Annotation of imach/src/imach.c, revision 1.251
1.251 ! brouard 1: /* $Id: imach.c,v 1.250 2016/09/08 16:07:27 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.251 ! brouard 4: Revision 1.250 2016/09/08 16:07:27 brouard
! 5: Summary: continue
! 6:
1.250 brouard 7: Revision 1.249 2016/09/07 17:14:18 brouard
8: Summary: Starting values from frequencies
9:
1.249 brouard 10: Revision 1.248 2016/09/07 14:10:18 brouard
11: *** empty log message ***
12:
1.248 brouard 13: Revision 1.247 2016/09/02 11:11:21 brouard
14: *** empty log message ***
15:
1.247 brouard 16: Revision 1.246 2016/09/02 08:49:22 brouard
17: *** empty log message ***
18:
1.246 brouard 19: Revision 1.245 2016/09/02 07:25:01 brouard
20: *** empty log message ***
21:
1.245 brouard 22: Revision 1.244 2016/09/02 07:17:34 brouard
23: *** empty log message ***
24:
1.244 brouard 25: Revision 1.243 2016/09/02 06:45:35 brouard
26: *** empty log message ***
27:
1.243 brouard 28: Revision 1.242 2016/08/30 15:01:20 brouard
29: Summary: Fixing a lots
30:
1.242 brouard 31: Revision 1.241 2016/08/29 17:17:25 brouard
32: Summary: gnuplot problem in Back projection to fix
33:
1.241 brouard 34: Revision 1.240 2016/08/29 07:53:18 brouard
35: Summary: Better
36:
1.240 brouard 37: Revision 1.239 2016/08/26 15:51:03 brouard
38: Summary: Improvement in Powell output in order to copy and paste
39:
40: Author:
41:
1.239 brouard 42: Revision 1.238 2016/08/26 14:23:35 brouard
43: Summary: Starting tests of 0.99
44:
1.238 brouard 45: Revision 1.237 2016/08/26 09:20:19 brouard
46: Summary: to valgrind
47:
1.237 brouard 48: Revision 1.236 2016/08/25 10:50:18 brouard
49: *** empty log message ***
50:
1.236 brouard 51: Revision 1.235 2016/08/25 06:59:23 brouard
52: *** empty log message ***
53:
1.235 brouard 54: Revision 1.234 2016/08/23 16:51:20 brouard
55: *** empty log message ***
56:
1.234 brouard 57: Revision 1.233 2016/08/23 07:40:50 brouard
58: Summary: not working
59:
1.233 brouard 60: Revision 1.232 2016/08/22 14:20:21 brouard
61: Summary: not working
62:
1.232 brouard 63: Revision 1.231 2016/08/22 07:17:15 brouard
64: Summary: not working
65:
1.231 brouard 66: Revision 1.230 2016/08/22 06:55:53 brouard
67: Summary: Not working
68:
1.230 brouard 69: Revision 1.229 2016/07/23 09:45:53 brouard
70: Summary: Completing for func too
71:
1.229 brouard 72: Revision 1.228 2016/07/22 17:45:30 brouard
73: Summary: Fixing some arrays, still debugging
74:
1.227 brouard 75: Revision 1.226 2016/07/12 18:42:34 brouard
76: Summary: temp
77:
1.226 brouard 78: Revision 1.225 2016/07/12 08:40:03 brouard
79: Summary: saving but not running
80:
1.225 brouard 81: Revision 1.224 2016/07/01 13:16:01 brouard
82: Summary: Fixes
83:
1.224 brouard 84: Revision 1.223 2016/02/19 09:23:35 brouard
85: Summary: temporary
86:
1.223 brouard 87: Revision 1.222 2016/02/17 08:14:50 brouard
88: Summary: Probably last 0.98 stable version 0.98r6
89:
1.222 brouard 90: Revision 1.221 2016/02/15 23:35:36 brouard
91: Summary: minor bug
92:
1.220 brouard 93: Revision 1.219 2016/02/15 00:48:12 brouard
94: *** empty log message ***
95:
1.219 brouard 96: Revision 1.218 2016/02/12 11:29:23 brouard
97: Summary: 0.99 Back projections
98:
1.218 brouard 99: Revision 1.217 2015/12/23 17:18:31 brouard
100: Summary: Experimental backcast
101:
1.217 brouard 102: Revision 1.216 2015/12/18 17:32:11 brouard
103: Summary: 0.98r4 Warning and status=-2
104:
105: Version 0.98r4 is now:
106: - displaying an error when status is -1, date of interview unknown and date of death known;
107: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
108: Older changes concerning s=-2, dating from 2005 have been supersed.
109:
1.216 brouard 110: Revision 1.215 2015/12/16 08:52:24 brouard
111: Summary: 0.98r4 working
112:
1.215 brouard 113: Revision 1.214 2015/12/16 06:57:54 brouard
114: Summary: temporary not working
115:
1.214 brouard 116: Revision 1.213 2015/12/11 18:22:17 brouard
117: Summary: 0.98r4
118:
1.213 brouard 119: Revision 1.212 2015/11/21 12:47:24 brouard
120: Summary: minor typo
121:
1.212 brouard 122: Revision 1.211 2015/11/21 12:41:11 brouard
123: Summary: 0.98r3 with some graph of projected cross-sectional
124:
125: Author: Nicolas Brouard
126:
1.211 brouard 127: Revision 1.210 2015/11/18 17:41:20 brouard
128: Summary: Start working on projected prevalences
129:
1.210 brouard 130: Revision 1.209 2015/11/17 22:12:03 brouard
131: Summary: Adding ftolpl parameter
132: Author: N Brouard
133:
134: We had difficulties to get smoothed confidence intervals. It was due
135: to the period prevalence which wasn't computed accurately. The inner
136: parameter ftolpl is now an outer parameter of the .imach parameter
137: file after estepm. If ftolpl is small 1.e-4 and estepm too,
138: computation are long.
139:
1.209 brouard 140: Revision 1.208 2015/11/17 14:31:57 brouard
141: Summary: temporary
142:
1.208 brouard 143: Revision 1.207 2015/10/27 17:36:57 brouard
144: *** empty log message ***
145:
1.207 brouard 146: Revision 1.206 2015/10/24 07:14:11 brouard
147: *** empty log message ***
148:
1.206 brouard 149: Revision 1.205 2015/10/23 15:50:53 brouard
150: Summary: 0.98r3 some clarification for graphs on likelihood contributions
151:
1.205 brouard 152: Revision 1.204 2015/10/01 16:20:26 brouard
153: Summary: Some new graphs of contribution to likelihood
154:
1.204 brouard 155: Revision 1.203 2015/09/30 17:45:14 brouard
156: Summary: looking at better estimation of the hessian
157:
158: Also a better criteria for convergence to the period prevalence And
159: therefore adding the number of years needed to converge. (The
160: prevalence in any alive state shold sum to one
161:
1.203 brouard 162: Revision 1.202 2015/09/22 19:45:16 brouard
163: Summary: Adding some overall graph on contribution to likelihood. Might change
164:
1.202 brouard 165: Revision 1.201 2015/09/15 17:34:58 brouard
166: Summary: 0.98r0
167:
168: - Some new graphs like suvival functions
169: - Some bugs fixed like model=1+age+V2.
170:
1.201 brouard 171: Revision 1.200 2015/09/09 16:53:55 brouard
172: Summary: Big bug thanks to Flavia
173:
174: Even model=1+age+V2. did not work anymore
175:
1.200 brouard 176: Revision 1.199 2015/09/07 14:09:23 brouard
177: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
178:
1.199 brouard 179: Revision 1.198 2015/09/03 07:14:39 brouard
180: Summary: 0.98q5 Flavia
181:
1.198 brouard 182: Revision 1.197 2015/09/01 18:24:39 brouard
183: *** empty log message ***
184:
1.197 brouard 185: Revision 1.196 2015/08/18 23:17:52 brouard
186: Summary: 0.98q5
187:
1.196 brouard 188: Revision 1.195 2015/08/18 16:28:39 brouard
189: Summary: Adding a hack for testing purpose
190:
191: After reading the title, ftol and model lines, if the comment line has
192: a q, starting with #q, the answer at the end of the run is quit. It
193: permits to run test files in batch with ctest. The former workaround was
194: $ echo q | imach foo.imach
195:
1.195 brouard 196: Revision 1.194 2015/08/18 13:32:00 brouard
197: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
198:
1.194 brouard 199: Revision 1.193 2015/08/04 07:17:42 brouard
200: Summary: 0.98q4
201:
1.193 brouard 202: Revision 1.192 2015/07/16 16:49:02 brouard
203: Summary: Fixing some outputs
204:
1.192 brouard 205: Revision 1.191 2015/07/14 10:00:33 brouard
206: Summary: Some fixes
207:
1.191 brouard 208: Revision 1.190 2015/05/05 08:51:13 brouard
209: Summary: Adding digits in output parameters (7 digits instead of 6)
210:
211: Fix 1+age+.
212:
1.190 brouard 213: Revision 1.189 2015/04/30 14:45:16 brouard
214: Summary: 0.98q2
215:
1.189 brouard 216: Revision 1.188 2015/04/30 08:27:53 brouard
217: *** empty log message ***
218:
1.188 brouard 219: Revision 1.187 2015/04/29 09:11:15 brouard
220: *** empty log message ***
221:
1.187 brouard 222: Revision 1.186 2015/04/23 12:01:52 brouard
223: Summary: V1*age is working now, version 0.98q1
224:
225: Some codes had been disabled in order to simplify and Vn*age was
226: working in the optimization phase, ie, giving correct MLE parameters,
227: but, as usual, outputs were not correct and program core dumped.
228:
1.186 brouard 229: Revision 1.185 2015/03/11 13:26:42 brouard
230: Summary: Inclusion of compile and links command line for Intel Compiler
231:
1.185 brouard 232: Revision 1.184 2015/03/11 11:52:39 brouard
233: Summary: Back from Windows 8. Intel Compiler
234:
1.184 brouard 235: Revision 1.183 2015/03/10 20:34:32 brouard
236: Summary: 0.98q0, trying with directest, mnbrak fixed
237:
238: We use directest instead of original Powell test; probably no
239: incidence on the results, but better justifications;
240: We fixed Numerical Recipes mnbrak routine which was wrong and gave
241: wrong results.
242:
1.183 brouard 243: Revision 1.182 2015/02/12 08:19:57 brouard
244: Summary: Trying to keep directest which seems simpler and more general
245: Author: Nicolas Brouard
246:
1.182 brouard 247: Revision 1.181 2015/02/11 23:22:24 brouard
248: Summary: Comments on Powell added
249:
250: Author:
251:
1.181 brouard 252: Revision 1.180 2015/02/11 17:33:45 brouard
253: Summary: Finishing move from main to function (hpijx and prevalence_limit)
254:
1.180 brouard 255: Revision 1.179 2015/01/04 09:57:06 brouard
256: Summary: back to OS/X
257:
1.179 brouard 258: Revision 1.178 2015/01/04 09:35:48 brouard
259: *** empty log message ***
260:
1.178 brouard 261: Revision 1.177 2015/01/03 18:40:56 brouard
262: Summary: Still testing ilc32 on OSX
263:
1.177 brouard 264: Revision 1.176 2015/01/03 16:45:04 brouard
265: *** empty log message ***
266:
1.176 brouard 267: Revision 1.175 2015/01/03 16:33:42 brouard
268: *** empty log message ***
269:
1.175 brouard 270: Revision 1.174 2015/01/03 16:15:49 brouard
271: Summary: Still in cross-compilation
272:
1.174 brouard 273: Revision 1.173 2015/01/03 12:06:26 brouard
274: Summary: trying to detect cross-compilation
275:
1.173 brouard 276: Revision 1.172 2014/12/27 12:07:47 brouard
277: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
278:
1.172 brouard 279: Revision 1.171 2014/12/23 13:26:59 brouard
280: Summary: Back from Visual C
281:
282: Still problem with utsname.h on Windows
283:
1.171 brouard 284: Revision 1.170 2014/12/23 11:17:12 brouard
285: Summary: Cleaning some \%% back to %%
286:
287: The escape was mandatory for a specific compiler (which one?), but too many warnings.
288:
1.170 brouard 289: Revision 1.169 2014/12/22 23:08:31 brouard
290: Summary: 0.98p
291:
292: Outputs some informations on compiler used, OS etc. Testing on different platforms.
293:
1.169 brouard 294: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 295: Summary: update
1.169 brouard 296:
1.168 brouard 297: Revision 1.167 2014/12/22 13:50:56 brouard
298: Summary: Testing uname and compiler version and if compiled 32 or 64
299:
300: Testing on Linux 64
301:
1.167 brouard 302: Revision 1.166 2014/12/22 11:40:47 brouard
303: *** empty log message ***
304:
1.166 brouard 305: Revision 1.165 2014/12/16 11:20:36 brouard
306: Summary: After compiling on Visual C
307:
308: * imach.c (Module): Merging 1.61 to 1.162
309:
1.165 brouard 310: Revision 1.164 2014/12/16 10:52:11 brouard
311: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
312:
313: * imach.c (Module): Merging 1.61 to 1.162
314:
1.164 brouard 315: Revision 1.163 2014/12/16 10:30:11 brouard
316: * imach.c (Module): Merging 1.61 to 1.162
317:
1.163 brouard 318: Revision 1.162 2014/09/25 11:43:39 brouard
319: Summary: temporary backup 0.99!
320:
1.162 brouard 321: Revision 1.1 2014/09/16 11:06:58 brouard
322: Summary: With some code (wrong) for nlopt
323:
324: Author:
325:
326: Revision 1.161 2014/09/15 20:41:41 brouard
327: Summary: Problem with macro SQR on Intel compiler
328:
1.161 brouard 329: Revision 1.160 2014/09/02 09:24:05 brouard
330: *** empty log message ***
331:
1.160 brouard 332: Revision 1.159 2014/09/01 10:34:10 brouard
333: Summary: WIN32
334: Author: Brouard
335:
1.159 brouard 336: Revision 1.158 2014/08/27 17:11:51 brouard
337: *** empty log message ***
338:
1.158 brouard 339: Revision 1.157 2014/08/27 16:26:55 brouard
340: Summary: Preparing windows Visual studio version
341: Author: Brouard
342:
343: In order to compile on Visual studio, time.h is now correct and time_t
344: and tm struct should be used. difftime should be used but sometimes I
345: just make the differences in raw time format (time(&now).
346: Trying to suppress #ifdef LINUX
347: Add xdg-open for __linux in order to open default browser.
348:
1.157 brouard 349: Revision 1.156 2014/08/25 20:10:10 brouard
350: *** empty log message ***
351:
1.156 brouard 352: Revision 1.155 2014/08/25 18:32:34 brouard
353: Summary: New compile, minor changes
354: Author: Brouard
355:
1.155 brouard 356: Revision 1.154 2014/06/20 17:32:08 brouard
357: Summary: Outputs now all graphs of convergence to period prevalence
358:
1.154 brouard 359: Revision 1.153 2014/06/20 16:45:46 brouard
360: Summary: If 3 live state, convergence to period prevalence on same graph
361: Author: Brouard
362:
1.153 brouard 363: Revision 1.152 2014/06/18 17:54:09 brouard
364: Summary: open browser, use gnuplot on same dir than imach if not found in the path
365:
1.152 brouard 366: Revision 1.151 2014/06/18 16:43:30 brouard
367: *** empty log message ***
368:
1.151 brouard 369: Revision 1.150 2014/06/18 16:42:35 brouard
370: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
371: Author: brouard
372:
1.150 brouard 373: Revision 1.149 2014/06/18 15:51:14 brouard
374: Summary: Some fixes in parameter files errors
375: Author: Nicolas Brouard
376:
1.149 brouard 377: Revision 1.148 2014/06/17 17:38:48 brouard
378: Summary: Nothing new
379: Author: Brouard
380:
381: Just a new packaging for OS/X version 0.98nS
382:
1.148 brouard 383: Revision 1.147 2014/06/16 10:33:11 brouard
384: *** empty log message ***
385:
1.147 brouard 386: Revision 1.146 2014/06/16 10:20:28 brouard
387: Summary: Merge
388: Author: Brouard
389:
390: Merge, before building revised version.
391:
1.146 brouard 392: Revision 1.145 2014/06/10 21:23:15 brouard
393: Summary: Debugging with valgrind
394: Author: Nicolas Brouard
395:
396: Lot of changes in order to output the results with some covariates
397: After the Edimburgh REVES conference 2014, it seems mandatory to
398: improve the code.
399: No more memory valgrind error but a lot has to be done in order to
400: continue the work of splitting the code into subroutines.
401: Also, decodemodel has been improved. Tricode is still not
402: optimal. nbcode should be improved. Documentation has been added in
403: the source code.
404:
1.144 brouard 405: Revision 1.143 2014/01/26 09:45:38 brouard
406: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
407:
408: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
409: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
410:
1.143 brouard 411: Revision 1.142 2014/01/26 03:57:36 brouard
412: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
413:
414: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
415:
1.142 brouard 416: Revision 1.141 2014/01/26 02:42:01 brouard
417: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
418:
1.141 brouard 419: Revision 1.140 2011/09/02 10:37:54 brouard
420: Summary: times.h is ok with mingw32 now.
421:
1.140 brouard 422: Revision 1.139 2010/06/14 07:50:17 brouard
423: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
424: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
425:
1.139 brouard 426: Revision 1.138 2010/04/30 18:19:40 brouard
427: *** empty log message ***
428:
1.138 brouard 429: Revision 1.137 2010/04/29 18:11:38 brouard
430: (Module): Checking covariates for more complex models
431: than V1+V2. A lot of change to be done. Unstable.
432:
1.137 brouard 433: Revision 1.136 2010/04/26 20:30:53 brouard
434: (Module): merging some libgsl code. Fixing computation
435: of likelione (using inter/intrapolation if mle = 0) in order to
436: get same likelihood as if mle=1.
437: Some cleaning of code and comments added.
438:
1.136 brouard 439: Revision 1.135 2009/10/29 15:33:14 brouard
440: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
441:
1.135 brouard 442: Revision 1.134 2009/10/29 13:18:53 brouard
443: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
444:
1.134 brouard 445: Revision 1.133 2009/07/06 10:21:25 brouard
446: just nforces
447:
1.133 brouard 448: Revision 1.132 2009/07/06 08:22:05 brouard
449: Many tings
450:
1.132 brouard 451: Revision 1.131 2009/06/20 16:22:47 brouard
452: Some dimensions resccaled
453:
1.131 brouard 454: Revision 1.130 2009/05/26 06:44:34 brouard
455: (Module): Max Covariate is now set to 20 instead of 8. A
456: lot of cleaning with variables initialized to 0. Trying to make
457: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
458:
1.130 brouard 459: Revision 1.129 2007/08/31 13:49:27 lievre
460: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
461:
1.129 lievre 462: Revision 1.128 2006/06/30 13:02:05 brouard
463: (Module): Clarifications on computing e.j
464:
1.128 brouard 465: Revision 1.127 2006/04/28 18:11:50 brouard
466: (Module): Yes the sum of survivors was wrong since
467: imach-114 because nhstepm was no more computed in the age
468: loop. Now we define nhstepma in the age loop.
469: (Module): In order to speed up (in case of numerous covariates) we
470: compute health expectancies (without variances) in a first step
471: and then all the health expectancies with variances or standard
472: deviation (needs data from the Hessian matrices) which slows the
473: computation.
474: In the future we should be able to stop the program is only health
475: expectancies and graph are needed without standard deviations.
476:
1.127 brouard 477: Revision 1.126 2006/04/28 17:23:28 brouard
478: (Module): Yes the sum of survivors was wrong since
479: imach-114 because nhstepm was no more computed in the age
480: loop. Now we define nhstepma in the age loop.
481: Version 0.98h
482:
1.126 brouard 483: Revision 1.125 2006/04/04 15:20:31 lievre
484: Errors in calculation of health expectancies. Age was not initialized.
485: Forecasting file added.
486:
487: Revision 1.124 2006/03/22 17:13:53 lievre
488: Parameters are printed with %lf instead of %f (more numbers after the comma).
489: The log-likelihood is printed in the log file
490:
491: Revision 1.123 2006/03/20 10:52:43 brouard
492: * imach.c (Module): <title> changed, corresponds to .htm file
493: name. <head> headers where missing.
494:
495: * imach.c (Module): Weights can have a decimal point as for
496: English (a comma might work with a correct LC_NUMERIC environment,
497: otherwise the weight is truncated).
498: Modification of warning when the covariates values are not 0 or
499: 1.
500: Version 0.98g
501:
502: Revision 1.122 2006/03/20 09:45:41 brouard
503: (Module): Weights can have a decimal point as for
504: English (a comma might work with a correct LC_NUMERIC environment,
505: otherwise the weight is truncated).
506: Modification of warning when the covariates values are not 0 or
507: 1.
508: Version 0.98g
509:
510: Revision 1.121 2006/03/16 17:45:01 lievre
511: * imach.c (Module): Comments concerning covariates added
512:
513: * imach.c (Module): refinements in the computation of lli if
514: status=-2 in order to have more reliable computation if stepm is
515: not 1 month. Version 0.98f
516:
517: Revision 1.120 2006/03/16 15:10:38 lievre
518: (Module): refinements in the computation of lli if
519: status=-2 in order to have more reliable computation if stepm is
520: not 1 month. Version 0.98f
521:
522: Revision 1.119 2006/03/15 17:42:26 brouard
523: (Module): Bug if status = -2, the loglikelihood was
524: computed as likelihood omitting the logarithm. Version O.98e
525:
526: Revision 1.118 2006/03/14 18:20:07 brouard
527: (Module): varevsij Comments added explaining the second
528: table of variances if popbased=1 .
529: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
530: (Module): Function pstamp added
531: (Module): Version 0.98d
532:
533: Revision 1.117 2006/03/14 17:16:22 brouard
534: (Module): varevsij Comments added explaining the second
535: table of variances if popbased=1 .
536: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
537: (Module): Function pstamp added
538: (Module): Version 0.98d
539:
540: Revision 1.116 2006/03/06 10:29:27 brouard
541: (Module): Variance-covariance wrong links and
542: varian-covariance of ej. is needed (Saito).
543:
544: Revision 1.115 2006/02/27 12:17:45 brouard
545: (Module): One freematrix added in mlikeli! 0.98c
546:
547: Revision 1.114 2006/02/26 12:57:58 brouard
548: (Module): Some improvements in processing parameter
549: filename with strsep.
550:
551: Revision 1.113 2006/02/24 14:20:24 brouard
552: (Module): Memory leaks checks with valgrind and:
553: datafile was not closed, some imatrix were not freed and on matrix
554: allocation too.
555:
556: Revision 1.112 2006/01/30 09:55:26 brouard
557: (Module): Back to gnuplot.exe instead of wgnuplot.exe
558:
559: Revision 1.111 2006/01/25 20:38:18 brouard
560: (Module): Lots of cleaning and bugs added (Gompertz)
561: (Module): Comments can be added in data file. Missing date values
562: can be a simple dot '.'.
563:
564: Revision 1.110 2006/01/25 00:51:50 brouard
565: (Module): Lots of cleaning and bugs added (Gompertz)
566:
567: Revision 1.109 2006/01/24 19:37:15 brouard
568: (Module): Comments (lines starting with a #) are allowed in data.
569:
570: Revision 1.108 2006/01/19 18:05:42 lievre
571: Gnuplot problem appeared...
572: To be fixed
573:
574: Revision 1.107 2006/01/19 16:20:37 brouard
575: Test existence of gnuplot in imach path
576:
577: Revision 1.106 2006/01/19 13:24:36 brouard
578: Some cleaning and links added in html output
579:
580: Revision 1.105 2006/01/05 20:23:19 lievre
581: *** empty log message ***
582:
583: Revision 1.104 2005/09/30 16:11:43 lievre
584: (Module): sump fixed, loop imx fixed, and simplifications.
585: (Module): If the status is missing at the last wave but we know
586: that the person is alive, then we can code his/her status as -2
587: (instead of missing=-1 in earlier versions) and his/her
588: contributions to the likelihood is 1 - Prob of dying from last
589: health status (= 1-p13= p11+p12 in the easiest case of somebody in
590: the healthy state at last known wave). Version is 0.98
591:
592: Revision 1.103 2005/09/30 15:54:49 lievre
593: (Module): sump fixed, loop imx fixed, and simplifications.
594:
595: Revision 1.102 2004/09/15 17:31:30 brouard
596: Add the possibility to read data file including tab characters.
597:
598: Revision 1.101 2004/09/15 10:38:38 brouard
599: Fix on curr_time
600:
601: Revision 1.100 2004/07/12 18:29:06 brouard
602: Add version for Mac OS X. Just define UNIX in Makefile
603:
604: Revision 1.99 2004/06/05 08:57:40 brouard
605: *** empty log message ***
606:
607: Revision 1.98 2004/05/16 15:05:56 brouard
608: New version 0.97 . First attempt to estimate force of mortality
609: directly from the data i.e. without the need of knowing the health
610: state at each age, but using a Gompertz model: log u =a + b*age .
611: This is the basic analysis of mortality and should be done before any
612: other analysis, in order to test if the mortality estimated from the
613: cross-longitudinal survey is different from the mortality estimated
614: from other sources like vital statistic data.
615:
616: The same imach parameter file can be used but the option for mle should be -3.
617:
1.133 brouard 618: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 619: former routines in order to include the new code within the former code.
620:
621: The output is very simple: only an estimate of the intercept and of
622: the slope with 95% confident intervals.
623:
624: Current limitations:
625: A) Even if you enter covariates, i.e. with the
626: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
627: B) There is no computation of Life Expectancy nor Life Table.
628:
629: Revision 1.97 2004/02/20 13:25:42 lievre
630: Version 0.96d. Population forecasting command line is (temporarily)
631: suppressed.
632:
633: Revision 1.96 2003/07/15 15:38:55 brouard
634: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
635: rewritten within the same printf. Workaround: many printfs.
636:
637: Revision 1.95 2003/07/08 07:54:34 brouard
638: * imach.c (Repository):
639: (Repository): Using imachwizard code to output a more meaningful covariance
640: matrix (cov(a12,c31) instead of numbers.
641:
642: Revision 1.94 2003/06/27 13:00:02 brouard
643: Just cleaning
644:
645: Revision 1.93 2003/06/25 16:33:55 brouard
646: (Module): On windows (cygwin) function asctime_r doesn't
647: exist so I changed back to asctime which exists.
648: (Module): Version 0.96b
649:
650: Revision 1.92 2003/06/25 16:30:45 brouard
651: (Module): On windows (cygwin) function asctime_r doesn't
652: exist so I changed back to asctime which exists.
653:
654: Revision 1.91 2003/06/25 15:30:29 brouard
655: * imach.c (Repository): Duplicated warning errors corrected.
656: (Repository): Elapsed time after each iteration is now output. It
657: helps to forecast when convergence will be reached. Elapsed time
658: is stamped in powell. We created a new html file for the graphs
659: concerning matrix of covariance. It has extension -cov.htm.
660:
661: Revision 1.90 2003/06/24 12:34:15 brouard
662: (Module): Some bugs corrected for windows. Also, when
663: mle=-1 a template is output in file "or"mypar.txt with the design
664: of the covariance matrix to be input.
