 7519:
1.227 brouard 7520: /* hstepm=1; */
7521: /* hstepm=hstepm/stepm; */
1.218 brouard 7522:
1.227 brouard 7523: /* if (popforecast==1) { */
7524: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7525: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7526: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7527: /* } */
7528: /* popage=ivector(0,AGESUP); */
7529: /* popeffectif=vector(0,AGESUP); */
7530: /* popcount=vector(0,AGESUP); */
1.126 brouard 7531:
1.227 brouard 7532: /* i=1; */
7533: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7534:
1.227 brouard 7535: /* imx=i; */
7536: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7537: /* } */
1.218 brouard 7538:
1.227 brouard 7539: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7540: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7541: /* k=k+1; */
7542: /* fprintf(ficrespop,"\n#******"); */
7543: /* for(j=1;j<=cptcoveff;j++) { */
7544: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7545: /* } */
7546: /* fprintf(ficrespop,"******\n"); */
7547: /* fprintf(ficrespop,"# Age"); */
7548: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7549: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7550:
1.227 brouard 7551: /* for (cpt=0; cpt<=0;cpt++) { */
7552: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7553:
1.227 brouard 7554: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7555: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7556: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7557:
1.227 brouard 7558: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7559: /* oldm=oldms;savm=savms; */
7560: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7561:
1.227 brouard 7562: /* for (h=0; h<=nhstepm; h++){ */
7563: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7564: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7565: /* } */
7566: /* for(j=1; j<=nlstate+ndeath;j++) { */
7567: /* kk1=0.;kk2=0; */
7568: /* for(i=1; i<=nlstate;i++) { */
7569: /* if (mobilav==1) */
7570: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7571: /* else { */
7572: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7573: /* } */
7574: /* } */
7575: /* if (h==(int)(calagedatem+12*cpt)){ */
7576: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7577: /* /\*fprintf(ficrespop," %.3f", kk1); */
7578: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7579: /* } */
7580: /* } */
7581: /* for(i=1; i<=nlstate;i++){ */
7582: /* kk1=0.; */
7583: /* for(j=1; j<=nlstate;j++){ */
7584: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7585: /* } */
7586: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7587: /* } */
1.218 brouard 7588:
1.227 brouard 7589: /* if (h==(int)(calagedatem+12*cpt)) */
7590: /* for(j=1; j<=nlstate;j++) */
7591: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7592: /* } */
7593: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7594: /* } */
7595: /* } */
1.218 brouard 7596:
1.227 brouard 7597: /* /\******\/ */
1.218 brouard 7598:
1.227 brouard 7599: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7600: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7601: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7602: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7603: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7604:
1.227 brouard 7605: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7606: /* oldm=oldms;savm=savms; */
7607: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7608: /* for (h=0; h<=nhstepm; h++){ */
7609: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7610: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7611: /* } */
7612: /* for(j=1; j<=nlstate+ndeath;j++) { */
7613: /* kk1=0.;kk2=0; */
7614: /* for(i=1; i<=nlstate;i++) { */
7615: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7616: /* } */
7617: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7618: /* } */
7619: /* } */
7620: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7621: /* } */
7622: /* } */
7623: /* } */
7624: /* } */
1.218 brouard 7625:
1.227 brouard 7626: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7627:
1.227 brouard 7628: /* if (popforecast==1) { */
7629: /* free_ivector(popage,0,AGESUP); */
7630: /* free_vector(popeffectif,0,AGESUP); */
7631: /* free_vector(popcount,0,AGESUP); */
7632: /* } */
7633: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7634: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7635: /* fclose(ficrespop); */
7636: /* } /\* End of popforecast *\/ */
1.218 brouard 7637:
1.126 brouard 7638: int fileappend(FILE *fichier, char *optionfich)
7639: {
7640: if((fichier=fopen(optionfich,"a"))==NULL) {
7641: printf("Problem with file: %s\n", optionfich);
7642: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7643: return (0);
7644: }
7645: fflush(fichier);
7646: return (1);
7647: }
7648:
7649:
7650: /**************** function prwizard **********************/
7651: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7652: {
7653:
7654: /* Wizard to print covariance matrix template */
7655:
1.164 brouard 7656: char ca[32], cb[32];
7657: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7658: int numlinepar;
7659:
7660: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7661: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7662: for(i=1; i <=nlstate; i++){
7663: jj=0;
7664: for(j=1; j <=nlstate+ndeath; j++){
7665: if(j==i) continue;
7666: jj++;
7667: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7668: printf("%1d%1d",i,j);
7669: fprintf(ficparo,"%1d%1d",i,j);
7670: for(k=1; k<=ncovmodel;k++){
7671: /* printf(" %lf",param[i][j][k]); */
7672: /* fprintf(ficparo," %lf",param[i][j][k]); */
7673: printf(" 0.");
7674: fprintf(ficparo," 0.");
7675: }
7676: printf("\n");
7677: fprintf(ficparo,"\n");
7678: }
7679: }
7680: printf("# Scales (for hessian or gradient estimation)\n");
7681: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7682: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7683: for(i=1; i <=nlstate; i++){
7684: jj=0;
7685: for(j=1; j <=nlstate+ndeath; j++){
7686: if(j==i) continue;
7687: jj++;
7688: fprintf(ficparo,"%1d%1d",i,j);
7689: printf("%1d%1d",i,j);
7690: fflush(stdout);
7691: for(k=1; k<=ncovmodel;k++){
7692: /* printf(" %le",delti3[i][j][k]); */
7693: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7694: printf(" 0.");
7695: fprintf(ficparo," 0.");
7696: }
7697: numlinepar++;
7698: printf("\n");
7699: fprintf(ficparo,"\n");
7700: }
7701: }
7702: printf("# Covariance matrix\n");
7703: /* # 121 Var(a12)\n\ */
7704: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7705: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7706: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7707: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7708: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7709: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7710: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7711: fflush(stdout);
7712: fprintf(ficparo,"# Covariance matrix\n");
7713: /* # 121 Var(a12)\n\ */
7714: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7715: /* # ...\n\ */
7716: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7717:
7718: for(itimes=1;itimes<=2;itimes++){
7719: jj=0;
7720: for(i=1; i <=nlstate; i++){
7721: for(j=1; j <=nlstate+ndeath; j++){
7722: if(j==i) continue;
7723: for(k=1; k<=ncovmodel;k++){
7724: jj++;
7725: ca[0]= k+'a'-1;ca[1]='\0';
7726: if(itimes==1){
7727: printf("#%1d%1d%d",i,j,k);
7728: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7729: }else{
7730: printf("%1d%1d%d",i,j,k);
7731: fprintf(ficparo,"%1d%1d%d",i,j,k);
7732: /* printf(" %.5le",matcov[i][j]); */
7733: }
7734: ll=0;
7735: for(li=1;li <=nlstate; li++){
7736: for(lj=1;lj <=nlstate+ndeath; lj++){
7737: if(lj==li) continue;
7738: for(lk=1;lk<=ncovmodel;lk++){
7739: ll++;
7740: if(ll<=jj){
7741: cb[0]= lk +'a'-1;cb[1]='\0';
7742: if(ll<jj){
7743: if(itimes==1){
7744: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7745: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7746: }else{
7747: printf(" 0.");
7748: fprintf(ficparo," 0.");
7749: }
7750: }else{
7751: if(itimes==1){
7752: printf(" Var(%s%1d%1d)",ca,i,j);
7753: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7754: }else{
7755: printf(" 0.");
7756: fprintf(ficparo," 0.");
7757: }
7758: }
7759: }
7760: } /* end lk */
7761: } /* end lj */
7762: } /* end li */
7763: printf("\n");
7764: fprintf(ficparo,"\n");
7765: numlinepar++;
7766: } /* end k*/
7767: } /*end j */
7768: } /* end i */
7769: } /* end itimes */
7770:
7771: } /* end of prwizard */
7772: /******************* Gompertz Likelihood ******************************/
7773: double gompertz(double x[])
7774: {
7775: double A,B,L=0.0,sump=0.,num=0.;
7776: int i,n=0; /* n is the size of the sample */
7777:
1.220 brouard 7778: for (i=1;i<=imx ; i++) {
1.126 brouard 7779: sump=sump+weight[i];
7780: /* sump=sump+1;*/
7781: num=num+1;
7782: }
7783:
7784:
7785: /* for (i=0; i<=imx; i++)
7786: 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]);*/
7787:
7788: for (i=1;i<=imx ; i++)
7789: {
7790: if (cens[i] == 1 && wav[i]>1)
7791: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7792:
7793: if (cens[i] == 0 && wav[i]>1)
7794: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7795: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7796:
7797: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7798: if (wav[i] > 1 ) { /* ??? */
7799: L=L+A*weight[i];
7800: /* 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]);*/
7801: }
7802: }
7803:
7804: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7805:
7806: return -2*L*num/sump;
7807: }
7808:
1.136 brouard 7809: #ifdef GSL
7810: /******************* Gompertz_f Likelihood ******************************/
7811: double gompertz_f(const gsl_vector *v, void *params)
7812: {
7813: double A,B,LL=0.0,sump=0.,num=0.;
7814: double *x= (double *) v->data;
7815: int i,n=0; /* n is the size of the sample */
7816:
7817: for (i=0;i<=imx-1 ; i++) {
7818: sump=sump+weight[i];
7819: /* sump=sump+1;*/
7820: num=num+1;
7821: }
7822:
7823:
7824: /* for (i=0; i<=imx; i++)
7825: 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]);*/
7826: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7827: for (i=1;i<=imx ; i++)
7828: {
7829: if (cens[i] == 1 && wav[i]>1)
7830: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7831:
7832: if (cens[i] == 0 && wav[i]>1)
7833: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7834: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7835:
7836: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7837: if (wav[i] > 1 ) { /* ??? */
7838: LL=LL+A*weight[i];
7839: /* 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]);*/
7840: }
7841: }
7842:
7843: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7844: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7845:
7846: return -2*LL*num/sump;
7847: }
7848: #endif
7849:
1.126 brouard 7850: /******************* Printing html file ***********/
1.201 brouard 7851: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7852: int lastpass, int stepm, int weightopt, char model[],\
7853: int imx, double p[],double **matcov,double agemortsup){
7854: int i,k;
7855:
7856: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7857: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7858: for (i=1;i<=2;i++)
7859: 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 7860: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7861: fprintf(fichtm,"</ul>");
7862:
7863: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7864:
7865: 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>");
7866:
7867: for (k=agegomp;k<(agemortsup-2);k++)
7868: 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]);
7869:
7870:
7871: fflush(fichtm);
7872: }
7873:
7874: /******************* Gnuplot file **************/
1.201 brouard 7875: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7876:
7877: char dirfileres[132],optfileres[132];
1.164 brouard 7878:
1.126 brouard 7879: int ng;
7880:
7881:
7882: /*#ifdef windows */
7883: fprintf(ficgp,"cd \"%s\" \n",pathc);
7884: /*#endif */
7885:
7886:
7887: strcpy(dirfileres,optionfilefiname);
7888: strcpy(optfileres,"vpl");
1.199 brouard 7889: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7890: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7891: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7892: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7893: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7894:
7895: }
7896:
1.136 brouard 7897: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7898: {
1.126 brouard 7899:
1.136 brouard 7900: /*-------- data file ----------*/
7901: FILE *fic;
7902: char dummy[]=" ";
1.240 brouard 7903: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7904: int lstra;
1.136 brouard 7905: int linei, month, year,iout;
7906: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7907: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7908: char *stratrunc;
1.223 brouard 7909:
1.240 brouard 7910: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7911: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7912:
1.240 brouard 7913: for(v=1; v <=ncovcol;v++){
7914: DummyV[v]=0;
7915: FixedV[v]=0;
7916: }
7917: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7918: DummyV[v]=1;
7919: FixedV[v]=0;
7920: }
7921: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7922: DummyV[v]=0;
7923: FixedV[v]=1;
7924: }
7925: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7926: DummyV[v]=1;
7927: FixedV[v]=1;
7928: }
7929: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
7930: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
7931: 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]);
7932: }
1.126 brouard 7933:
1.136 brouard 7934: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 7935: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
7936: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 7937: }
1.126 brouard 7938:
1.136 brouard 7939: i=1;
7940: linei=0;
7941: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
7942: linei=linei+1;
7943: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
7944: if(line[j] == '\t')
7945: line[j] = ' ';
7946: }
7947: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
7948: ;
7949: };
7950: line[j+1]=0; /* Trims blanks at end of line */
7951: if(line[0]=='#'){
7952: fprintf(ficlog,"Comment line\n%s\n",line);
7953: printf("Comment line\n%s\n",line);
7954: continue;
7955: }
7956: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 7957: strcpy(line, linetmp);
1.223 brouard 7958:
7959: /* Loops on waves */
7960: for (j=maxwav;j>=1;j--){
7961: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 7962: cutv(stra, strb, line, ' ');
7963: if(strb[0]=='.') { /* Missing value */
7964: lval=-1;
7965: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
7966: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
7967: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
7968: 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);
7969: 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);
7970: return 1;
7971: }
7972: }else{
7973: errno=0;
7974: /* what_kind_of_number(strb); */
7975: dval=strtod(strb,&endptr);
7976: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
7977: /* if(strb != endptr && *endptr == '\0') */
7978: /* dval=dlval; */
7979: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
7980: if( strb[0]=='\0' || (*endptr != '\0')){
7981: 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);
7982: 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);
7983: return 1;
7984: }
7985: cotqvar[j][iv][i]=dval;
7986: cotvar[j][ntv+iv][i]=dval;
7987: }
7988: strcpy(line,stra);
1.223 brouard 7989: }/* end loop ntqv */
1.225 brouard 7990:
1.223 brouard 7991: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 7992: cutv(stra, strb, line, ' ');
7993: if(strb[0]=='.') { /* Missing value */
7994: lval=-1;
7995: }else{
7996: errno=0;
7997: lval=strtol(strb,&endptr,10);
7998: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
7999: if( strb[0]=='\0' || (*endptr != '\0')){
8000: 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);
8001: 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);
8002: return 1;
8003: }
8004: }
8005: if(lval <-1 || lval >1){
8006: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8007: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8008: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8009: For example, for multinomial values like 1, 2 and 3,\n \
8010: build V1=0 V2=0 for the reference value (1),\n \
8011: V1=1 V2=0 for (2) \n \
1.223 brouard 8012: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8013: output of IMaCh is often meaningless.\n \