665:
666: Revision 1.89 2003/06/24 12:30:52 brouard
667: (Module): Some bugs corrected for windows. Also, when
668: mle=-1 a template is output in file "or"mypar.txt with the design
669: of the covariance matrix to be input.
670:
671: Revision 1.88 2003/06/23 17:54:56 brouard
672: * 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.
673:
674: Revision 1.87 2003/06/18 12:26:01 brouard
675: Version 0.96
676:
677: Revision 1.86 2003/06/17 20:04:08 brouard
678: (Module): Change position of html and gnuplot routines and added
679: routine fileappend.
680:
681: Revision 1.85 2003/06/17 13:12:43 brouard
682: * imach.c (Repository): Check when date of death was earlier that
683: current date of interview. It may happen when the death was just
684: prior to the death. In this case, dh was negative and likelihood
685: was wrong (infinity). We still send an "Error" but patch by
686: assuming that the date of death was just one stepm after the
687: interview.
688: (Repository): Because some people have very long ID (first column)
689: we changed int to long in num[] and we added a new lvector for
690: memory allocation. But we also truncated to 8 characters (left
691: truncation)
692: (Repository): No more line truncation errors.
693:
694: Revision 1.84 2003/06/13 21:44:43 brouard
695: * imach.c (Repository): Replace "freqsummary" at a correct
696: place. It differs from routine "prevalence" which may be called
697: many times. Probs is memory consuming and must be used with
698: parcimony.
699: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
700:
701: Revision 1.83 2003/06/10 13:39:11 lievre
702: *** empty log message ***
703:
704: Revision 1.82 2003/06/05 15:57:20 brouard
705: Add log in imach.c and fullversion number is now printed.
706:
707: */
708: /*
709: Interpolated Markov Chain
710:
711: Short summary of the programme:
712:
1.227 brouard 713: This program computes Healthy Life Expectancies or State-specific
714: (if states aren't health statuses) Expectancies from
715: cross-longitudinal data. Cross-longitudinal data consist in:
716:
717: -1- a first survey ("cross") where individuals from different ages
718: are interviewed on their health status or degree of disability (in
719: the case of a health survey which is our main interest)
720:
721: -2- at least a second wave of interviews ("longitudinal") which
722: measure each change (if any) in individual health status. Health
723: expectancies are computed from the time spent in each health state
724: according to a model. More health states you consider, more time is
725: necessary to reach the Maximum Likelihood of the parameters involved
726: in the model. The simplest model is the multinomial logistic model
727: where pij is the probability to be observed in state j at the second
728: wave conditional to be observed in state i at the first
729: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
730: etc , where 'age' is age and 'sex' is a covariate. If you want to
731: have a more complex model than "constant and age", you should modify
732: the program where the markup *Covariates have to be included here
733: again* invites you to do it. More covariates you add, slower the
1.126 brouard 734: convergence.
735:
736: The advantage of this computer programme, compared to a simple
737: multinomial logistic model, is clear when the delay between waves is not
738: identical for each individual. Also, if a individual missed an
739: intermediate interview, the information is lost, but taken into
740: account using an interpolation or extrapolation.
741:
742: hPijx is the probability to be observed in state i at age x+h
743: conditional to the observed state i at age x. The delay 'h' can be
744: split into an exact number (nh*stepm) of unobserved intermediate
745: states. This elementary transition (by month, quarter,
746: semester or year) is modelled as a multinomial logistic. The hPx
747: matrix is simply the matrix product of nh*stepm elementary matrices
748: and the contribution of each individual to the likelihood is simply
749: hPijx.
750:
751: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 752: of the life expectancies. It also computes the period (stable) prevalence.
753:
754: Back prevalence and projections:
1.227 brouard 755:
756: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
757: double agemaxpar, double ftolpl, int *ncvyearp, double
758: dateprev1,double dateprev2, int firstpass, int lastpass, int
759: mobilavproj)
760:
761: Computes the back prevalence limit for any combination of
762: covariate values k at any age between ageminpar and agemaxpar and
763: returns it in **bprlim. In the loops,
764:
765: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
766: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
767:
768: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 769: Computes for any combination of covariates k and any age between bage and fage
770: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
771: oldm=oldms;savm=savms;
1.227 brouard 772:
773: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 774: Computes the transition matrix starting at age 'age' over
775: 'nhstepm*hstepm*stepm' months (i.e. until
776: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 777: nhstepm*hstepm matrices.
778:
779: Returns p3mat[i][j][h] after calling
780: p3mat[i][j][h]=matprod2(newm,
781: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
782: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
783: oldm);
1.226 brouard 784:
785: Important routines
786:
787: - func (or funcone), computes logit (pij) distinguishing
788: o fixed variables (single or product dummies or quantitative);
789: o varying variables by:
790: (1) wave (single, product dummies, quantitative),
791: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
792: % fixed dummy (treated) or quantitative (not done because time-consuming);
793: % varying dummy (not done) or quantitative (not done);
794: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
795: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
796: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
797: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
798: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 799:
1.226 brouard 800:
801:
1.133 brouard 802: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
803: Institut national d'études démographiques, Paris.
1.126 brouard 804: This software have been partly granted by Euro-REVES, a concerted action
805: from the European Union.
806: It is copyrighted identically to a GNU software product, ie programme and
807: software can be distributed freely for non commercial use. Latest version
808: can be accessed at http://euroreves.ined.fr/imach .
809:
810: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
811: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
812:
813: **********************************************************************/
814: /*
815: main
816: read parameterfile
817: read datafile
818: concatwav
819: freqsummary
820: if (mle >= 1)
821: mlikeli
822: print results files
823: if mle==1
824: computes hessian
825: read end of parameter file: agemin, agemax, bage, fage, estepm
826: begin-prev-date,...
827: open gnuplot file
828: open html file
1.145 brouard 829: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
830: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
831: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
832: freexexit2 possible for memory heap.
833:
834: h Pij x | pij_nom ficrestpij
835: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
836: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
837: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
838:
839: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
840: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
841: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
842: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
843: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
844:
1.126 brouard 845: forecasting if prevfcast==1 prevforecast call prevalence()
846: health expectancies
847: Variance-covariance of DFLE
848: prevalence()
849: movingaverage()
850: varevsij()
851: if popbased==1 varevsij(,popbased)
852: total life expectancies
853: Variance of period (stable) prevalence
854: end
855: */
856:
1.187 brouard 857: /* #define DEBUG */
858: /* #define DEBUGBRENT */
1.203 brouard 859: /* #define DEBUGLINMIN */
860: /* #define DEBUGHESS */
861: #define DEBUGHESSIJ
1.224 brouard 862: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 863: #define POWELL /* Instead of NLOPT */
1.224 brouard 864: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 865: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
866: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 867:
868: #include <math.h>
869: #include <stdio.h>
870: #include <stdlib.h>
871: #include <string.h>
1.226 brouard 872: #include <ctype.h>
1.159 brouard 873:
874: #ifdef _WIN32
875: #include <io.h>
1.172 brouard 876: #include <windows.h>
877: #include <tchar.h>
1.159 brouard 878: #else
1.126 brouard 879: #include <unistd.h>
1.159 brouard 880: #endif
1.126 brouard 881:
882: #include <limits.h>
883: #include <sys/types.h>
1.171 brouard 884:
885: #if defined(__GNUC__)
886: #include <sys/utsname.h> /* Doesn't work on Windows */
887: #endif
888:
1.126 brouard 889: #include <sys/stat.h>
890: #include <errno.h>
1.159 brouard 891: /* extern int errno; */
1.126 brouard 892:
1.157 brouard 893: /* #ifdef LINUX */
894: /* #include <time.h> */
895: /* #include "timeval.h" */
896: /* #else */
897: /* #include <sys/time.h> */
898: /* #endif */
899:
1.126 brouard 900: #include <time.h>
901:
1.136 brouard 902: #ifdef GSL
903: #include <gsl/gsl_errno.h>
904: #include <gsl/gsl_multimin.h>
905: #endif
906:
1.167 brouard 907:
1.162 brouard 908: #ifdef NLOPT
909: #include <nlopt.h>
910: typedef struct {
911: double (* function)(double [] );
912: } myfunc_data ;
913: #endif
914:
1.126 brouard 915: /* #include <libintl.h> */
916: /* #define _(String) gettext (String) */
917:
1.251 ! brouard 918: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 919:
920: #define GNUPLOTPROGRAM "gnuplot"
921: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
922: #define FILENAMELENGTH 132
923:
924: #define GLOCK_ERROR_NOPATH -1 /* empty path */
925: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
926:
1.144 brouard 927: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
928: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 929:
930: #define NINTERVMAX 8
1.144 brouard 931: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
932: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
933: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 934: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 935: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
936: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 937: #define MAXN 20000
1.144 brouard 938: #define YEARM 12. /**< Number of months per year */
1.218 brouard 939: /* #define AGESUP 130 */
940: #define AGESUP 150
941: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 942: #define AGEBASE 40
1.194 brouard 943: #define AGEOVERFLOW 1.e20
1.164 brouard 944: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 945: #ifdef _WIN32
946: #define DIRSEPARATOR '\\'
947: #define CHARSEPARATOR "\\"
948: #define ODIRSEPARATOR '/'
949: #else
1.126 brouard 950: #define DIRSEPARATOR '/'
951: #define CHARSEPARATOR "/"
952: #define ODIRSEPARATOR '\\'
953: #endif
954:
1.251 ! brouard 955: /* $Id: imach.c,v 1.250 2016/09/08 16:07:27 brouard Exp $ */
1.126 brouard 956: /* $State: Exp $ */
1.196 brouard 957: #include "version.h"
958: char version[]=__IMACH_VERSION__;
1.224 brouard 959: 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";
1.251 ! brouard 960: char fullversion[]="$Revision: 1.250 $ $Date: 2016/09/08 16:07:27 $";
1.126 brouard 961: char strstart[80];
962: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 963: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 964: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 965: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
966: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
967: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 968: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
969: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 970: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
971: int cptcovprodnoage=0; /**< Number of covariate products without age */
972: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 973: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
974: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 975: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 976: int nsd=0; /**< Total number of single dummy variables (output) */
977: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 978: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 979: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 980: int ntveff=0; /**< ntveff number of effective time varying variables */
981: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 982: int cptcov=0; /* Working variable */
1.218 brouard 983: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 984: int npar=NPARMAX;
985: int nlstate=2; /* Number of live states */
986: int ndeath=1; /* Number of dead states */
1.130 brouard 987: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 988: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 989: int popbased=0;
990:
991: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 992: int maxwav=0; /* Maxim number of waves */
993: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
994: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
995: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 996: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 997: int mle=1, weightopt=0;
1.126 brouard 998: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
999: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1000: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1001: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1002: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1003: int selected(int kvar); /* Is covariate kvar selected for printing results */
1004:
1.130 brouard 1005: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1006: double **matprod2(); /* test */
1.126 brouard 1007: double **oldm, **newm, **savm; /* Working pointers to matrices */
1008: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1009: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1010:
1.136 brouard 1011: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1012: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1013: FILE *ficlog, *ficrespow;
1.130 brouard 1014: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1015: double fretone; /* Only one call to likelihood */
1.130 brouard 1016: long ipmx=0; /* Number of contributions */
1.126 brouard 1017: double sw; /* Sum of weights */
1018: char filerespow[FILENAMELENGTH];
1019: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1020: FILE *ficresilk;
1021: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1022: FILE *ficresprobmorprev;
1023: FILE *fichtm, *fichtmcov; /* Html File */
1024: FILE *ficreseij;
1025: char filerese[FILENAMELENGTH];
1026: FILE *ficresstdeij;
1027: char fileresstde[FILENAMELENGTH];
1028: FILE *ficrescveij;
1029: char filerescve[FILENAMELENGTH];
1030: FILE *ficresvij;
1031: char fileresv[FILENAMELENGTH];
1032: FILE *ficresvpl;
1033: char fileresvpl[FILENAMELENGTH];
1034: char title[MAXLINE];
1.234 brouard 1035: char model[MAXLINE]; /**< The model line */
1.217 brouard 1036: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1037: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1038: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1039: char command[FILENAMELENGTH];
1040: int outcmd=0;
1041:
1.217 brouard 1042: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1043: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1044: char filelog[FILENAMELENGTH]; /* Log file */
1045: char filerest[FILENAMELENGTH];
1046: char fileregp[FILENAMELENGTH];
1047: char popfile[FILENAMELENGTH];
1048:
1049: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1050:
1.157 brouard 1051: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1052: /* struct timezone tzp; */
1053: /* extern int gettimeofday(); */
1054: struct tm tml, *gmtime(), *localtime();
1055:
1056: extern time_t time();
1057:
1058: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1059: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1060: struct tm tm;
1061:
1.126 brouard 1062: char strcurr[80], strfor[80];
1063:
1064: char *endptr;
1065: long lval;
1066: double dval;
1067:
1068: #define NR_END 1
1069: #define FREE_ARG char*
1070: #define FTOL 1.0e-10
1071:
1072: #define NRANSI
1.240 brouard 1073: #define ITMAX 200
1074: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1075:
1076: #define TOL 2.0e-4
1077:
1078: #define CGOLD 0.3819660
1079: #define ZEPS 1.0e-10
1080: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1081:
1082: #define GOLD 1.618034
1083: #define GLIMIT 100.0
1084: #define TINY 1.0e-20
1085:
1086: static double maxarg1,maxarg2;
1087: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1088: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1089:
1090: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1091: #define rint(a) floor(a+0.5)
1.166 brouard 1092: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1093: #define mytinydouble 1.0e-16
1.166 brouard 1094: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1095: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1096: /* static double dsqrarg; */
1097: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1098: static double sqrarg;
1099: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1100: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1101: int agegomp= AGEGOMP;
1102:
1103: int imx;
1104: int stepm=1;
1105: /* Stepm, step in month: minimum step interpolation*/
1106:
1107: int estepm;
1108: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1109:
1110: int m,nb;
1111: long *num;
1.197 brouard 1112: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1113: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1114: covariate for which somebody answered excluding
1115: undefined. Usually 2: 0 and 1. */
1116: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1117: covariate for which somebody answered including
1118: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1119: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1120: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1121: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1122: double *ageexmed,*agecens;
1123: double dateintmean=0;
1124:
1125: double *weight;
1126: int **s; /* Status */
1.141 brouard 1127: double *agedc;
1.145 brouard 1128: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1129: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1130: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1131: double **coqvar; /* Fixed quantitative covariate iqv */
1132: double ***cotvar; /* Time varying covariate itv */
1133: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1134: double idx;
1135: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1136: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1137: /*k 1 2 3 4 5 6 7 8 9 */
1138: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1139: /* Tndvar[k] 1 2 3 4 5 */
1140: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1141: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1142: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1143: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1144: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1145: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1146: /* Tprod[i]=k 4 7 */
1147: /* Tage[i]=k 5 8 */
1148: /* */
1149: /* Type */
1150: /* V 1 2 3 4 5 */
1151: /* F F V V V */
1152: /* D Q D D Q */
1153: /* */
1154: int *TvarsD;
1155: int *TvarsDind;
1156: int *TvarsQ;
1157: int *TvarsQind;
1158:
1.235 brouard 1159: #define MAXRESULTLINES 10
1160: int nresult=0;
1161: int TKresult[MAXRESULTLINES];
1.237 brouard 1162: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1163: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1164: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1165: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1166: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1167: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1168:
1.234 brouard 1169: /* 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 1170: 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 */
1171: 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 */
1172: 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 */
1173: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1174: 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 */
1175: 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 1176: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1177: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1178: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1179: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1180: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1181: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1182: 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 */
1183: 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 */
1184:
1.230 brouard 1185: int *Tvarsel; /**< Selected covariates for output */
1186: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1187: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1188: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1189: 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 1190: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1191: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1192: int *Tage;
1.227 brouard 1193: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1194: 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 1195: 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*/
1196: 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 1197: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1198: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1199: int **Tvard;
1200: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1201: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1202: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1203: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1204: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1205: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1206: double *lsurv, *lpop, *tpop;
1207:
1.231 brouard 1208: #define FD 1; /* Fixed dummy covariate */
1209: #define FQ 2; /* Fixed quantitative covariate */
1210: #define FP 3; /* Fixed product covariate */
1211: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1212: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1213: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1214: #define VD 10; /* Varying dummy covariate */
1215: #define VQ 11; /* Varying quantitative covariate */
1216: #define VP 12; /* Varying product covariate */
1217: #define VPDD 13; /* Varying product dummy*dummy covariate */
1218: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1219: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1220: #define APFD 16; /* Age product * fixed dummy covariate */
1221: #define APFQ 17; /* Age product * fixed quantitative covariate */
1222: #define APVD 18; /* Age product * varying dummy covariate */
1223: #define APVQ 19; /* Age product * varying quantitative covariate */
1224:
1225: #define FTYPE 1; /* Fixed covariate */
1226: #define VTYPE 2; /* Varying covariate (loop in wave) */
1227: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1228:
1229: struct kmodel{
1230: int maintype; /* main type */
1231: int subtype; /* subtype */
1232: };
1233: struct kmodel modell[NCOVMAX];
1234:
1.143 brouard 1235: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1236: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1237:
1238: /**************** split *************************/
1239: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1240: {
1241: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1242: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1243: */
1244: char *ss; /* pointer */
1.186 brouard 1245: int l1=0, l2=0; /* length counters */
1.126 brouard 1246:
1247: l1 = strlen(path ); /* length of path */
1248: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1249: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1250: if ( ss == NULL ) { /* no directory, so determine current directory */
1251: strcpy( name, path ); /* we got the fullname name because no directory */
1252: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1253: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1254: /* get current working directory */
1255: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1256: #ifdef WIN32
1257: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1258: #else
1259: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1260: #endif
1.126 brouard 1261: return( GLOCK_ERROR_GETCWD );
1262: }
1263: /* got dirc from getcwd*/
1264: printf(" DIRC = %s \n",dirc);
1.205 brouard 1265: } else { /* strip directory from path */
1.126 brouard 1266: ss++; /* after this, the filename */
1267: l2 = strlen( ss ); /* length of filename */
1268: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1269: strcpy( name, ss ); /* save file name */
1270: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1271: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1272: printf(" DIRC2 = %s \n",dirc);
1273: }
1274: /* We add a separator at the end of dirc if not exists */
1275: l1 = strlen( dirc ); /* length of directory */
1276: if( dirc[l1-1] != DIRSEPARATOR ){
1277: dirc[l1] = DIRSEPARATOR;
1278: dirc[l1+1] = 0;
1279: printf(" DIRC3 = %s \n",dirc);
1280: }
1281: ss = strrchr( name, '.' ); /* find last / */
1282: if (ss >0){
1283: ss++;
1284: strcpy(ext,ss); /* save extension */
1285: l1= strlen( name);
1286: l2= strlen(ss)+1;
1287: strncpy( finame, name, l1-l2);
1288: finame[l1-l2]= 0;
1289: }
1290:
1291: return( 0 ); /* we're done */
1292: }
1293:
1294:
1295: /******************************************/
1296:
1297: void replace_back_to_slash(char *s, char*t)
1298: {
1299: int i;
1300: int lg=0;
1301: i=0;
1302: lg=strlen(t);
1303: for(i=0; i<= lg; i++) {
1304: (s[i] = t[i]);
1305: if (t[i]== '\\') s[i]='/';
1306: }
1307: }
1308:
1.132 brouard 1309: char *trimbb(char *out, char *in)
1.137 brouard 1310: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1311: char *s;
1312: s=out;
1313: while (*in != '\0'){
1.137 brouard 1314: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1315: in++;
1316: }
1317: *out++ = *in++;
1318: }
1319: *out='\0';
1320: return s;
1321: }
1322:
1.187 brouard 1323: /* char *substrchaine(char *out, char *in, char *chain) */
1324: /* { */
1325: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1326: /* char *s, *t; */
1327: /* t=in;s=out; */
1328: /* while ((*in != *chain) && (*in != '\0')){ */
1329: /* *out++ = *in++; */
1330: /* } */
1331:
1332: /* /\* *in matches *chain *\/ */
1333: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1334: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1335: /* } */
1336: /* in--; chain--; */
1337: /* while ( (*in != '\0')){ */
1338: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1339: /* *out++ = *in++; */
1340: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1341: /* } */
1342: /* *out='\0'; */
1343: /* out=s; */
1344: /* return out; */
1345: /* } */
1346: char *substrchaine(char *out, char *in, char *chain)
1347: {
1348: /* Substract chain 'chain' from 'in', return and output 'out' */
1349: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1350:
1351: char *strloc;
1352:
1353: strcpy (out, in);
1354: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1355: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1356: if(strloc != NULL){
1357: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1358: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1359: /* strcpy (strloc, strloc +strlen(chain));*/
1360: }
1361: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1362: return out;
1363: }
1364:
1365:
1.145 brouard 1366: char *cutl(char *blocc, char *alocc, char *in, char occ)
1367: {
1.187 brouard 1368: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1369: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1370: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1371: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1372: */
1.160 brouard 1373: char *s, *t;
1.145 brouard 1374: t=in;s=in;
1375: while ((*in != occ) && (*in != '\0')){
1376: *alocc++ = *in++;
1377: }
1378: if( *in == occ){
1379: *(alocc)='\0';
1380: s=++in;
1381: }
1382:
1383: if (s == t) {/* occ not found */
1384: *(alocc-(in-s))='\0';
1385: in=s;
1386: }
1387: while ( *in != '\0'){
1388: *blocc++ = *in++;
1389: }
1390:
1391: *blocc='\0';
1392: return t;
1393: }
1.137 brouard 1394: char *cutv(char *blocc, char *alocc, char *in, char occ)
1395: {
1.187 brouard 1396: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1397: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1398: gives blocc="abcdef2ghi" and alocc="j".
1399: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1400: */
1401: char *s, *t;
1402: t=in;s=in;
1403: while (*in != '\0'){
1404: while( *in == occ){
1405: *blocc++ = *in++;
1406: s=in;
1407: }
1408: *blocc++ = *in++;
1409: }
1410: if (s == t) /* occ not found */
1411: *(blocc-(in-s))='\0';
1412: else
1413: *(blocc-(in-s)-1)='\0';
1414: in=s;
1415: while ( *in != '\0'){
1416: *alocc++ = *in++;
1417: }
1418:
1419: *alocc='\0';
1420: return s;
1421: }
1422:
1.126 brouard 1423: int nbocc(char *s, char occ)
1424: {
1425: int i,j=0;
1426: int lg=20;
1427: i=0;
1428: lg=strlen(s);
1429: for(i=0; i<= lg; i++) {
1.234 brouard 1430: if (s[i] == occ ) j++;
1.126 brouard 1431: }
1432: return j;
1433: }
1434:
1.137 brouard 1435: /* void cutv(char *u,char *v, char*t, char occ) */
1436: /* { */
1437: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1438: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1439: /* gives u="abcdef2ghi" and v="j" *\/ */
1440: /* int i,lg,j,p=0; */
1441: /* i=0; */
1442: /* lg=strlen(t); */
1443: /* for(j=0; j<=lg-1; j++) { */
1444: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1445: /* } */
1.126 brouard 1446:
1.137 brouard 1447: /* for(j=0; j<p; j++) { */
1448: /* (u[j] = t[j]); */
1449: /* } */
1450: /* u[p]='\0'; */
1.126 brouard 1451:
1.137 brouard 1452: /* for(j=0; j<= lg; j++) { */
1453: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1454: /* } */
1455: /* } */
1.126 brouard 1456:
1.160 brouard 1457: #ifdef _WIN32
1458: char * strsep(char **pp, const char *delim)
1459: {
1460: char *p, *q;
1461:
1462: if ((p = *pp) == NULL)
1463: return 0;
1464: if ((q = strpbrk (p, delim)) != NULL)
1465: {
1466: *pp = q + 1;
1467: *q = '\0';
1468: }
1469: else
1470: *pp = 0;
1471: return p;
1472: }
1473: #endif
1474:
1.126 brouard 1475: /********************** nrerror ********************/
1476:
1477: void nrerror(char error_text[])
1478: {
1479: fprintf(stderr,"ERREUR ...\n");
1480: fprintf(stderr,"%s\n",error_text);
1481: exit(EXIT_FAILURE);
1482: }
1483: /*********************** vector *******************/
1484: double *vector(int nl, int nh)
1485: {
1486: double *v;
1487: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1488: if (!v) nrerror("allocation failure in vector");
1489: return v-nl+NR_END;
1490: }
1491:
1492: /************************ free vector ******************/
1493: void free_vector(double*v, int nl, int nh)
1494: {
1495: free((FREE_ARG)(v+nl-NR_END));
1496: }
1497:
1498: /************************ivector *******************************/
1499: int *ivector(long nl,long nh)
1500: {
1501: int *v;
1502: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1503: if (!v) nrerror("allocation failure in ivector");
1504: return v-nl+NR_END;
1505: }
1506:
1507: /******************free ivector **************************/
1508: void free_ivector(int *v, long nl, long nh)
1509: {
1510: free((FREE_ARG)(v+nl-NR_END));
1511: }
1512:
1513: /************************lvector *******************************/
1514: long *lvector(long nl,long nh)
1515: {
1516: long *v;
1517: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1518: if (!v) nrerror("allocation failure in ivector");
1519: return v-nl+NR_END;
1520: }
1521:
1522: /******************free lvector **************************/
1523: void free_lvector(long *v, long nl, long nh)
1524: {
1525: free((FREE_ARG)(v+nl-NR_END));
1526: }
1527:
1528: /******************* imatrix *******************************/
1529: int **imatrix(long nrl, long nrh, long ncl, long nch)
1530: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1531: {
1532: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1533: int **m;
1534:
1535: /* allocate pointers to rows */
1536: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1537: if (!m) nrerror("allocation failure 1 in matrix()");
1538: m += NR_END;
1539: m -= nrl;
1540:
1541:
1542: /* allocate rows and set pointers to them */
1543: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1544: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1545: m[nrl] += NR_END;
1546: m[nrl] -= ncl;
1547:
1548: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1549:
1550: /* return pointer to array of pointers to rows */
1551: return m;
1552: }
1553:
1554: /****************** free_imatrix *************************/
1555: void free_imatrix(m,nrl,nrh,ncl,nch)
1556: int **m;
1557: long nch,ncl,nrh,nrl;
1558: /* free an int matrix allocated by imatrix() */
1559: {
1560: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1561: free((FREE_ARG) (m+nrl-NR_END));
1562: }
1563:
1564: /******************* matrix *******************************/
1565: double **matrix(long nrl, long nrh, long ncl, long nch)
1566: {
1567: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1568: double **m;
1569:
1570: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1571: if (!m) nrerror("allocation failure 1 in matrix()");
1572: m += NR_END;
1573: m -= nrl;
1574:
1575: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1576: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1577: m[nrl] += NR_END;
1578: m[nrl] -= ncl;
1579:
1580: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1581: return m;
1.145 brouard 1582: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1583: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1584: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1585: */
1586: }
1587:
1588: /*************************free matrix ************************/
1589: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1590: {
1591: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1592: free((FREE_ARG)(m+nrl-NR_END));
1593: }
1594:
1595: /******************* ma3x *******************************/
1596: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1597: {
1598: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1599: double ***m;
1600:
1601: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1602: if (!m) nrerror("allocation failure 1 in matrix()");
1603: m += NR_END;
1604: m -= nrl;
1605:
1606: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1607: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1608: m[nrl] += NR_END;
1609: m[nrl] -= ncl;
1610:
1611: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1612:
1613: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1614: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1615: m[nrl][ncl] += NR_END;
1616: m[nrl][ncl] -= nll;
1617: for (j=ncl+1; j<=nch; j++)
1618: m[nrl][j]=m[nrl][j-1]+nlay;
1619:
1620: for (i=nrl+1; i<=nrh; i++) {
1621: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1622: for (j=ncl+1; j<=nch; j++)
1623: m[i][j]=m[i][j-1]+nlay;
1624: }
1625: return m;
1626: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1627: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1628: */
1629: }
1630:
1631: /*************************free ma3x ************************/
1632: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1633: {
1634: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1635: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1636: free((FREE_ARG)(m+nrl-NR_END));
1637: }
1638:
1639: /*************** function subdirf ***********/
1640: char *subdirf(char fileres[])
1641: {
1642: /* Caution optionfilefiname is hidden */
1643: strcpy(tmpout,optionfilefiname);
1644: strcat(tmpout,"/"); /* Add to the right */
1645: strcat(tmpout,fileres);
1646: return tmpout;
1647: }
1648:
1649: /*************** function subdirf2 ***********/
1650: char *subdirf2(char fileres[], char *preop)
1651: {
1652:
1653: /* Caution optionfilefiname is hidden */
1654: strcpy(tmpout,optionfilefiname);
1655: strcat(tmpout,"/");
1656: strcat(tmpout,preop);
1657: strcat(tmpout,fileres);
1658: return tmpout;
1659: }
1660:
1661: /*************** function subdirf3 ***********/
1662: char *subdirf3(char fileres[], char *preop, char *preop2)
1663: {
1664:
1665: /* Caution optionfilefiname is hidden */
1666: strcpy(tmpout,optionfilefiname);
1667: strcat(tmpout,"/");
1668: strcat(tmpout,preop);
1669: strcat(tmpout,preop2);
1670: strcat(tmpout,fileres);
1671: return tmpout;
1672: }
1.213 brouard 1673:
1674: /*************** function subdirfext ***********/
1675: char *subdirfext(char fileres[], char *preop, char *postop)
1676: {
1677:
1678: strcpy(tmpout,preop);
1679: strcat(tmpout,fileres);
1680: strcat(tmpout,postop);
1681: return tmpout;
1682: }
1.126 brouard 1683:
1.213 brouard 1684: /*************** function subdirfext3 ***********/
1685: char *subdirfext3(char fileres[], char *preop, char *postop)
1686: {
1687:
1688: /* Caution optionfilefiname is hidden */
1689: strcpy(tmpout,optionfilefiname);
1690: strcat(tmpout,"/");
1691: strcat(tmpout,preop);
1692: strcat(tmpout,fileres);
1693: strcat(tmpout,postop);
1694: return tmpout;
1695: }
1696:
1.162 brouard 1697: char *asc_diff_time(long time_sec, char ascdiff[])
1698: {
1699: long sec_left, days, hours, minutes;
1700: days = (time_sec) / (60*60*24);
1701: sec_left = (time_sec) % (60*60*24);
1702: hours = (sec_left) / (60*60) ;
1703: sec_left = (sec_left) %(60*60);
1704: minutes = (sec_left) /60;
1705: sec_left = (sec_left) % (60);
1706: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1707: return ascdiff;
1708: }
1709:
1.126 brouard 1710: /***************** f1dim *************************/
1711: extern int ncom;
1712: extern double *pcom,*xicom;
1713: extern double (*nrfunc)(double []);
1714:
1715: double f1dim(double x)
1716: {
1717: int j;
1718: double f;
1719: double *xt;
1720:
1721: xt=vector(1,ncom);
1722: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1723: f=(*nrfunc)(xt);
1724: free_vector(xt,1,ncom);
1725: return f;
1726: }
1727:
1728: /*****************brent *************************/
1729: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1730: {
1731: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1732: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1733: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1734: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1735: * returned function value.
1736: */
1.126 brouard 1737: int iter;
1738: double a,b,d,etemp;
1.159 brouard 1739: double fu=0,fv,fw,fx;
1.164 brouard 1740: double ftemp=0.;
1.126 brouard 1741: double p,q,r,tol1,tol2,u,v,w,x,xm;
1742: double e=0.0;
1743:
1744: a=(ax < cx ? ax : cx);
1745: b=(ax > cx ? ax : cx);
1746: x=w=v=bx;
1747: fw=fv=fx=(*f)(x);
1748: for (iter=1;iter<=ITMAX;iter++) {
1749: xm=0.5*(a+b);
1750: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1751: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1752: printf(".");fflush(stdout);
1753: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1754: #ifdef DEBUGBRENT
1.126 brouard 1755: 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);
1756: 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);
1757: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1758: #endif
1759: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1760: *xmin=x;
1761: return fx;
1762: }
1763: ftemp=fu;
1764: if (fabs(e) > tol1) {
1765: r=(x-w)*(fx-fv);
1766: q=(x-v)*(fx-fw);
1767: p=(x-v)*q-(x-w)*r;
1768: q=2.0*(q-r);
1769: if (q > 0.0) p = -p;
1770: q=fabs(q);
1771: etemp=e;
1772: e=d;
1773: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1774: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1775: else {
1.224 brouard 1776: d=p/q;
1777: u=x+d;
1778: if (u-a < tol2 || b-u < tol2)
1779: d=SIGN(tol1,xm-x);
1.126 brouard 1780: }
1781: } else {
1782: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1783: }
1784: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1785: fu=(*f)(u);
1786: if (fu <= fx) {
1787: if (u >= x) a=x; else b=x;
1788: SHFT(v,w,x,u)
1.183 brouard 1789: SHFT(fv,fw,fx,fu)
1790: } else {
1791: if (u < x) a=u; else b=u;
1792: if (fu <= fw || w == x) {
1.224 brouard 1793: v=w;
1794: w=u;
1795: fv=fw;
1796: fw=fu;
1.183 brouard 1797: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1798: v=u;
1799: fv=fu;
1.183 brouard 1800: }
1801: }
1.126 brouard 1802: }
1803: nrerror("Too many iterations in brent");
1804: *xmin=x;
1805: return fx;
1806: }
1807:
1808: /****************** mnbrak ***********************/
1809:
1810: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1811: double (*func)(double))
1.183 brouard 1812: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1813: the downhill direction (defined by the function as evaluated at the initial points) and returns
1814: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1815: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1816: */
1.126 brouard 1817: double ulim,u,r,q, dum;
1818: double fu;
1.187 brouard 1819:
1820: double scale=10.;
1821: int iterscale=0;
1822:
1823: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1824: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1825:
1826:
1827: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1828: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1829: /* *bx = *ax - (*ax - *bx)/scale; */
1830: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1831: /* } */
1832:
1.126 brouard 1833: if (*fb > *fa) {
1834: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1835: SHFT(dum,*fb,*fa,dum)
1836: }
1.126 brouard 1837: *cx=(*bx)+GOLD*(*bx-*ax);
1838: *fc=(*func)(*cx);
1.183 brouard 1839: #ifdef DEBUG
1.224 brouard 1840: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1841: 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 1842: #endif
1.224 brouard 1843: 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 1844: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1845: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1846: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1847: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1848: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1849: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1850: fu=(*func)(u);
1.163 brouard 1851: #ifdef DEBUG
1852: /* f(x)=A(x-u)**2+f(u) */
1853: double A, fparabu;
1854: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1855: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1856: 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);
1857: 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 1858: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1859: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1860: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1861: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1862: #endif
1.184 brouard 1863: #ifdef MNBRAKORIGINAL
1.183 brouard 1864: #else
1.191 brouard 1865: /* if (fu > *fc) { */
1866: /* #ifdef DEBUG */
1867: /* printf("mnbrak4 fu > fc \n"); */
1868: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1869: /* #endif */
1870: /* /\* 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 *\\/ *\/ */
1871: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1872: /* dum=u; /\* Shifting c and u *\/ */
1873: /* u = *cx; */
1874: /* *cx = dum; */
1875: /* dum = fu; */
1876: /* fu = *fc; */
1877: /* *fc =dum; */
1878: /* } else { /\* end *\/ */
1879: /* #ifdef DEBUG */
1880: /* printf("mnbrak3 fu < fc \n"); */
1881: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1882: /* #endif */
1883: /* dum=u; /\* Shifting c and u *\/ */
1884: /* u = *cx; */
1885: /* *cx = dum; */
1886: /* dum = fu; */
1887: /* fu = *fc; */
1888: /* *fc =dum; */
1889: /* } */
1.224 brouard 1890: #ifdef DEBUGMNBRAK
1891: double A, fparabu;
1892: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1893: fparabu= *fa - A*(*ax-u)*(*ax-u);
1894: 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);
1895: 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 1896: #endif
1.191 brouard 1897: dum=u; /* Shifting c and u */
1898: u = *cx;
1899: *cx = dum;
1900: dum = fu;
1901: fu = *fc;
1902: *fc =dum;
1.183 brouard 1903: #endif
1.162 brouard 1904: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1905: #ifdef DEBUG
1.224 brouard 1906: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1907: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1908: #endif
1.126 brouard 1909: fu=(*func)(u);
1910: if (fu < *fc) {
1.183 brouard 1911: #ifdef DEBUG
1.224 brouard 1912: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1913: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1914: #endif
1915: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1916: SHFT(*fb,*fc,fu,(*func)(u))
1917: #ifdef DEBUG
1918: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1919: #endif
1920: }
1.162 brouard 1921: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1922: #ifdef DEBUG
1.224 brouard 1923: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1924: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1925: #endif
1.126 brouard 1926: u=ulim;
1927: fu=(*func)(u);
1.183 brouard 1928: } else { /* u could be left to b (if r > q parabola has a maximum) */
1929: #ifdef DEBUG
1.224 brouard 1930: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1931: 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 1932: #endif
1.126 brouard 1933: u=(*cx)+GOLD*(*cx-*bx);
1934: fu=(*func)(u);
1.224 brouard 1935: #ifdef DEBUG
1936: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1937: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1938: #endif
1.183 brouard 1939: } /* end tests */
1.126 brouard 1940: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1941: SHFT(*fa,*fb,*fc,fu)
1942: #ifdef DEBUG
1.224 brouard 1943: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1944: 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 1945: #endif
1946: } /* 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 1947: }
1948:
1949: /*************** linmin ************************/
1.162 brouard 1950: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1951: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1952: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1953: the value of func at the returned location p . This is actually all accomplished by calling the
1954: routines mnbrak and brent .*/
1.126 brouard 1955: int ncom;
1956: double *pcom,*xicom;
1957: double (*nrfunc)(double []);
1958:
1.224 brouard 1959: #ifdef LINMINORIGINAL
1.126 brouard 1960: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1961: #else
1962: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1963: #endif
1.126 brouard 1964: {
1965: double brent(double ax, double bx, double cx,
1966: double (*f)(double), double tol, double *xmin);
1967: double f1dim(double x);
1968: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1969: double *fc, double (*func)(double));
1970: int j;
1971: double xx,xmin,bx,ax;
1972: double fx,fb,fa;
1.187 brouard 1973:
1.203 brouard 1974: #ifdef LINMINORIGINAL
1975: #else
1976: double scale=10., axs, xxs; /* Scale added for infinity */
1977: #endif
1978:
1.126 brouard 1979: ncom=n;
1980: pcom=vector(1,n);
1981: xicom=vector(1,n);
1982: nrfunc=func;
1983: for (j=1;j<=n;j++) {
1984: pcom[j]=p[j];
1.202 brouard 1985: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1986: }
1.187 brouard 1987:
1.203 brouard 1988: #ifdef LINMINORIGINAL
1989: xx=1.;
1990: #else
1991: axs=0.0;
1992: xxs=1.;
1993: do{
1994: xx= xxs;
1995: #endif
1.187 brouard 1996: ax=0.;
1997: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1998: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1999: /* 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)) */
2000: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2001: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2002: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2003: /* 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 2004: #ifdef LINMINORIGINAL
2005: #else
2006: if (fx != fx){
1.224 brouard 2007: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2008: printf("|");
2009: fprintf(ficlog,"|");
1.203 brouard 2010: #ifdef DEBUGLINMIN
1.224 brouard 2011: 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 2012: #endif
2013: }
1.224 brouard 2014: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2015: #endif
2016:
1.191 brouard 2017: #ifdef DEBUGLINMIN
2018: 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 2019: 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 2020: #endif
1.224 brouard 2021: #ifdef LINMINORIGINAL
2022: #else
2023: if(fb == fx){ /* Flat function in the direction */
2024: xmin=xx;
2025: *flat=1;
2026: }else{
2027: *flat=0;
2028: #endif
2029: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2030: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2031: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2032: /* fmin = f(p[j] + xmin * xi[j]) */
2033: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2034: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2035: #ifdef DEBUG
1.224 brouard 2036: 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);
2037: 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);
2038: #endif
2039: #ifdef LINMINORIGINAL
2040: #else
2041: }
1.126 brouard 2042: #endif
1.191 brouard 2043: #ifdef DEBUGLINMIN
2044: printf("linmin end ");
1.202 brouard 2045: fprintf(ficlog,"linmin end ");
1.191 brouard 2046: #endif
1.126 brouard 2047: for (j=1;j<=n;j++) {
1.203 brouard 2048: #ifdef LINMINORIGINAL
2049: xi[j] *= xmin;
2050: #else
2051: #ifdef DEBUGLINMIN
2052: if(xxs <1.0)
2053: printf(" before xi[%d]=%12.8f", j,xi[j]);
2054: #endif
2055: 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) */
2056: #ifdef DEBUGLINMIN
2057: if(xxs <1.0)
2058: 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 );
2059: #endif
2060: #endif
1.187 brouard 2061: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2062: }
1.191 brouard 2063: #ifdef DEBUGLINMIN
1.203 brouard 2064: printf("\n");
1.191 brouard 2065: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2066: 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 2067: for (j=1;j<=n;j++) {
1.202 brouard 2068: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2069: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2070: if(j % ncovmodel == 0){
1.191 brouard 2071: printf("\n");
1.202 brouard 2072: fprintf(ficlog,"\n");
2073: }
1.191 brouard 2074: }
1.203 brouard 2075: #else
1.191 brouard 2076: #endif
1.126 brouard 2077: free_vector(xicom,1,n);
2078: free_vector(pcom,1,n);
2079: }
2080:
2081:
2082: /*************** powell ************************/
1.162 brouard 2083: /*
2084: Minimization of a function func of n variables. Input consists of an initial starting point
2085: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2086: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2087: such that failure to decrease by more than this amount on one iteration signals doneness. On
2088: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2089: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2090: */
1.224 brouard 2091: #ifdef LINMINORIGINAL
2092: #else
2093: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2094: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2095: #endif
1.126 brouard 2096: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2097: double (*func)(double []))
2098: {
1.224 brouard 2099: #ifdef LINMINORIGINAL
2100: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2101: double (*func)(double []));
1.224 brouard 2102: #else
1.241 brouard 2103: void linmin(double p[], double xi[], int n, double *fret,
2104: double (*func)(double []),int *flat);
1.224 brouard 2105: #endif
1.239 brouard 2106: int i,ibig,j,jk,k;
1.126 brouard 2107: double del,t,*pt,*ptt,*xit;
1.181 brouard 2108: double directest;
1.126 brouard 2109: double fp,fptt;
2110: double *xits;
2111: int niterf, itmp;
1.224 brouard 2112: #ifdef LINMINORIGINAL
2113: #else
2114:
2115: flatdir=ivector(1,n);
2116: for (j=1;j<=n;j++) flatdir[j]=0;
2117: #endif
1.126 brouard 2118:
2119: pt=vector(1,n);
2120: ptt=vector(1,n);
2121: xit=vector(1,n);
2122: xits=vector(1,n);
2123: *fret=(*func)(p);
2124: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2125: rcurr_time = time(NULL);
1.126 brouard 2126: for (*iter=1;;++(*iter)) {
1.187 brouard 2127: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2128: ibig=0;
2129: del=0.0;
1.157 brouard 2130: rlast_time=rcurr_time;
2131: /* (void) gettimeofday(&curr_time,&tzp); */
2132: rcurr_time = time(NULL);
2133: curr_time = *localtime(&rcurr_time);
2134: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2135: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2136: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2137: for (i=1;i<=n;i++) {
1.126 brouard 2138: fprintf(ficrespow," %.12lf", p[i]);
2139: }
1.239 brouard 2140: fprintf(ficrespow,"\n");fflush(ficrespow);
2141: printf("\n#model= 1 + age ");
2142: fprintf(ficlog,"\n#model= 1 + age ");
2143: if(nagesqr==1){
1.241 brouard 2144: printf(" + age*age ");
2145: fprintf(ficlog," + age*age ");
1.239 brouard 2146: }
2147: for(j=1;j <=ncovmodel-2;j++){
2148: if(Typevar[j]==0) {
2149: printf(" + V%d ",Tvar[j]);
2150: fprintf(ficlog," + V%d ",Tvar[j]);
2151: }else if(Typevar[j]==1) {
2152: printf(" + V%d*age ",Tvar[j]);
2153: fprintf(ficlog," + V%d*age ",Tvar[j]);
2154: }else if(Typevar[j]==2) {
2155: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2156: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2157: }
2158: }
1.126 brouard 2159: printf("\n");
1.239 brouard 2160: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2161: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2162: fprintf(ficlog,"\n");
1.239 brouard 2163: for(i=1,jk=1; i <=nlstate; i++){
2164: for(k=1; k <=(nlstate+ndeath); k++){
2165: if (k != i) {
2166: printf("%d%d ",i,k);
2167: fprintf(ficlog,"%d%d ",i,k);
2168: for(j=1; j <=ncovmodel; j++){
2169: printf("%12.7f ",p[jk]);
2170: fprintf(ficlog,"%12.7f ",p[jk]);
2171: jk++;
2172: }
2173: printf("\n");
2174: fprintf(ficlog,"\n");
2175: }
2176: }
2177: }
1.241 brouard 2178: if(*iter <=3 && *iter >1){
1.157 brouard 2179: tml = *localtime(&rcurr_time);
2180: strcpy(strcurr,asctime(&tml));
2181: rforecast_time=rcurr_time;
1.126 brouard 2182: itmp = strlen(strcurr);
2183: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2184: strcurr[itmp-1]='\0';
1.162 brouard 2185: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2186: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2187: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2188: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2189: forecast_time = *localtime(&rforecast_time);
2190: strcpy(strfor,asctime(&forecast_time));
2191: itmp = strlen(strfor);
2192: if(strfor[itmp-1]=='\n')
2193: strfor[itmp-1]='\0';
2194: 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);
2195: 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 2196: }
2197: }
1.187 brouard 2198: for (i=1;i<=n;i++) { /* For each direction i */
2199: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2200: fptt=(*fret);
2201: #ifdef DEBUG
1.203 brouard 2202: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2203: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2204: #endif
1.203 brouard 2205: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2206: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2207: #ifdef LINMINORIGINAL
1.188 brouard 2208: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2209: #else
2210: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2211: flatdir[i]=flat; /* Function is vanishing in that direction i */
2212: #endif
2213: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2214: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2215: /* because that direction will be replaced unless the gain del is small */
2216: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2217: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2218: /* with the new direction. */
2219: del=fabs(fptt-(*fret));
2220: ibig=i;
1.126 brouard 2221: }
2222: #ifdef DEBUG
2223: printf("%d %.12e",i,(*fret));
2224: fprintf(ficlog,"%d %.12e",i,(*fret));
2225: for (j=1;j<=n;j++) {
1.224 brouard 2226: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2227: printf(" x(%d)=%.12e",j,xit[j]);
2228: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2229: }
2230: for(j=1;j<=n;j++) {
1.225 brouard 2231: printf(" p(%d)=%.12e",j,p[j]);
2232: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2233: }
2234: printf("\n");
2235: fprintf(ficlog,"\n");
2236: #endif
1.187 brouard 2237: } /* end loop on each direction i */
2238: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2239: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2240: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2241: for(j=1;j<=n;j++) {
1.225 brouard 2242: if(flatdir[j] >0){
2243: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2244: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2245: }
2246: /* printf("\n"); */
2247: /* fprintf(ficlog,"\n"); */
2248: }
1.243 brouard 2249: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2250: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2251: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2252: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2253: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2254: /* decreased of more than 3.84 */
2255: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2256: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2257: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2258:
1.188 brouard 2259: /* Starting the program with initial values given by a former maximization will simply change */
2260: /* the scales of the directions and the directions, because the are reset to canonical directions */
2261: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2262: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2263: #ifdef DEBUG
2264: int k[2],l;
2265: k[0]=1;
2266: k[1]=-1;
2267: printf("Max: %.12e",(*func)(p));
2268: fprintf(ficlog,"Max: %.12e",(*func)(p));
2269: for (j=1;j<=n;j++) {
2270: printf(" %.12e",p[j]);
2271: fprintf(ficlog," %.12e",p[j]);
2272: }
2273: printf("\n");
2274: fprintf(ficlog,"\n");
2275: for(l=0;l<=1;l++) {
2276: for (j=1;j<=n;j++) {
2277: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2278: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2279: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2280: }
2281: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2282: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2283: }
2284: #endif
2285:
1.224 brouard 2286: #ifdef LINMINORIGINAL
2287: #else
2288: free_ivector(flatdir,1,n);
2289: #endif
1.126 brouard 2290: free_vector(xit,1,n);
2291: free_vector(xits,1,n);
2292: free_vector(ptt,1,n);
2293: free_vector(pt,1,n);
2294: return;
1.192 brouard 2295: } /* enough precision */
1.240 brouard 2296: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2297: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2298: ptt[j]=2.0*p[j]-pt[j];
2299: xit[j]=p[j]-pt[j];
2300: pt[j]=p[j];
2301: }
1.181 brouard 2302: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2303: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2304: if (*iter <=4) {
1.225 brouard 2305: #else
2306: #endif
1.224 brouard 2307: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2308: #else
1.161 brouard 2309: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2310: #endif
1.162 brouard 2311: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2312: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2313: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2314: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2315: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2316: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2317: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2318: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2319: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2320: /* Even if f3 <f1, directest can be negative and t >0 */
2321: /* mu² and del² are equal when f3=f1 */
2322: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2323: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2324: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2325: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2326: #ifdef NRCORIGINAL
2327: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2328: #else
2329: 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 2330: t= t- del*SQR(fp-fptt);
1.183 brouard 2331: #endif
1.202 brouard 2332: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2333: #ifdef DEBUG
1.181 brouard 2334: 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);
2335: 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 2336: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2337: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2338: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2339: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2340: 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);
2341: 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);
2342: #endif
1.183 brouard 2343: #ifdef POWELLORIGINAL
2344: if (t < 0.0) { /* Then we use it for new direction */
2345: #else
1.182 brouard 2346: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2347: 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 2348: 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 2349: 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 2350: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2351: }
1.181 brouard 2352: if (directest < 0.0) { /* Then we use it for new direction */
2353: #endif
1.191 brouard 2354: #ifdef DEBUGLINMIN
1.234 brouard 2355: printf("Before linmin in direction P%d-P0\n",n);
2356: for (j=1;j<=n;j++) {
2357: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2358: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2359: if(j % ncovmodel == 0){
2360: printf("\n");
2361: fprintf(ficlog,"\n");
2362: }
2363: }
1.224 brouard 2364: #endif
2365: #ifdef LINMINORIGINAL
1.234 brouard 2366: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2367: #else
1.234 brouard 2368: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2369: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2370: #endif
1.234 brouard 2371:
1.191 brouard 2372: #ifdef DEBUGLINMIN
1.234 brouard 2373: for (j=1;j<=n;j++) {
2374: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2375: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2376: if(j % ncovmodel == 0){
2377: printf("\n");
2378: fprintf(ficlog,"\n");
2379: }
2380: }
1.224 brouard 2381: #endif
1.234 brouard 2382: for (j=1;j<=n;j++) {
2383: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2384: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2385: }
1.224 brouard 2386: #ifdef LINMINORIGINAL
2387: #else
1.234 brouard 2388: for (j=1, flatd=0;j<=n;j++) {
2389: if(flatdir[j]>0)
2390: flatd++;
2391: }
2392: if(flatd >0){
2393: printf("%d flat directions\n",flatd);
2394: fprintf(ficlog,"%d flat directions\n",flatd);
2395: for (j=1;j<=n;j++) {
2396: if(flatdir[j]>0){
2397: printf("%d ",j);
2398: fprintf(ficlog,"%d ",j);
2399: }
2400: }
2401: printf("\n");
2402: fprintf(ficlog,"\n");
2403: }
1.191 brouard 2404: #endif
1.234 brouard 2405: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2406: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2407:
1.126 brouard 2408: #ifdef DEBUG
1.234 brouard 2409: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2410: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2411: for(j=1;j<=n;j++){
2412: printf(" %lf",xit[j]);
2413: fprintf(ficlog," %lf",xit[j]);
2414: }
2415: printf("\n");
2416: fprintf(ficlog,"\n");
1.126 brouard 2417: #endif
1.192 brouard 2418: } /* end of t or directest negative */
1.224 brouard 2419: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2420: #else
1.234 brouard 2421: } /* end if (fptt < fp) */
1.192 brouard 2422: #endif
1.225 brouard 2423: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2424: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2425: #else
1.224 brouard 2426: #endif
1.234 brouard 2427: } /* loop iteration */
1.126 brouard 2428: }
1.234 brouard 2429:
1.126 brouard 2430: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2431:
1.235 brouard 2432: 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 2433: {
1.235 brouard 2434: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2435: (and selected quantitative values in nres)
2436: by left multiplying the unit
1.234 brouard 2437: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2438: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2439: /* Wx is row vector: population in state 1, population in state 2, population dead */
2440: /* or prevalence in state 1, prevalence in state 2, 0 */
2441: /* newm is the matrix after multiplications, its rows are identical at a factor */
2442: /* Initial matrix pimij */
2443: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2444: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2445: /* 0, 0 , 1} */
2446: /*
2447: * and after some iteration: */
2448: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2449: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2450: /* 0, 0 , 1} */
2451: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2452: /* {0.51571254859325999, 0.4842874514067399, */
2453: /* 0.51326036147820708, 0.48673963852179264} */
2454: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2455:
1.126 brouard 2456: int i, ii,j,k;
1.209 brouard 2457: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2458: /* double **matprod2(); */ /* test */
1.218 brouard 2459: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2460: double **newm;
1.209 brouard 2461: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2462: int ncvloop=0;
1.169 brouard 2463:
1.209 brouard 2464: min=vector(1,nlstate);
2465: max=vector(1,nlstate);
2466: meandiff=vector(1,nlstate);
2467:
1.218 brouard 2468: /* Starting with matrix unity */
1.126 brouard 2469: for (ii=1;ii<=nlstate+ndeath;ii++)
2470: for (j=1;j<=nlstate+ndeath;j++){
2471: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2472: }
1.169 brouard 2473:
2474: cov[1]=1.;
2475:
2476: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2477: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2478: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2479: ncvloop++;
1.126 brouard 2480: newm=savm;
2481: /* Covariates have to be included here again */
1.138 brouard 2482: cov[2]=agefin;
1.187 brouard 2483: if(nagesqr==1)
2484: cov[3]= agefin*agefin;;
1.234 brouard 2485: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2486: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2487: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2488: /* 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 2489: }
2490: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2491: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2492: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2493: /* 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 2494: }
1.237 brouard 2495: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2496: if(Dummy[Tvar[Tage[k]]]){
2497: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2498: } else{
1.235 brouard 2499: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2500: }
1.235 brouard 2501: /* 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 2502: }
1.237 brouard 2503: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2504: /* 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 2505: if(Dummy[Tvard[k][1]==0]){
2506: if(Dummy[Tvard[k][2]==0]){
2507: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2508: }else{
2509: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2510: }
2511: }else{
2512: if(Dummy[Tvard[k][2]==0]){
2513: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2514: }else{
2515: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2516: }
2517: }
1.234 brouard 2518: }
1.138 brouard 2519: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2520: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2521: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2522: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2523: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2524: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2525: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2526:
1.126 brouard 2527: savm=oldm;
2528: oldm=newm;
1.209 brouard 2529:
2530: for(j=1; j<=nlstate; j++){
2531: max[j]=0.;
2532: min[j]=1.;
2533: }
2534: for(i=1;i<=nlstate;i++){
2535: sumnew=0;
2536: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2537: for(j=1; j<=nlstate; j++){
2538: prlim[i][j]= newm[i][j]/(1-sumnew);
2539: max[j]=FMAX(max[j],prlim[i][j]);
2540: min[j]=FMIN(min[j],prlim[i][j]);
2541: }
2542: }
2543:
1.126 brouard 2544: maxmax=0.;
1.209 brouard 2545: for(j=1; j<=nlstate; j++){
2546: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2547: maxmax=FMAX(maxmax,meandiff[j]);
2548: /* 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 2549: } /* j loop */
1.203 brouard 2550: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2551: /* 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 2552: if(maxmax < ftolpl){
1.209 brouard 2553: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2554: free_vector(min,1,nlstate);
2555: free_vector(max,1,nlstate);
2556: free_vector(meandiff,1,nlstate);
1.126 brouard 2557: return prlim;
2558: }
1.169 brouard 2559: } /* age loop */
1.208 brouard 2560: /* After some age loop it doesn't converge */
1.209 brouard 2561: 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 2562: 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 2563: /* 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); */
2564: free_vector(min,1,nlstate);
2565: free_vector(max,1,nlstate);
2566: free_vector(meandiff,1,nlstate);
1.208 brouard 2567:
1.169 brouard 2568: return prlim; /* should not reach here */
1.126 brouard 2569: }
2570:
1.217 brouard 2571:
2572: /**** Back Prevalence limit (stable or period prevalence) ****************/
2573:
1.218 brouard 2574: /* 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) */
2575: /* 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 2576: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2577: {
1.218 brouard 2578: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2579: matrix by transitions matrix until convergence is reached with precision ftolpl */
2580: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2581: /* Wx is row vector: population in state 1, population in state 2, population dead */
2582: /* or prevalence in state 1, prevalence in state 2, 0 */
2583: /* newm is the matrix after multiplications, its rows are identical at a factor */
2584: /* Initial matrix pimij */
2585: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2586: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2587: /* 0, 0 , 1} */
2588: /*
2589: * and after some iteration: */
2590: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2591: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2592: /* 0, 0 , 1} */
2593: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2594: /* {0.51571254859325999, 0.4842874514067399, */
2595: /* 0.51326036147820708, 0.48673963852179264} */
2596: /* If we start from prlim again, prlim tends to a constant matrix */
2597:
2598: int i, ii,j,k;
1.247 brouard 2599: int first=0;
1.217 brouard 2600: double *min, *max, *meandiff, maxmax,sumnew=0.;
2601: /* double **matprod2(); */ /* test */
2602: double **out, cov[NCOVMAX+1], **bmij();
2603: double **newm;
1.218 brouard 2604: double **dnewm, **doldm, **dsavm; /* for use */
2605: double **oldm, **savm; /* for use */
2606:
1.217 brouard 2607: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2608: int ncvloop=0;
2609:
2610: min=vector(1,nlstate);
2611: max=vector(1,nlstate);
2612: meandiff=vector(1,nlstate);
2613:
1.218 brouard 2614: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2615: oldm=oldms; savm=savms;
2616:
2617: /* Starting with matrix unity */
2618: for (ii=1;ii<=nlstate+ndeath;ii++)
2619: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2620: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2621: }
2622:
2623: cov[1]=1.;
2624:
2625: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2626: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2627: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2628: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2629: ncvloop++;
1.218 brouard 2630: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2631: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2632: /* Covariates have to be included here again */
2633: cov[2]=agefin;
2634: if(nagesqr==1)
2635: cov[3]= agefin*agefin;;
1.242 brouard 2636: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2637: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2638: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2639: /* 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)); */
2640: }
2641: /* for (k=1; k<=cptcovn;k++) { */
2642: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2643: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2644: /* /\* 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])]); *\/ */
2645: /* } */
2646: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2647: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2648: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2649: /* 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]); */
2650: }
2651: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2652: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2653: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2654: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2655: for (k=1; k<=cptcovage;k++){ /* For product with age */
2656: if(Dummy[Tvar[Tage[k]]]){
2657: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2658: } else{
2659: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2660: }
2661: /* 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]); */
2662: }
2663: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2664: /* 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]); */
2665: if(Dummy[Tvard[k][1]==0]){
2666: if(Dummy[Tvard[k][2]==0]){
2667: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2668: }else{
2669: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2670: }
2671: }else{
2672: if(Dummy[Tvard[k][2]==0]){
2673: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2674: }else{
2675: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2676: }
2677: }
1.217 brouard 2678: }
2679:
2680: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2681: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2682: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2683: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2684: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2685: /* ij should be linked to the correct index of cov */
2686: /* age and covariate values ij are in 'cov', but we need to pass
2687: * ij for the observed prevalence at age and status and covariate
2688: * number: prevacurrent[(int)agefin][ii][ij]
2689: */
2690: /* 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 *\/ */
2691: /* 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 *\/ */
2692: 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 2693: savm=oldm;
2694: oldm=newm;
2695: for(j=1; j<=nlstate; j++){
2696: max[j]=0.;
2697: min[j]=1.;
2698: }
2699: for(j=1; j<=nlstate; j++){
2700: for(i=1;i<=nlstate;i++){
1.234 brouard 2701: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2702: bprlim[i][j]= newm[i][j];
2703: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2704: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2705: }
2706: }
1.218 brouard 2707:
1.217 brouard 2708: maxmax=0.;
2709: for(i=1; i<=nlstate; i++){
2710: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2711: maxmax=FMAX(maxmax,meandiff[i]);
2712: /* 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); */
2713: } /* j loop */
2714: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2715: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2716: if(maxmax < ftolpl){
1.220 brouard 2717: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2718: free_vector(min,1,nlstate);
2719: free_vector(max,1,nlstate);
2720: free_vector(meandiff,1,nlstate);
2721: return bprlim;
2722: }
2723: } /* age loop */
2724: /* After some age loop it doesn't converge */
1.247 brouard 2725: if(first){
2726: first=1;
2727: printf("Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. Others in log file only...\n\
2728: 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);
2729: }
2730: fprintf(ficlog,"Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.217 brouard 2731: 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);
2732: /* 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); */
2733: free_vector(min,1,nlstate);
2734: free_vector(max,1,nlstate);
2735: free_vector(meandiff,1,nlstate);
2736:
2737: return bprlim; /* should not reach here */
2738: }
2739:
1.126 brouard 2740: /*************** transition probabilities ***************/
2741:
2742: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2743: {
1.138 brouard 2744: /* According to parameters values stored in x and the covariate's values stored in cov,
2745: computes the probability to be observed in state j being in state i by appying the
2746: model to the ncovmodel covariates (including constant and age).