1.223 brouard 8014: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8015: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8016: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8017: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8018: For example, for multinomial values like 1, 2 and 3,\n \
8019: build V1=0 V2=0 for the reference value (1),\n \
8020: V1=1 V2=0 for (2) \n \
1.223 brouard 8021: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8022: output of IMaCh is often meaningless.\n \
1.223 brouard 8023: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8024: return 1;
8025: }
8026: cotvar[j][iv][i]=(double)(lval);
8027: strcpy(line,stra);
1.223 brouard 8028: }/* end loop ntv */
1.225 brouard 8029:
1.223 brouard 8030: /* Statuses at wave */
1.137 brouard 8031: cutv(stra, strb, line, ' ');
1.223 brouard 8032: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8033: lval=-1;
1.136 brouard 8034: }else{
1.238 brouard 8035: errno=0;
8036: lval=strtol(strb,&endptr,10);
8037: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8038: if( strb[0]=='\0' || (*endptr != '\0')){
8039: 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);
8040: 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);
8041: return 1;
8042: }
1.136 brouard 8043: }
1.225 brouard 8044:
1.136 brouard 8045: s[j][i]=lval;
1.225 brouard 8046:
1.223 brouard 8047: /* Date of Interview */
1.136 brouard 8048: strcpy(line,stra);
8049: cutv(stra, strb,line,' ');
1.169 brouard 8050: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8051: }
1.169 brouard 8052: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8053: month=99;
8054: year=9999;
1.136 brouard 8055: }else{
1.225 brouard 8056: 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);
8057: 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);
8058: return 1;
1.136 brouard 8059: }
8060: anint[j][i]= (double) year;
8061: mint[j][i]= (double)month;
8062: strcpy(line,stra);
1.223 brouard 8063: } /* End loop on waves */
1.225 brouard 8064:
1.223 brouard 8065: /* Date of death */
1.136 brouard 8066: cutv(stra, strb,line,' ');
1.169 brouard 8067: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8068: }
1.169 brouard 8069: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8070: month=99;
8071: year=9999;
8072: }else{
1.141 brouard 8073: 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 8074: 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);
8075: return 1;
1.136 brouard 8076: }
8077: andc[i]=(double) year;
8078: moisdc[i]=(double) month;
8079: strcpy(line,stra);
8080:
1.223 brouard 8081: /* Date of birth */
1.136 brouard 8082: cutv(stra, strb,line,' ');
1.169 brouard 8083: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8084: }
1.169 brouard 8085: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8086: month=99;
8087: year=9999;
8088: }else{
1.141 brouard 8089: 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);
8090: 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 8091: return 1;
1.136 brouard 8092: }
8093: if (year==9999) {
1.141 brouard 8094: 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);
8095: 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 8096: return 1;
8097:
1.136 brouard 8098: }
8099: annais[i]=(double)(year);
8100: moisnais[i]=(double)(month);
8101: strcpy(line,stra);
1.225 brouard 8102:
1.223 brouard 8103: /* Sample weight */
1.136 brouard 8104: cutv(stra, strb,line,' ');
8105: errno=0;
8106: dval=strtod(strb,&endptr);
8107: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8108: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8109: 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 8110: fflush(ficlog);
8111: return 1;
8112: }
8113: weight[i]=dval;
8114: strcpy(line,stra);
1.225 brouard 8115:
1.223 brouard 8116: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8117: cutv(stra, strb, line, ' ');
8118: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8119: lval=-1;
1.223 brouard 8120: }else{
1.225 brouard 8121: errno=0;
8122: /* what_kind_of_number(strb); */
8123: dval=strtod(strb,&endptr);
8124: /* if(strb != endptr && *endptr == '\0') */
8125: /* dval=dlval; */
8126: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8127: if( strb[0]=='\0' || (*endptr != '\0')){
8128: 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);
8129: 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);
8130: return 1;
8131: }
8132: coqvar[iv][i]=dval;
1.226 brouard 8133: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8134: }
8135: strcpy(line,stra);
8136: }/* end loop nqv */
1.136 brouard 8137:
1.223 brouard 8138: /* Covariate values */
1.136 brouard 8139: for (j=ncovcol;j>=1;j--){
8140: cutv(stra, strb,line,' ');
1.223 brouard 8141: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8142: lval=-1;
1.136 brouard 8143: }else{
1.225 brouard 8144: errno=0;
8145: lval=strtol(strb,&endptr,10);
8146: if( strb[0]=='\0' || (*endptr != '\0')){
8147: 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);
8148: 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);
8149: return 1;
8150: }
1.136 brouard 8151: }
8152: if(lval <-1 || lval >1){
1.225 brouard 8153: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8154: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8155: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8156: For example, for multinomial values like 1, 2 and 3,\n \
8157: build V1=0 V2=0 for the reference value (1),\n \
8158: V1=1 V2=0 for (2) \n \
1.136 brouard 8159: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8160: output of IMaCh is often meaningless.\n \
1.136 brouard 8161: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8162: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8163: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8164: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8165: For example, for multinomial values like 1, 2 and 3,\n \
8166: build V1=0 V2=0 for the reference value (1),\n \
8167: V1=1 V2=0 for (2) \n \
1.136 brouard 8168: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8169: output of IMaCh is often meaningless.\n \
1.136 brouard 8170: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8171: return 1;
1.136 brouard 8172: }
8173: covar[j][i]=(double)(lval);
8174: strcpy(line,stra);
8175: }
8176: lstra=strlen(stra);
1.225 brouard 8177:
1.136 brouard 8178: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8179: stratrunc = &(stra[lstra-9]);
8180: num[i]=atol(stratrunc);
8181: }
8182: else
8183: num[i]=atol(stra);
8184: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8185: 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;}*/
8186:
8187: i=i+1;
8188: } /* End loop reading data */
1.225 brouard 8189:
1.136 brouard 8190: *imax=i-1; /* Number of individuals */
8191: fclose(fic);
1.225 brouard 8192:
1.136 brouard 8193: return (0);
1.164 brouard 8194: /* endread: */
1.225 brouard 8195: printf("Exiting readdata: ");
8196: fclose(fic);
8197: return (1);
1.223 brouard 8198: }
1.126 brouard 8199:
1.234 brouard 8200: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8201: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8202: while (*p2 == ' ')
1.234 brouard 8203: p2++;
8204: /* while ((*p1++ = *p2++) !=0) */
8205: /* ; */
8206: /* do */
8207: /* while (*p2 == ' ') */
8208: /* p2++; */
8209: /* while (*p1++ == *p2++); */
8210: *stri=p2;
1.145 brouard 8211: }
8212:
1.235 brouard 8213: int decoderesult ( char resultline[], int nres)
1.230 brouard 8214: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8215: {
1.235 brouard 8216: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8217: char resultsav[MAXLINE];
1.234 brouard 8218: int resultmodel[MAXLINE];
8219: int modelresult[MAXLINE];
1.230 brouard 8220: char stra[80], strb[80], strc[80], strd[80],stre[80];
8221:
1.234 brouard 8222: removefirstspace(&resultline);
1.233 brouard 8223: printf("decoderesult:%s\n",resultline);
1.230 brouard 8224:
8225: if (strstr(resultline,"v") !=0){
8226: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8227: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8228: return 1;
8229: }
8230: trimbb(resultsav, resultline);
8231: if (strlen(resultsav) >1){
8232: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8233: }
1.234 brouard 8234: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8235: 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);
8236: 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);
8237: }
8238: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8239: if(nbocc(resultsav,'=') >1){
8240: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8241: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8242: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8243: }else
8244: cutl(strc,strd,resultsav,'=');
1.230 brouard 8245: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8246:
1.230 brouard 8247: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8248: Tvarsel[k]=atoi(strc);
8249: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8250: /* cptcovsel++; */
8251: if (nbocc(stra,'=') >0)
8252: strcpy(resultsav,stra); /* and analyzes it */
8253: }
1.235 brouard 8254: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8255: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8256: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8257: match=0;
1.236 brouard 8258: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8259: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8260: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8261: match=1;
8262: break;
8263: }
8264: }
8265: if(match == 0){
8266: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8267: }
8268: }
8269: }
1.235 brouard 8270: /* Checking for missing or useless values in comparison of current model needs */
8271: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8272: match=0;
1.235 brouard 8273: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8274: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8275: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8276: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8277: ++match;
8278: }
8279: }
8280: }
8281: if(match == 0){
8282: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8283: }else if(match > 1){
8284: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8285: }
8286: }
1.235 brouard 8287:
1.234 brouard 8288: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8289: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8290: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8291: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8292: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8293: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8294: /* 1 0 0 0 */
8295: /* 2 1 0 0 */
8296: /* 3 0 1 0 */
8297: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8298: /* 5 0 0 1 */
8299: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8300: /* 7 0 1 1 */
8301: /* 8 1 1 1 */
1.237 brouard 8302: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8303: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8304: /* V5*age V5 known which value for nres? */
8305: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8306: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8307: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8308: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8309: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8310: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8311: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8312: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8313: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8314: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8315: k4++;;
8316: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8317: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8318: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8319: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8320: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8321: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8322: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8323: k4q++;;
8324: }
8325: }
1.234 brouard 8326:
1.235 brouard 8327: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8328: return (0);
8329: }
1.235 brouard 8330:
1.230 brouard 8331: int decodemodel( char model[], int lastobs)
8332: /**< This routine decodes the model and returns:
1.224 brouard 8333: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8334: * - nagesqr = 1 if age*age in the model, otherwise 0.
8335: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8336: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8337: * - cptcovage number of covariates with age*products =2
8338: * - cptcovs number of simple covariates
8339: * - 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
8340: * which is a new column after the 9 (ncovcol) variables.
8341: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8342: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8343: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8344: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8345: */
1.136 brouard 8346: {
1.238 brouard 8347: int i, j, k, ks, v;
1.227 brouard 8348: int j1, k1, k2, k3, k4;
1.136 brouard 8349: char modelsav[80];
1.145 brouard 8350: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8351: char *strpt;
1.136 brouard 8352:
1.145 brouard 8353: /*removespace(model);*/
1.136 brouard 8354: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8355: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8356: if (strstr(model,"AGE") !=0){
1.192 brouard 8357: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8358: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8359: return 1;
8360: }
1.141 brouard 8361: if (strstr(model,"v") !=0){
8362: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8363: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8364: return 1;
8365: }
1.187 brouard 8366: strcpy(modelsav,model);
8367: if ((strpt=strstr(model,"age*age")) !=0){
8368: printf(" strpt=%s, model=%s\n",strpt, model);
8369: if(strpt != model){
1.234 brouard 8370: printf("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);
1.234 brouard 8373: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8374: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8375: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8376: return 1;
1.225 brouard 8377: }
1.187 brouard 8378: nagesqr=1;
8379: if (strstr(model,"+age*age") !=0)
1.234 brouard 8380: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8381: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8382: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8383: else
1.234 brouard 8384: substrchaine(modelsav, model, "age*age");
1.187 brouard 8385: }else
8386: nagesqr=0;
8387: if (strlen(modelsav) >1){
8388: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8389: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8390: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8391: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8392: * cst, age and age*age
8393: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8394: /* including age products which are counted in cptcovage.
8395: * but the covariates which are products must be treated
8396: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8397: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8398: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8399:
8400:
1.187 brouard 8401: /* Design
8402: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8403: * < ncovcol=8 >
8404: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8405: * k= 1 2 3 4 5 6 7 8
8406: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8407: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8408: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8409: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8410: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8411: * Tage[++cptcovage]=k
8412: * if products, new covar are created after ncovcol with k1
8413: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8414: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8415: * 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
8416: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8417: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8418: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8419: * < ncovcol=8 >
8420: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8421: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8422: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8423: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8424: * p Tprod[1]@2={ 6, 5}
8425: *p Tvard[1][1]@4= {7, 8, 5, 6}
8426: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8427: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8428: *How to reorganize?