2747: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2748: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2749: ncth covariate in the global vector x is given by the formula:
2750: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2751: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2752: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2753: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2754: Outputs ps[i][j] the probability to be observed in j being in j according to
2755: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2756: */
2757: double s1, lnpijopii;
1.126 brouard 2758: /*double t34;*/
1.164 brouard 2759: int i,j, nc, ii, jj;
1.126 brouard 2760:
1.223 brouard 2761: for(i=1; i<= nlstate; i++){
2762: for(j=1; j<i;j++){
2763: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2764: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2765: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2766: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2767: }
2768: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2769: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2770: }
2771: for(j=i+1; j<=nlstate+ndeath;j++){
2772: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2773: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2774: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2775: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2776: }
2777: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2778: }
2779: }
1.218 brouard 2780:
1.223 brouard 2781: for(i=1; i<= nlstate; i++){
2782: s1=0;
2783: for(j=1; j<i; j++){
2784: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2785: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2786: }
2787: for(j=i+1; j<=nlstate+ndeath; j++){
2788: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2789: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2790: }
2791: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2792: ps[i][i]=1./(s1+1.);
2793: /* Computing other pijs */
2794: for(j=1; j<i; j++)
2795: ps[i][j]= exp(ps[i][j])*ps[i][i];
2796: for(j=i+1; j<=nlstate+ndeath; j++)
2797: ps[i][j]= exp(ps[i][j])*ps[i][i];
2798: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2799: } /* end i */
1.218 brouard 2800:
1.223 brouard 2801: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2802: for(jj=1; jj<= nlstate+ndeath; jj++){
2803: ps[ii][jj]=0;
2804: ps[ii][ii]=1;
2805: }
2806: }
1.218 brouard 2807:
2808:
1.223 brouard 2809: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2810: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2811: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2812: /* } */
2813: /* printf("\n "); */
2814: /* } */
2815: /* printf("\n ");printf("%lf ",cov[2]);*/
2816: /*
2817: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2818: goto end;*/
1.223 brouard 2819: return ps;
1.126 brouard 2820: }
2821:
1.218 brouard 2822: /*************** backward transition probabilities ***************/
2823:
2824: /* 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 ) */
2825: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2826: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2827: {
1.222 brouard 2828: /* Computes the backward probability at age agefin and covariate ij
2829: * and returns in **ps as well as **bmij.
2830: */
1.218 brouard 2831: int i, ii, j,k;
1.222 brouard 2832:
2833: double **out, **pmij();
2834: double sumnew=0.;
1.218 brouard 2835: double agefin;
1.222 brouard 2836:
2837: double **dnewm, **dsavm, **doldm;
2838: double **bbmij;
2839:
1.218 brouard 2840: doldm=ddoldms; /* global pointers */
1.222 brouard 2841: dnewm=ddnewms;
2842: dsavm=ddsavms;
2843:
2844: agefin=cov[2];
2845: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2846: the observed prevalence (with this covariate ij) */
2847: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2848: /* We do have the matrix Px in savm and we need pij */
2849: for (j=1;j<=nlstate+ndeath;j++){
2850: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2851: for (ii=1;ii<=nlstate;ii++){
2852: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2853: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2854: for (ii=1;ii<=nlstate+ndeath;ii++){
2855: if(sumnew >= 1.e-10){
2856: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2857: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2858: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2859: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2860: /* }else */
2861: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2862: }else{
1.242 brouard 2863: ;
2864: /* 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 2865: }
2866: } /*End ii */
2867: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2868: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2869: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2870: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2871: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2872: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2873: /* left Product of this matrix by diag matrix of prevalences (savm) */
2874: for (j=1;j<=nlstate+ndeath;j++){
2875: for (ii=1;ii<=nlstate+ndeath;ii++){
2876: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2877: }
2878: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2879: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2880: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2881: /* end bmij */
2882: return ps;
1.218 brouard 2883: }
1.217 brouard 2884: /*************** transition probabilities ***************/
2885:
1.218 brouard 2886: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2887: {
2888: /* According to parameters values stored in x and the covariate's values stored in cov,
2889: computes the probability to be observed in state j being in state i by appying the
2890: model to the ncovmodel covariates (including constant and age).
2891: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2892: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2893: ncth covariate in the global vector x is given by the formula:
2894: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2895: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2896: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2897: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2898: Outputs ps[i][j] the probability to be observed in j being in j according to
2899: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2900: */
2901: double s1, lnpijopii;
2902: /*double t34;*/
2903: int i,j, nc, ii, jj;
2904:
1.234 brouard 2905: for(i=1; i<= nlstate; i++){
2906: for(j=1; j<i;j++){
2907: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2908: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2909: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2910: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2911: }
2912: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2913: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2914: }
2915: for(j=i+1; j<=nlstate+ndeath;j++){
2916: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2917: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2918: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2919: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2920: }
2921: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2922: }
2923: }
2924:
2925: for(i=1; i<= nlstate; i++){
2926: s1=0;
2927: for(j=1; j<i; j++){
2928: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2929: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2930: }
2931: for(j=i+1; j<=nlstate+ndeath; j++){
2932: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2933: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2934: }
2935: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2936: ps[i][i]=1./(s1+1.);
2937: /* Computing other pijs */
2938: for(j=1; j<i; j++)
2939: ps[i][j]= exp(ps[i][j])*ps[i][i];
2940: for(j=i+1; j<=nlstate+ndeath; j++)
2941: ps[i][j]= exp(ps[i][j])*ps[i][i];
2942: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2943: } /* end i */
2944:
2945: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2946: for(jj=1; jj<= nlstate+ndeath; jj++){
2947: ps[ii][jj]=0;
2948: ps[ii][ii]=1;
2949: }
2950: }
2951: /* Added for backcast */ /* Transposed matrix too */
2952: for(jj=1; jj<= nlstate+ndeath; jj++){
2953: s1=0.;
2954: for(ii=1; ii<= nlstate+ndeath; ii++){
2955: s1+=ps[ii][jj];
2956: }
2957: for(ii=1; ii<= nlstate; ii++){
2958: ps[ii][jj]=ps[ii][jj]/s1;
2959: }
2960: }
2961: /* Transposition */
2962: for(jj=1; jj<= nlstate+ndeath; jj++){
2963: for(ii=jj; ii<= nlstate+ndeath; ii++){
2964: s1=ps[ii][jj];
2965: ps[ii][jj]=ps[jj][ii];
2966: ps[jj][ii]=s1;
2967: }
2968: }
2969: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2970: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2971: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2972: /* } */
2973: /* printf("\n "); */
2974: /* } */
2975: /* printf("\n ");printf("%lf ",cov[2]);*/
2976: /*
2977: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2978: goto end;*/
2979: return ps;
1.217 brouard 2980: }
2981:
2982:
1.126 brouard 2983: /**************** Product of 2 matrices ******************/
2984:
1.145 brouard 2985: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2986: {
2987: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2988: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2989: /* in, b, out are matrice of pointers which should have been initialized
2990: before: only the contents of out is modified. The function returns
2991: a pointer to pointers identical to out */
1.145 brouard 2992: int i, j, k;
1.126 brouard 2993: for(i=nrl; i<= nrh; i++)
1.145 brouard 2994: for(k=ncolol; k<=ncoloh; k++){
2995: out[i][k]=0.;
2996: for(j=ncl; j<=nch; j++)
2997: out[i][k] +=in[i][j]*b[j][k];
2998: }
1.126 brouard 2999: return out;
3000: }
3001:
3002:
3003: /************* Higher Matrix Product ***************/
3004:
1.235 brouard 3005: 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 3006: {
1.218 brouard 3007: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3008: 'nhstepm*hstepm*stepm' months (i.e. until
3009: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3010: nhstepm*hstepm matrices.
3011: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3012: (typically every 2 years instead of every month which is too big
3013: for the memory).
3014: Model is determined by parameters x and covariates have to be
3015: included manually here.
3016:
3017: */
3018:
3019: int i, j, d, h, k;
1.131 brouard 3020: double **out, cov[NCOVMAX+1];
1.126 brouard 3021: double **newm;
1.187 brouard 3022: double agexact;
1.214 brouard 3023: double agebegin, ageend;
1.126 brouard 3024:
3025: /* Hstepm could be zero and should return the unit matrix */
3026: for (i=1;i<=nlstate+ndeath;i++)
3027: for (j=1;j<=nlstate+ndeath;j++){
3028: oldm[i][j]=(i==j ? 1.0 : 0.0);
3029: po[i][j][0]=(i==j ? 1.0 : 0.0);
3030: }
3031: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3032: for(h=1; h <=nhstepm; h++){
3033: for(d=1; d <=hstepm; d++){
3034: newm=savm;
3035: /* Covariates have to be included here again */
3036: cov[1]=1.;
1.214 brouard 3037: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3038: cov[2]=agexact;
3039: if(nagesqr==1)
1.227 brouard 3040: cov[3]= agexact*agexact;
1.235 brouard 3041: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3042: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3043: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3044: /* 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)); */
3045: }
3046: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3047: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3048: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3049: /* 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]); */
3050: }
3051: for (k=1; k<=cptcovage;k++){
3052: if(Dummy[Tvar[Tage[k]]]){
3053: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3054: } else{
3055: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3056: }
3057: /* 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]); */
3058: }
3059: for (k=1; k<=cptcovprod;k++){ /* */
3060: /* 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]); */
3061: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3062: }
3063: /* for (k=1; k<=cptcovn;k++) */
3064: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3065: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3066: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3067: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3068: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3069:
3070:
1.126 brouard 3071: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3072: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3073: /* right multiplication of oldm by the current matrix */
1.126 brouard 3074: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3075: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3076: /* if((int)age == 70){ */
3077: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3078: /* for(i=1; i<=nlstate+ndeath; i++) { */
3079: /* printf("%d pmmij ",i); */
3080: /* for(j=1;j<=nlstate+ndeath;j++) { */
3081: /* printf("%f ",pmmij[i][j]); */
3082: /* } */
3083: /* printf(" oldm "); */
3084: /* for(j=1;j<=nlstate+ndeath;j++) { */
3085: /* printf("%f ",oldm[i][j]); */
3086: /* } */
3087: /* printf("\n"); */
3088: /* } */
3089: /* } */
1.126 brouard 3090: savm=oldm;
3091: oldm=newm;
3092: }
3093: for(i=1; i<=nlstate+ndeath; i++)
3094: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3095: po[i][j][h]=newm[i][j];
3096: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3097: }
1.128 brouard 3098: /*printf("h=%d ",h);*/
1.126 brouard 3099: } /* end h */
1.218 brouard 3100: /* printf("\n H=%d \n",h); */
1.126 brouard 3101: return po;
3102: }
3103:
1.217 brouard 3104: /************* Higher Back Matrix Product ***************/
1.218 brouard 3105: /* 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 3106: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3107: {
1.218 brouard 3108: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3109: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3110: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3111: nhstepm*hstepm matrices.
3112: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3113: (typically every 2 years instead of every month which is too big
1.217 brouard 3114: for the memory).
1.218 brouard 3115: Model is determined by parameters x and covariates have to be
3116: included manually here.
1.217 brouard 3117:
1.222 brouard 3118: */
1.217 brouard 3119:
3120: int i, j, d, h, k;
3121: double **out, cov[NCOVMAX+1];
3122: double **newm;
3123: double agexact;
3124: double agebegin, ageend;
1.222 brouard 3125: double **oldm, **savm;
1.217 brouard 3126:
1.222 brouard 3127: oldm=oldms;savm=savms;
1.217 brouard 3128: /* Hstepm could be zero and should return the unit matrix */
3129: for (i=1;i<=nlstate+ndeath;i++)
3130: for (j=1;j<=nlstate+ndeath;j++){
3131: oldm[i][j]=(i==j ? 1.0 : 0.0);
3132: po[i][j][0]=(i==j ? 1.0 : 0.0);
3133: }
3134: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3135: for(h=1; h <=nhstepm; h++){
3136: for(d=1; d <=hstepm; d++){
3137: newm=savm;
3138: /* Covariates have to be included here again */
3139: cov[1]=1.;
3140: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3141: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3142: cov[2]=agexact;
3143: if(nagesqr==1)
1.222 brouard 3144: cov[3]= agexact*agexact;
1.218 brouard 3145: for (k=1; k<=cptcovn;k++)
1.222 brouard 3146: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3147: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3148: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3149: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3150: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3151: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3152: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3153: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3154: /* 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 3155:
3156:
1.217 brouard 3157: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3158: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3159: /* Careful transposed matrix */
1.222 brouard 3160: /* age is in cov[2] */
1.218 brouard 3161: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3162: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3163: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3164: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3165: /* if((int)age == 70){ */
3166: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3167: /* for(i=1; i<=nlstate+ndeath; i++) { */
3168: /* printf("%d pmmij ",i); */
3169: /* for(j=1;j<=nlstate+ndeath;j++) { */
3170: /* printf("%f ",pmmij[i][j]); */
3171: /* } */
3172: /* printf(" oldm "); */
3173: /* for(j=1;j<=nlstate+ndeath;j++) { */
3174: /* printf("%f ",oldm[i][j]); */
3175: /* } */
3176: /* printf("\n"); */
3177: /* } */
3178: /* } */
3179: savm=oldm;
3180: oldm=newm;
3181: }
3182: for(i=1; i<=nlstate+ndeath; i++)
3183: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3184: po[i][j][h]=newm[i][j];
3185: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3186: }
3187: /*printf("h=%d ",h);*/
3188: } /* end h */
1.222 brouard 3189: /* printf("\n H=%d \n",h); */
1.217 brouard 3190: return po;
3191: }
3192:
3193:
1.162 brouard 3194: #ifdef NLOPT
3195: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3196: double fret;
3197: double *xt;
3198: int j;
3199: myfunc_data *d2 = (myfunc_data *) pd;
3200: /* xt = (p1-1); */
3201: xt=vector(1,n);
3202: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3203:
3204: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3205: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3206: printf("Function = %.12lf ",fret);
3207: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3208: printf("\n");
3209: free_vector(xt,1,n);
3210: return fret;
3211: }
3212: #endif
1.126 brouard 3213:
3214: /*************** log-likelihood *************/
3215: double func( double *x)
3216: {
1.226 brouard 3217: int i, ii, j, k, mi, d, kk;
3218: int ioffset=0;
3219: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3220: double **out;
3221: double lli; /* Individual log likelihood */
3222: int s1, s2;
1.228 brouard 3223: 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 3224: double bbh, survp;
3225: long ipmx;
3226: double agexact;
3227: /*extern weight */
3228: /* We are differentiating ll according to initial status */
3229: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3230: /*for(i=1;i<imx;i++)
3231: printf(" %d\n",s[4][i]);
3232: */
1.162 brouard 3233:
1.226 brouard 3234: ++countcallfunc;
1.162 brouard 3235:
1.226 brouard 3236: cov[1]=1.;
1.126 brouard 3237:
1.226 brouard 3238: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3239: ioffset=0;
1.226 brouard 3240: if(mle==1){
3241: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3242: /* Computes the values of the ncovmodel covariates of the model
3243: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3244: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3245: to be observed in j being in i according to the model.
3246: */
1.243 brouard 3247: ioffset=2+nagesqr ;
1.233 brouard 3248: /* Fixed */
1.234 brouard 3249: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3250: 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)*/
3251: }
1.226 brouard 3252: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3253: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3254: has been calculated etc */
3255: /* For an individual i, wav[i] gives the number of effective waves */
3256: /* We compute the contribution to Likelihood of each effective transition
3257: mw[mi][i] is real wave of the mi th effectve wave */
3258: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3259: s2=s[mw[mi+1][i]][i];
3260: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3261: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3262: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3263: */
3264: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3265: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3266: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3267: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3268: }
3269: for (ii=1;ii<=nlstate+ndeath;ii++)
3270: for (j=1;j<=nlstate+ndeath;j++){
3271: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3272: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3273: }
3274: for(d=0; d<dh[mi][i]; d++){
3275: newm=savm;
3276: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3277: cov[2]=agexact;
3278: if(nagesqr==1)
3279: cov[3]= agexact*agexact; /* Should be changed here */
3280: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3281: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3282: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3283: else
3284: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3285: }
3286: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3287: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3288: savm=oldm;
3289: oldm=newm;
3290: } /* end mult */
3291:
3292: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3293: /* But now since version 0.9 we anticipate for bias at large stepm.
3294: * If stepm is larger than one month (smallest stepm) and if the exact delay
3295: * (in months) between two waves is not a multiple of stepm, we rounded to
3296: * the nearest (and in case of equal distance, to the lowest) interval but now
3297: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3298: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3299: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3300: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3301: * -stepm/2 to stepm/2 .
3302: * For stepm=1 the results are the same as for previous versions of Imach.
3303: * For stepm > 1 the results are less biased than in previous versions.
3304: */
1.234 brouard 3305: s1=s[mw[mi][i]][i];
3306: s2=s[mw[mi+1][i]][i];
3307: bbh=(double)bh[mi][i]/(double)stepm;
3308: /* bias bh is positive if real duration
3309: * is higher than the multiple of stepm and negative otherwise.
3310: */
3311: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3312: if( s2 > nlstate){
3313: /* i.e. if s2 is a death state and if the date of death is known
3314: then the contribution to the likelihood is the probability to
3315: die between last step unit time and current step unit time,
3316: which is also equal to probability to die before dh
3317: minus probability to die before dh-stepm .
3318: In version up to 0.92 likelihood was computed
3319: as if date of death was unknown. Death was treated as any other
3320: health state: the date of the interview describes the actual state
3321: and not the date of a change in health state. The former idea was
3322: to consider that at each interview the state was recorded
3323: (healthy, disable or death) and IMaCh was corrected; but when we
3324: introduced the exact date of death then we should have modified
3325: the contribution of an exact death to the likelihood. This new
3326: contribution is smaller and very dependent of the step unit
3327: stepm. It is no more the probability to die between last interview
3328: and month of death but the probability to survive from last
3329: interview up to one month before death multiplied by the
3330: probability to die within a month. Thanks to Chris
3331: Jackson for correcting this bug. Former versions increased
3332: mortality artificially. The bad side is that we add another loop
3333: which slows down the processing. The difference can be up to 10%
3334: lower mortality.
3335: */
3336: /* If, at the beginning of the maximization mostly, the
3337: cumulative probability or probability to be dead is
3338: constant (ie = 1) over time d, the difference is equal to
3339: 0. out[s1][3] = savm[s1][3]: probability, being at state
3340: s1 at precedent wave, to be dead a month before current
3341: wave is equal to probability, being at state s1 at
3342: precedent wave, to be dead at mont of the current
3343: wave. Then the observed probability (that this person died)
3344: is null according to current estimated parameter. In fact,
3345: it should be very low but not zero otherwise the log go to
3346: infinity.