8429: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8430: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8431: * {2, 1, 4, 8, 5, 6, 3, 7}
8432: * Struct []
8433: */
1.225 brouard 8434:
1.187 brouard 8435: /* This loop fills the array Tvar from the string 'model'.*/
8436: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8437: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8438: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8439: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8440: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8441: /* k=1 Tvar[1]=2 (from V2) */
8442: /* k=5 Tvar[5] */
8443: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8444: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8445: /* } */
1.198 brouard 8446: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8447: /*
8448: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8449: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8450: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8451: }
1.187 brouard 8452: cptcovage=0;
8453: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8454: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8455: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8456: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8457: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8458: /*scanf("%d",i);*/
8459: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8460: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8461: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8462: /* covar is not filled and then is empty */
8463: cptcovprod--;
8464: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8465: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8466: Typevar[k]=1; /* 1 for age product */
8467: cptcovage++; /* Sums the number of covariates which include age as a product */
8468: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8469: /*printf("stre=%s ", stre);*/
8470: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8471: cptcovprod--;
8472: cutl(stre,strb,strc,'V');
8473: Tvar[k]=atoi(stre);
8474: Typevar[k]=1; /* 1 for age product */
8475: cptcovage++;
8476: Tage[cptcovage]=k;
8477: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8478: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8479: cptcovn++;
8480: cptcovprodnoage++;k1++;
8481: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8482: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8483: because this model-covariate is a construction we invent a new column
8484: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8485: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8486: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8487: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8488: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8489: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8490: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8491: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8492: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8493: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8494: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8495: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8496: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8497: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8498: for (i=1; i<=lastobs;i++){
8499: /* Computes the new covariate which is a product of
8500: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8501: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8502: }
8503: } /* End age is not in the model */
8504: } /* End if model includes a product */
8505: else { /* no more sum */
8506: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8507: /* scanf("%d",i);*/
8508: cutl(strd,strc,strb,'V');
8509: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8510: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8511: Tvar[k]=atoi(strd);
8512: Typevar[k]=0; /* 0 for simple covariates */
8513: }
8514: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8515: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8516: scanf("%d",i);*/
1.187 brouard 8517: } /* end of loop + on total covariates */
8518: } /* end if strlen(modelsave == 0) age*age might exist */
8519: } /* end if strlen(model == 0) */
1.136 brouard 8520:
8521: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8522: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8523:
1.136 brouard 8524: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8525: printf("cptcovprod=%d ", cptcovprod);
8526: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8527: scanf("%d ",i);*/
8528:
8529:
1.230 brouard 8530: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8531: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8532: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8533: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8534: k = 1 2 3 4 5 6 7 8 9
8535: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8536: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8537: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8538: Dummy[k] 1 0 0 0 3 1 1 2 3
8539: Tmodelind[combination of covar]=k;
1.225 brouard 8540: */
8541: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8542: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8543: /* 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 8544: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8545: printf("Model=%s\n\
8546: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8547: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8548: 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);
8549: fprintf(ficlog,"Model=%s\n\
8550: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8551: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8552: 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 8553: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8554: 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 */
8555: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8556: Fixed[k]= 0;
8557: Dummy[k]= 0;
1.225 brouard 8558: ncoveff++;
1.232 brouard 8559: ncovf++;
1.234 brouard 8560: nsd++;
8561: modell[k].maintype= FTYPE;
8562: TvarsD[nsd]=Tvar[k];
8563: TvarsDind[nsd]=k;
8564: TvarF[ncovf]=Tvar[k];
8565: TvarFind[ncovf]=k;
8566: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8567: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8568: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8569: Fixed[k]= 0;
8570: Dummy[k]= 0;
8571: ncoveff++;
8572: ncovf++;
8573: modell[k].maintype= FTYPE;
8574: TvarF[ncovf]=Tvar[k];
8575: TvarFind[ncovf]=k;
1.230 brouard 8576: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8577: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8578: }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 8579: Fixed[k]= 0;
8580: Dummy[k]= 1;
1.230 brouard 8581: nqfveff++;
1.234 brouard 8582: modell[k].maintype= FTYPE;
8583: modell[k].subtype= FQ;
8584: nsq++;
8585: TvarsQ[nsq]=Tvar[k];
8586: TvarsQind[nsq]=k;
1.232 brouard 8587: ncovf++;
1.234 brouard 8588: TvarF[ncovf]=Tvar[k];
8589: TvarFind[ncovf]=k;
1.231 brouard 8590: 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 8591: 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 8592: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8593: Fixed[k]= 1;
8594: Dummy[k]= 0;
1.225 brouard 8595: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8596: modell[k].maintype= VTYPE;
8597: modell[k].subtype= VD;
8598: nsd++;
8599: TvarsD[nsd]=Tvar[k];
8600: TvarsDind[nsd]=k;
8601: ncovv++; /* Only simple time varying variables */
8602: TvarV[ncovv]=Tvar[k];
1.242 brouard 8603: 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 8604: 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 */
8605: 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 8606: 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);
8607: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8608: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8609: Fixed[k]= 1;
8610: Dummy[k]= 1;
8611: nqtveff++;
8612: modell[k].maintype= VTYPE;
8613: modell[k].subtype= VQ;
8614: ncovv++; /* Only simple time varying variables */
8615: nsq++;
8616: TvarsQ[nsq]=Tvar[k];
8617: TvarsQind[nsq]=k;
8618: TvarV[ncovv]=Tvar[k];
1.242 brouard 8619: 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 8620: 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 */
8621: 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 8622: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8623: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8624: 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 8625: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8626: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8627: ncova++;
8628: TvarA[ncova]=Tvar[k];
8629: TvarAind[ncova]=k;
1.231 brouard 8630: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8631: Fixed[k]= 2;
8632: Dummy[k]= 2;
8633: modell[k].maintype= ATYPE;
8634: modell[k].subtype= APFD;
8635: /* ncoveff++; */
1.227 brouard 8636: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8637: Fixed[k]= 2;
8638: Dummy[k]= 3;
8639: modell[k].maintype= ATYPE;
8640: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8641: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8642: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8643: Fixed[k]= 3;
8644: Dummy[k]= 2;
8645: modell[k].maintype= ATYPE;
8646: modell[k].subtype= APVD; /* Product age * varying dummy */
8647: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8648: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8649: Fixed[k]= 3;
8650: Dummy[k]= 3;
8651: modell[k].maintype= ATYPE;
8652: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8653: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8654: }
8655: }else if (Typevar[k] == 2) { /* product without age */
8656: k1=Tposprod[k];
8657: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8658: if(Tvard[k1][2] <=ncovcol){
8659: Fixed[k]= 1;
8660: Dummy[k]= 0;
8661: modell[k].maintype= FTYPE;
8662: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8663: ncovf++; /* Fixed variables without age */
8664: TvarF[ncovf]=Tvar[k];
8665: TvarFind[ncovf]=k;
8666: }else if(Tvard[k1][2] <=ncovcol+nqv){
8667: Fixed[k]= 0; /* or 2 ?*/
8668: Dummy[k]= 1;
8669: modell[k].maintype= FTYPE;
8670: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8671: ncovf++; /* Varying variables without age */
8672: TvarF[ncovf]=Tvar[k];
8673: TvarFind[ncovf]=k;
8674: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8675: Fixed[k]= 1;
8676: Dummy[k]= 0;
8677: modell[k].maintype= VTYPE;
8678: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8679: ncovv++; /* Varying variables without age */
8680: TvarV[ncovv]=Tvar[k];
8681: TvarVind[ncovv]=k;
8682: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8683: Fixed[k]= 1;
8684: Dummy[k]= 1;
8685: modell[k].maintype= VTYPE;
8686: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8687: ncovv++; /* Varying variables without age */
8688: TvarV[ncovv]=Tvar[k];
8689: TvarVind[ncovv]=k;
8690: }
1.227 brouard 8691: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8692: if(Tvard[k1][2] <=ncovcol){
8693: Fixed[k]= 0; /* or 2 ?*/
8694: Dummy[k]= 1;
8695: modell[k].maintype= FTYPE;
8696: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8697: ncovf++; /* Fixed variables without age */
8698: TvarF[ncovf]=Tvar[k];
8699: TvarFind[ncovf]=k;
8700: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8701: Fixed[k]= 1;
8702: Dummy[k]= 1;
8703: modell[k].maintype= VTYPE;
8704: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8705: ncovv++; /* Varying variables without age */
8706: TvarV[ncovv]=Tvar[k];
8707: TvarVind[ncovv]=k;
8708: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8709: Fixed[k]= 1;
8710: Dummy[k]= 1;
8711: modell[k].maintype= VTYPE;
8712: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8713: ncovv++; /* Varying variables without age */
8714: TvarV[ncovv]=Tvar[k];
8715: TvarVind[ncovv]=k;
8716: ncovv++; /* Varying variables without age */
8717: TvarV[ncovv]=Tvar[k];
8718: TvarVind[ncovv]=k;
8719: }
1.227 brouard 8720: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8721: if(Tvard[k1][2] <=ncovcol){
8722: Fixed[k]= 1;
8723: Dummy[k]= 1;
8724: modell[k].maintype= VTYPE;
8725: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8726: ncovv++; /* Varying variables without age */
8727: TvarV[ncovv]=Tvar[k];
8728: TvarVind[ncovv]=k;
8729: }else if(Tvard[k1][2] <=ncovcol+nqv){
8730: Fixed[k]= 1;
8731: Dummy[k]= 1;
8732: modell[k].maintype= VTYPE;
8733: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8734: ncovv++; /* Varying variables without age */
8735: TvarV[ncovv]=Tvar[k];
8736: TvarVind[ncovv]=k;
8737: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8738: Fixed[k]= 1;
8739: Dummy[k]= 0;
8740: modell[k].maintype= VTYPE;
8741: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8742: ncovv++; /* Varying variables without age */
8743: TvarV[ncovv]=Tvar[k];
8744: TvarVind[ncovv]=k;
8745: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8746: Fixed[k]= 1;
8747: Dummy[k]= 1;
8748: modell[k].maintype= VTYPE;
8749: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8750: ncovv++; /* Varying variables without age */
8751: TvarV[ncovv]=Tvar[k];
8752: TvarVind[ncovv]=k;
8753: }
1.227 brouard 8754: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8755: if(Tvard[k1][2] <=ncovcol){
8756: Fixed[k]= 1;
8757: Dummy[k]= 1;
8758: modell[k].maintype= VTYPE;
8759: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8760: ncovv++; /* Varying variables without age */
8761: TvarV[ncovv]=Tvar[k];
8762: TvarVind[ncovv]=k;
8763: }else if(Tvard[k1][2] <=ncovcol+nqv){
8764: Fixed[k]= 1;
8765: Dummy[k]= 1;
8766: modell[k].maintype= VTYPE;
8767: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8768: ncovv++; /* Varying variables without age */
8769: TvarV[ncovv]=Tvar[k];
8770: TvarVind[ncovv]=k;
8771: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8772: Fixed[k]= 1;
8773: Dummy[k]= 1;
8774: modell[k].maintype= VTYPE;
8775: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8776: ncovv++; /* Varying variables without age */
8777: TvarV[ncovv]=Tvar[k];
8778: TvarVind[ncovv]=k;
8779: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8780: Fixed[k]= 1;
8781: Dummy[k]= 1;
8782: modell[k].maintype= VTYPE;
8783: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8784: ncovv++; /* Varying variables without age */
8785: TvarV[ncovv]=Tvar[k];
8786: TvarVind[ncovv]=k;
8787: }
1.227 brouard 8788: }else{
1.240 brouard 8789: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8790: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8791: } /*end k1*/
1.225 brouard 8792: }else{
1.226 brouard 8793: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8794: 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 8795: }
1.227 brouard 8796: 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 8797: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8798: 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]);
8799: }
8800: /* Searching for doublons in the model */
8801: for(k1=1; k1<= cptcovt;k1++){
8802: for(k2=1; k2 <k1;k2++){
8803: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8804: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8805: if(Tvar[k1]==Tvar[k2]){
8806: 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]]);
8807: 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);
8808: return(1);
8809: }
8810: }else if (Typevar[k1] ==2){
8811: k3=Tposprod[k1];
8812: k4=Tposprod[k2];
8813: 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])) ){
8814: 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]]);
8815: 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);
8816: return(1);
8817: }
8818: }
1.227 brouard 8819: }
8820: }
1.225 brouard 8821: }
8822: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8823: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8824: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8825: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8826: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8827: /*endread:*/
1.225 brouard 8828: printf("Exiting decodemodel: ");
8829: return (1);
1.136 brouard 8830: }
8831:
1.169 brouard 8832: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.136 brouard 8833: {
8834: int i, m;
1.218 brouard 8835: int firstone=0;
8836:
1.136 brouard 8837: for (i=1; i<=imx; i++) {
8838: for(m=2; (m<= maxwav); m++) {
8839: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8840: anint[m][i]=9999;
1.216 brouard 8841: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8842: s[m][i]=-1;
1.136 brouard 8843: }
8844: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8845: *nberr = *nberr + 1;
1.218 brouard 8846: if(firstone == 0){
8847: firstone=1;
8848: 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);
8849: }
8850: 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 8851: s[m][i]=-1;
8852: }
8853: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8854: (*nberr)++;
1.136 brouard 8855: 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]);
8856: 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]);
8857: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8858: }
8859: }
8860: }
8861:
8862: for (i=1; i<=imx; i++) {
8863: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8864: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8865: 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 8866: if (s[m][i] >= nlstate+1) {
1.169 brouard 8867: if(agedc[i]>0){
8868: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8869: agev[m][i]=agedc[i];
1.214 brouard 8870: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8871: }else {
1.136 brouard 8872: if ((int)andc[i]!=9999){
8873: nbwarn++;
8874: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8875: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8876: agev[m][i]=-1;
8877: }
8878: }
1.169 brouard 8879: } /* agedc > 0 */
1.214 brouard 8880: } /* end if */
1.136 brouard 8881: else if(s[m][i] !=9){ /* Standard case, age in fractional
8882: years but with the precision of a month */
8883: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8884: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8885: agev[m][i]=1;
8886: else if(agev[m][i] < *agemin){
8887: *agemin=agev[m][i];
8888: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8889: }
8890: else if(agev[m][i] >*agemax){
8891: *agemax=agev[m][i];
1.156 brouard 8892: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8893: }
8894: /*agev[m][i]=anint[m][i]-annais[i];*/
8895: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8896: } /* en if 9*/
1.136 brouard 8897: else { /* =9 */
1.214 brouard 8898: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8899: agev[m][i]=1;
8900: s[m][i]=-1;
8901: }
8902: }
1.214 brouard 8903: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8904: agev[m][i]=1;
1.214 brouard 8905: else{
8906: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8907: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8908: agev[m][i]=0;
8909: }
8910: } /* End for lastpass */
8911: }
1.136 brouard 8912:
8913: for (i=1; i<=imx; i++) {
8914: for(m=firstpass; (m<=lastpass); m++){
8915: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8916: (*nberr)++;
1.136 brouard 8917: 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);
8918: 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);
8919: return 1;
8920: }
8921: }
8922: }
8923:
8924: /*for (i=1; i<=imx; i++){
8925: for (m=firstpass; (m<lastpass); m++){
8926: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8927: }
8928:
8929: }*/
8930:
8931:
1.139 brouard 8932: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
8933: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 8934:
8935: return (0);
1.164 brouard 8936: /* endread:*/
1.136 brouard 8937: printf("Exiting calandcheckages: ");
8938: return (1);
8939: }
8940:
1.172 brouard 8941: #if defined(_MSC_VER)