3347: */
1.183 brouard 3348: /* #ifdef INFINITYORIGINAL */
3349: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3350: /* #else */
3351: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3352: /* lli=log(mytinydouble); */
3353: /* else */
3354: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3355: /* #endif */
1.226 brouard 3356: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3357:
1.226 brouard 3358: } else if ( s2==-1 ) { /* alive */
3359: for (j=1,survp=0. ; j<=nlstate; j++)
3360: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3361: /*survp += out[s1][j]; */
3362: lli= log(survp);
3363: }
3364: else if (s2==-4) {
3365: for (j=3,survp=0. ; j<=nlstate; j++)
3366: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3367: lli= log(survp);
3368: }
3369: else if (s2==-5) {
3370: for (j=1,survp=0. ; j<=2; j++)
3371: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3372: lli= log(survp);
3373: }
3374: else{
3375: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3376: /* 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 */
3377: }
3378: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3379: /*if(lli ==000.0)*/
3380: /*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); */
3381: ipmx +=1;
3382: sw += weight[i];
3383: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3384: /* if (lli < log(mytinydouble)){ */
3385: /* 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); */
3386: /* 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]); */
3387: /* } */
3388: } /* end of wave */
3389: } /* end of individual */
3390: } else if(mle==2){
3391: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3392: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3393: for(mi=1; mi<= wav[i]-1; mi++){
3394: for (ii=1;ii<=nlstate+ndeath;ii++)
3395: for (j=1;j<=nlstate+ndeath;j++){
3396: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3397: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3398: }
3399: for(d=0; d<=dh[mi][i]; d++){
3400: newm=savm;
3401: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3402: cov[2]=agexact;
3403: if(nagesqr==1)
3404: cov[3]= agexact*agexact;
3405: for (kk=1; kk<=cptcovage;kk++) {
3406: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3407: }
3408: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3409: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3410: savm=oldm;
3411: oldm=newm;
3412: } /* end mult */
3413:
3414: s1=s[mw[mi][i]][i];
3415: s2=s[mw[mi+1][i]][i];
3416: bbh=(double)bh[mi][i]/(double)stepm;
3417: 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 */
3418: ipmx +=1;
3419: sw += weight[i];
3420: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3421: } /* end of wave */
3422: } /* end of individual */
3423: } else if(mle==3){ /* exponential inter-extrapolation */
3424: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3425: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3426: for(mi=1; mi<= wav[i]-1; mi++){
3427: for (ii=1;ii<=nlstate+ndeath;ii++)
3428: for (j=1;j<=nlstate+ndeath;j++){
3429: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3430: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3431: }
3432: for(d=0; d<dh[mi][i]; d++){
3433: newm=savm;
3434: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3435: cov[2]=agexact;
3436: if(nagesqr==1)
3437: cov[3]= agexact*agexact;
3438: for (kk=1; kk<=cptcovage;kk++) {
3439: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3440: }
3441: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3442: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3443: savm=oldm;
3444: oldm=newm;
3445: } /* end mult */
3446:
3447: s1=s[mw[mi][i]][i];
3448: s2=s[mw[mi+1][i]][i];
3449: bbh=(double)bh[mi][i]/(double)stepm;
3450: 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 */
3451: ipmx +=1;
3452: sw += weight[i];
3453: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3454: } /* end of wave */
3455: } /* end of individual */
3456: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3457: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3458: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3459: for(mi=1; mi<= wav[i]-1; mi++){
3460: for (ii=1;ii<=nlstate+ndeath;ii++)
3461: for (j=1;j<=nlstate+ndeath;j++){
3462: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3463: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3464: }
3465: for(d=0; d<dh[mi][i]; d++){
3466: newm=savm;
3467: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3468: cov[2]=agexact;
3469: if(nagesqr==1)
3470: cov[3]= agexact*agexact;
3471: for (kk=1; kk<=cptcovage;kk++) {
3472: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3473: }
1.126 brouard 3474:
1.226 brouard 3475: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3476: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3477: savm=oldm;
3478: oldm=newm;
3479: } /* end mult */
3480:
3481: s1=s[mw[mi][i]][i];
3482: s2=s[mw[mi+1][i]][i];
3483: if( s2 > nlstate){
3484: lli=log(out[s1][s2] - savm[s1][s2]);
3485: } else if ( s2==-1 ) { /* alive */
3486: for (j=1,survp=0. ; j<=nlstate; j++)
3487: survp += out[s1][j];
3488: lli= log(survp);
3489: }else{
3490: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3491: }
3492: ipmx +=1;
3493: sw += weight[i];
3494: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3495: /* 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 3496: } /* end of wave */
3497: } /* end of individual */
3498: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3499: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3500: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3501: for(mi=1; mi<= wav[i]-1; mi++){
3502: for (ii=1;ii<=nlstate+ndeath;ii++)
3503: for (j=1;j<=nlstate+ndeath;j++){
3504: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3505: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3506: }
3507: for(d=0; d<dh[mi][i]; d++){
3508: newm=savm;
3509: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3510: cov[2]=agexact;
3511: if(nagesqr==1)
3512: cov[3]= agexact*agexact;
3513: for (kk=1; kk<=cptcovage;kk++) {
3514: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3515: }
1.126 brouard 3516:
1.226 brouard 3517: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3518: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3519: savm=oldm;
3520: oldm=newm;
3521: } /* end mult */
3522:
3523: s1=s[mw[mi][i]][i];
3524: s2=s[mw[mi+1][i]][i];
3525: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3526: ipmx +=1;
3527: sw += weight[i];
3528: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3529: /*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]);*/
3530: } /* end of wave */
3531: } /* end of individual */
3532: } /* End of if */
3533: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3534: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3535: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3536: return -l;
1.126 brouard 3537: }
3538:
3539: /*************** log-likelihood *************/
3540: double funcone( double *x)
3541: {
1.228 brouard 3542: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3543: int i, ii, j, k, mi, d, kk;
1.228 brouard 3544: int ioffset=0;
1.131 brouard 3545: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3546: double **out;
3547: double lli; /* Individual log likelihood */
3548: double llt;
3549: int s1, s2;
1.228 brouard 3550: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3551:
1.126 brouard 3552: double bbh, survp;
1.187 brouard 3553: double agexact;
1.214 brouard 3554: double agebegin, ageend;
1.126 brouard 3555: /*extern weight */
3556: /* We are differentiating ll according to initial status */
3557: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3558: /*for(i=1;i<imx;i++)
3559: printf(" %d\n",s[4][i]);
3560: */
3561: cov[1]=1.;
3562:
3563: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3564: ioffset=0;
3565: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3566: /* ioffset=2+nagesqr+cptcovage; */
3567: ioffset=2+nagesqr;
1.232 brouard 3568: /* Fixed */
1.224 brouard 3569: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3570: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3571: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3572: 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)*/
3573: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3574: /* cov[2+6]=covar[Tvar[6]][i]; */
3575: /* cov[2+6]=covar[2][i]; V2 */
3576: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3577: /* cov[2+7]=covar[Tvar[7]][i]; */
3578: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3579: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3580: /* cov[2+9]=covar[Tvar[9]][i]; */
3581: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3582: }
1.232 brouard 3583: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3584: /* 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?)*\/ */
3585: /* } */
1.231 brouard 3586: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3587: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3588: /* } */
1.225 brouard 3589:
1.233 brouard 3590:
3591: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3592: /* Wave varying (but not age varying) */
3593: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3594: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3595: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3596: }
1.232 brouard 3597: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3598: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3599: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3600: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3601: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3602: /* 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 3603: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3604: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3605: /* /\* 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]); *\/ */
3606: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3607: /* } */
1.126 brouard 3608: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3609: for (j=1;j<=nlstate+ndeath;j++){
3610: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3611: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3612: }
1.214 brouard 3613:
3614: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3615: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3616: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3617: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3618: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3619: and mw[mi+1][i]. dh depends on stepm.*/
3620: newm=savm;
1.247 brouard 3621: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3622: cov[2]=agexact;
3623: if(nagesqr==1)
3624: cov[3]= agexact*agexact;
3625: for (kk=1; kk<=cptcovage;kk++) {
3626: if(!FixedV[Tvar[Tage[kk]]])
3627: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3628: else
3629: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3630: }
3631: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3632: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3633: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3634: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3635: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3636: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3637: savm=oldm;
3638: oldm=newm;
1.126 brouard 3639: } /* end mult */
3640:
3641: s1=s[mw[mi][i]][i];
3642: s2=s[mw[mi+1][i]][i];
1.217 brouard 3643: /* if(s2==-1){ */
3644: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3645: /* /\* exit(1); *\/ */
3646: /* } */
1.126 brouard 3647: bbh=(double)bh[mi][i]/(double)stepm;
3648: /* bias is positive if real duration
3649: * is higher than the multiple of stepm and negative otherwise.
3650: */
3651: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3652: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3653: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3654: for (j=1,survp=0. ; j<=nlstate; j++)
3655: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3656: lli= log(survp);
1.126 brouard 3657: }else if (mle==1){
1.242 brouard 3658: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3659: } else if(mle==2){
1.242 brouard 3660: 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 3661: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3662: 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 3663: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3664: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3665: } else{ /* mle=0 back to 1 */
1.242 brouard 3666: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3667: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3668: } /* End of if */
3669: ipmx +=1;
3670: sw += weight[i];
3671: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3672: /*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 3673: if(globpr){
1.246 brouard 3674: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3675: %11.6f %11.6f %11.6f ", \
1.242 brouard 3676: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3677: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3678: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3679: llt +=ll[k]*gipmx/gsw;
3680: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3681: }
3682: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3683: }
1.232 brouard 3684: } /* end of wave */
3685: } /* end of individual */
3686: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3687: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3688: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3689: if(globpr==0){ /* First time we count the contributions and weights */
3690: gipmx=ipmx;
3691: gsw=sw;
3692: }
3693: return -l;
1.126 brouard 3694: }
3695:
3696:
3697: /*************** function likelione ***********/
3698: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3699: {
3700: /* This routine should help understanding what is done with
3701: the selection of individuals/waves and
3702: to check the exact contribution to the likelihood.
3703: Plotting could be done.
3704: */
3705: int k;
3706:
3707: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3708: strcpy(fileresilk,"ILK_");
1.202 brouard 3709: strcat(fileresilk,fileresu);
1.126 brouard 3710: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3711: printf("Problem with resultfile: %s\n", fileresilk);
3712: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3713: }
1.214 brouard 3714: 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");
3715: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3716: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3717: for(k=1; k<=nlstate; k++)
3718: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3719: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3720: }
3721:
3722: *fretone=(*funcone)(p);
3723: if(*globpri !=0){
3724: fclose(ficresilk);
1.205 brouard 3725: if (mle ==0)
3726: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3727: else if(mle >=1)
3728: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3729: 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 3730:
1.208 brouard 3731:
3732: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3733: 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 3734: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3735: }
1.207 brouard 3736: 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 3737: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3738: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3739: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3740: fflush(fichtm);
1.205 brouard 3741: }
1.126 brouard 3742: return;
3743: }
3744:
3745:
3746: /*********** Maximum Likelihood Estimation ***************/
3747:
3748: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3749: {
1.165 brouard 3750: int i,j, iter=0;
1.126 brouard 3751: double **xi;
3752: double fret;
3753: double fretone; /* Only one call to likelihood */
3754: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3755:
3756: #ifdef NLOPT
3757: int creturn;
3758: nlopt_opt opt;
3759: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3760: double *lb;
3761: double minf; /* the minimum objective value, upon return */
3762: double * p1; /* Shifted parameters from 0 instead of 1 */
3763: myfunc_data dinst, *d = &dinst;
3764: #endif
3765:
3766:
1.126 brouard 3767: xi=matrix(1,npar,1,npar);
3768: for (i=1;i<=npar;i++)
3769: for (j=1;j<=npar;j++)
3770: xi[i][j]=(i==j ? 1.0 : 0.0);
3771: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3772: strcpy(filerespow,"POW_");
1.126 brouard 3773: strcat(filerespow,fileres);
3774: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3775: printf("Problem with resultfile: %s\n", filerespow);
3776: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3777: }
3778: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3779: for (i=1;i<=nlstate;i++)
3780: for(j=1;j<=nlstate+ndeath;j++)
3781: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3782: fprintf(ficrespow,"\n");
1.162 brouard 3783: #ifdef POWELL
1.126 brouard 3784: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3785: #endif
1.126 brouard 3786:
1.162 brouard 3787: #ifdef NLOPT
3788: #ifdef NEWUOA
3789: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3790: #else
3791: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3792: #endif
3793: lb=vector(0,npar-1);
3794: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3795: nlopt_set_lower_bounds(opt, lb);
3796: nlopt_set_initial_step1(opt, 0.1);
3797:
3798: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3799: d->function = func;
3800: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3801: nlopt_set_min_objective(opt, myfunc, d);
3802: nlopt_set_xtol_rel(opt, ftol);
3803: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3804: printf("nlopt failed! %d\n",creturn);
3805: }
3806: else {
3807: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3808: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3809: iter=1; /* not equal */
3810: }
3811: nlopt_destroy(opt);
3812: #endif
1.126 brouard 3813: free_matrix(xi,1,npar,1,npar);
3814: fclose(ficrespow);
1.203 brouard 3815: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3816: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3817: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3818:
3819: }
3820:
3821: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3822: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3823: {
3824: double **a,**y,*x,pd;
1.203 brouard 3825: /* double **hess; */
1.164 brouard 3826: int i, j;
1.126 brouard 3827: int *indx;
3828:
3829: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3830: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3831: void lubksb(double **a, int npar, int *indx, double b[]) ;
3832: void ludcmp(double **a, int npar, int *indx, double *d) ;
3833: double gompertz(double p[]);
1.203 brouard 3834: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3835:
3836: printf("\nCalculation of the hessian matrix. Wait...\n");
3837: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3838: for (i=1;i<=npar;i++){
1.203 brouard 3839: printf("%d-",i);fflush(stdout);
3840: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3841:
3842: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3843:
3844: /* printf(" %f ",p[i]);
3845: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3846: }
3847:
3848: for (i=1;i<=npar;i++) {
3849: for (j=1;j<=npar;j++) {
3850: if (j>i) {
1.203 brouard 3851: printf(".%d-%d",i,j);fflush(stdout);
3852: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3853: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3854:
3855: hess[j][i]=hess[i][j];
3856: /*printf(" %lf ",hess[i][j]);*/
3857: }
3858: }
3859: }
3860: printf("\n");
3861: fprintf(ficlog,"\n");
3862:
3863: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3864: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3865:
3866: a=matrix(1,npar,1,npar);
3867: y=matrix(1,npar,1,npar);
3868: x=vector(1,npar);
3869: indx=ivector(1,npar);
3870: for (i=1;i<=npar;i++)
3871: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3872: ludcmp(a,npar,indx,&pd);
3873:
3874: for (j=1;j<=npar;j++) {
3875: for (i=1;i<=npar;i++) x[i]=0;
3876: x[j]=1;
3877: lubksb(a,npar,indx,x);
3878: for (i=1;i<=npar;i++){
3879: matcov[i][j]=x[i];
3880: }
3881: }
3882:
3883: printf("\n#Hessian matrix#\n");
3884: fprintf(ficlog,"\n#Hessian matrix#\n");
3885: for (i=1;i<=npar;i++) {
3886: for (j=1;j<=npar;j++) {
1.203 brouard 3887: printf("%.6e ",hess[i][j]);
3888: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3889: }
3890: printf("\n");
3891: fprintf(ficlog,"\n");
3892: }
3893:
1.203 brouard 3894: /* printf("\n#Covariance matrix#\n"); */
3895: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3896: /* for (i=1;i<=npar;i++) { */
3897: /* for (j=1;j<=npar;j++) { */
3898: /* printf("%.6e ",matcov[i][j]); */
3899: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3900: /* } */
3901: /* printf("\n"); */
3902: /* fprintf(ficlog,"\n"); */
3903: /* } */
3904:
1.126 brouard 3905: /* Recompute Inverse */
1.203 brouard 3906: /* for (i=1;i<=npar;i++) */
3907: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3908: /* ludcmp(a,npar,indx,&pd); */
3909:
3910: /* printf("\n#Hessian matrix recomputed#\n"); */
3911:
3912: /* for (j=1;j<=npar;j++) { */
3913: /* for (i=1;i<=npar;i++) x[i]=0; */
3914: /* x[j]=1; */
3915: /* lubksb(a,npar,indx,x); */
3916: /* for (i=1;i<=npar;i++){ */
3917: /* y[i][j]=x[i]; */
3918: /* printf("%.3e ",y[i][j]); */
3919: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3920: /* } */
3921: /* printf("\n"); */
3922: /* fprintf(ficlog,"\n"); */
3923: /* } */
3924:
3925: /* Verifying the inverse matrix */
3926: #ifdef DEBUGHESS
3927: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3928:
1.203 brouard 3929: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3930: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3931:
3932: for (j=1;j<=npar;j++) {
3933: for (i=1;i<=npar;i++){
1.203 brouard 3934: printf("%.2f ",y[i][j]);
3935: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3936: }
3937: printf("\n");
3938: fprintf(ficlog,"\n");
3939: }
1.203 brouard 3940: #endif
1.126 brouard 3941:
3942: free_matrix(a,1,npar,1,npar);
3943: free_matrix(y,1,npar,1,npar);
3944: free_vector(x,1,npar);
3945: free_ivector(indx,1,npar);
1.203 brouard 3946: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3947:
3948:
3949: }
3950:
3951: /*************** hessian matrix ****************/
3952: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3953: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3954: int i;
3955: int l=1, lmax=20;
1.203 brouard 3956: double k1,k2, res, fx;
1.132 brouard 3957: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3958: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3959: int k=0,kmax=10;
3960: double l1;
3961:
3962: fx=func(x);
3963: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3964: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3965: l1=pow(10,l);
3966: delts=delt;
3967: for(k=1 ; k <kmax; k=k+1){
3968: delt = delta*(l1*k);
3969: p2[theta]=x[theta] +delt;
1.145 brouard 3970: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3971: p2[theta]=x[theta]-delt;
3972: k2=func(p2)-fx;
3973: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3974: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3975:
1.203 brouard 3976: #ifdef DEBUGHESSII
1.126 brouard 3977: 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);
3978: 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);
3979: #endif
3980: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3981: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3982: k=kmax;
3983: }
3984: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3985: k=kmax; l=lmax*10;
1.126 brouard 3986: }
3987: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3988: delts=delt;
3989: }
1.203 brouard 3990: } /* End loop k */
1.126 brouard 3991: }
3992: delti[theta]=delts;
3993: return res;
3994:
3995: }
3996:
1.203 brouard 3997: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3998: {
3999: int i;
1.164 brouard 4000: int l=1, lmax=20;
1.126 brouard 4001: double k1,k2,k3,k4,res,fx;
1.132 brouard 4002: double p2[MAXPARM+1];
1.203 brouard 4003: int k, kmax=1;
4004: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4005:
4006: int firstime=0;
1.203 brouard 4007:
1.126 brouard 4008: fx=func(x);
1.203 brouard 4009: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4010: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4011: p2[thetai]=x[thetai]+delti[thetai]*k;
4012: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4013: k1=func(p2)-fx;
4014:
1.203 brouard 4015: p2[thetai]=x[thetai]+delti[thetai]*k;
4016: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4017: k2=func(p2)-fx;
4018:
1.203 brouard 4019: p2[thetai]=x[thetai]-delti[thetai]*k;
4020: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4021: k3=func(p2)-fx;
4022:
1.203 brouard 4023: p2[thetai]=x[thetai]-delti[thetai]*k;
4024: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4025: k4=func(p2)-fx;
1.203 brouard 4026: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4027: if(k1*k2*k3*k4 <0.){
1.208 brouard 4028: firstime=1;
1.203 brouard 4029: kmax=kmax+10;
1.208 brouard 4030: }
4031: if(kmax >=10 || firstime ==1){
1.246 brouard 4032: 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);
4033: 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 4034: 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);
4035: 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);
4036: }
4037: #ifdef DEBUGHESSIJ
4038: v1=hess[thetai][thetai];
4039: v2=hess[thetaj][thetaj];
4040: cv12=res;
4041: /* Computing eigen value of Hessian matrix */
4042: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4043: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4044: if ((lc2 <0) || (lc1 <0) ){
4045: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4046: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4047: 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);
4048: 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);
4049: }
1.126 brouard 4050: #endif
4051: }
4052: return res;
4053: }
4054:
1.203 brouard 4055: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4056: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4057: /* { */
4058: /* int i; */
4059: /* int l=1, lmax=20; */
4060: /* double k1,k2,k3,k4,res,fx; */
4061: /* double p2[MAXPARM+1]; */
4062: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4063: /* int k=0,kmax=10; */
4064: /* double l1; */
4065:
4066: /* fx=func(x); */
4067: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4068: /* l1=pow(10,l); */
4069: /* delts=delt; */
4070: /* for(k=1 ; k <kmax; k=k+1){ */
4071: /* delt = delti*(l1*k); */
4072: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4073: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4074: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4075: /* k1=func(p2)-fx; */
4076:
4077: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4078: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4079: /* k2=func(p2)-fx; */
4080:
4081: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4082: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4083: /* k3=func(p2)-fx; */
4084:
4085: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4086: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4087: /* k4=func(p2)-fx; */
4088: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4089: /* #ifdef DEBUGHESSIJ */
4090: /* 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); */
4091: /* 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); */
4092: /* #endif */
4093: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4094: /* k=kmax; */
4095: /* } */
4096: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4097: /* k=kmax; l=lmax*10; */
4098: /* } */
4099: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4100: /* delts=delt; */
4101: /* } */
4102: /* } /\* End loop k *\/ */
4103: /* } */
4104: /* delti[theta]=delts; */
4105: /* return res; */
4106: /* } */
4107:
4108:
1.126 brouard 4109: /************** Inverse of matrix **************/
4110: void ludcmp(double **a, int n, int *indx, double *d)
4111: {
4112: int i,imax,j,k;
4113: double big,dum,sum,temp;
4114: double *vv;
4115:
4116: vv=vector(1,n);
4117: *d=1.0;
4118: for (i=1;i<=n;i++) {
4119: big=0.0;
4120: for (j=1;j<=n;j++)
4121: if ((temp=fabs(a[i][j])) > big) big=temp;
4122: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4123: vv[i]=1.0/big;
4124: }
4125: for (j=1;j<=n;j++) {
4126: for (i=1;i<j;i++) {
4127: sum=a[i][j];
4128: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4129: a[i][j]=sum;
4130: }
4131: big=0.0;
4132: for (i=j;i<=n;i++) {
4133: sum=a[i][j];
4134: for (k=1;k<j;k++)
4135: sum -= a[i][k]*a[k][j];
4136: a[i][j]=sum;
4137: if ( (dum=vv[i]*fabs(sum)) >= big) {
4138: big=dum;
4139: imax=i;
4140: }
4141: }
4142: if (j != imax) {
4143: for (k=1;k<=n;k++) {
4144: dum=a[imax][k];
4145: a[imax][k]=a[j][k];
4146: a[j][k]=dum;
4147: }
4148: *d = -(*d);
4149: vv[imax]=vv[j];
4150: }
4151: indx[j]=imax;
4152: if (a[j][j] == 0.0) a[j][j]=TINY;
4153: if (j != n) {
4154: dum=1.0/(a[j][j]);
4155: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4156: }
4157: }
4158: free_vector(vv,1,n); /* Doesn't work */
4159: ;
4160: }
4161:
4162: void lubksb(double **a, int n, int *indx, double b[])
4163: {
4164: int i,ii=0,ip,j;
4165: double sum;
4166:
4167: for (i=1;i<=n;i++) {
4168: ip=indx[i];
4169: sum=b[ip];
4170: b[ip]=b[i];
4171: if (ii)
4172: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4173: else if (sum) ii=i;
4174: b[i]=sum;
4175: }
4176: for (i=n;i>=1;i--) {
4177: sum=b[i];
4178: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4179: b[i]=sum/a[i][i];
4180: }
4181: }
4182:
4183: void pstamp(FILE *fichier)
4184: {
1.196 brouard 4185: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4186: }
4187:
4188: /************ Frequencies ********************/
1.251 ! brouard 4189: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4190: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4191: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4192: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4193:
1.250 brouard 4194: int i, m, jk, j1, bool, z1,j, k, iv, jj=0;
1.226 brouard 4195: int iind=0, iage=0;
4196: int mi; /* Effective wave */
4197: int first;
4198: double ***freq; /* Frequencies */
4199: double *meanq;
4200: double **meanqt;
4201: double *pp, **prop, *posprop, *pospropt;
4202: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4203: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4204: double agebegin, ageend;
4205:
4206: pp=vector(1,nlstate);
1.251 ! brouard 4207: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4208: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4209: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4210: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4211: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4212: meanqt=matrix(1,lastpass,1,nqtveff);
4213: strcpy(fileresp,"P_");
4214: strcat(fileresp,fileresu);
4215: /*strcat(fileresphtm,fileresu);*/
4216: if((ficresp=fopen(fileresp,"w"))==NULL) {
4217: printf("Problem with prevalence resultfile: %s\n", fileresp);
4218: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4219: exit(0);
4220: }
1.240 brouard 4221:
1.226 brouard 4222: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4223: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4224: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4225: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4226: fflush(ficlog);
4227: exit(70);
4228: }
4229: else{
4230: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4231: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4232: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4233: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4234: }
1.237 brouard 4235: 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 4236:
1.226 brouard 4237: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4238: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4239: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4240: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4241: fflush(ficlog);
4242: exit(70);
1.240 brouard 4243: } else{
1.226 brouard 4244: 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 4245: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4246: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4247: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4248: }
1.240 brouard 4249: 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);
4250:
1.251 ! brouard 4251: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4252: j1=0;
1.126 brouard 4253:
1.227 brouard 4254: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4255: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4256: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4257:
4258:
1.226 brouard 4259: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4260: reference=low_education V1=0,V2=0
4261: med_educ V1=1 V2=0,
4262: high_educ V1=0 V2=1
4263: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4264: */
1.249 brouard 4265: dateintsum=0;
4266: k2cpt=0;
4267:
1.251 ! brouard 4268: for (j = 0; j <= cptcoveff; j+=cptcoveff){ /* j= 0 constant model */
! 4269: first=1;
! 4270: 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 */
! 4271: posproptt=0.;
! 4272: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
! 4273: scanf("%d", i);*/
! 4274: for (i=-5; i<=nlstate+ndeath; i++)
! 4275: for (jk=-5; jk<=nlstate+ndeath; jk++)
! 4276: for(m=iagemin; m <= iagemax+3; m++)
! 4277: freq[i][jk][m]=0;
! 4278:
! 4279: for (i=1; i<=nlstate; i++) {
1.240 brouard 4280: for(m=iagemin; m <= iagemax+3; m++)
1.251 ! brouard 4281: prop[i][m]=0;
! 4282: posprop[i]=0;
! 4283: pospropt[i]=0;
! 4284: }
! 4285: /* for (z1=1; z1<= nqfveff; z1++) { */
! 4286: /* meanq[z1]+=0.; */
! 4287: /* for(m=1;m<=lastpass;m++){ */
! 4288: /* meanqt[m][z1]=0.; */
! 4289: /* } */
! 4290: /* } */
! 4291:
! 4292: /* dateintsum=0; */
! 4293: /* k2cpt=0; */
! 4294:
! 4295: /* For that combination of covariate j1, we count and print the frequencies in one pass */
! 4296: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
! 4297: bool=1;
! 4298: if(j !=0){
! 4299: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
! 4300: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
! 4301: /* for (z1=1; z1<= nqfveff; z1++) { */
! 4302: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
! 4303: /* } */
! 4304: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
! 4305: /* if(Tvaraff[z1] ==-20){ */
! 4306: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
! 4307: /* }else if(Tvaraff[z1] ==-10){ */
! 4308: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
! 4309: /* }else */
! 4310: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
! 4311: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