8942: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8943: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8944: //#include "stdafx.h"
8945: //#include <stdio.h>
8946: //#include <tchar.h>
8947: //#include <windows.h>
8948: //#include <iostream>
8949: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
8950:
8951: LPFN_ISWOW64PROCESS fnIsWow64Process;
8952:
8953: BOOL IsWow64()
8954: {
8955: BOOL bIsWow64 = FALSE;
8956:
8957: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
8958: // (HANDLE, PBOOL);
8959:
8960: //LPFN_ISWOW64PROCESS fnIsWow64Process;
8961:
8962: HMODULE module = GetModuleHandle(_T("kernel32"));
8963: const char funcName[] = "IsWow64Process";
8964: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
8965: GetProcAddress(module, funcName);
8966:
8967: if (NULL != fnIsWow64Process)
8968: {
8969: if (!fnIsWow64Process(GetCurrentProcess(),
8970: &bIsWow64))
8971: //throw std::exception("Unknown error");
8972: printf("Unknown error\n");
8973: }
8974: return bIsWow64 != FALSE;
8975: }
8976: #endif
1.177 brouard 8977:
1.191 brouard 8978: void syscompilerinfo(int logged)
1.167 brouard 8979: {
8980: /* #include "syscompilerinfo.h"*/
1.185 brouard 8981: /* command line Intel compiler 32bit windows, XP compatible:*/
8982: /* /GS /W3 /Gy
8983: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
8984: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
8985: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 8986: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
8987: */
8988: /* 64 bits */
1.185 brouard 8989: /*
8990: /GS /W3 /Gy
8991: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
8992: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
8993: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
8994: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
8995: /* Optimization are useless and O3 is slower than O2 */
8996: /*
8997: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
8998: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
8999: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9000: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9001: */
1.186 brouard 9002: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9003: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9004: /PDB:"visual studio
9005: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9006: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9007: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9008: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9009: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9010: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9011: uiAccess='false'"
9012: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9013: /NOLOGO /TLBID:1
9014: */
1.177 brouard 9015: #if defined __INTEL_COMPILER
1.178 brouard 9016: #if defined(__GNUC__)
9017: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9018: #endif
1.177 brouard 9019: #elif defined(__GNUC__)
1.179 brouard 9020: #ifndef __APPLE__
1.174 brouard 9021: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9022: #endif
1.177 brouard 9023: struct utsname sysInfo;
1.178 brouard 9024: int cross = CROSS;
9025: if (cross){
9026: printf("Cross-");
1.191 brouard 9027: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9028: }
1.174 brouard 9029: #endif
9030:
1.171 brouard 9031: #include <stdint.h>
1.178 brouard 9032:
1.191 brouard 9033: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9034: #if defined(__clang__)
1.191 brouard 9035: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9036: #endif
9037: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9038: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9039: #endif
9040: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9041: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9042: #endif
9043: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9044: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9045: #endif
9046: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9047: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9048: #endif
9049: #if defined(_MSC_VER)
1.191 brouard 9050: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9051: #endif
9052: #if defined(__PGI)
1.191 brouard 9053: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9054: #endif
9055: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9056: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9057: #endif
1.191 brouard 9058: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9059:
1.167 brouard 9060: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9061: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9062: // Windows (x64 and x86)
1.191 brouard 9063: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9064: #elif __unix__ // all unices, not all compilers
9065: // Unix
1.191 brouard 9066: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9067: #elif __linux__
9068: // linux
1.191 brouard 9069: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9070: #elif __APPLE__
1.174 brouard 9071: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9072: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9073: #endif
9074:
9075: /* __MINGW32__ */
9076: /* __CYGWIN__ */
9077: /* __MINGW64__ */
9078: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9079: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9080: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9081: /* _WIN64 // Defined for applications for Win64. */
9082: /* _M_X64 // Defined for compilations that target x64 processors. */
9083: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9084:
1.167 brouard 9085: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9086: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9087: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9088: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9089: #else
1.191 brouard 9090: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9091: #endif
9092:
1.169 brouard 9093: #if defined(__GNUC__)
9094: # if defined(__GNUC_PATCHLEVEL__)
9095: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9096: + __GNUC_MINOR__ * 100 \
9097: + __GNUC_PATCHLEVEL__)
9098: # else
9099: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9100: + __GNUC_MINOR__ * 100)
9101: # endif
1.174 brouard 9102: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9103: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9104:
9105: if (uname(&sysInfo) != -1) {
9106: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9107: 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 9108: }
9109: else
9110: perror("uname() error");
1.179 brouard 9111: //#ifndef __INTEL_COMPILER
9112: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9113: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9114: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9115: #endif
1.169 brouard 9116: #endif
1.172 brouard 9117:
9118: // void main()
9119: // {
1.169 brouard 9120: #if defined(_MSC_VER)
1.174 brouard 9121: if (IsWow64()){
1.191 brouard 9122: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9123: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9124: }
9125: else{
1.191 brouard 9126: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9127: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9128: }
1.172 brouard 9129: // printf("\nPress Enter to continue...");
9130: // getchar();
9131: // }
9132:
1.169 brouard 9133: #endif
9134:
1.167 brouard 9135:
1.219 brouard 9136: }
1.136 brouard 9137:
1.219 brouard 9138: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9139: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9140: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9141: /* double ftolpl = 1.e-10; */
1.180 brouard 9142: double age, agebase, agelim;
1.203 brouard 9143: double tot;
1.180 brouard 9144:
1.202 brouard 9145: strcpy(filerespl,"PL_");
9146: strcat(filerespl,fileresu);
9147: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9148: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9149: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9150: }
1.227 brouard 9151: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9152: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9153: pstamp(ficrespl);
1.203 brouard 9154: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9155: fprintf(ficrespl,"#Age ");
9156: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9157: fprintf(ficrespl,"\n");
1.180 brouard 9158:
1.219 brouard 9159: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9160:
1.219 brouard 9161: agebase=ageminpar;
9162: agelim=agemaxpar;
1.180 brouard 9163:
1.227 brouard 9164: /* i1=pow(2,ncoveff); */
1.234 brouard 9165: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9166: if (cptcovn < 1){i1=1;}
1.180 brouard 9167:
1.238 brouard 9168: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9169: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9170: if(TKresult[nres]!= k)
9171: continue;
1.235 brouard 9172:
1.238 brouard 9173: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9174: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9175: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9176: /* k=k+1; */
9177: /* to clean */
9178: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9179: fprintf(ficrespl,"#******");
9180: printf("#******");
9181: fprintf(ficlog,"#******");
9182: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9183: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9184: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9185: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9186: }
9187: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9188: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9189: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9190: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9191: }
9192: fprintf(ficrespl,"******\n");
9193: printf("******\n");
9194: fprintf(ficlog,"******\n");
9195: if(invalidvarcomb[k]){
9196: printf("\nCombination (%d) ignored because no case \n",k);
9197: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9198: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9199: continue;
9200: }
1.219 brouard 9201:
1.238 brouard 9202: fprintf(ficrespl,"#Age ");
9203: for(j=1;j<=cptcoveff;j++) {
9204: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9205: }
9206: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9207: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9208:
1.238 brouard 9209: for (age=agebase; age<=agelim; age++){
9210: /* for (age=agebase; age<=agebase; age++){ */
9211: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9212: fprintf(ficrespl,"%.0f ",age );
9213: for(j=1;j<=cptcoveff;j++)
9214: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9215: tot=0.;
9216: for(i=1; i<=nlstate;i++){
9217: tot += prlim[i][i];
9218: fprintf(ficrespl," %.5f", prlim[i][i]);
9219: }
9220: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9221: } /* Age */
9222: /* was end of cptcod */
9223: } /* cptcov */
9224: } /* nres */
1.219 brouard 9225: return 0;
1.180 brouard 9226: }
9227:
1.218 brouard 9228: 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){
9229: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9230:
9231: /* Computes the back prevalence limit for any combination of covariate values
9232: * at any age between ageminpar and agemaxpar
9233: */
1.235 brouard 9234: int i, j, k, i1, nres=0 ;
1.217 brouard 9235: /* double ftolpl = 1.e-10; */
9236: double age, agebase, agelim;
9237: double tot;
1.218 brouard 9238: /* double ***mobaverage; */
9239: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9240:
9241: strcpy(fileresplb,"PLB_");
9242: strcat(fileresplb,fileresu);
9243: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9244: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9245: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9246: }
9247: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9248: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9249: pstamp(ficresplb);
9250: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9251: fprintf(ficresplb,"#Age ");
9252: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9253: fprintf(ficresplb,"\n");
9254:
1.218 brouard 9255:
9256: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9257:
9258: agebase=ageminpar;
9259: agelim=agemaxpar;
9260:
9261:
1.227 brouard 9262: i1=pow(2,cptcoveff);
1.218 brouard 9263: if (cptcovn < 1){i1=1;}
1.227 brouard 9264:
1.238 brouard 9265: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9266: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9267: if(TKresult[nres]!= k)
9268: continue;
9269: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9270: fprintf(ficresplb,"#******");
9271: printf("#******");
9272: fprintf(ficlog,"#******");
9273: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9274: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9275: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9276: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9277: }
9278: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9279: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9280: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9281: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9282: }
9283: fprintf(ficresplb,"******\n");
9284: printf("******\n");
9285: fprintf(ficlog,"******\n");
9286: if(invalidvarcomb[k]){
9287: printf("\nCombination (%d) ignored because no cases \n",k);
9288: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9289: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9290: continue;
9291: }
1.218 brouard 9292:
1.238 brouard 9293: fprintf(ficresplb,"#Age ");
9294: for(j=1;j<=cptcoveff;j++) {
9295: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9296: }
9297: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9298: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9299:
9300:
1.238 brouard 9301: for (age=agebase; age<=agelim; age++){
9302: /* for (age=agebase; age<=agebase; age++){ */
9303: if(mobilavproj > 0){
9304: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9305: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9306: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9307: }else if (mobilavproj == 0){
9308: 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);
9309: 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);
9310: exit(1);
9311: }else{
9312: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9313: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9314: }
9315: fprintf(ficresplb,"%.0f ",age );
9316: for(j=1;j<=cptcoveff;j++)
9317: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9318: tot=0.;
9319: for(i=1; i<=nlstate;i++){
9320: tot += bprlim[i][i];
9321: fprintf(ficresplb," %.5f", bprlim[i][i]);
9322: }
9323: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9324: } /* Age */
9325: /* was end of cptcod */
9326: } /* end of any combination */
9327: } /* end of nres */
1.218 brouard 9328: /* hBijx(p, bage, fage); */
9329: /* fclose(ficrespijb); */
9330:
9331: return 0;
1.217 brouard 9332: }
1.218 brouard 9333:
1.180 brouard 9334: int hPijx(double *p, int bage, int fage){
9335: /*------------- h Pij x at various ages ------------*/
9336:
9337: int stepsize;
9338: int agelim;
9339: int hstepm;
9340: int nhstepm;
1.235 brouard 9341: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9342:
9343: double agedeb;
9344: double ***p3mat;
9345:
1.201 brouard 9346: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9347: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9348: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9349: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9350: }
9351: printf("Computing pij: result on file '%s' \n", filerespij);
9352: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9353:
9354: stepsize=(int) (stepm+YEARM-1)/YEARM;
9355: /*if (stepm<=24) stepsize=2;*/
9356:
9357: agelim=AGESUP;
9358: hstepm=stepsize*YEARM; /* Every year of age */
9359: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9360:
1.180 brouard 9361: /* hstepm=1; aff par mois*/
9362: pstamp(ficrespij);
9363: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9364: i1= pow(2,cptcoveff);
1.218 brouard 9365: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9366: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9367: /* k=k+1; */
1.235 brouard 9368: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9369: for(k=1; k<=i1;k++){
9370: if(TKresult[nres]!= k)
9371: continue;
1.183 brouard 9372: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9373: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9374: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9375: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9376: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9377: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9378: }
1.183 brouard 9379: fprintf(ficrespij,"******\n");
9380:
9381: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9382: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9383: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9384:
9385: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9386:
1.183 brouard 9387: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9388: oldm=oldms;savm=savms;
1.235 brouard 9389: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9390: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9391: for(i=1; i<=nlstate;i++)
9392: for(j=1; j<=nlstate+ndeath;j++)
9393: fprintf(ficrespij," %1d-%1d",i,j);
9394: fprintf(ficrespij,"\n");
9395: for (h=0; h<=nhstepm; h++){
9396: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9397: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9398: for(i=1; i<=nlstate;i++)
9399: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9400: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9401: fprintf(ficrespij,"\n");
9402: }
1.183 brouard 9403: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9404: fprintf(ficrespij,"\n");
9405: }
1.180 brouard 9406: /*}*/
9407: }
1.218 brouard 9408: return 0;
1.180 brouard 9409: }
1.218 brouard 9410:
9411: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9412: /*------------- h Bij x at various ages ------------*/
9413:
9414: int stepsize;
1.218 brouard 9415: /* int agelim; */
9416: int ageminl;
1.217 brouard 9417: int hstepm;
9418: int nhstepm;
1.238 brouard 9419: int h, i, i1, j, k, nres;
1.218 brouard 9420:
1.217 brouard 9421: double agedeb;
9422: double ***p3mat;
1.218 brouard 9423:
9424: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9425: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9426: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9427: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9428: }
9429: printf("Computing pij back: result on file '%s' \n", filerespijb);
9430: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9431:
9432: stepsize=(int) (stepm+YEARM-1)/YEARM;
9433: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9434:
1.218 brouard 9435: /* agelim=AGESUP; */
9436: ageminl=30;
9437: hstepm=stepsize*YEARM; /* Every year of age */
9438: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9439:
9440: /* hstepm=1; aff par mois*/
9441: pstamp(ficrespijb);
9442: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9443: i1= pow(2,cptcoveff);
1.218 brouard 9444: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9445: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9446: /* k=k+1; */
1.238 brouard 9447: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9448: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9449: if(TKresult[nres]!= k)
9450: continue;
9451: fprintf(ficrespijb,"\n#****** ");
9452: for(j=1;j<=cptcoveff;j++)
9453: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9454: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9455: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9456: }
9457: fprintf(ficrespijb,"******\n");
9458: if(invalidvarcomb[k]){
9459: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9460: continue;