! 4312: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
! 4313: /* 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",
! 4314: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
! 4315: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
! 4316: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
! 4317: } /* Onlyf fixed */
! 4318: } /* end z1 */
! 4319: } /* cptcovn > 0 */
! 4320: } /* end any */
! 4321: }/* end j==0 */
! 4322: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
! 4323: /* for(m=firstpass; m<=lastpass; m++){ */
! 4324: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
! 4325: m=mw[mi][iind];
! 4326: if(j!=0){
! 4327: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
! 4328: for (z1=1; z1<=cptcoveff; z1++) {
! 4329: if( Fixed[Tmodelind[z1]]==1){
! 4330: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
! 4331: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
! 4332: value is -1, we don't select. It differs from the
! 4333: constant and age model which counts them. */
! 4334: bool=0; /* not selected */
! 4335: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
! 4336: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
! 4337: bool=0;
! 4338: }
! 4339: }
! 4340: }
! 4341: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
! 4342: } /* end j==0 */
! 4343: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
! 4344: if(bool==1){
! 4345: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
! 4346: and mw[mi+1][iind]. dh depends on stepm. */
! 4347: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
! 4348: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
! 4349: if(m >=firstpass && m <=lastpass){
! 4350: k2=anint[m][iind]+(mint[m][iind]/12.);
! 4351: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
! 4352: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
! 4353: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
! 4354: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
! 4355: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
! 4356: if (m<lastpass) {
! 4357: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
! 4358: /* 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]); */
! 4359: if(s[m][iind]==-1)
! 4360: 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.));
! 4361: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
! 4362: /* if((int)agev[m][iind] == 55) */
! 4363: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
! 4364: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
! 4365: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 4366: }
1.251 ! brouard 4367: } /* end if between passes */
! 4368: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
! 4369: dateintsum=dateintsum+k2; /* on all covariates ?*/
! 4370: k2cpt++;
! 4371: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4372: }
1.251 ! brouard 4373: }else{
! 4374: bool=1;
! 4375: }/* end bool 2 */
! 4376: } /* end m */
! 4377: } /* end bool */
! 4378: } /* end iind = 1 to imx */
! 4379: /* prop[s][age] is feeded for any initial and valid live state as well as
! 4380: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
! 4381:
! 4382:
! 4383: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
! 4384: pstamp(ficresp);
! 4385: if (cptcoveff>0 && j!=0){
! 4386: printf( "\n#********** Variable ");
! 4387: fprintf(ficresp, "\n#********** Variable ");
! 4388: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
! 4389: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
! 4390: fprintf(ficlog, "\n#********** Variable ");
! 4391: for (z1=1; z1<=cptcoveff; z1++){
! 4392: if(!FixedV[Tvaraff[z1]]){
! 4393: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4394: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4395: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4396: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4397: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4398: }else{
1.251 ! brouard 4399: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4400: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4401: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4402: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4403: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
! 4404: }
! 4405: }
! 4406: printf( "**********\n#");
! 4407: fprintf(ficresp, "**********\n#");
! 4408: fprintf(ficresphtm, "**********</h3>\n");
! 4409: fprintf(ficresphtmfr, "**********</h3>\n");
! 4410: fprintf(ficlog, "**********\n");
! 4411: }
! 4412: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
! 4413: for(i=1; i<=nlstate;i++) {
! 4414: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
! 4415: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
! 4416: }
! 4417: fprintf(ficresp, "\n");
! 4418: fprintf(ficresphtm, "\n");
! 4419:
! 4420: /* Header of frequency table by age */
! 4421: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
! 4422: fprintf(ficresphtmfr,"<th>Age</th> ");
! 4423: for(jk=-1; jk <=nlstate+ndeath; jk++){
! 4424: for(m=-1; m <=nlstate+ndeath; m++){
! 4425: if(jk!=0 && m!=0)
! 4426: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4427: }
1.226 brouard 4428: }
1.251 ! brouard 4429: fprintf(ficresphtmfr, "\n");
! 4430:
! 4431: /* For each age */
! 4432: for(iage=iagemin; iage <= iagemax+3; iage++){
! 4433: fprintf(ficresphtm,"<tr>");
! 4434: if(iage==iagemax+1){
! 4435: fprintf(ficlog,"1");
! 4436: fprintf(ficresphtmfr,"<tr><th>0</th> ");
! 4437: }else if(iage==iagemax+2){
! 4438: fprintf(ficlog,"0");
! 4439: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
! 4440: }else if(iage==iagemax+3){
! 4441: fprintf(ficlog,"Total");
! 4442: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
! 4443: }else{
1.240 brouard 4444: if(first==1){
1.251 ! brouard 4445: first=0;
! 4446: printf("See log file for details...\n");
! 4447: }
! 4448: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
! 4449: fprintf(ficlog,"Age %d", iage);
! 4450: }
! 4451: for(jk=1; jk <=nlstate ; jk++){
! 4452: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
! 4453: pp[jk] += freq[jk][m][iage];
! 4454: }
! 4455: for(jk=1; jk <=nlstate ; jk++){
! 4456: for(m=-1, pos=0; m <=0 ; m++)
! 4457: pos += freq[jk][m][iage];
! 4458: if(pp[jk]>=1.e-10){
! 4459: if(first==1){
! 4460: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
! 4461: }
! 4462: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
! 4463: }else{
! 4464: if(first==1)
! 4465: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
! 4466: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4467: }
4468: }
4469:
1.251 ! brouard 4470: for(jk=1; jk <=nlstate ; jk++){
! 4471: /* posprop[jk]=0; */
! 4472: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
! 4473: pp[jk] += freq[jk][m][iage];
! 4474: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
! 4475:
! 4476: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
! 4477: pos += pp[jk]; /* pos is the total number of transitions until this age */
! 4478: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
! 4479: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
! 4480: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4481: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4482: }
1.251 ! brouard 4483: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4484: if(pos>=1.e-5){
1.251 ! brouard 4485: if(first==1)
! 4486: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
! 4487: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
! 4488: }else{
! 4489: if(first==1)
! 4490: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
! 4491: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
! 4492: }
! 4493: if( iage <= iagemax){
! 4494: if(pos>=1.e-5){
! 4495: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
! 4496: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
! 4497: /*probs[iage][jk][j1]= pp[jk]/pos;*/
! 4498: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
! 4499: }
! 4500: else{
! 4501: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
! 4502: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
! 4503: }
1.240 brouard 4504: }
1.251 ! brouard 4505: pospropt[jk] +=posprop[jk];
! 4506: } /* end loop jk */
! 4507: /* pospropt=0.; */
! 4508: for(jk=-1; jk <=nlstate+ndeath; jk++){
! 4509: for(m=-1; m <=nlstate+ndeath; m++){
! 4510: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
! 4511: if(first==1){
! 4512: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
! 4513: }
! 4514: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
! 4515: }
! 4516: if(jk!=0 && m!=0)
! 4517: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4518: }
1.251 ! brouard 4519: } /* end loop jk */
! 4520: posproptt=0.;
! 4521: for(jk=1; jk <=nlstate; jk++){
! 4522: posproptt += pospropt[jk];
! 4523: }
! 4524: fprintf(ficresphtmfr,"</tr>\n ");
! 4525: if(iage <= iagemax){
! 4526: fprintf(ficresp,"\n");
! 4527: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4528: }
1.251 ! brouard 4529: if(first==1)
! 4530: printf("Others in log...\n");
! 4531: fprintf(ficlog,"\n");
! 4532: } /* end loop age iage */
! 4533: fprintf(ficresphtm,"<tr><th>Tot</th>");
! 4534: for(jk=1; jk <=nlstate ; jk++){
! 4535: if(posproptt < 1.e-5){
! 4536: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
! 4537: }else{
! 4538: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4539: }
1.226 brouard 4540: }
1.251 ! brouard 4541: fprintf(ficresphtm,"</tr>\n");
! 4542: fprintf(ficresphtm,"</table>\n");
! 4543: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4544: if(posproptt < 1.e-5){
1.251 ! brouard 4545: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
! 4546: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
! 4547: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
! 4548: invalidvarcomb[j1]=1;
1.226 brouard 4549: }else{
1.251 ! brouard 4550: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
! 4551: invalidvarcomb[j1]=0;
1.226 brouard 4552: }
1.251 ! brouard 4553: fprintf(ficresphtmfr,"</table>\n");
! 4554: fprintf(ficlog,"\n");
! 4555: if(j!=0){
! 4556: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
! 4557: for(i=1,jk=1; i <=nlstate; i++){
! 4558: for(k=1; k <=(nlstate+ndeath); k++){
! 4559: if (k != i) {
! 4560: printf("%d%d ",i,k);
! 4561: fprintf(ficlog,"%d%d ",i,k);
! 4562: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
! 4563: if(jj==1){ /* Constant case */
! 4564: if(j1==1){ /* All dummy covariates to zero */
! 4565: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
! 4566: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
! 4567: }
! 4568: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
! 4569: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
! 4570: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
! 4571: }else if( (log(j1-1)/log(2)+1 == jj -2 -nagesqr) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
! 4572: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
! 4573: printf("jk=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",jk,i,k,jk,p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
! 4574: }else if(jj==2 || nagesqr==1){ /* age or age*age parameter */
! 4575: ;
! 4576: }else{ /* Other cases, like quantitative fixed or varying covariates */
! 4577: ;
! 4578: }
! 4579: /* printf("%12.7f )", param[i][jj][k]); */
! 4580: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
! 4581: jk++;
! 4582: } /* end jj */
! 4583: printf("\n");
! 4584: fprintf(ficlog,"\n");
! 4585: } /* end k!= i */
! 4586: } /* end k */
! 4587: } /* end i, jk */
! 4588: } /* end j !=0 */
! 4589: } /* end selected combination of covariate j1 */
! 4590: if(j==0){ /* We can estimate starting values from the occurences in each case */
! 4591: printf("#Freqsummary: Starting values for the constants:\n");
! 4592: fprintf(ficlog,"\n");
! 4593: for(i=1,jk=1; i <=nlstate; i++){
! 4594: for(k=1; k <=(nlstate+ndeath); k++){
! 4595: if (k != i) {
! 4596: printf("%d%d ",i,k);
! 4597: fprintf(ficlog,"%d%d ",i,k);
! 4598: for(jj=1; jj <=ncovmodel; jj++){
! 4599: if(jj==1){
! 4600: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
! 4601: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
! 4602: }
! 4603: /* printf("%12.7f )", param[i][jj][k]); */
! 4604: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
! 4605: jk++;
1.250 brouard 4606: }
1.251 ! brouard 4607: printf("\n");
! 4608: fprintf(ficlog,"\n");
1.250 brouard 4609: }
4610: }
4611: }
1.251 ! brouard 4612: printf("#Freqsummary\n");
! 4613: fprintf(ficlog,"\n");
! 4614: for(jk=-1; jk <=nlstate+ndeath; jk++){
! 4615: for(m=-1; m <=nlstate+ndeath; m++){
! 4616: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4617: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4618: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 ! brouard 4619: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
! 4620: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
! 4621: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
! 4622: /* } */
! 4623: }
! 4624: } /* end loop jk */
! 4625:
! 4626: printf("\n");
! 4627: fprintf(ficlog,"\n");
! 4628: } /* end j=0 */
1.249 brouard 4629: } /* end j */
1.226 brouard 4630: dateintmean=dateintsum/k2cpt;
1.240 brouard 4631:
1.226 brouard 4632: fclose(ficresp);
4633: fclose(ficresphtm);
4634: fclose(ficresphtmfr);
4635: free_vector(meanq,1,nqfveff);
4636: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.251 ! brouard 4637: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4638: free_vector(pospropt,1,nlstate);
4639: free_vector(posprop,1,nlstate);
1.251 ! brouard 4640: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4641: free_vector(pp,1,nlstate);
4642: /* End of freqsummary */
4643: }
1.126 brouard 4644:
4645: /************ Prevalence ********************/
1.227 brouard 4646: 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)
4647: {
4648: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4649: in each health status at the date of interview (if between dateprev1 and dateprev2).
4650: We still use firstpass and lastpass as another selection.
4651: */
1.126 brouard 4652:
1.227 brouard 4653: int i, m, jk, j1, bool, z1,j, iv;
4654: int mi; /* Effective wave */
4655: int iage;
4656: double agebegin, ageend;
4657:
4658: double **prop;
4659: double posprop;
4660: double y2; /* in fractional years */
4661: int iagemin, iagemax;
4662: int first; /** to stop verbosity which is redirected to log file */
4663:
4664: iagemin= (int) agemin;
4665: iagemax= (int) agemax;
4666: /*pp=vector(1,nlstate);*/
1.251 ! brouard 4667: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4668: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4669: j1=0;
1.222 brouard 4670:
1.227 brouard 4671: /*j=cptcoveff;*/
4672: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4673:
1.227 brouard 4674: first=1;
4675: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4676: for (i=1; i<=nlstate; i++)
1.251 ! brouard 4677: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4678: prop[i][iage]=0.0;
4679: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4680: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4681: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4682:
4683: for (i=1; i<=imx; i++) { /* Each individual */
4684: bool=1;
4685: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4686: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4687: m=mw[mi][i];
4688: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4689: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4690: for (z1=1; z1<=cptcoveff; z1++){
4691: if( Fixed[Tmodelind[z1]]==1){
4692: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4693: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4694: bool=0;
4695: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4696: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4697: bool=0;
4698: }
4699: }
4700: if(bool==1){ /* Otherwise we skip that wave/person */
4701: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4702: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4703: if(m >=firstpass && m <=lastpass){
4704: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4705: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4706: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4707: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 ! brouard 4708: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4709: 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);
4710: exit(1);
4711: }
4712: if (s[m][i]>0 && s[m][i]<=nlstate) {
4713: /*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]]);*/
4714: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4715: prop[s[m][i]][iagemax+3] += weight[i];
4716: } /* end valid statuses */
4717: } /* end selection of dates */
4718: } /* end selection of waves */
4719: } /* end bool */
4720: } /* end wave */
4721: } /* end individual */
4722: for(i=iagemin; i <= iagemax+3; i++){
4723: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4724: posprop += prop[jk][i];
4725: }
4726:
4727: for(jk=1; jk <=nlstate ; jk++){
4728: if( i <= iagemax){
4729: if(posprop>=1.e-5){
4730: probs[i][jk][j1]= prop[jk][i]/posprop;
4731: } else{
4732: if(first==1){
4733: first=0;
4734: 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]);
4735: }
4736: }
4737: }
4738: }/* end jk */
4739: }/* end i */
1.222 brouard 4740: /*} *//* end i1 */
1.227 brouard 4741: } /* end j1 */
1.222 brouard 4742:
1.227 brouard 4743: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4744: /*free_vector(pp,1,nlstate);*/
1.251 ! brouard 4745: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4746: } /* End of prevalence */
1.126 brouard 4747:
4748: /************* Waves Concatenation ***************/
4749:
4750: 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)
4751: {
4752: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4753: Death is a valid wave (if date is known).
4754: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4755: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4756: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4757: */
1.126 brouard 4758:
1.224 brouard 4759: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4760: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4761: double sum=0., jmean=0.;*/
1.224 brouard 4762: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4763: int j, k=0,jk, ju, jl;
4764: double sum=0.;
4765: first=0;
1.214 brouard 4766: firstwo=0;
1.217 brouard 4767: firsthree=0;
1.218 brouard 4768: firstfour=0;
1.164 brouard 4769: jmin=100000;
1.126 brouard 4770: jmax=-1;
4771: jmean=0.;
1.224 brouard 4772:
4773: /* Treating live states */
1.214 brouard 4774: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4775: mi=0; /* First valid wave */
1.227 brouard 4776: mli=0; /* Last valid wave */
1.126 brouard 4777: m=firstpass;
1.214 brouard 4778: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4779: 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 */
4780: mli=m-1;/* mw[++mi][i]=m-1; */
4781: }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 */
4782: mw[++mi][i]=m;
4783: mli=m;
1.224 brouard 4784: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4785: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4786: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4787: }
1.227 brouard 4788: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4789: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4790: break;
1.224 brouard 4791: #else
1.227 brouard 4792: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4793: if(firsthree == 0){
4794: 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);
4795: firsthree=1;
4796: }
4797: 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);
4798: mw[++mi][i]=m;
4799: mli=m;
4800: }
4801: if(s[m][i]==-2){ /* Vital status is really unknown */
4802: nbwarn++;
4803: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4804: 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);
4805: 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);
4806: }
4807: break;
4808: }
4809: break;
1.224 brouard 4810: #endif
1.227 brouard 4811: }/* End m >= lastpass */
1.126 brouard 4812: }/* end while */
1.224 brouard 4813:
1.227 brouard 4814: /* 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 4815: /* After last pass */
1.224 brouard 4816: /* Treating death states */
1.214 brouard 4817: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4818: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4819: /* } */
1.126 brouard 4820: mi++; /* Death is another wave */
4821: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4822: /* Only death is a correct wave */
1.126 brouard 4823: mw[mi][i]=m;
1.224 brouard 4824: }
4825: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4826: 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 4827: /* m++; */
4828: /* mi++; */
4829: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4830: /* mw[mi][i]=m; */
1.218 brouard 4831: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4832: 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 */
4833: nbwarn++;
4834: if(firstfiv==0){
4835: 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 );
4836: firstfiv=1;
4837: }else{
4838: 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 );
4839: }
4840: }else{ /* Death occured afer last wave potential bias */
4841: nberr++;
4842: if(firstwo==0){
4843: 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 );
4844: firstwo=1;
4845: }
4846: 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 );
4847: }
1.218 brouard 4848: }else{ /* end date of interview is known */
1.227 brouard 4849: /* death is known but not confirmed by death status at any wave */
4850: if(firstfour==0){
4851: 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 );
4852: firstfour=1;
4853: }
4854: 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 4855: }
1.224 brouard 4856: } /* end if date of death is known */
4857: #endif
4858: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4859: /* wav[i]=mw[mi][i]; */
1.126 brouard 4860: if(mi==0){
4861: nbwarn++;
4862: if(first==0){
1.227 brouard 4863: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4864: first=1;
1.126 brouard 4865: }
4866: if(first==1){
1.227 brouard 4867: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4868: }
4869: } /* end mi==0 */
4870: } /* End individuals */
1.214 brouard 4871: /* wav and mw are no more changed */
1.223 brouard 4872:
1.214 brouard 4873:
1.126 brouard 4874: for(i=1; i<=imx; i++){
4875: for(mi=1; mi<wav[i];mi++){
4876: if (stepm <=0)
1.227 brouard 4877: dh[mi][i]=1;
1.126 brouard 4878: else{
1.227 brouard 4879: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4880: if (agedc[i] < 2*AGESUP) {
4881: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4882: if(j==0) j=1; /* Survives at least one month after exam */
4883: else if(j<0){
4884: nberr++;
4885: 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]);
4886: j=1; /* Temporary Dangerous patch */
4887: 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);
4888: 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]);
4889: 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);
4890: }
4891: k=k+1;
4892: if (j >= jmax){
4893: jmax=j;
4894: ijmax=i;
4895: }
4896: if (j <= jmin){
4897: jmin=j;
4898: ijmin=i;
4899: }
4900: sum=sum+j;
4901: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4902: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4903: }
4904: }
4905: else{
4906: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4907: /* 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 4908:
1.227 brouard 4909: k=k+1;
4910: if (j >= jmax) {
4911: jmax=j;
4912: ijmax=i;
4913: }
4914: else if (j <= jmin){
4915: jmin=j;
4916: ijmin=i;
4917: }
4918: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4919: /*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]);*/
4920: if(j<0){
4921: nberr++;
4922: 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]);
4923: 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]);
4924: }
4925: sum=sum+j;
4926: }
4927: jk= j/stepm;
4928: jl= j -jk*stepm;
4929: ju= j -(jk+1)*stepm;
4930: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4931: if(jl==0){
4932: dh[mi][i]=jk;
4933: bh[mi][i]=0;
4934: }else{ /* We want a negative bias in order to only have interpolation ie
4935: * to avoid the price of an extra matrix product in likelihood */
4936: dh[mi][i]=jk+1;
4937: bh[mi][i]=ju;
4938: }
4939: }else{
4940: if(jl <= -ju){
4941: dh[mi][i]=jk;
4942: bh[mi][i]=jl; /* bias is positive if real duration
4943: * is higher than the multiple of stepm and negative otherwise.
4944: */
4945: }
4946: else{
4947: dh[mi][i]=jk+1;
4948: bh[mi][i]=ju;
4949: }
4950: if(dh[mi][i]==0){
4951: dh[mi][i]=1; /* At least one step */
4952: bh[mi][i]=ju; /* At least one step */
4953: /* 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);*/
4954: }
4955: } /* end if mle */
1.126 brouard 4956: }
4957: } /* end wave */
4958: }
4959: jmean=sum/k;
4960: 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 4961: 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 4962: }
1.126 brouard 4963:
4964: /*********** Tricode ****************************/
1.220 brouard 4965: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4966: {
4967: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4968: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4969: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4970: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4971: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4972: */
1.130 brouard 4973:
1.242 brouard 4974: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
4975: int modmaxcovj=0; /* Modality max of covariates j */
4976: int cptcode=0; /* Modality max of covariates j */
4977: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4978:
4979:
1.242 brouard 4980: /* cptcoveff=0; */
4981: /* *cptcov=0; */
1.126 brouard 4982:
1.242 brouard 4983: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4984:
1.242 brouard 4985: /* Loop on covariates without age and products and no quantitative variable */
4986: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
4987: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4988: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4989: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4990: switch(Fixed[k]) {
4991: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
4992: 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*/
4993: ij=(int)(covar[Tvar[k]][i]);
4994: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4995: * If product of Vn*Vm, still boolean *:
4996: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4997: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4998: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4999: modality of the nth covariate of individual i. */
5000: if (ij > modmaxcovj)
5001: modmaxcovj=ij;
5002: else if (ij < modmincovj)
5003: modmincovj=ij;
5004: if ((ij < -1) && (ij > NCOVMAX)){
5005: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5006: exit(1);
5007: }else
5008: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5009: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5010: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5011: /* getting the maximum value of the modality of the covariate
5012: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5013: female ies 1, then modmaxcovj=1.
5014: */
5015: } /* end for loop on individuals i */
5016: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5017: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5018: cptcode=modmaxcovj;
5019: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5020: /*for (i=0; i<=cptcode; i++) {*/
5021: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5022: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5023: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5024: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5025: if( j != -1){
5026: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5027: covariate for which somebody answered excluding
5028: undefined. Usually 2: 0 and 1. */
5029: }
5030: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5031: covariate for which somebody answered including
5032: undefined. Usually 3: -1, 0 and 1. */
5033: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5034: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5035: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5036:
1.242 brouard 5037: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5038: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5039: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5040: /* modmincovj=3; modmaxcovj = 7; */
5041: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5042: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5043: /* defining two dummy variables: variables V1_1 and V1_2.*/
5044: /* nbcode[Tvar[j]][ij]=k; */
5045: /* nbcode[Tvar[j]][1]=0; */
5046: /* nbcode[Tvar[j]][2]=1; */
5047: /* nbcode[Tvar[j]][3]=2; */
5048: /* To be continued (not working yet). */
5049: ij=0; /* ij is similar to i but can jump over null modalities */
5050: 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*/
5051: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5052: break;
5053: }
5054: ij++;
5055: 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*/
5056: cptcode = ij; /* New max modality for covar j */
5057: } /* end of loop on modality i=-1 to 1 or more */
5058: break;
5059: case 1: /* Testing on varying covariate, could be simple and
5060: * should look at waves or product of fixed *
5061: * varying. No time to test -1, assuming 0 and 1 only */
5062: ij=0;
5063: for(i=0; i<=1;i++){
5064: nbcode[Tvar[k]][++ij]=i;
5065: }
5066: break;
5067: default:
5068: break;
5069: } /* end switch */
5070: } /* end dummy test */
5071:
5072: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5073: /* /\*recode from 0 *\/ */
5074: /* k is a modality. If we have model=V1+V1*sex */
5075: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5076: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5077: /* } */
5078: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5079: /* if (ij > ncodemax[j]) { */
5080: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5081: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5082: /* break; */
5083: /* } */
5084: /* } /\* end of loop on modality k *\/ */
5085: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5086:
5087: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5088: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5089: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5090: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5091: 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 */
5092: 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 */
5093: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5094: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5095:
5096: ij=0;
5097: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5098: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5099: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5100: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5101: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5102: /* If product not in single variable we don't print results */
5103: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5104: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5105: 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*/
5106: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5107: 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 */
5108: if(Fixed[k]!=0)
5109: anyvaryingduminmodel=1;
5110: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5111: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5112: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5113: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5114: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5115: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5116: }
5117: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5118: /* ij--; */
5119: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5120: *cptcov=ij; /*Number of total real effective covariates: effective
5121: * because they can be excluded from the model and real
5122: * if in the model but excluded because missing values, but how to get k from ij?*/
5123: for(j=ij+1; j<= cptcovt; j++){
5124: Tvaraff[j]=0;
5125: Tmodelind[j]=0;
5126: }
5127: for(j=ntveff+1; j<= cptcovt; j++){
5128: TmodelInvind[j]=0;
5129: }
5130: /* To be sorted */
5131: ;
5132: }
1.126 brouard 5133:
1.145 brouard 5134:
1.126 brouard 5135: /*********** Health Expectancies ****************/
5136:
1.235 brouard 5137: 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 5138:
5139: {
5140: /* Health expectancies, no variances */
1.164 brouard 5141: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5142: int nhstepma, nstepma; /* Decreasing with age */
5143: double age, agelim, hf;
5144: double ***p3mat;
5145: double eip;
5146:
1.238 brouard 5147: /* pstamp(ficreseij); */
1.126 brouard 5148: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5149: fprintf(ficreseij,"# Age");
5150: for(i=1; i<=nlstate;i++){
5151: for(j=1; j<=nlstate;j++){
5152: fprintf(ficreseij," e%1d%1d ",i,j);
5153: }
5154: fprintf(ficreseij," e%1d. ",i);
5155: }
5156: fprintf(ficreseij,"\n");
5157:
5158:
5159: if(estepm < stepm){
5160: printf ("Problem %d lower than %d\n",estepm, stepm);
5161: }
5162: else hstepm=estepm;
5163: /* We compute the life expectancy from trapezoids spaced every estepm months
5164: * This is mainly to measure the difference between two models: for example
5165: * if stepm=24 months pijx are given only every 2 years and by summing them
5166: * we are calculating an estimate of the Life Expectancy assuming a linear
5167: * progression in between and thus overestimating or underestimating according
5168: * to the curvature of the survival function. If, for the same date, we
5169: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5170: * to compare the new estimate of Life expectancy with the same linear
5171: * hypothesis. A more precise result, taking into account a more precise
5172: * curvature will be obtained if estepm is as small as stepm. */
5173:
5174: /* For example we decided to compute the life expectancy with the smallest unit */
5175: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5176: nhstepm is the number of hstepm from age to agelim
5177: nstepm is the number of stepm from age to agelin.
5178: Look at hpijx to understand the reason of that which relies in memory size
5179: and note for a fixed period like estepm months */
5180: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5181: survival function given by stepm (the optimization length). Unfortunately it
5182: means that if the survival funtion is printed only each two years of age and if
5183: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5184: results. So we changed our mind and took the option of the best precision.