9461: }
9462:
9463: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9464: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9465: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9466: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9467: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9468:
9469: /* nhstepm=nhstepm*YEARM; aff par mois*/
9470:
9471: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9472: /* oldm=oldms;savm=savms; */
9473: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9474: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9475: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9476: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9477: for(i=1; i<=nlstate;i++)
9478: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9479: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9480: fprintf(ficrespijb,"\n");
1.238 brouard 9481: for (h=0; h<=nhstepm; h++){
9482: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9483: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9484: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9485: for(i=1; i<=nlstate;i++)
9486: for(j=1; j<=nlstate+ndeath;j++)
9487: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9488: fprintf(ficrespijb,"\n");
9489: }
9490: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9491: fprintf(ficrespijb,"\n");
9492: } /* end age deb */
9493: } /* end combination */
9494: } /* end nres */
1.218 brouard 9495: return 0;
9496: } /* hBijx */
1.217 brouard 9497:
1.180 brouard 9498:
1.136 brouard 9499: /***********************************************/
9500: /**************** Main Program *****************/
9501: /***********************************************/
9502:
9503: int main(int argc, char *argv[])
9504: {
9505: #ifdef GSL
9506: const gsl_multimin_fminimizer_type *T;
9507: size_t iteri = 0, it;
9508: int rval = GSL_CONTINUE;
9509: int status = GSL_SUCCESS;
9510: double ssval;
9511: #endif
9512: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9513: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9514: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9515: int jj, ll, li, lj, lk;
1.136 brouard 9516: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9517: int num_filled;
1.136 brouard 9518: int itimes;
9519: int NDIM=2;
9520: int vpopbased=0;
1.235 brouard 9521: int nres=0;
1.136 brouard 9522:
1.164 brouard 9523: char ca[32], cb[32];
1.136 brouard 9524: /* FILE *fichtm; *//* Html File */
9525: /* FILE *ficgp;*/ /*Gnuplot File */
9526: struct stat info;
1.191 brouard 9527: double agedeb=0.;
1.194 brouard 9528:
9529: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9530: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9531:
1.165 brouard 9532: double fret;
1.191 brouard 9533: double dum=0.; /* Dummy variable */
1.136 brouard 9534: double ***p3mat;
1.218 brouard 9535: /* double ***mobaverage; */
1.164 brouard 9536:
9537: char line[MAXLINE];
1.197 brouard 9538: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9539:
1.234 brouard 9540: char modeltemp[MAXLINE];
1.230 brouard 9541: char resultline[MAXLINE];
9542:
1.136 brouard 9543: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9544: char *tok, *val; /* pathtot */
1.136 brouard 9545: int firstobs=1, lastobs=10;
1.195 brouard 9546: int c, h , cpt, c2;
1.191 brouard 9547: int jl=0;
9548: int i1, j1, jk, stepsize=0;
1.194 brouard 9549: int count=0;
9550:
1.164 brouard 9551: int *tab;
1.136 brouard 9552: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9553: int backcast=0;
1.136 brouard 9554: int mobilav=0,popforecast=0;
1.191 brouard 9555: int hstepm=0, nhstepm=0;
1.136 brouard 9556: int agemortsup;
9557: float sumlpop=0.;
9558: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9559: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9560:
1.191 brouard 9561: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9562: double ftolpl=FTOL;
9563: double **prlim;
1.217 brouard 9564: double **bprlim;
1.136 brouard 9565: double ***param; /* Matrix of parameters */
9566: double *p;
9567: double **matcov; /* Matrix of covariance */
1.203 brouard 9568: double **hess; /* Hessian matrix */
1.136 brouard 9569: double ***delti3; /* Scale */
9570: double *delti; /* Scale */
9571: double ***eij, ***vareij;
9572: double **varpl; /* Variances of prevalence limits by age */
9573: double *epj, vepp;
1.164 brouard 9574:
1.136 brouard 9575: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9576: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9577:
1.136 brouard 9578: double **ximort;
1.145 brouard 9579: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9580: int *dcwave;
9581:
1.164 brouard 9582: char z[1]="c";
1.136 brouard 9583:
9584: /*char *strt;*/
9585: char strtend[80];
1.126 brouard 9586:
1.164 brouard 9587:
1.126 brouard 9588: /* setlocale (LC_ALL, ""); */
9589: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9590: /* textdomain (PACKAGE); */
9591: /* setlocale (LC_CTYPE, ""); */
9592: /* setlocale (LC_MESSAGES, ""); */
9593:
9594: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9595: rstart_time = time(NULL);
9596: /* (void) gettimeofday(&start_time,&tzp);*/
9597: start_time = *localtime(&rstart_time);
1.126 brouard 9598: curr_time=start_time;
1.157 brouard 9599: /*tml = *localtime(&start_time.tm_sec);*/
9600: /* strcpy(strstart,asctime(&tml)); */
9601: strcpy(strstart,asctime(&start_time));
1.126 brouard 9602:
9603: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9604: /* tp.tm_sec = tp.tm_sec +86400; */
9605: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9606: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9607: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9608: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9609: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9610: /* strt=asctime(&tmg); */
9611: /* printf("Time(after) =%s",strstart); */
9612: /* (void) time (&time_value);
9613: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9614: * tm = *localtime(&time_value);
9615: * strstart=asctime(&tm);
9616: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9617: */
9618:
9619: nberr=0; /* Number of errors and warnings */
9620: nbwarn=0;
1.184 brouard 9621: #ifdef WIN32
9622: _getcwd(pathcd, size);
9623: #else
1.126 brouard 9624: getcwd(pathcd, size);
1.184 brouard 9625: #endif
1.191 brouard 9626: syscompilerinfo(0);
1.196 brouard 9627: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9628: if(argc <=1){
9629: printf("\nEnter the parameter file name: ");
1.205 brouard 9630: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9631: printf("ERROR Empty parameter file name\n");
9632: goto end;
9633: }
1.126 brouard 9634: i=strlen(pathr);
9635: if(pathr[i-1]=='\n')
9636: pathr[i-1]='\0';
1.156 brouard 9637: i=strlen(pathr);
1.205 brouard 9638: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9639: pathr[i-1]='\0';
1.205 brouard 9640: }
9641: i=strlen(pathr);
9642: if( i==0 ){
9643: printf("ERROR Empty parameter file name\n");
9644: goto end;
9645: }
9646: for (tok = pathr; tok != NULL; ){
1.126 brouard 9647: printf("Pathr |%s|\n",pathr);
9648: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9649: printf("val= |%s| pathr=%s\n",val,pathr);
9650: strcpy (pathtot, val);
9651: if(pathr[0] == '\0') break; /* Dirty */
9652: }
9653: }
9654: else{
9655: strcpy(pathtot,argv[1]);
9656: }
9657: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9658: /*cygwin_split_path(pathtot,path,optionfile);
9659: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9660: /* cutv(path,optionfile,pathtot,'\\');*/
9661:
9662: /* Split argv[0], imach program to get pathimach */
9663: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9664: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9665: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9666: /* strcpy(pathimach,argv[0]); */
9667: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9668: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9669: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9670: #ifdef WIN32
9671: _chdir(path); /* Can be a relative path */
9672: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9673: #else
1.126 brouard 9674: chdir(path); /* Can be a relative path */
1.184 brouard 9675: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9676: #endif
9677: printf("Current directory %s!\n",pathcd);
1.126 brouard 9678: strcpy(command,"mkdir ");
9679: strcat(command,optionfilefiname);
9680: if((outcmd=system(command)) != 0){
1.169 brouard 9681: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9682: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9683: /* fclose(ficlog); */
9684: /* exit(1); */
9685: }
9686: /* if((imk=mkdir(optionfilefiname))<0){ */
9687: /* perror("mkdir"); */
9688: /* } */
9689:
9690: /*-------- arguments in the command line --------*/
9691:
1.186 brouard 9692: /* Main Log file */
1.126 brouard 9693: strcat(filelog, optionfilefiname);
9694: strcat(filelog,".log"); /* */
9695: if((ficlog=fopen(filelog,"w"))==NULL) {
9696: printf("Problem with logfile %s\n",filelog);
9697: goto end;
9698: }
9699: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9700: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9701: fprintf(ficlog,"\nEnter the parameter file name: \n");
9702: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9703: path=%s \n\
9704: optionfile=%s\n\
9705: optionfilext=%s\n\
1.156 brouard 9706: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9707:
1.197 brouard 9708: syscompilerinfo(1);
1.167 brouard 9709:
1.126 brouard 9710: printf("Local time (at start):%s",strstart);
9711: fprintf(ficlog,"Local time (at start): %s",strstart);
9712: fflush(ficlog);
9713: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9714: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9715:
9716: /* */
9717: strcpy(fileres,"r");
9718: strcat(fileres, optionfilefiname);
1.201 brouard 9719: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9720: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9721: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9722:
1.186 brouard 9723: /* Main ---------arguments file --------*/
1.126 brouard 9724:
9725: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9726: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9727: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9728: fflush(ficlog);
1.149 brouard 9729: /* goto end; */
9730: exit(70);
1.126 brouard 9731: }
9732:
9733:
9734:
9735: strcpy(filereso,"o");
1.201 brouard 9736: strcat(filereso,fileresu);
1.126 brouard 9737: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9738: printf("Problem with Output resultfile: %s\n", filereso);
9739: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9740: fflush(ficlog);
9741: goto end;
9742: }
9743:
9744: /* Reads comments: lines beginning with '#' */
9745: numlinepar=0;
1.197 brouard 9746:
9747: /* First parameter line */
9748: while(fgets(line, MAXLINE, ficpar)) {
9749: /* If line starts with a # it is a comment */
9750: if (line[0] == '#') {
9751: numlinepar++;
9752: fputs(line,stdout);
9753: fputs(line,ficparo);
9754: fputs(line,ficlog);
9755: continue;
9756: }else
9757: break;
9758: }
9759: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9760: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9761: if (num_filled != 5) {
9762: printf("Should be 5 parameters\n");
9763: }
1.126 brouard 9764: numlinepar++;
1.197 brouard 9765: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9766: }
9767: /* Second parameter line */
9768: while(fgets(line, MAXLINE, ficpar)) {
9769: /* If line starts with a # it is a comment */
9770: if (line[0] == '#') {
9771: numlinepar++;
9772: fputs(line,stdout);
9773: fputs(line,ficparo);
9774: fputs(line,ficlog);
9775: continue;
9776: }else
9777: break;
9778: }
1.223 brouard 9779: 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", \
9780: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9781: if (num_filled != 11) {
9782: 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 9783: printf("but line=%s\n",line);
1.197 brouard 9784: }
1.223 brouard 9785: 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 9786: }
1.203 brouard 9787: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9788: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9789: /* Third parameter line */
9790: while(fgets(line, MAXLINE, ficpar)) {
9791: /* If line starts with a # it is a comment */
9792: if (line[0] == '#') {
9793: numlinepar++;
9794: fputs(line,stdout);
9795: fputs(line,ficparo);
9796: fputs(line,ficlog);
9797: continue;
9798: }else
9799: break;
9800: }
1.201 brouard 9801: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9802: if (num_filled == 0)
9803: model[0]='\0';
9804: else if (num_filled != 1){
1.197 brouard 9805: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9806: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9807: model[0]='\0';
9808: goto end;
9809: }
9810: else{
9811: if (model[0]=='+'){
9812: for(i=1; i<=strlen(model);i++)
9813: modeltemp[i-1]=model[i];
1.201 brouard 9814: strcpy(model,modeltemp);
1.197 brouard 9815: }
9816: }
1.199 brouard 9817: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9818: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9819: }
9820: /* 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); */
9821: /* numlinepar=numlinepar+3; /\* In general *\/ */
9822: /* 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 9823: 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);
9824: 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 9825: fflush(ficlog);
1.190 brouard 9826: /* if(model[0]=='#'|| model[0]== '\0'){ */
9827: if(model[0]=='#'){
1.187 brouard 9828: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9829: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9830: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9831: if(mle != -1){
9832: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9833: exit(1);
9834: }
9835: }
1.126 brouard 9836: while((c=getc(ficpar))=='#' && c!= EOF){
9837: ungetc(c,ficpar);
9838: fgets(line, MAXLINE, ficpar);
9839: numlinepar++;
1.195 brouard 9840: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9841: z[0]=line[1];
9842: }
9843: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9844: fputs(line, stdout);
9845: //puts(line);
1.126 brouard 9846: fputs(line,ficparo);
9847: fputs(line,ficlog);
9848: }
9849: ungetc(c,ficpar);
9850:
9851:
1.145 brouard 9852: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9853: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9854: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9855: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9856: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9857: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9858: v1+v2*age+v2*v3 makes cptcovn = 3
9859: */
9860: if (strlen(model)>1)
1.187 brouard 9861: 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 9862: else
1.187 brouard 9863: ncovmodel=2; /* Constant and age */
1.133 brouard 9864: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9865: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9866: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9867: 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);
9868: 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);
9869: fflush(stdout);
9870: fclose (ficlog);
9871: goto end;
9872: }
1.126 brouard 9873: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9874: delti=delti3[1][1];
9875: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9876: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
9877: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9878: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9879: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9880: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9881: fclose (ficparo);
9882: fclose (ficlog);
9883: goto end;
9884: exit(0);
1.220 brouard 9885: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9886: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9887: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9888: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9889: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9890: matcov=matrix(1,npar,1,npar);
1.203 brouard 9891: hess=matrix(1,npar,1,npar);
1.220 brouard 9892: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9893: /* Read guessed parameters */
1.126 brouard 9894: /* Reads comments: lines beginning with '#' */
9895: while((c=getc(ficpar))=='#' && c!= EOF){
9896: ungetc(c,ficpar);
9897: fgets(line, MAXLINE, ficpar);
9898: numlinepar++;
1.141 brouard 9899: fputs(line,stdout);
1.126 brouard 9900: fputs(line,ficparo);
9901: fputs(line,ficlog);
9902: }
9903: ungetc(c,ficpar);
9904:
9905: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9906: for(i=1; i <=nlstate; i++){
1.234 brouard 9907: j=0;
1.126 brouard 9908: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9909: if(jj==i) continue;
9910: j++;
9911: fscanf(ficpar,"%1d%1d",&i1,&j1);
9912: if ((i1 != i) || (j1 != jj)){
9913: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9914: It might be a problem of design; if ncovcol and the model are correct\n \
9915: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9916: exit(1);
9917: }
9918: fprintf(ficparo,"%1d%1d",i1,j1);
9919: if(mle==1)
9920: printf("%1d%1d",i,jj);
9921: fprintf(ficlog,"%1d%1d",i,jj);
9922: for(k=1; k<=ncovmodel;k++){
9923: fscanf(ficpar," %lf",¶m[i][j][k]);
9924: if(mle==1){
9925: printf(" %lf",param[i][j][k]);
9926: fprintf(ficlog," %lf",param[i][j][k]);
9927: }
9928: else
9929: fprintf(ficlog," %lf",param[i][j][k]);
9930: fprintf(ficparo," %lf",param[i][j][k]);
9931: }
9932: fscanf(ficpar,"\n");
9933: numlinepar++;
9934: if(mle==1)
9935: printf("\n");
9936: fprintf(ficlog,"\n");
9937: fprintf(ficparo,"\n");
1.126 brouard 9938: }
9939: }
9940: fflush(ficlog);
1.234 brouard 9941:
1.145 brouard 9942: /* Reads scales values */
1.126 brouard 9943: p=param[1][1];
9944:
9945: /* Reads comments: lines beginning with '#' */
9946: while((c=getc(ficpar))=='#' && c!= EOF){
9947: ungetc(c,ficpar);
9948: fgets(line, MAXLINE, ficpar);
9949: numlinepar++;
1.141 brouard 9950: fputs(line,stdout);
1.126 brouard 9951: fputs(line,ficparo);
9952: fputs(line,ficlog);
9953: }
9954: ungetc(c,ficpar);
9955:
9956: for(i=1; i <=nlstate; i++){
9957: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 9958: fscanf(ficpar,"%1d%1d",&i1,&j1);
9959: if ( (i1-i) * (j1-j) != 0){
9960: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
9961: exit(1);
9962: }
9963: printf("%1d%1d",i,j);
9964: fprintf(ficparo,"%1d%1d",i1,j1);
9965: fprintf(ficlog,"%1d%1d",i1,j1);
9966: for(k=1; k<=ncovmodel;k++){
9967: fscanf(ficpar,"%le",&delti3[i][j][k]);
9968: printf(" %le",delti3[i][j][k]);
9969: fprintf(ficparo," %le",delti3[i][j][k]);
9970: fprintf(ficlog," %le",delti3[i][j][k]);
9971: }
9972: fscanf(ficpar,"\n");
9973: numlinepar++;
9974: printf("\n");
9975: fprintf(ficparo,"\n");
9976: fprintf(ficlog,"\n");
1.126 brouard 9977: }
9978: }
9979: fflush(ficlog);
1.234 brouard 9980:
1.145 brouard 9981: /* Reads covariance matrix */
1.126 brouard 9982: delti=delti3[1][1];
1.220 brouard 9983:
9984:
1.126 brouard 9985: /* 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 9986:
1.126 brouard 9987: /* Reads comments: lines beginning with '#' */
9988: while((c=getc(ficpar))=='#' && c!= EOF){
9989: ungetc(c,ficpar);
9990: fgets(line, MAXLINE, ficpar);
9991: numlinepar++;
1.141 brouard 9992: fputs(line,stdout);
1.126 brouard 9993: fputs(line,ficparo);
9994: fputs(line,ficlog);
9995: }
9996: ungetc(c,ficpar);
1.220 brouard 9997:
1.126 brouard 9998: matcov=matrix(1,npar,1,npar);
1.203 brouard 9999: hess=matrix(1,npar,1,npar);
1.131 brouard 10000: for(i=1; i <=npar; i++)
10001: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10002:
1.194 brouard 10003: /* Scans npar lines */
1.126 brouard 10004: for(i=1; i <=npar; i++){
1.226 brouard 10005: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10006: if(count != 3){