5185: */
5186: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5187:
5188: agelim=AGESUP;
5189: /* If stepm=6 months */
5190: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5191: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5192:
5193: /* nhstepm age range expressed in number of stepm */
5194: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5195: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5196: /* if (stepm >= YEARM) hstepm=1;*/
5197: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5198: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5199:
5200: for (age=bage; age<=fage; age ++){
5201: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5202: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5203: /* if (stepm >= YEARM) hstepm=1;*/
5204: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5205:
5206: /* If stepm=6 months */
5207: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5208: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5209:
1.235 brouard 5210: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5211:
5212: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5213:
5214: printf("%d|",(int)age);fflush(stdout);
5215: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5216:
5217: /* Computing expectancies */
5218: for(i=1; i<=nlstate;i++)
5219: for(j=1; j<=nlstate;j++)
5220: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5221: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5222:
5223: /* 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]);*/
5224:
5225: }
5226:
5227: fprintf(ficreseij,"%3.0f",age );
5228: for(i=1; i<=nlstate;i++){
5229: eip=0;
5230: for(j=1; j<=nlstate;j++){
5231: eip +=eij[i][j][(int)age];
5232: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5233: }
5234: fprintf(ficreseij,"%9.4f", eip );
5235: }
5236: fprintf(ficreseij,"\n");
5237:
5238: }
5239: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5240: printf("\n");
5241: fprintf(ficlog,"\n");
5242:
5243: }
5244:
1.235 brouard 5245: 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 5246:
5247: {
5248: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5249: to initial status i, ei. .
1.126 brouard 5250: */
5251: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5252: int nhstepma, nstepma; /* Decreasing with age */
5253: double age, agelim, hf;
5254: double ***p3matp, ***p3matm, ***varhe;
5255: double **dnewm,**doldm;
5256: double *xp, *xm;
5257: double **gp, **gm;
5258: double ***gradg, ***trgradg;
5259: int theta;
5260:
5261: double eip, vip;
5262:
5263: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5264: xp=vector(1,npar);
5265: xm=vector(1,npar);
5266: dnewm=matrix(1,nlstate*nlstate,1,npar);
5267: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5268:
5269: pstamp(ficresstdeij);
5270: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5271: fprintf(ficresstdeij,"# Age");
5272: for(i=1; i<=nlstate;i++){
5273: for(j=1; j<=nlstate;j++)
5274: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5275: fprintf(ficresstdeij," e%1d. ",i);
5276: }
5277: fprintf(ficresstdeij,"\n");
5278:
5279: pstamp(ficrescveij);
5280: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5281: fprintf(ficrescveij,"# Age");
5282: for(i=1; i<=nlstate;i++)
5283: for(j=1; j<=nlstate;j++){
5284: cptj= (j-1)*nlstate+i;
5285: for(i2=1; i2<=nlstate;i2++)
5286: for(j2=1; j2<=nlstate;j2++){
5287: cptj2= (j2-1)*nlstate+i2;
5288: if(cptj2 <= cptj)
5289: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5290: }
5291: }
5292: fprintf(ficrescveij,"\n");
5293:
5294: if(estepm < stepm){
5295: printf ("Problem %d lower than %d\n",estepm, stepm);
5296: }
5297: else hstepm=estepm;
5298: /* We compute the life expectancy from trapezoids spaced every estepm months
5299: * This is mainly to measure the difference between two models: for example
5300: * if stepm=24 months pijx are given only every 2 years and by summing them
5301: * we are calculating an estimate of the Life Expectancy assuming a linear
5302: * progression in between and thus overestimating or underestimating according
5303: * to the curvature of the survival function. If, for the same date, we
5304: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5305: * to compare the new estimate of Life expectancy with the same linear
5306: * hypothesis. A more precise result, taking into account a more precise
5307: * curvature will be obtained if estepm is as small as stepm. */
5308:
5309: /* For example we decided to compute the life expectancy with the smallest unit */
5310: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5311: nhstepm is the number of hstepm from age to agelim
5312: nstepm is the number of stepm from age to agelin.
5313: Look at hpijx to understand the reason of that which relies in memory size
5314: and note for a fixed period like estepm months */
5315: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5316: survival function given by stepm (the optimization length). Unfortunately it
5317: means that if the survival funtion is printed only each two years of age and if
5318: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5319: results. So we changed our mind and took the option of the best precision.
5320: */
5321: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5322:
5323: /* If stepm=6 months */
5324: /* nhstepm age range expressed in number of stepm */
5325: agelim=AGESUP;
5326: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5327: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5328: /* if (stepm >= YEARM) hstepm=1;*/
5329: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5330:
5331: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5332: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5333: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5334: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5335: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5336: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5337:
5338: for (age=bage; age<=fage; age ++){
5339: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5340: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5341: /* if (stepm >= YEARM) hstepm=1;*/
5342: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5343:
1.126 brouard 5344: /* If stepm=6 months */
5345: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5346: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5347:
5348: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5349:
1.126 brouard 5350: /* Computing Variances of health expectancies */
5351: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5352: decrease memory allocation */
5353: for(theta=1; theta <=npar; theta++){
5354: for(i=1; i<=npar; i++){
1.222 brouard 5355: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5356: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5357: }
1.235 brouard 5358: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5359: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5360:
1.126 brouard 5361: for(j=1; j<= nlstate; j++){
1.222 brouard 5362: for(i=1; i<=nlstate; i++){
5363: for(h=0; h<=nhstepm-1; h++){
5364: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5365: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5366: }
5367: }
1.126 brouard 5368: }
1.218 brouard 5369:
1.126 brouard 5370: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5371: for(h=0; h<=nhstepm-1; h++){
5372: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5373: }
1.126 brouard 5374: }/* End theta */
5375:
5376:
5377: for(h=0; h<=nhstepm-1; h++)
5378: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5379: for(theta=1; theta <=npar; theta++)
5380: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5381:
1.218 brouard 5382:
1.222 brouard 5383: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5384: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5385: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5386:
1.222 brouard 5387: printf("%d|",(int)age);fflush(stdout);
5388: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5389: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5390: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5391: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5392: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5393: for(ij=1;ij<=nlstate*nlstate;ij++)
5394: for(ji=1;ji<=nlstate*nlstate;ji++)
5395: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5396: }
5397: }
1.218 brouard 5398:
1.126 brouard 5399: /* Computing expectancies */
1.235 brouard 5400: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5401: for(i=1; i<=nlstate;i++)
5402: for(j=1; j<=nlstate;j++)
1.222 brouard 5403: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5404: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5405:
1.222 brouard 5406: /* 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 5407:
1.222 brouard 5408: }
1.218 brouard 5409:
1.126 brouard 5410: fprintf(ficresstdeij,"%3.0f",age );
5411: for(i=1; i<=nlstate;i++){
5412: eip=0.;
5413: vip=0.;
5414: for(j=1; j<=nlstate;j++){
1.222 brouard 5415: eip += eij[i][j][(int)age];
5416: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5417: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5418: 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 5419: }
5420: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5421: }
5422: fprintf(ficresstdeij,"\n");
1.218 brouard 5423:
1.126 brouard 5424: fprintf(ficrescveij,"%3.0f",age );
5425: for(i=1; i<=nlstate;i++)
5426: for(j=1; j<=nlstate;j++){
1.222 brouard 5427: cptj= (j-1)*nlstate+i;
5428: for(i2=1; i2<=nlstate;i2++)
5429: for(j2=1; j2<=nlstate;j2++){
5430: cptj2= (j2-1)*nlstate+i2;
5431: if(cptj2 <= cptj)
5432: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5433: }
1.126 brouard 5434: }
5435: fprintf(ficrescveij,"\n");
1.218 brouard 5436:
1.126 brouard 5437: }
5438: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5439: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5440: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5441: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5442: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5443: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5444: printf("\n");
5445: fprintf(ficlog,"\n");
1.218 brouard 5446:
1.126 brouard 5447: free_vector(xm,1,npar);
5448: free_vector(xp,1,npar);
5449: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5450: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5451: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5452: }
1.218 brouard 5453:
1.126 brouard 5454: /************ Variance ******************/
1.235 brouard 5455: 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 5456: {
5457: /* Variance of health expectancies */
5458: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5459: /* double **newm;*/
5460: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5461:
5462: /* int movingaverage(); */
5463: double **dnewm,**doldm;
5464: double **dnewmp,**doldmp;
5465: int i, j, nhstepm, hstepm, h, nstepm ;
5466: int k;
5467: double *xp;
5468: double **gp, **gm; /* for var eij */
5469: double ***gradg, ***trgradg; /*for var eij */
5470: double **gradgp, **trgradgp; /* for var p point j */
5471: double *gpp, *gmp; /* for var p point j */
5472: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5473: double ***p3mat;
5474: double age,agelim, hf;
5475: /* double ***mobaverage; */
5476: int theta;
5477: char digit[4];
5478: char digitp[25];
5479:
5480: char fileresprobmorprev[FILENAMELENGTH];
5481:
5482: if(popbased==1){
5483: if(mobilav!=0)
5484: strcpy(digitp,"-POPULBASED-MOBILAV_");
5485: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5486: }
5487: else
5488: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5489:
1.218 brouard 5490: /* if (mobilav!=0) { */
5491: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5492: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5493: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5494: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5495: /* } */
5496: /* } */
5497:
5498: strcpy(fileresprobmorprev,"PRMORPREV-");
5499: sprintf(digit,"%-d",ij);
5500: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5501: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5502: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5503: strcat(fileresprobmorprev,fileresu);
5504: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5505: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5506: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5507: }
5508: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5509: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5510: pstamp(ficresprobmorprev);
5511: 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 5512: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5513: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5514: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5515: }
5516: for(j=1;j<=cptcoveff;j++)
5517: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5518: fprintf(ficresprobmorprev,"\n");
5519:
1.218 brouard 5520: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5521: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5522: fprintf(ficresprobmorprev," p.%-d SE",j);
5523: for(i=1; i<=nlstate;i++)
5524: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5525: }
5526: fprintf(ficresprobmorprev,"\n");
5527:
5528: fprintf(ficgp,"\n# Routine varevsij");
5529: fprintf(ficgp,"\nunset title \n");
5530: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5531: 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");
5532: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5533: /* } */
5534: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5535: pstamp(ficresvij);
5536: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5537: if(popbased==1)
5538: 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);
5539: else
5540: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5541: fprintf(ficresvij,"# Age");
5542: for(i=1; i<=nlstate;i++)
5543: for(j=1; j<=nlstate;j++)
5544: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5545: fprintf(ficresvij,"\n");
5546:
5547: xp=vector(1,npar);
5548: dnewm=matrix(1,nlstate,1,npar);
5549: doldm=matrix(1,nlstate,1,nlstate);
5550: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5551: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5552:
5553: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5554: gpp=vector(nlstate+1,nlstate+ndeath);
5555: gmp=vector(nlstate+1,nlstate+ndeath);
5556: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5557:
1.218 brouard 5558: if(estepm < stepm){
5559: printf ("Problem %d lower than %d\n",estepm, stepm);
5560: }
5561: else hstepm=estepm;
5562: /* For example we decided to compute the life expectancy with the smallest unit */
5563: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5564: nhstepm is the number of hstepm from age to agelim
5565: nstepm is the number of stepm from age to agelim.
5566: Look at function hpijx to understand why because of memory size limitations,
5567: we decided (b) to get a life expectancy respecting the most precise curvature of the
5568: survival function given by stepm (the optimization length). Unfortunately it
5569: means that if the survival funtion is printed every two years of age and if
5570: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5571: results. So we changed our mind and took the option of the best precision.
5572: */
5573: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5574: agelim = AGESUP;
5575: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5576: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5577: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5578: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5579: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5580: gp=matrix(0,nhstepm,1,nlstate);
5581: gm=matrix(0,nhstepm,1,nlstate);
5582:
5583:
5584: for(theta=1; theta <=npar; theta++){
5585: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5586: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5587: }
5588:
1.242 brouard 5589: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5590:
5591: if (popbased==1) {
5592: if(mobilav ==0){
5593: for(i=1; i<=nlstate;i++)
5594: prlim[i][i]=probs[(int)age][i][ij];
5595: }else{ /* mobilav */
5596: for(i=1; i<=nlstate;i++)
5597: prlim[i][i]=mobaverage[(int)age][i][ij];
5598: }
5599: }
5600:
1.235 brouard 5601: 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 5602: for(j=1; j<= nlstate; j++){
5603: for(h=0; h<=nhstepm; h++){
5604: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5605: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5606: }
5607: }
5608: /* Next for computing probability of death (h=1 means
5609: computed over hstepm matrices product = hstepm*stepm months)
5610: as a weighted average of prlim.
5611: */
5612: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5613: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5614: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5615: }
5616: /* end probability of death */
5617:
5618: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5619: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5620:
1.242 brouard 5621: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5622:
5623: if (popbased==1) {
5624: if(mobilav ==0){
5625: for(i=1; i<=nlstate;i++)
5626: prlim[i][i]=probs[(int)age][i][ij];
5627: }else{ /* mobilav */
5628: for(i=1; i<=nlstate;i++)
5629: prlim[i][i]=mobaverage[(int)age][i][ij];
5630: }
5631: }
5632:
1.235 brouard 5633: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5634:
5635: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5636: for(h=0; h<=nhstepm; h++){
5637: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5638: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5639: }
5640: }
5641: /* This for computing probability of death (h=1 means
5642: computed over hstepm matrices product = hstepm*stepm months)
5643: as a weighted average of prlim.
5644: */
5645: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5646: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5647: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5648: }
5649: /* end probability of death */
5650:
5651: for(j=1; j<= nlstate; j++) /* vareij */
5652: for(h=0; h<=nhstepm; h++){
5653: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5654: }
5655:
5656: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5657: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5658: }
5659:
5660: } /* End theta */
5661:
5662: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5663:
5664: for(h=0; h<=nhstepm; h++) /* veij */
5665: for(j=1; j<=nlstate;j++)
5666: for(theta=1; theta <=npar; theta++)
5667: trgradg[h][j][theta]=gradg[h][theta][j];
5668:
5669: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5670: for(theta=1; theta <=npar; theta++)
5671: trgradgp[j][theta]=gradgp[theta][j];
5672:
5673:
5674: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5675: for(i=1;i<=nlstate;i++)
5676: for(j=1;j<=nlstate;j++)
5677: vareij[i][j][(int)age] =0.;
5678:
5679: for(h=0;h<=nhstepm;h++){
5680: for(k=0;k<=nhstepm;k++){
5681: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5682: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5683: for(i=1;i<=nlstate;i++)
5684: for(j=1;j<=nlstate;j++)
5685: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5686: }
5687: }
5688:
5689: /* pptj */
5690: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5691: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5692: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5693: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5694: varppt[j][i]=doldmp[j][i];
5695: /* end ppptj */
5696: /* x centered again */
5697:
1.242 brouard 5698: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5699:
5700: if (popbased==1) {
5701: if(mobilav ==0){
5702: for(i=1; i<=nlstate;i++)
5703: prlim[i][i]=probs[(int)age][i][ij];
5704: }else{ /* mobilav */
5705: for(i=1; i<=nlstate;i++)
5706: prlim[i][i]=mobaverage[(int)age][i][ij];
5707: }
5708: }
5709:
5710: /* This for computing probability of death (h=1 means
5711: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5712: as a weighted average of prlim.
5713: */
1.235 brouard 5714: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5715: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5716: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5717: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5718: }
5719: /* end probability of death */
5720:
5721: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5722: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5723: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5724: for(i=1; i<=nlstate;i++){
5725: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5726: }
5727: }
5728: fprintf(ficresprobmorprev,"\n");
5729:
5730: fprintf(ficresvij,"%.0f ",age );
5731: for(i=1; i<=nlstate;i++)
5732: for(j=1; j<=nlstate;j++){
5733: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5734: }
5735: fprintf(ficresvij,"\n");
5736: free_matrix(gp,0,nhstepm,1,nlstate);
5737: free_matrix(gm,0,nhstepm,1,nlstate);
5738: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5739: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5740: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5741: } /* End age */
5742: free_vector(gpp,nlstate+1,nlstate+ndeath);
5743: free_vector(gmp,nlstate+1,nlstate+ndeath);
5744: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5745: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5746: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5747: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5748: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5749: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5750: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5751: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5752: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5753: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5754: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5755: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5756: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5757: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5758: 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);
5759: /* 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 5760: */
1.218 brouard 5761: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5762: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5763:
1.218 brouard 5764: free_vector(xp,1,npar);
5765: free_matrix(doldm,1,nlstate,1,nlstate);
5766: free_matrix(dnewm,1,nlstate,1,npar);
5767: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5768: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5769: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5770: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5771: fclose(ficresprobmorprev);
5772: fflush(ficgp);
5773: fflush(fichtm);
5774: } /* end varevsij */
1.126 brouard 5775:
5776: /************ Variance of prevlim ******************/
1.235 brouard 5777: 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 5778: {
1.205 brouard 5779: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5780: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5781:
1.126 brouard 5782: double **dnewm,**doldm;
5783: int i, j, nhstepm, hstepm;
5784: double *xp;
5785: double *gp, *gm;
5786: double **gradg, **trgradg;
1.208 brouard 5787: double **mgm, **mgp;
1.126 brouard 5788: double age,agelim;
5789: int theta;
5790:
5791: pstamp(ficresvpl);
5792: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5793: fprintf(ficresvpl,"# Age ");
5794: if(nresult >=1)
5795: fprintf(ficresvpl," Result# ");
1.126 brouard 5796: for(i=1; i<=nlstate;i++)
5797: fprintf(ficresvpl," %1d-%1d",i,i);
5798: fprintf(ficresvpl,"\n");
5799:
5800: xp=vector(1,npar);
5801: dnewm=matrix(1,nlstate,1,npar);
5802: doldm=matrix(1,nlstate,1,nlstate);
5803:
5804: hstepm=1*YEARM; /* Every year of age */
5805: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5806: agelim = AGESUP;
5807: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5808: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5809: if (stepm >= YEARM) hstepm=1;
5810: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5811: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5812: mgp=matrix(1,npar,1,nlstate);
5813: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5814: gp=vector(1,nlstate);
5815: gm=vector(1,nlstate);
5816:
5817: for(theta=1; theta <=npar; theta++){
5818: for(i=1; i<=npar; i++){ /* Computes gradient */
5819: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5820: }
1.209 brouard 5821: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5822: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5823: else
1.235 brouard 5824: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5825: for(i=1;i<=nlstate;i++){
1.126 brouard 5826: gp[i] = prlim[i][i];
1.208 brouard 5827: mgp[theta][i] = prlim[i][i];
5828: }
1.126 brouard 5829: for(i=1; i<=npar; i++) /* Computes gradient */
5830: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5831: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5832: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5833: else
1.235 brouard 5834: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5835: for(i=1;i<=nlstate;i++){
1.126 brouard 5836: gm[i] = prlim[i][i];
1.208 brouard 5837: mgm[theta][i] = prlim[i][i];
5838: }
1.126 brouard 5839: for(i=1;i<=nlstate;i++)
5840: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5841: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5842: } /* End theta */
5843:
5844: trgradg =matrix(1,nlstate,1,npar);
5845:
5846: for(j=1; j<=nlstate;j++)
5847: for(theta=1; theta <=npar; theta++)
5848: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5849: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5850: /* printf("\nmgm mgp %d ",(int)age); */
5851: /* for(j=1; j<=nlstate;j++){ */
5852: /* printf(" %d ",j); */
5853: /* for(theta=1; theta <=npar; theta++) */
5854: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5855: /* printf("\n "); */
5856: /* } */
5857: /* } */
5858: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5859: /* printf("\n gradg %d ",(int)age); */
5860: /* for(j=1; j<=nlstate;j++){ */
5861: /* printf("%d ",j); */
5862: /* for(theta=1; theta <=npar; theta++) */
5863: /* printf("%d %lf ",theta,gradg[theta][j]); */
5864: /* printf("\n "); */
5865: /* } */
5866: /* } */
1.126 brouard 5867:
5868: for(i=1;i<=nlstate;i++)
5869: varpl[i][(int)age] =0.;
1.209 brouard 5870: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5871: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5872: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5873: }else{
1.126 brouard 5874: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5875: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5876: }
1.126 brouard 5877: for(i=1;i<=nlstate;i++)
5878: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5879:
5880: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5881: if(nresult >=1)
5882: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5883: for(i=1; i<=nlstate;i++)
5884: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5885: fprintf(ficresvpl,"\n");
5886: free_vector(gp,1,nlstate);
5887: free_vector(gm,1,nlstate);
1.208 brouard 5888: free_matrix(mgm,1,npar,1,nlstate);
5889: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5890: free_matrix(gradg,1,npar,1,nlstate);
5891: free_matrix(trgradg,1,nlstate,1,npar);
5892: } /* End age */
5893:
5894: free_vector(xp,1,npar);
5895: free_matrix(doldm,1,nlstate,1,npar);
5896: free_matrix(dnewm,1,nlstate,1,nlstate);
5897:
5898: }
5899:
5900: /************ Variance of one-step probabilities ******************/
5901: 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 5902: {
5903: int i, j=0, k1, l1, tj;
5904: int k2, l2, j1, z1;
5905: int k=0, l;
5906: int first=1, first1, first2;
5907: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5908: double **dnewm,**doldm;
5909: double *xp;
5910: double *gp, *gm;
5911: double **gradg, **trgradg;
5912: double **mu;
5913: double age, cov[NCOVMAX+1];
5914: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5915: int theta;
5916: char fileresprob[FILENAMELENGTH];
5917: char fileresprobcov[FILENAMELENGTH];
5918: char fileresprobcor[FILENAMELENGTH];
5919: double ***varpij;
5920:
5921: strcpy(fileresprob,"PROB_");
5922: strcat(fileresprob,fileres);
5923: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5924: printf("Problem with resultfile: %s\n", fileresprob);
5925: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5926: }
5927: strcpy(fileresprobcov,"PROBCOV_");
5928: strcat(fileresprobcov,fileresu);
5929: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5930: printf("Problem with resultfile: %s\n", fileresprobcov);
5931: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5932: }
5933: strcpy(fileresprobcor,"PROBCOR_");
5934: strcat(fileresprobcor,fileresu);
5935: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5936: printf("Problem with resultfile: %s\n", fileresprobcor);
5937: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5938: }
5939: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5940: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5941: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5942: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5943: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5944: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5945: pstamp(ficresprob);
5946: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5947: fprintf(ficresprob,"# Age");
5948: pstamp(ficresprobcov);
5949: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5950: fprintf(ficresprobcov,"# Age");
5951: pstamp(ficresprobcor);
5952: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5953: fprintf(ficresprobcor,"# Age");
1.126 brouard 5954:
5955:
1.222 brouard 5956: for(i=1; i<=nlstate;i++)
5957: for(j=1; j<=(nlstate+ndeath);j++){
5958: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5959: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5960: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5961: }
5962: /* fprintf(ficresprob,"\n");
5963: fprintf(ficresprobcov,"\n");
5964: fprintf(ficresprobcor,"\n");
5965: */
5966: xp=vector(1,npar);
5967: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5968: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5969: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5970: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5971: first=1;
5972: fprintf(ficgp,"\n# Routine varprob");
5973: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5974: fprintf(fichtm,"\n");
5975:
5976: 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);
5977: 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);
5978: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5979: and drawn. It helps understanding how is the covariance between two incidences.\
5980: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5981: 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 5982: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5983: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5984: standard deviations wide on each axis. <br>\
5985: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5986: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5987: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5988:
1.222 brouard 5989: cov[1]=1;
5990: /* tj=cptcoveff; */
1.225 brouard 5991: tj = (int) pow(2,cptcoveff);
1.222 brouard 5992: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5993: j1=0;
1.224 brouard 5994: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5995: if (cptcovn>0) {
5996: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5997: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5998: fprintf(ficresprob, "**********\n#\n");
5999: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6000: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6001: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6002:
1.222 brouard 6003: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6004: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6005: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6006:
6007:
1.222 brouard 6008: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6009: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6010: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6011:
1.222 brouard 6012: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6013: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6014: fprintf(ficresprobcor, "**********\n#");
6015: if(invalidvarcomb[j1]){
6016: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6017: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6018: continue;
6019: }
6020: }
6021: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6022: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6023: gp=vector(1,(nlstate)*(nlstate+ndeath));
6024: gm=vector(1,(nlstate)*(nlstate+ndeath));
6025: for (age=bage; age<=fage; age ++){
6026: cov[2]=age;
6027: if(nagesqr==1)
6028: cov[3]= age*age;
6029: for (k=1; k<=cptcovn;k++) {
6030: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6031: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6032: * 1 1 1 1 1
6033: * 2 2 1 1 1
6034: * 3 1 2 1 1
6035: */
6036: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6037: }
6038: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6039: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6040: for (k=1; k<=cptcovprod;k++)
6041: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6042:
6043:
1.222 brouard 6044: for(theta=1; theta <=npar; theta++){
6045: for(i=1; i<=npar; i++)
6046: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6047:
1.222 brouard 6048: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6049:
1.222 brouard 6050: k=0;
6051: for(i=1; i<= (nlstate); i++){
6052: for(j=1; j<=(nlstate+ndeath);j++){
6053: k=k+1;
6054: gp[k]=pmmij[i][j];
6055: }
6056: }
1.220 brouard 6057:
1.222 brouard 6058: for(i=1; i<=npar; i++)
6059: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6060:
1.222 brouard 6061: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6062: k=0;
6063: for(i=1; i<=(nlstate); i++){
6064: for(j=1; j<=(nlstate+ndeath);j++){
6065: k=k+1;
6066: gm[k]=pmmij[i][j];
6067: }
6068: }
1.220 brouard 6069:
1.222 brouard 6070: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6071: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6072: }
1.126 brouard 6073:
1.222 brouard 6074: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6075: for(theta=1; theta <=npar; theta++)
6076: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6077:
1.222 brouard 6078: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6079: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6080:
1.222 brouard 6081: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6082:
1.222 brouard 6083: k=0;
6084: for(i=1; i<=(nlstate); i++){
6085: for(j=1; j<=(nlstate+ndeath);j++){
6086: k=k+1;
6087: mu[k][(int) age]=pmmij[i][j];
6088: }
6089: }
6090: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6091: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6092: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6093:
1.222 brouard 6094: /*printf("\n%d ",(int)age);
6095: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6096: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6097: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6098: }*/
1.220 brouard 6099:
1.222 brouard 6100: fprintf(ficresprob,"\n%d ",(int)age);
6101: fprintf(ficresprobcov,"\n%d ",(int)age);
6102: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6103:
1.222 brouard 6104: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6105: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6106: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6107: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6108: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6109: }
6110: i=0;
6111: for (k=1; k<=(nlstate);k++){
6112: for (l=1; l<=(nlstate+ndeath);l++){
6113: i++;
6114: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6115: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6116: for (j=1; j<=i;j++){
6117: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6118: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6119: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6120: }
6121: }
6122: }/* end of loop for state */
6123: } /* end of loop for age */
6124: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6125: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6126: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6127: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6128:
6129: /* Confidence intervalle of pij */
6130: /*
6131: fprintf(ficgp,"\nunset parametric;unset label");
6132: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6133: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6134: 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);
6135: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6136: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6137: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6138: */
6139:
6140: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6141: first1=1;first2=2;
6142: for (k2=1; k2<=(nlstate);k2++){
6143: for (l2=1; l2<=(nlstate+ndeath);l2++){
6144: if(l2==k2) continue;
6145: j=(k2-1)*(nlstate+ndeath)+l2;
6146: for (k1=1; k1<=(nlstate);k1++){
6147: for (l1=1; l1<=(nlstate+ndeath);l1++){
6148: if(l1==k1) continue;
6149: i=(k1-1)*(nlstate+ndeath)+l1;
6150: if(i<=j) continue;
6151: for (age=bage; age<=fage; age ++){
6152: if ((int)age %5==0){
6153: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6154: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6155: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6156: mu1=mu[i][(int) age]/stepm*YEARM ;
6157: mu2=mu[j][(int) age]/stepm*YEARM;
6158: c12=cv12/sqrt(v1*v2);
6159: /* Computing eigen value of matrix of covariance */
6160: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6161: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6162: if ((lc2 <0) || (lc1 <0) ){
6163: if(first2==1){
6164: first1=0;
6165: 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);
6166: }
6167: 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);
6168: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6169: /* lc2=fabs(lc2); */
6170: }
1.220 brouard 6171:
1.222 brouard 6172: /* Eigen vectors */
6173: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6174: /*v21=sqrt(1.-v11*v11); *//* error */
6175: v21=(lc1-v1)/cv12*v11;
6176: v12=-v21;
6177: v22=v11;
6178: tnalp=v21/v11;
6179: if(first1==1){
6180: first1=0;
6181: 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);
6182: }
6183: 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);
6184: /*printf(fignu*/
6185: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6186: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6187: if(first==1){
6188: first=0;
6189: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6190: fprintf(ficgp,"\nset parametric;unset label");
6191: 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);
6192: fprintf(ficgp,"\nset ter svg size 640, 480");
6193: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6194: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6195: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6196: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6197: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6198: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6199: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6200: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6201: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6202: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6203: 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", \
6204: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6205: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6206: }else{
6207: first=0;
6208: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6209: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6210: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6211: 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", \
6212: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6213: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6214: }/* if first */
6215: } /* age mod 5 */
6216: } /* end loop age */
6217: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6218: first=1;
6219: } /*l12 */
6220: } /* k12 */
6221: } /*l1 */
6222: }/* k1 */
6223: } /* loop on combination of covariates j1 */
6224: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6225: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6226: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6227: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6228: free_vector(xp,1,npar);
6229: fclose(ficresprob);
6230: fclose(ficresprobcov);
6231: fclose(ficresprobcor);
6232: fflush(ficgp);
6233: fflush(fichtmcov);
6234: }
1.126 brouard 6235:
6236:
6237: /******************* Printing html file ***********/
1.201 brouard 6238: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6239: int lastpass, int stepm, int weightopt, char model[],\