1.226 brouard 10007: printf("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: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10011: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10012: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10013: exit(1);
1.220 brouard 10014: }else{
1.226 brouard 10015: if(mle==1)
10016: printf("%1d%1d%d",i1,j1,jk);
10017: }
10018: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10019: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10020: for(j=1; j <=i; j++){
1.226 brouard 10021: fscanf(ficpar," %le",&matcov[i][j]);
10022: if(mle==1){
10023: printf(" %.5le",matcov[i][j]);
10024: }
10025: fprintf(ficlog," %.5le",matcov[i][j]);
10026: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10027: }
10028: fscanf(ficpar,"\n");
10029: numlinepar++;
10030: if(mle==1)
1.220 brouard 10031: printf("\n");
1.126 brouard 10032: fprintf(ficlog,"\n");
10033: fprintf(ficparo,"\n");
10034: }
1.194 brouard 10035: /* End of read covariance matrix npar lines */
1.126 brouard 10036: for(i=1; i <=npar; i++)
10037: for(j=i+1;j<=npar;j++)
1.226 brouard 10038: matcov[i][j]=matcov[j][i];
1.126 brouard 10039:
10040: if(mle==1)
10041: printf("\n");
10042: fprintf(ficlog,"\n");
10043:
10044: fflush(ficlog);
10045:
10046: /*-------- Rewriting parameter file ----------*/
10047: strcpy(rfileres,"r"); /* "Rparameterfile */
10048: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10049: strcat(rfileres,"."); /* */
10050: strcat(rfileres,optionfilext); /* Other files have txt extension */
10051: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10052: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10053: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10054: }
10055: fprintf(ficres,"#%s\n",version);
10056: } /* End of mle != -3 */
1.218 brouard 10057:
1.186 brouard 10058: /* Main data
10059: */
1.126 brouard 10060: n= lastobs;
10061: num=lvector(1,n);
10062: moisnais=vector(1,n);
10063: annais=vector(1,n);
10064: moisdc=vector(1,n);
10065: andc=vector(1,n);
1.220 brouard 10066: weight=vector(1,n);
1.126 brouard 10067: agedc=vector(1,n);
10068: cod=ivector(1,n);
1.220 brouard 10069: for(i=1;i<=n;i++){
1.234 brouard 10070: num[i]=0;
10071: moisnais[i]=0;
10072: annais[i]=0;
10073: moisdc[i]=0;
10074: andc[i]=0;
10075: agedc[i]=0;
10076: cod[i]=0;
10077: weight[i]=1.0; /* Equal weights, 1 by default */
10078: }
1.126 brouard 10079: mint=matrix(1,maxwav,1,n);
10080: anint=matrix(1,maxwav,1,n);
1.131 brouard 10081: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10082: tab=ivector(1,NCOVMAX);
1.144 brouard 10083: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10084: 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 10085:
1.136 brouard 10086: /* Reads data from file datafile */
10087: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10088: goto end;
10089:
10090: /* Calculation of the number of parameters from char model */
1.234 brouard 10091: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10092: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10093: k=3 V4 Tvar[k=3]= 4 (from V4)
10094: k=2 V1 Tvar[k=2]= 1 (from V1)
10095: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10096: */
10097:
10098: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10099: TvarsDind=ivector(1,NCOVMAX); /* */
10100: TvarsD=ivector(1,NCOVMAX); /* */
10101: TvarsQind=ivector(1,NCOVMAX); /* */
10102: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10103: TvarF=ivector(1,NCOVMAX); /* */
10104: TvarFind=ivector(1,NCOVMAX); /* */
10105: TvarV=ivector(1,NCOVMAX); /* */
10106: TvarVind=ivector(1,NCOVMAX); /* */
10107: TvarA=ivector(1,NCOVMAX); /* */
10108: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10109: TvarFD=ivector(1,NCOVMAX); /* */
10110: TvarFDind=ivector(1,NCOVMAX); /* */
10111: TvarFQ=ivector(1,NCOVMAX); /* */
10112: TvarFQind=ivector(1,NCOVMAX); /* */
10113: TvarVD=ivector(1,NCOVMAX); /* */
10114: TvarVDind=ivector(1,NCOVMAX); /* */
10115: TvarVQ=ivector(1,NCOVMAX); /* */
10116: TvarVQind=ivector(1,NCOVMAX); /* */
10117:
1.230 brouard 10118: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10119: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10120: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10121: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10122: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10123: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10124: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10125: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10126: */
10127: /* For model-covariate k tells which data-covariate to use but
10128: because this model-covariate is a construction we invent a new column
10129: ncovcol + k1
10130: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10131: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10132: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10133: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10134: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10135: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10136: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10137: */
1.145 brouard 10138: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10139: 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 10140: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10141: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10142: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10143: 4 covariates (3 plus signs)
10144: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10145: */
1.230 brouard 10146: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10147: * individual dummy, fixed or varying:
10148: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10149: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10150: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10151: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10152: * Tmodelind[1]@9={9,0,3,2,}*/
10153: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10154: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10155: * individual quantitative, fixed or varying:
10156: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10157: * 3, 1, 0, 0, 0, 0, 0, 0},
10158: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10159: /* Main decodemodel */
10160:
1.187 brouard 10161:
1.223 brouard 10162: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10163: goto end;
10164:
1.137 brouard 10165: if((double)(lastobs-imx)/(double)imx > 1.10){
10166: nbwarn++;
10167: 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);
10168: 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);
10169: }
1.136 brouard 10170: /* if(mle==1){*/
1.137 brouard 10171: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10172: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10173: }
10174:
10175: /*-calculation of age at interview from date of interview and age at death -*/
10176: agev=matrix(1,maxwav,1,imx);
10177:
10178: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10179: goto end;
10180:
1.126 brouard 10181:
1.136 brouard 10182: agegomp=(int)agemin;
10183: free_vector(moisnais,1,n);
10184: free_vector(annais,1,n);
1.126 brouard 10185: /* free_matrix(mint,1,maxwav,1,n);
10186: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10187: /* free_vector(moisdc,1,n); */
10188: /* free_vector(andc,1,n); */
1.145 brouard 10189: /* */
10190:
1.126 brouard 10191: wav=ivector(1,imx);
1.214 brouard 10192: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10193: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10194: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10195: 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.*/
10196: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10197: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10198:
10199: /* Concatenates waves */
1.214 brouard 10200: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10201: Death is a valid wave (if date is known).
10202: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10203: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10204: and mw[mi+1][i]. dh depends on stepm.
10205: */
10206:
1.126 brouard 10207: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.145 brouard 10208: /* */
10209:
1.215 brouard 10210: free_vector(moisdc,1,n);
10211: free_vector(andc,1,n);
10212:
1.126 brouard 10213: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10214: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10215: ncodemax[1]=1;
1.145 brouard 10216: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10217: cptcoveff=0;
1.220 brouard 10218: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10219: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10220: }
10221:
10222: ncovcombmax=pow(2,cptcoveff);
10223: invalidvarcomb=ivector(1, ncovcombmax);
10224: for(i=1;i<ncovcombmax;i++)
10225: invalidvarcomb[i]=0;
10226:
1.211 brouard 10227: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10228: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10229: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10230:
1.200 brouard 10231: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10232: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10233: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10234: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10235: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10236: * (currently 0 or 1) in the data.
10237: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10238: * corresponding modality (h,j).
10239: */
10240:
1.145 brouard 10241: h=0;
10242: /*if (cptcovn > 0) */
1.126 brouard 10243: m=pow(2,cptcoveff);
10244:
1.144 brouard 10245: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10246: * For k=4 covariates, h goes from 1 to m=2**k
10247: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10248: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10249: * h\k 1 2 3 4
1.143 brouard 10250: *______________________________
10251: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10252: * 2 2 1 1 1
10253: * 3 i=2 1 2 1 1
10254: * 4 2 2 1 1
10255: * 5 i=3 1 i=2 1 2 1
10256: * 6 2 1 2 1
10257: * 7 i=4 1 2 2 1
10258: * 8 2 2 2 1
1.197 brouard 10259: * 9 i=5 1 i=3 1 i=2 1 2
10260: * 10 2 1 1 2
10261: * 11 i=6 1 2 1 2
10262: * 12 2 2 1 2
10263: * 13 i=7 1 i=4 1 2 2
10264: * 14 2 1 2 2
10265: * 15 i=8 1 2 2 2
10266: * 16 2 2 2 2
1.143 brouard 10267: */
1.212 brouard 10268: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10269: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10270: * and the value of each covariate?
10271: * V1=1, V2=1, V3=2, V4=1 ?
10272: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10273: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10274: * In order to get the real value in the data, we use nbcode
10275: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10276: * We are keeping this crazy system in order to be able (in the future?)
10277: * to have more than 2 values (0 or 1) for a covariate.
10278: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10279: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10280: * bbbbbbbb
10281: * 76543210
10282: * h-1 00000101 (6-1=5)
1.219 brouard 10283: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10284: * &
10285: * 1 00000001 (1)
1.219 brouard 10286: * 00000000 = 1 & ((h-1) >> (k-1))
10287: * +1= 00000001 =1
1.211 brouard 10288: *
10289: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10290: * h' 1101 =2^3+2^2+0x2^1+2^0
10291: * >>k' 11
10292: * & 00000001
10293: * = 00000001
10294: * +1 = 00000010=2 = codtabm(14,3)
10295: * Reverse h=6 and m=16?
10296: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10297: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10298: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10299: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10300: * V3=decodtabm(14,3,2**4)=2
10301: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10302: *(h-1) >> (j-1) 0011 =13 >> 2
10303: * &1 000000001
10304: * = 000000001
10305: * +1= 000000010 =2
10306: * 2211
10307: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10308: * V3=2
1.220 brouard 10309: * codtabm and decodtabm are identical
1.211 brouard 10310: */
10311:
1.145 brouard 10312:
10313: free_ivector(Ndum,-1,NCOVMAX);
10314:
10315:
1.126 brouard 10316:
1.186 brouard 10317: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10318: strcpy(optionfilegnuplot,optionfilefiname);
10319: if(mle==-3)
1.201 brouard 10320: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10321: strcat(optionfilegnuplot,".gp");
10322:
10323: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10324: printf("Problem with file %s",optionfilegnuplot);
10325: }
10326: else{
1.204 brouard 10327: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10328: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10329: //fprintf(ficgp,"set missing 'NaNq'\n");
10330: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10331: }
10332: /* fclose(ficgp);*/
1.186 brouard 10333:
10334:
10335: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10336:
10337: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10338: if(mle==-3)
1.201 brouard 10339: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10340: strcat(optionfilehtm,".htm");
10341: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10342: printf("Problem with %s \n",optionfilehtm);
10343: exit(0);
1.126 brouard 10344: }
10345:
10346: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10347: strcat(optionfilehtmcov,"-cov.htm");
10348: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10349: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10350: }
10351: else{
10352: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10353: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10354: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10355: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10356: }
10357:
1.213 brouard 10358: 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 10359: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10360: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10361: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10362: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10363: \n\
10364: <hr size=\"2\" color=\"#EC5E5E\">\
10365: <ul><li><h4>Parameter files</h4>\n\
10366: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10367: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10368: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10369: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10370: - Date and time at start: %s</ul>\n",\
10371: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10372: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10373: fileres,fileres,\
10374: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10375: fflush(fichtm);
10376:
10377: strcpy(pathr,path);
10378: strcat(pathr,optionfilefiname);
1.184 brouard 10379: #ifdef WIN32
10380: _chdir(optionfilefiname); /* Move to directory named optionfile */
10381: #else
1.126 brouard 10382: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10383: #endif
10384:
1.126 brouard 10385:
1.220 brouard 10386: /* Calculates basic frequencies. Computes observed prevalence at single age
10387: and for any valid combination of covariates
1.126 brouard 10388: and prints on file fileres'p'. */
1.227 brouard 10389: freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
10390: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10391:
10392: fprintf(fichtm,"\n");
10393: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10394: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10395: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10396: imx,agemin,agemax,jmin,jmax,jmean);
10397: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10398: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10399: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10400: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10401: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10402:
1.126 brouard 10403: /* For Powell, parameters are in a vector p[] starting at p[1]
10404: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10405: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10406:
10407: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10408: /* For mortality only */
1.126 brouard 10409: if (mle==-3){
1.136 brouard 10410: ximort=matrix(1,NDIM,1,NDIM);
1.220 brouard 10411: for(i=1;i<=NDIM;i++)
10412: for(j=1;j<=NDIM;j++)
10413: ximort[i][j]=0.;
1.186 brouard 10414: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10415: cens=ivector(1,n);
10416: ageexmed=vector(1,n);
10417: agecens=vector(1,n);
10418: dcwave=ivector(1,n);
1.223 brouard 10419:
1.126 brouard 10420: for (i=1; i<=imx; i++){
10421: dcwave[i]=-1;
10422: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10423: if (s[m][i]>nlstate) {
10424: dcwave[i]=m;
10425: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10426: break;
10427: }
1.126 brouard 10428: }
1.226 brouard 10429:
1.126 brouard 10430: for (i=1; i<=imx; i++) {
10431: if (wav[i]>0){
1.226 brouard 10432: ageexmed[i]=agev[mw[1][i]][i];
10433: j=wav[i];
10434: agecens[i]=1.;
10435:
10436: if (ageexmed[i]> 1 && wav[i] > 0){
10437: agecens[i]=agev[mw[j][i]][i];
10438: cens[i]= 1;
10439: }else if (ageexmed[i]< 1)
10440: cens[i]= -1;
10441: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10442: cens[i]=0 ;
1.126 brouard 10443: }
10444: else cens[i]=-1;
10445: }
10446:
10447: for (i=1;i<=NDIM;i++) {
10448: for (j=1;j<=NDIM;j++)
1.226 brouard 10449: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10450: }
10451:
1.145 brouard 10452: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10453: /*printf("%lf %lf", p[1], p[2]);*/
10454:
10455:
1.136 brouard 10456: #ifdef GSL
10457: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10458: #else
1.126 brouard 10459: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10460: #endif
1.201 brouard 10461: strcpy(filerespow,"POW-MORT_");
10462: strcat(filerespow,fileresu);
1.126 brouard 10463: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10464: printf("Problem with resultfile: %s\n", filerespow);
10465: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10466: }
1.136 brouard 10467: #ifdef GSL
10468: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10469: #else
1.126 brouard 10470: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10471: #endif
1.126 brouard 10472: /* for (i=1;i<=nlstate;i++)
10473: for(j=1;j<=nlstate+ndeath;j++)
10474: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10475: */
10476: fprintf(ficrespow,"\n");
1.136 brouard 10477: #ifdef GSL
10478: /* gsl starts here */
10479: T = gsl_multimin_fminimizer_nmsimplex;
10480: gsl_multimin_fminimizer *sfm = NULL;
10481: gsl_vector *ss, *x;
10482: gsl_multimin_function minex_func;
10483:
10484: /* Initial vertex size vector */
10485: ss = gsl_vector_alloc (NDIM);
10486:
10487: if (ss == NULL){
10488: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10489: }
10490: /* Set all step sizes to 1 */
10491: gsl_vector_set_all (ss, 0.001);
10492:
10493: /* Starting point */
1.126 brouard 10494:
1.136 brouard 10495: x = gsl_vector_alloc (NDIM);
10496:
10497: if (x == NULL){
10498: gsl_vector_free(ss);
10499: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10500: }
10501:
10502: /* Initialize method and iterate */
10503: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10504: /* gsl_vector_set(x, 0, 0.0268); */
10505: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10506: gsl_vector_set(x, 0, p[1]);
10507: gsl_vector_set(x, 1, p[2]);
10508:
10509: minex_func.f = &gompertz_f;
10510: minex_func.n = NDIM;
10511: minex_func.params = (void *)&p; /* ??? */
10512:
10513: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10514: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10515:
10516: printf("Iterations beginning .....\n\n");
10517: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10518:
10519: iteri=0;
10520: while (rval == GSL_CONTINUE){
10521: iteri++;
10522: status = gsl_multimin_fminimizer_iterate(sfm);
10523:
10524: if (status) printf("error: %s\n", gsl_strerror (status));
10525: fflush(0);
10526:
10527: if (status)
10528: break;
10529:
10530: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10531: ssval = gsl_multimin_fminimizer_size (sfm);
10532:
10533: if (rval == GSL_SUCCESS)
10534: printf ("converged to a local maximum at\n");
10535:
10536: printf("%5d ", iteri);
10537: for (it = 0; it < NDIM; it++){
10538: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10539: }
10540: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10541: }
10542:
10543: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10544:
10545: gsl_vector_free(x); /* initial values */
10546: gsl_vector_free(ss); /* inital step size */
10547: for (it=0; it<NDIM; it++){
10548: p[it+1]=gsl_vector_get(sfm->x,it);
10549: fprintf(ficrespow," %.12lf", p[it]);
10550: }
10551: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10552: #endif
10553: #ifdef POWELL
10554: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10555: #endif
1.126 brouard 10556: fclose(ficrespow);
10557:
1.203 brouard 10558: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10559:
10560: for(i=1; i <=NDIM; i++)
10561: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10562: matcov[i][j]=matcov[j][i];
1.126 brouard 10563:
10564: printf("\nCovariance matrix\n ");
1.203 brouard 10565: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10566: for(i=1; i <=NDIM; i++) {
10567: for(j=1;j<=NDIM;j++){
1.220 brouard 10568: printf("%f ",matcov[i][j]);
10569: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10570: }
1.203 brouard 10571: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10572: }
10573:
10574: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10575: for (i=1;i<=NDIM;i++) {
1.126 brouard 10576: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10577: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10578: }
1.126 brouard 10579: lsurv=vector(1,AGESUP);
10580: lpop=vector(1,AGESUP);
10581: tpop=vector(1,AGESUP);
10582: lsurv[agegomp]=100000;
10583:
10584: for (k=agegomp;k<=AGESUP;k++) {
10585: agemortsup=k;
10586: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10587: }
10588:
10589: for (k=agegomp;k<agemortsup;k++)
10590: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10591:
10592: for (k=agegomp;k<agemortsup;k++){
10593: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10594: sumlpop=sumlpop+lpop[k];
10595: }
10596:
10597: tpop[agegomp]=sumlpop;
10598: for (k=agegomp;k<(agemortsup-3);k++){
10599: /* tpop[k+1]=2;*/
10600: tpop[k+1]=tpop[k]-lpop[k];
10601: }
10602:
10603:
10604: printf("\nAge lx qx dx Lx Tx e(x)\n");
10605: for (k=agegomp;k<(agemortsup-2);k++)
10606: 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]);
10607:
10608:
10609: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10610: ageminpar=50;
10611: agemaxpar=100;
1.194 brouard 10612: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10613: printf("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);
10616: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10617: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10618: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10619: }else{
10620: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10621: 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 10622: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10623: }
1.201 brouard 10624: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10625: stepm, weightopt,\
10626: model,imx,p,matcov,agemortsup);
10627:
10628: free_vector(lsurv,1,AGESUP);
10629: free_vector(lpop,1,AGESUP);
10630: free_vector(tpop,1,AGESUP);
1.220 brouard 10631: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10632: free_ivector(cens,1,n);
10633: free_vector(agecens,1,n);
10634: free_ivector(dcwave,1,n);
1.220 brouard 10635: #ifdef GSL
1.136 brouard 10636: #endif
1.186 brouard 10637: } /* Endof if mle==-3 mortality only */
1.205 brouard 10638: /* Standard */
10639: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10640: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10641: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10642: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10643: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10644: for (k=1; k<=npar;k++)
10645: printf(" %d %8.5f",k,p[k]);
10646: printf("\n");
1.205 brouard 10647: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10648: /* mlikeli uses func not funcone */
10649: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10650: }
10651: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10652: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10653: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10654: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10655: }
10656: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10657: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10658: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10659: for (k=1; k<=npar;k++)
10660: printf(" %d %8.5f",k,p[k]);
10661: printf("\n");
10662:
10663: /*--------- results files --------------*/
1.224 brouard 10664: 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 10665:
10666:
10667: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10668: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10669: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10670: for(i=1,jk=1; i <=nlstate; i++){
10671: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10672: if (k != i) {
10673: printf("%d%d ",i,k);
10674: fprintf(ficlog,"%d%d ",i,k);
10675: fprintf(ficres,"%1d%1d ",i,k);
10676: for(j=1; j <=ncovmodel; j++){
10677: printf("%12.7f ",p[jk]);
10678: fprintf(ficlog,"%12.7f ",p[jk]);
10679: fprintf(ficres,"%12.7f ",p[jk]);
10680: jk++;
10681: }
10682: printf("\n");
10683: fprintf(ficlog,"\n");
10684: fprintf(ficres,"\n");
10685: }
1.126 brouard 10686: }
10687: }
1.203 brouard 10688: if(mle != 0){
10689: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10690: ftolhess=ftol; /* Usually correct */
1.203 brouard 10691: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10692: 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");
10693: 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");
10694: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10695: for(k=1; k <=(nlstate+ndeath); k++){
10696: if (k != i) {
10697: printf("%d%d ",i,k);
10698: fprintf(ficlog,"%d%d ",i,k);
10699: for(j=1; j <=ncovmodel; j++){
10700: 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]));
10701: 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]));
10702: jk++;
10703: }
10704: printf("\n");
10705: fprintf(ficlog,"\n");
10706: }
10707: }
1.193 brouard 10708: }
1.203 brouard 10709: } /* end of hesscov and Wald tests */
1.225 brouard 10710:
1.203 brouard 10711: /* */
1.126 brouard 10712: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10713: printf("# Scales (for hessian or gradient estimation)\n");
10714: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10715: for(i=1,jk=1; i <=nlstate; i++){
10716: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10717: if (j!=i) {
10718: fprintf(ficres,"%1d%1d",i,j);
10719: printf("%1d%1d",i,j);
10720: fprintf(ficlog,"%1d%1d",i,j);
10721: for(k=1; k<=ncovmodel;k++){
10722: printf(" %.5e",delti[jk]);
10723: fprintf(ficlog," %.5e",delti[jk]);
10724: fprintf(ficres," %.5e",delti[jk]);
10725: jk++;
10726: }
10727: printf("\n");
10728: fprintf(ficlog,"\n");
10729: fprintf(ficres,"\n");
10730: }
1.126 brouard 10731: }
10732: }
10733:
10734: 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 10735: if(mle >= 1) /* To big for the screen */
1.126 brouard 10736: 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");
10737: 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");
10738: /* # 121 Var(a12)\n\ */
10739: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10740: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10741: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10742: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10743: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10744: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10745: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10746:
10747:
10748: /* Just to have a covariance matrix which will be more understandable
10749: even is we still don't want to manage dictionary of variables
10750: */
10751: for(itimes=1;itimes<=2;itimes++){
10752: jj=0;
10753: for(i=1; i <=nlstate; i++){
1.225 brouard 10754: for(j=1; j <=nlstate+ndeath; j++){
10755: if(j==i) continue;
10756: for(k=1; k<=ncovmodel;k++){
10757: jj++;
10758: ca[0]= k+'a'-1;ca[1]='\0';
10759: if(itimes==1){
10760: if(mle>=1)
10761: printf("#%1d%1d%d",i,j,k);
10762: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10763: fprintf(ficres,"#%1d%1d%d",i,j,k);
10764: }else{
10765: if(mle>=1)
10766: printf("%1d%1d%d",i,j,k);
10767: fprintf(ficlog,"%1d%1d%d",i,j,k);
10768: fprintf(ficres,"%1d%1d%d",i,j,k);
10769: }
10770: ll=0;
10771: for(li=1;li <=nlstate; li++){
10772: for(lj=1;lj <=nlstate+ndeath; lj++){
10773: if(lj==li) continue;
10774: for(lk=1;lk<=ncovmodel;lk++){
10775: ll++;
10776: if(ll<=jj){
10777: cb[0]= lk +'a'-1;cb[1]='\0';
10778: if(ll<jj){
10779: if(itimes==1){
10780: if(mle>=1)
10781: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10782: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10783: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10784: }else{
10785: if(mle>=1)
10786: printf(" %.5e",matcov[jj][ll]);
10787: fprintf(ficlog," %.5e",matcov[jj][ll]);
10788: fprintf(ficres," %.5e",matcov[jj][ll]);
10789: }
10790: }else{
10791: if(itimes==1){
10792: if(mle>=1)
10793: printf(" Var(%s%1d%1d)",ca,i,j);
10794: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10795: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10796: }else{
10797: if(mle>=1)
10798: printf(" %.7e",matcov[jj][ll]);
10799: fprintf(ficlog," %.7e",matcov[jj][ll]);
10800: fprintf(ficres," %.7e",matcov[jj][ll]);
10801: }
10802: }
10803: }
10804: } /* end lk */
10805: } /* end lj */
10806: } /* end li */
10807: if(mle>=1)
10808: printf("\n");
10809: fprintf(ficlog,"\n");
10810: fprintf(ficres,"\n");
10811: numlinepar++;
10812: } /* end k*/
10813: } /*end j */
1.126 brouard 10814: } /* end i */
10815: } /* end itimes */
10816:
10817: fflush(ficlog);
10818: fflush(ficres);
1.225 brouard 10819: while(fgets(line, MAXLINE, ficpar)) {
10820: /* If line starts with a # it is a comment */
10821: if (line[0] == '#') {
10822: numlinepar++;
10823: fputs(line,stdout);
10824: fputs(line,ficparo);
10825: fputs(line,ficlog);
10826: continue;
10827: }else
10828: break;
10829: }
10830:
1.209 brouard 10831: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10832: /* ungetc(c,ficpar); */
10833: /* fgets(line, MAXLINE, ficpar); */
10834: /* fputs(line,stdout); */
10835: /* fputs(line,ficparo); */
10836: /* } */
10837: /* ungetc(c,ficpar); */
1.126 brouard 10838:
10839: estepm=0;
1.209 brouard 10840: 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 10841:
10842: if (num_filled != 6) {
10843: 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);
10844: 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);
10845: goto end;
10846: }
10847: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10848: }
10849: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10850: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10851:
1.209 brouard 10852: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10853: if (estepm==0 || estepm < stepm) estepm=stepm;
10854: if (fage <= 2) {
10855: bage = ageminpar;
10856: fage = agemaxpar;
10857: }
10858:
10859: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10860: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10861: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10862:
1.186 brouard 10863: /* Other stuffs, more or less useful */
1.126 brouard 10864: while((c=getc(ficpar))=='#' && c!= EOF){
10865: ungetc(c,ficpar);
10866: fgets(line, MAXLINE, ficpar);
1.141 brouard 10867: fputs(line,stdout);
1.126 brouard 10868: fputs(line,ficparo);
10869: }
10870: ungetc(c,ficpar);
10871:
10872: 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);
10873: 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);
10874: 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);
10875: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10876: 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);
10877:
10878: while((c=getc(ficpar))=='#' && c!= EOF){
10879: ungetc(c,ficpar);
10880: fgets(line, MAXLINE, ficpar);
1.141 brouard 10881: fputs(line,stdout);
1.126 brouard 10882: fputs(line,ficparo);
10883: }
10884: ungetc(c,ficpar);
10885:
10886:
10887: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10888: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10889:
10890: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10891: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10892: fprintf(ficparo,"pop_based=%d\n",popbased);
10893: fprintf(ficres,"pop_based=%d\n",popbased);
10894:
10895: while((c=getc(ficpar))=='#' && c!= EOF){
10896: ungetc(c,ficpar);
10897: fgets(line, MAXLINE, ficpar);
1.141 brouard 10898: fputs(line,stdout);
1.238 brouard 10899: fputs(line,ficres);
1.126 brouard 10900: fputs(line,ficparo);
10901: }
10902: ungetc(c,ficpar);
10903:
10904: 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);
10905: 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);
10906: 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);
10907: 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);
10908: 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);
10909: /* day and month of proj2 are not used but only year anproj2.*/
10910:
1.217 brouard 10911: while((c=getc(ficpar))=='#' && c!= EOF){
10912: ungetc(c,ficpar);
10913: fgets(line, MAXLINE, ficpar);
10914: fputs(line,stdout);
10915: fputs(line,ficparo);
1.238 brouard 10916: fputs(line,ficres);
1.217 brouard 10917: }
10918: ungetc(c,ficpar);
10919:
10920: 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 10921: 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);
10922: 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);
10923: 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 10924: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 10925:
1.230 brouard 10926: /* Results */
1.235 brouard 10927: nresult=0;
1.230 brouard 10928: while(fgets(line, MAXLINE, ficpar)) {
10929: /* If line starts with a # it is a comment */
10930: if (line[0] == '#') {
10931: numlinepar++;
10932: fputs(line,stdout);
10933: fputs(line,ficparo);
10934: fputs(line,ficlog);
1.238 brouard 10935: fputs(line,ficres);
1.230 brouard 10936: continue;
10937: }else
10938: break;
10939: }
1.240 brouard 10940: if (!feof(ficpar))
1.230 brouard 10941: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 10942: if (num_filled == 0){
1.230 brouard 10943: resultline[0]='\0';
1.240 brouard 10944: break;
10945: } else if (num_filled != 1){
1.230 brouard 10946: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
10947: }
1.235 brouard 10948: nresult++; /* Sum of resultlines */
10949: printf("Result %d: result=%s\n",nresult, resultline);
10950: if(nresult > MAXRESULTLINES){
10951: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10952: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10953: goto end;
10954: }
10955: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 10956: fprintf(ficparo,"result: %s\n",resultline);
10957: fprintf(ficres,"result: %s\n",resultline);
10958: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 10959: while(fgets(line, MAXLINE, ficpar)) {
10960: /* If line starts with a # it is a comment */
10961: if (line[0] == '#') {
10962: numlinepar++;
10963: fputs(line,stdout);
10964: fputs(line,ficparo);
1.238 brouard 10965: fputs(line,ficres);
1.230 brouard 10966: fputs(line,ficlog);
10967: continue;
10968: }else
10969: break;
10970: }
10971: if (feof(ficpar))
10972: break;
10973: else{ /* Processess output results for this combination of covariate values */
10974: }
1.240 brouard 10975: } /* end while */
1.230 brouard 10976:
10977:
1.126 brouard 10978:
1.230 brouard 10979: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 10980: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 10981:
10982: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 10983: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 10984: printf("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);
1.230 brouard 10987: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 10988: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10989: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10990: }else{
1.218 brouard 10991: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 10992: }
10993: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 10994: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
10995: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 10996:
1.225 brouard 10997: /*------------ free_vector -------------*/
10998: /* chdir(path); */
1.220 brouard 10999:
1.215 brouard 11000: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11001: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11002: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11003: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11004: free_lvector(num,1,n);
11005: free_vector(agedc,1,n);
11006: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11007: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11008: fclose(ficparo);
11009: fclose(ficres);
1.220 brouard 11010:
11011:
1.186 brouard 11012: /* Other results (useful)*/
1.220 brouard 11013:
11014:
1.126 brouard 11015: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11016: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11017: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11018: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11019: fclose(ficrespl);
11020:
11021: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11022: /*#include "hpijx.h"*/
11023: hPijx(p, bage, fage);
1.145 brouard 11024: fclose(ficrespij);
1.227 brouard 11025:
1.220 brouard 11026: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11027: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11028: k=1;
1.126 brouard 11029: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11030:
1.219 brouard 11031: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11032: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11033: for(i=1;i<=AGESUP;i++)
1.219 brouard 11034: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11035: for(k=1;k<=ncovcombmax;k++)
11036: probs[i][j][k]=0.;
1.219 brouard 11037: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11038: if (mobilav!=0 ||mobilavproj !=0 ) {
11039: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11040: for(i=1;i<=AGESUP;i++)
11041: for(j=1;j<=nlstate;j++)
11042: for(k=1;k<=ncovcombmax;k++)
11043: mobaverages[i][j][k]=0.;
1.219 brouard 11044: mobaverage=mobaverages;
11045: if (mobilav!=0) {
1.235 brouard 11046: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11047: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11048: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11049: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11050: }
1.219 brouard 11051: }
11052: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11053: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11054: else if (mobilavproj !=0) {
1.235 brouard 11055: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11056: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11057: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11058: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11059: }
1.219 brouard 11060: }
11061: }/* end if moving average */
1.227 brouard 11062:
1.126 brouard 11063: /*---------- Forecasting ------------------*/
11064: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11065: if(prevfcast==1){
11066: /* if(stepm ==1){*/
1.225 brouard 11067: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11068: }
1.217 brouard 11069: if(backcast==1){
1.219 brouard 11070: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11071: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11072: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11073:
11074: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11075:
11076: bprlim=matrix(1,nlstate,1,nlstate);
11077: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11078: fclose(ficresplb);
11079:
1.222 brouard 11080: hBijx(p, bage, fage, mobaverage);
11081: fclose(ficrespijb);
1.219 brouard 11082: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11083:
11084: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11085: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11086: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11087: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11088: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11089: }
1.217 brouard 11090:
1.186 brouard 11091:
11092: /* ------ Other prevalence ratios------------ */
1.126 brouard 11093:
1.215 brouard 11094: free_ivector(wav,1,imx);
11095: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11096: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11097: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11098:
11099:
1.127 brouard 11100: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11101:
1.201 brouard 11102: strcpy(filerese,"E_");
11103: strcat(filerese,fileresu);
1.126 brouard 11104: if((ficreseij=fopen(filerese,"w"))==NULL) {
11105: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11106: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11107: }
1.208 brouard 11108: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11109: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11110:
11111: pstamp(ficreseij);
1.219 brouard 11112:
1.235 brouard 11113: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11114: if (cptcovn < 1){i1=1;}
11115:
11116: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11117: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11118: if(TKresult[nres]!= k)
11119: continue;
1.219 brouard 11120: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11121: printf("\n#****** ");
1.225 brouard 11122: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11123: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11124: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11125: }
11126: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11127: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11128: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11129: }
11130: fprintf(ficreseij,"******\n");
1.235 brouard 11131: printf("******\n");
1.219 brouard 11132:
11133: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11134: oldm=oldms;savm=savms;
1.235 brouard 11135: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11136:
1.219 brouard 11137: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11138: }
11139: fclose(ficreseij);
1.208 brouard 11140: printf("done evsij\n");fflush(stdout);
11141: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11142:
1.227 brouard 11143: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11144:
11145:
1.201 brouard 11146: strcpy(filerest,"T_");
11147: strcat(filerest,fileresu);
1.127 brouard 11148: if((ficrest=fopen(filerest,"w"))==NULL) {
11149: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11150: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11151: }
1.208 brouard 11152: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11153: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11154:
1.126 brouard 11155:
1.201 brouard 11156: strcpy(fileresstde,"STDE_");
11157: strcat(fileresstde,fileresu);
1.126 brouard 11158: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11159: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11160: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11161: }
1.227 brouard 11162: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11163: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11164:
1.201 brouard 11165: strcpy(filerescve,"CVE_");
11166: strcat(filerescve,fileresu);
1.126 brouard 11167: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11168: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11169: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11170: }
1.227 brouard 11171: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11172: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11173:
1.201 brouard 11174: strcpy(fileresv,"V_");
11175: strcat(fileresv,fileresu);
1.126 brouard 11176: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11177: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11178: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11179: }
1.227 brouard 11180: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11181: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11182:
1.145 brouard 11183: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11184: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11185:
1.235 brouard 11186: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11187: if (cptcovn < 1){i1=1;}
11188:
11189: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11190: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11191: if(TKresult[nres]!= k)
11192: continue;
1.242 brouard 11193: printf("\n#****** Result for:");
11194: fprintf(ficrest,"\n#****** Result for:");
11195: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11196: for(j=1;j<=cptcoveff;j++){
11197: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11198: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11199: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11200: }
1.235 brouard 11201: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11202: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11203: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11204: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11205: }
1.208 brouard 11206: fprintf(ficrest,"******\n");
1.227 brouard 11207: fprintf(ficlog,"******\n");
11208: printf("******\n");
1.208 brouard 11209:
11210: fprintf(ficresstdeij,"\n#****** ");
11211: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11212: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11213: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11214: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11215: }
1.235 brouard 11216: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11217: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11218: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11219: }
1.208 brouard 11220: fprintf(ficresstdeij,"******\n");
11221: fprintf(ficrescveij,"******\n");
11222:
11223: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11224: /* pstamp(ficresvij); */
1.225 brouard 11225: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11226: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11227: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11228: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11229: }
1.208 brouard 11230: fprintf(ficresvij,"******\n");
11231:
11232: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11233: oldm=oldms;savm=savms;
1.235 brouard 11234: printf(" cvevsij ");
11235: fprintf(ficlog, " cvevsij ");
11236: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11237: printf(" end cvevsij \n ");
11238: fprintf(ficlog, " end cvevsij \n ");
11239:
11240: /*
11241: */
11242: /* goto endfree; */
11243:
11244: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11245: pstamp(ficrest);
11246:
11247:
11248: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11249: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11250: cptcod= 0; /* To be deleted */
11251: printf("varevsij vpopbased=%d \n",vpopbased);
11252: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11253: 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 11254: 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 ");
11255: if(vpopbased==1)
11256: 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);
11257: else
11258: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11259: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11260: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11261: fprintf(ficrest,"\n");
11262: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11263: epj=vector(1,nlstate+1);
11264: printf("Computing age specific period (stable) prevalences in each health state \n");
11265: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11266: for(age=bage; age <=fage ;age++){
1.235 brouard 11267: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11268: if (vpopbased==1) {
11269: if(mobilav ==0){
11270: for(i=1; i<=nlstate;i++)
11271: prlim[i][i]=probs[(int)age][i][k];
11272: }else{ /* mobilav */
11273: for(i=1; i<=nlstate;i++)
11274: prlim[i][i]=mobaverage[(int)age][i][k];
11275: }
11276: }
1.219 brouard 11277:
1.227 brouard 11278: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11279: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11280: /* printf(" age %4.0f ",age); */
11281: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11282: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11283: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11284: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11285: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11286: }
11287: epj[nlstate+1] +=epj[j];
11288: }
11289: /* printf(" age %4.0f \n",age); */
1.219 brouard 11290:
1.227 brouard 11291: for(i=1, vepp=0.;i <=nlstate;i++)
11292: for(j=1;j <=nlstate;j++)
11293: vepp += vareij[i][j][(int)age];
11294: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11295: for(j=1;j <=nlstate;j++){
11296: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11297: }
11298: fprintf(ficrest,"\n");
11299: }
1.208 brouard 11300: } /* End vpopbased */
11301: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11302: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11303: free_vector(epj,1,nlstate+1);
1.235 brouard 11304: printf("done selection\n");fflush(stdout);
11305: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11306:
1.145 brouard 11307: /*}*/
1.235 brouard 11308: } /* End k selection */
1.227 brouard 11309:
11310: printf("done State-specific expectancies\n");fflush(stdout);
11311: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11312:
1.126 brouard 11313: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11314:
1.201 brouard 11315: strcpy(fileresvpl,"VPL_");
11316: strcat(fileresvpl,fileresu);
1.126 brouard 11317: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11318: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11319: exit(0);
11320: }
1.208 brouard 11321: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11322: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11323:
1.145 brouard 11324: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11325: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11326:
1.235 brouard 11327: i1=pow(2,cptcoveff);
11328: if (cptcovn < 1){i1=1;}
11329:
11330: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11331: for(k=1; k<=i1;k++){
11332: if(TKresult[nres]!= k)
11333: continue;
1.227 brouard 11334: fprintf(ficresvpl,"\n#****** ");
11335: printf("\n#****** ");
11336: fprintf(ficlog,"\n#****** ");
11337: for(j=1;j<=cptcoveff;j++) {
11338: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11339: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11340: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11341: }
1.235 brouard 11342: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11343: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11344: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11345: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11346: }
1.227 brouard 11347: fprintf(ficresvpl,"******\n");
11348: printf("******\n");
11349: fprintf(ficlog,"******\n");
11350:
11351: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11352: oldm=oldms;savm=savms;
1.235 brouard 11353: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11354: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11355: /*}*/
1.126 brouard 11356: }
1.227 brouard 11357:
1.126 brouard 11358: fclose(ficresvpl);
1.208 brouard 11359: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11360: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11361:
11362: free_vector(weight,1,n);
11363: free_imatrix(Tvard,1,NCOVMAX,1,2);
11364: free_imatrix(s,1,maxwav+1,1,n);
11365: free_matrix(anint,1,maxwav,1,n);
11366: free_matrix(mint,1,maxwav,1,n);
11367: free_ivector(cod,1,n);
11368: free_ivector(tab,1,NCOVMAX);
11369: fclose(ficresstdeij);
11370: fclose(ficrescveij);
11371: fclose(ficresvij);
11372: fclose(ficrest);
11373: fclose(ficpar);
11374:
11375:
1.126 brouard 11376: /*---------- End : free ----------------*/
1.219 brouard 11377: if (mobilav!=0 ||mobilavproj !=0)
11378: 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 11379: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11380: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11381: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11382: } /* mle==-3 arrives here for freeing */
1.227 brouard 11383: /* endfree:*/
11384: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11385: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11386: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11387: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11388: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11389: free_matrix(coqvar,1,maxwav,1,n);
11390: free_matrix(covar,0,NCOVMAX,1,n);
11391: free_matrix(matcov,1,npar,1,npar);
11392: free_matrix(hess,1,npar,1,npar);
11393: /*free_vector(delti,1,npar);*/
11394: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11395: free_matrix(agev,1,maxwav,1,imx);
11396: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11397:
11398: free_ivector(ncodemax,1,NCOVMAX);
11399: free_ivector(ncodemaxwundef,1,NCOVMAX);
11400: free_ivector(Dummy,-1,NCOVMAX);
11401: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11402: free_ivector(DummyV,1,NCOVMAX);
11403: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11404: free_ivector(Typevar,-1,NCOVMAX);
11405: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11406: free_ivector(TvarsQ,1,NCOVMAX);
11407: free_ivector(TvarsQind,1,NCOVMAX);
11408: free_ivector(TvarsD,1,NCOVMAX);
11409: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11410: free_ivector(TvarFD,1,NCOVMAX);
11411: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11412: free_ivector(TvarF,1,NCOVMAX);
11413: free_ivector(TvarFind,1,NCOVMAX);
11414: free_ivector(TvarV,1,NCOVMAX);
11415: free_ivector(TvarVind,1,NCOVMAX);
11416: free_ivector(TvarA,1,NCOVMAX);
11417: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11418: free_ivector(TvarFQ,1,NCOVMAX);
11419: free_ivector(TvarFQind,1,NCOVMAX);
11420: free_ivector(TvarVD,1,NCOVMAX);
11421: free_ivector(TvarVDind,1,NCOVMAX);
11422: free_ivector(TvarVQ,1,NCOVMAX);
11423: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11424: free_ivector(Tvarsel,1,NCOVMAX);
11425: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11426: free_ivector(Tposprod,1,NCOVMAX);
11427: free_ivector(Tprod,1,NCOVMAX);
11428: free_ivector(Tvaraff,1,NCOVMAX);
11429: free_ivector(invalidvarcomb,1,ncovcombmax);
11430: free_ivector(Tage,1,NCOVMAX);
11431: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11432: free_ivector(TmodelInvind,1,NCOVMAX);
11433: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11434:
11435: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11436: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11437: fflush(fichtm);
11438: fflush(ficgp);
11439:
1.227 brouard 11440:
1.126 brouard 11441: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11442: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11443: 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 11444: }else{
11445: printf("End of Imach\n");
11446: fprintf(ficlog,"End of Imach\n");
11447: }
11448: printf("See log file on %s\n",filelog);
11449: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11450: /*(void) gettimeofday(&end_time,&tzp);*/
11451: rend_time = time(NULL);
11452: end_time = *localtime(&rend_time);
11453: /* tml = *localtime(&end_time.tm_sec); */
11454: strcpy(strtend,asctime(&end_time));
1.126 brouard 11455: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11456: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11457: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11458:
1.157 brouard 11459: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11460: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11461: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11462: /* printf("Total time was %d uSec.\n", total_usecs);*/
11463: /* if(fileappend(fichtm,optionfilehtm)){ */
11464: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11465: fclose(fichtm);
11466: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11467: fclose(fichtmcov);
11468: fclose(ficgp);
11469: fclose(ficlog);
11470: /*------ End -----------*/
1.227 brouard 11471:
11472:
11473: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11474: #ifdef WIN32
1.227 brouard 11475: if (_chdir(pathcd) != 0)
11476: printf("Can't move to directory %s!\n",path);
11477: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11478: #else
1.227 brouard 11479: if(chdir(pathcd) != 0)
11480: printf("Can't move to directory %s!\n", path);
11481: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11482: #endif
1.126 brouard 11483: printf("Current directory %s!\n",pathcd);
11484: /*strcat(plotcmd,CHARSEPARATOR);*/
11485: sprintf(plotcmd,"gnuplot");
1.157 brouard 11486: #ifdef _WIN32
1.126 brouard 11487: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11488: #endif
11489: if(!stat(plotcmd,&info)){
1.158 brouard 11490: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11491: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11492: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11493: }else
11494: strcpy(pplotcmd,plotcmd);
1.157 brouard 11495: #ifdef __unix
1.126 brouard 11496: strcpy(plotcmd,GNUPLOTPROGRAM);
11497: if(!stat(plotcmd,&info)){
1.158 brouard 11498: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11499: }else
11500: strcpy(pplotcmd,plotcmd);
11501: #endif
11502: }else
11503: strcpy(pplotcmd,plotcmd);
11504:
11505: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11506: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11507:
1.126 brouard 11508: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11509: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11510: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11511: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11512: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11513: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11514: }
1.158 brouard 11515: printf(" Successful, please wait...");
1.126 brouard 11516: while (z[0] != 'q') {
11517: /* chdir(path); */
1.154 brouard 11518: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11519: scanf("%s",z);
11520: /* if (z[0] == 'c') system("./imach"); */
11521: if (z[0] == 'e') {
1.158 brouard 11522: #ifdef __APPLE__
1.152 brouard 11523: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11524: #elif __linux
11525: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11526: #else
1.152 brouard 11527: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11528: #endif
11529: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11530: system(pplotcmd);
1.126 brouard 11531: }
11532: else if (z[0] == 'g') system(plotcmd);
11533: else if (z[0] == 'q') exit(0);
11534: }
1.227 brouard 11535: end:
1.126 brouard 11536: while (z[0] != 'q') {
1.195 brouard 11537: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11538: scanf("%s",z);
11539: }
11540: }
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