6240: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6241: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6242: double jprev1, double mprev1,double anprev1, double dateprev1, \
6243: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6244: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6245:
6246: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6247: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6248: </ul>");
1.237 brouard 6249: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6250: </ul>", model);
1.214 brouard 6251: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6252: 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",
6253: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6254: 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 6255: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6256: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6257: fprintf(fichtm,"\
6258: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6259: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6260: fprintf(fichtm,"\
1.217 brouard 6261: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6262: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6263: fprintf(fichtm,"\
1.126 brouard 6264: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6265: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6266: fprintf(fichtm,"\
1.217 brouard 6267: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6268: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6269: fprintf(fichtm,"\
1.211 brouard 6270: - (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 6271: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6272: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6273: if(prevfcast==1){
6274: fprintf(fichtm,"\
6275: - Prevalence projections by age and states: \
1.201 brouard 6276: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6277: }
1.126 brouard 6278:
1.222 brouard 6279: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6280:
1.225 brouard 6281: m=pow(2,cptcoveff);
1.222 brouard 6282: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6283:
1.222 brouard 6284: jj1=0;
1.237 brouard 6285:
6286: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6287: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6288: if(TKresult[nres]!= k1)
6289: continue;
1.220 brouard 6290:
1.222 brouard 6291: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6292: jj1++;
6293: if (cptcovn > 0) {
6294: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6295: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6296: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6297: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6298: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6299: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6300: }
1.237 brouard 6301: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6302: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6303: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6304: }
6305:
1.230 brouard 6306: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6307: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6308: if(invalidvarcomb[k1]){
6309: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6310: printf("\nCombination (%d) ignored because no cases \n",k1);
6311: continue;
6312: }
6313: }
6314: /* aij, bij */
1.241 brouard 6315: 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> \
6316: <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 6317: /* Pij */
1.241 brouard 6318: 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> \
6319: <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 6320: /* Quasi-incidences */
6321: 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 6322: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6323: 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 6324: 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> \
6325: <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 6326: /* Survival functions (period) in state j */
6327: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6328: 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> \
6329: <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 6330: }
6331: /* State specific survival functions (period) */
6332: for(cpt=1; cpt<=nlstate;cpt++){
6333: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6334: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6335: <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 6336: }
6337: /* Period (stable) prevalence in each health state */
6338: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6339: 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> \
6340: <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 6341: }
6342: if(backcast==1){
6343: /* Period (stable) back prevalence in each health state */
6344: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6345: 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> \
6346: <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 6347: }
1.217 brouard 6348: }
1.222 brouard 6349: if(prevfcast==1){
6350: /* Projection of prevalence up to period (stable) prevalence in each health state */
6351: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6352: 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> \
6353: <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 6354: }
6355: }
1.220 brouard 6356:
1.222 brouard 6357: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6358: 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> \
6359: <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 6360: }
6361: /* } /\* end i1 *\/ */
6362: }/* End k1 */
6363: fprintf(fichtm,"</ul>");
1.126 brouard 6364:
1.222 brouard 6365: fprintf(fichtm,"\
1.126 brouard 6366: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6367: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6368: - 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 6369: But because parameters are usually highly correlated (a higher incidence of disability \
6370: and a higher incidence of recovery can give very close observed transition) it might \
6371: be very useful to look not only at linear confidence intervals estimated from the \
6372: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6373: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6374: covariance matrix of the one-step probabilities. \
6375: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6376:
1.222 brouard 6377: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6378: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6379: fprintf(fichtm,"\
1.126 brouard 6380: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6381: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6382:
1.222 brouard 6383: fprintf(fichtm,"\
1.126 brouard 6384: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6385: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6386: fprintf(fichtm,"\
1.126 brouard 6387: - 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): \
6388: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6389: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6390: fprintf(fichtm,"\
1.126 brouard 6391: - (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): \
6392: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6393: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6394: fprintf(fichtm,"\
1.128 brouard 6395: - 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 6396: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6397: fprintf(fichtm,"\
1.128 brouard 6398: - 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 6399: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6400: fprintf(fichtm,"\
1.126 brouard 6401: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6402: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6403:
6404: /* if(popforecast==1) fprintf(fichtm,"\n */
6405: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6406: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6407: /* <br>",fileres,fileres,fileres,fileres); */
6408: /* else */
6409: /* 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 6410: fflush(fichtm);
6411: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6412:
1.225 brouard 6413: m=pow(2,cptcoveff);
1.222 brouard 6414: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6415:
1.222 brouard 6416: jj1=0;
1.237 brouard 6417:
1.241 brouard 6418: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6419: for(k1=1; k1<=m;k1++){
1.237 brouard 6420: if(TKresult[nres]!= k1)
6421: continue;
1.222 brouard 6422: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6423: jj1++;
1.126 brouard 6424: if (cptcovn > 0) {
6425: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6426: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6427: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6428: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6429: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6430: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6431: }
6432:
1.126 brouard 6433: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6434:
1.222 brouard 6435: if(invalidvarcomb[k1]){
6436: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6437: continue;
6438: }
1.126 brouard 6439: }
6440: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6441: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6442: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6443: <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 6444: }
6445: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6446: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6447: true period expectancies (those weighted with period prevalences are also\
6448: drawn in addition to the population based expectancies computed using\
1.241 brouard 6449: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6450: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6451: /* } /\* end i1 *\/ */
6452: }/* End k1 */
1.241 brouard 6453: }/* End nres */
1.222 brouard 6454: fprintf(fichtm,"</ul>");
6455: fflush(fichtm);
1.126 brouard 6456: }
6457:
6458: /******************* Gnuplot file **************/
1.223 brouard 6459: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6460:
6461: char dirfileres[132],optfileres[132];
1.223 brouard 6462: char gplotcondition[132];
1.237 brouard 6463: 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 6464: int lv=0, vlv=0, kl=0;
1.130 brouard 6465: int ng=0;
1.201 brouard 6466: int vpopbased;
1.223 brouard 6467: int ioffset; /* variable offset for columns */
1.235 brouard 6468: int nres=0; /* Index of resultline */
1.219 brouard 6469:
1.126 brouard 6470: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6471: /* printf("Problem with file %s",optionfilegnuplot); */
6472: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6473: /* } */
6474:
6475: /*#ifdef windows */
6476: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6477: /*#endif */
1.225 brouard 6478: m=pow(2,cptcoveff);
1.126 brouard 6479:
1.202 brouard 6480: /* Contribution to likelihood */
6481: /* Plot the probability implied in the likelihood */
1.223 brouard 6482: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6483: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6484: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6485: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6486: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6487: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6488: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6489: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6490: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6491: 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));
6492: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6493: 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));
6494: for (i=1; i<= nlstate ; i ++) {
6495: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6496: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6497: 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);
6498: for (j=2; j<= nlstate+ndeath ; j ++) {
6499: 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);
6500: }
6501: fprintf(ficgp,";\nset out; unset ylabel;\n");
6502: }
6503: /* 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 */
6504: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6505: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6506: fprintf(ficgp,"\nset out;unset log\n");
6507: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6508:
1.126 brouard 6509: strcpy(dirfileres,optionfilefiname);
6510: strcpy(optfileres,"vpl");
1.223 brouard 6511: /* 1eme*/
1.238 brouard 6512: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6513: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6514: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6515: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6516: if(TKresult[nres]!= k1)
6517: continue;
6518: /* We are interested in selected combination by the resultline */
1.246 brouard 6519: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6520: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6521: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6522: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6523: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6524: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6525: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6526: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6527: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6528: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6529: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6530: }
6531: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6532: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6533: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6534: }
1.246 brouard 6535: /* printf("\n#\n"); */
1.238 brouard 6536: fprintf(ficgp,"\n#\n");
6537: if(invalidvarcomb[k1]){
6538: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6539: continue;
6540: }
1.235 brouard 6541:
1.241 brouard 6542: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6543: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6544: 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 6545:
1.238 brouard 6546: for (i=1; i<= nlstate ; i ++) {
6547: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6548: else fprintf(ficgp," %%*lf (%%*lf)");
6549: }
1.242 brouard 6550: 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 6551: for (i=1; i<= nlstate ; i ++) {
6552: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6553: else fprintf(ficgp," %%*lf (%%*lf)");
6554: }
1.242 brouard 6555: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6556: for (i=1; i<= nlstate ; i ++) {
6557: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6558: else fprintf(ficgp," %%*lf (%%*lf)");
6559: }
6560: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6561: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6562: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6563: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6564: if(cptcoveff ==0){
1.245 brouard 6565: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6566: }else{
6567: kl=0;
6568: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6569: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6570: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6571: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6572: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6573: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6574: kl++;
1.238 brouard 6575: /* 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 *\/ */
6576: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6577: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6578: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6579: if(k==cptcoveff){
1.245 brouard 6580: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6581: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6582: }else{
6583: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6584: kl++;
6585: }
6586: } /* end covariate */
6587: } /* end if no covariate */
6588: } /* end if backcast */
6589: fprintf(ficgp,"\nset out \n");
6590: } /* nres */
1.201 brouard 6591: } /* k1 */
6592: } /* cpt */
1.235 brouard 6593:
6594:
1.126 brouard 6595: /*2 eme*/
1.238 brouard 6596: for (k1=1; k1<= m ; k1 ++){
6597: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6598: if(TKresult[nres]!= k1)
6599: continue;
6600: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6601: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6602: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6603: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6604: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6605: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6606: vlv= nbcode[Tvaraff[k]][lv];
6607: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6608: }
1.237 brouard 6609: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6610: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6611: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6612: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6613: }
1.211 brouard 6614: fprintf(ficgp,"\n#\n");
1.223 brouard 6615: if(invalidvarcomb[k1]){
6616: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6617: continue;
6618: }
1.219 brouard 6619:
1.241 brouard 6620: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6621: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6622: if(vpopbased==0)
6623: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6624: else
6625: fprintf(ficgp,"\nreplot ");
6626: for (i=1; i<= nlstate+1 ; i ++) {
6627: k=2*i;
6628: 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);
6629: for (j=1; j<= nlstate+1 ; j ++) {
6630: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6631: else fprintf(ficgp," %%*lf (%%*lf)");
6632: }
6633: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6634: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6635: 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);
6636: for (j=1; j<= nlstate+1 ; j ++) {
6637: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6638: else fprintf(ficgp," %%*lf (%%*lf)");
6639: }
6640: fprintf(ficgp,"\" t\"\" w l lt 0,");
6641: 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);
6642: for (j=1; j<= nlstate+1 ; j ++) {
6643: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6644: else fprintf(ficgp," %%*lf (%%*lf)");
6645: }
6646: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6647: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6648: } /* state */
6649: } /* vpopbased */
1.244 brouard 6650: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6651: } /* end nres */
6652: } /* k1 end 2 eme*/
6653:
6654:
6655: /*3eme*/
6656: for (k1=1; k1<= m ; k1 ++){
6657: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6658: if(TKresult[nres]!= k1)
1.238 brouard 6659: continue;
6660:
6661: for (cpt=1; cpt<= nlstate ; cpt ++) {
6662: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6663: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6664: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6665: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6666: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6667: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6668: vlv= nbcode[Tvaraff[k]][lv];
6669: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6670: }
6671: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6672: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6673: }
6674: fprintf(ficgp,"\n#\n");
6675: if(invalidvarcomb[k1]){
6676: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6677: continue;
6678: }
6679:
6680: /* k=2+nlstate*(2*cpt-2); */
6681: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6682: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6683: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6684: 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 6685: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6686: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6687: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6688: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6689: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6690: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6691:
1.238 brouard 6692: */
6693: for (i=1; i< nlstate ; i ++) {
6694: 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);
6695: /* 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 6696:
1.238 brouard 6697: }
6698: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6699: }
6700: } /* end nres */
6701: } /* end kl 3eme */
1.126 brouard 6702:
1.223 brouard 6703: /* 4eme */
1.201 brouard 6704: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6705: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6706: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6707: if(TKresult[nres]!= k1)
1.223 brouard 6708: continue;
1.238 brouard 6709: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6710: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6711: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6712: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6713: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6714: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6715: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6716: vlv= nbcode[Tvaraff[k]][lv];
6717: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6718: }
6719: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6720: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6721: }
6722: fprintf(ficgp,"\n#\n");
6723: if(invalidvarcomb[k1]){
6724: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6725: continue;
1.223 brouard 6726: }
1.238 brouard 6727:
1.241 brouard 6728: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6729: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6730: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6731: k=3;
6732: for (i=1; i<= nlstate ; i ++){
6733: if(i==1){
6734: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6735: }else{
6736: fprintf(ficgp,", '' ");
6737: }
6738: l=(nlstate+ndeath)*(i-1)+1;
6739: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6740: for (j=2; j<= nlstate+ndeath ; j ++)
6741: fprintf(ficgp,"+$%d",k+l+j-1);
6742: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6743: } /* nlstate */
6744: fprintf(ficgp,"\nset out\n");
6745: } /* end cpt state*/
6746: } /* end nres */
6747: } /* end covariate k1 */
6748:
1.220 brouard 6749: /* 5eme */
1.201 brouard 6750: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6751: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6752: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6753: if(TKresult[nres]!= k1)
1.227 brouard 6754: continue;
1.238 brouard 6755: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6756: 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);
6757: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6758: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6759: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6760: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6761: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6762: vlv= nbcode[Tvaraff[k]][lv];
6763: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6764: }
6765: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6766: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6767: }
6768: fprintf(ficgp,"\n#\n");
6769: if(invalidvarcomb[k1]){
6770: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6771: continue;
6772: }
1.227 brouard 6773:
1.241 brouard 6774: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6775: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6776: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6777: k=3;
6778: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6779: if(j==1)
6780: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6781: else
6782: fprintf(ficgp,", '' ");
6783: l=(nlstate+ndeath)*(cpt-1) +j;
6784: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6785: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6786: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6787: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6788: } /* nlstate */
6789: fprintf(ficgp,", '' ");
6790: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6791: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6792: l=(nlstate+ndeath)*(cpt-1) +j;
6793: if(j < nlstate)
6794: fprintf(ficgp,"$%d +",k+l);
6795: else
6796: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6797: }
6798: fprintf(ficgp,"\nset out\n");
6799: } /* end cpt state*/
6800: } /* end covariate */
6801: } /* end nres */
1.227 brouard 6802:
1.220 brouard 6803: /* 6eme */
1.202 brouard 6804: /* CV preval stable (period) for each covariate */
1.237 brouard 6805: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6806: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6807: if(TKresult[nres]!= k1)
6808: continue;
1.153 brouard 6809: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6810:
1.211 brouard 6811: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6812: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6813: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6814: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6815: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6816: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6817: vlv= nbcode[Tvaraff[k]][lv];
6818: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6819: }
1.237 brouard 6820: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6821: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6822: }
1.211 brouard 6823: fprintf(ficgp,"\n#\n");
1.223 brouard 6824: if(invalidvarcomb[k1]){
1.227 brouard 6825: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6826: continue;
1.223 brouard 6827: }
1.227 brouard 6828:
1.241 brouard 6829: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6830: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6831: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6832: k=3; /* Offset */
1.153 brouard 6833: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6834: if(i==1)
6835: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6836: else
6837: fprintf(ficgp,", '' ");
6838: l=(nlstate+ndeath)*(i-1)+1;
6839: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6840: for (j=2; j<= nlstate ; j ++)
6841: fprintf(ficgp,"+$%d",k+l+j-1);
6842: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6843: } /* nlstate */
1.201 brouard 6844: fprintf(ficgp,"\nset out\n");
1.153 brouard 6845: } /* end cpt state*/
6846: } /* end covariate */
1.227 brouard 6847:
6848:
1.220 brouard 6849: /* 7eme */
1.218 brouard 6850: if(backcast == 1){
1.217 brouard 6851: /* CV back preval stable (period) for each covariate */
1.237 brouard 6852: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6853: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6854: if(TKresult[nres]!= k1)
6855: continue;
1.218 brouard 6856: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6857: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6858: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6859: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6860: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6861: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6862: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6863: vlv= nbcode[Tvaraff[k]][lv];
6864: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6865: }
1.237 brouard 6866: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6867: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6868: }
1.227 brouard 6869: fprintf(ficgp,"\n#\n");
6870: if(invalidvarcomb[k1]){
6871: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6872: continue;
6873: }
6874:
1.241 brouard 6875: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6876: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6877: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6878: k=3; /* Offset */
6879: for (i=1; i<= nlstate ; i ++){
6880: if(i==1)
6881: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6882: else
6883: fprintf(ficgp,", '' ");
6884: /* l=(nlstate+ndeath)*(i-1)+1; */
6885: l=(nlstate+ndeath)*(cpt-1)+1;
6886: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6887: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6888: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6889: /* for (j=2; j<= nlstate ; j ++) */
6890: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6891: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6892: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6893: } /* nlstate */
6894: fprintf(ficgp,"\nset out\n");
1.218 brouard 6895: } /* end cpt state*/
6896: } /* end covariate */
6897: } /* End if backcast */
6898:
1.223 brouard 6899: /* 8eme */
1.218 brouard 6900: if(prevfcast==1){
6901: /* Projection from cross-sectional to stable (period) for each covariate */
6902:
1.237 brouard 6903: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6904: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6905: if(TKresult[nres]!= k1)
6906: continue;
1.211 brouard 6907: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6908: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6909: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6910: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6911: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6912: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6913: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6914: vlv= nbcode[Tvaraff[k]][lv];
6915: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6916: }
1.237 brouard 6917: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6918: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6919: }
1.227 brouard 6920: fprintf(ficgp,"\n#\n");
6921: if(invalidvarcomb[k1]){
6922: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6923: continue;
6924: }
6925:
6926: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6927: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6928: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6929: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6930: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6931: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6932: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6933: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6934: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6935: if(i==1){
6936: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6937: }else{
6938: fprintf(ficgp,",\\\n '' ");
6939: }
6940: if(cptcoveff ==0){ /* No covariate */
6941: ioffset=2; /* Age is in 2 */
6942: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6943: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6944: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6945: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6946: fprintf(ficgp," u %d:(", ioffset);
6947: if(i==nlstate+1)
6948: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6949: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6950: else
6951: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6952: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6953: }else{ /* more than 2 covariates */
6954: if(cptcoveff ==1){
6955: ioffset=4; /* Age is in 4 */
6956: }else{
6957: ioffset=6; /* Age is in 6 */
6958: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6959: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6960: }
6961: fprintf(ficgp," u %d:(",ioffset);
6962: kl=0;
6963: strcpy(gplotcondition,"(");
6964: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6965: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6966: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6967: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6968: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6969: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6970: kl++;
6971: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6972: kl++;
6973: if(k <cptcoveff && cptcoveff>1)
6974: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6975: }
6976: strcpy(gplotcondition+strlen(gplotcondition),")");
6977: /* 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 *\/ */
6978: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6979: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6980: /* '' 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*/
6981: if(i==nlstate+1){
6982: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6983: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6984: }else{
6985: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6986: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6987: }
6988: } /* end if covariate */
6989: } /* nlstate */
6990: fprintf(ficgp,"\nset out\n");
1.223 brouard 6991: } /* end cpt state*/
6992: } /* end covariate */
6993: } /* End if prevfcast */
1.227 brouard 6994:
6995:
1.238 brouard 6996: /* 9eme writing MLE parameters */
6997: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6998: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6999: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7000: for(k=1; k <=(nlstate+ndeath); k++){
7001: if (k != i) {
1.227 brouard 7002: fprintf(ficgp,"# current state %d\n",k);
7003: for(j=1; j <=ncovmodel; j++){
7004: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7005: jk++;
7006: }
7007: fprintf(ficgp,"\n");
1.126 brouard 7008: }
7009: }
1.223 brouard 7010: }
1.187 brouard 7011: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7012:
1.145 brouard 7013: /*goto avoid;*/
1.238 brouard 7014: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7015: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7016: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7017: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7018: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7019: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7020: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7021: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7022: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7023: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7024: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7025: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7026: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7027: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7028: fprintf(ficgp,"#\n");
1.223 brouard 7029: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7030: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7031: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7032: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7033: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7034: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7035: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7036: if(TKresult[nres]!= jk)
7037: continue;
7038: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7039: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7040: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7041: }
7042: fprintf(ficgp,"\n#\n");
1.241 brouard 7043: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7044: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7045: if (ng==1){
7046: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7047: fprintf(ficgp,"\nunset log y");
7048: }else if (ng==2){
7049: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7050: fprintf(ficgp,"\nset log y");
7051: }else if (ng==3){
7052: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7053: fprintf(ficgp,"\nset log y");
7054: }else
7055: fprintf(ficgp,"\nunset title ");
7056: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7057: i=1;
7058: for(k2=1; k2<=nlstate; k2++) {
7059: k3=i;
7060: for(k=1; k<=(nlstate+ndeath); k++) {
7061: if (k != k2){
7062: switch( ng) {
7063: case 1:
7064: if(nagesqr==0)
7065: fprintf(ficgp," p%d+p%d*x",i,i+1);
7066: else /* nagesqr =1 */
7067: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7068: break;
7069: case 2: /* ng=2 */
7070: if(nagesqr==0)
7071: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7072: else /* nagesqr =1 */
7073: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7074: break;
7075: case 3:
7076: if(nagesqr==0)
7077: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7078: else /* nagesqr =1 */
7079: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7080: break;
7081: }
7082: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7083: ijp=1; /* product no age */
7084: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7085: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7086: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7087: if(j==Tage[ij]) { /* Product by age */
7088: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7089: if(DummyV[j]==0){
1.237 brouard 7090: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7091: }else{ /* quantitative */
7092: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7093: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7094: }
7095: ij++;
7096: }
7097: }else if(j==Tprod[ijp]) { /* */
7098: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7099: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7100: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7101: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7102: /* 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)]); */
7103: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7104: }else{ /* Vn is dummy and Vm is quanti */
7105: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7106: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7107: }
7108: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7109: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7110: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7111: }else{ /* Both quanti */
7112: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7113: }
7114: }
1.238 brouard 7115: ijp++;
1.237 brouard 7116: }
7117: } else{ /* simple covariate */
7118: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7119: if(Dummy[j]==0){
7120: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7121: }else{ /* quantitative */
7122: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7123: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7124: }
1.237 brouard 7125: } /* end simple */
7126: } /* end j */
1.223 brouard 7127: }else{
7128: i=i-ncovmodel;
7129: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7130: fprintf(ficgp," (1.");
7131: }
1.227 brouard 7132:
1.223 brouard 7133: if(ng != 1){
7134: fprintf(ficgp,")/(1");
1.227 brouard 7135:
1.223 brouard 7136: for(k1=1; k1 <=nlstate; k1++){
7137: if(nagesqr==0)
7138: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7139: else /* nagesqr =1 */
7140: 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 7141:
1.223 brouard 7142: ij=1;
7143: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7144: if((j-2)==Tage[ij]) { /* Bug valgrind */
7145: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7146: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7147: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7148: ij++;
7149: }
7150: }
7151: else
1.225 brouard 7152: 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 7153: }
7154: fprintf(ficgp,")");
7155: }
7156: fprintf(ficgp,")");
7157: if(ng ==2)
7158: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7159: else /* ng= 3 */
7160: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7161: }else{ /* end ng <> 1 */
7162: if( k !=k2) /* logit p11 is hard to draw */
7163: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7164: }
7165: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7166: fprintf(ficgp,",");
7167: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7168: fprintf(ficgp,",");
7169: i=i+ncovmodel;
7170: } /* end k */
7171: } /* end k2 */
7172: fprintf(ficgp,"\n set out\n");
7173: } /* end jk */
7174: } /* end ng */
7175: /* avoid: */
7176: fflush(ficgp);
1.126 brouard 7177: } /* end gnuplot */
7178:
7179:
7180: /*************** Moving average **************/
1.219 brouard 7181: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7182: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7183:
1.222 brouard 7184: int i, cpt, cptcod;
7185: int modcovmax =1;
7186: int mobilavrange, mob;
7187: int iage=0;
7188:
7189: double sum=0.;
7190: double age;
7191: double *sumnewp, *sumnewm;
7192: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7193:
7194:
1.225 brouard 7195: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7196: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7197:
7198: sumnewp = vector(1,ncovcombmax);
7199: sumnewm = vector(1,ncovcombmax);
7200: agemingood = vector(1,ncovcombmax);
7201: agemaxgood = vector(1,ncovcombmax);
7202:
7203: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7204: sumnewm[cptcod]=0.;
7205: sumnewp[cptcod]=0.;
7206: agemingood[cptcod]=0;
7207: agemaxgood[cptcod]=0;
7208: }
7209: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7210:
7211: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7212: if(mobilav==1) mobilavrange=5; /* default */
7213: else mobilavrange=mobilav;
7214: for (age=bage; age<=fage; age++)
7215: for (i=1; i<=nlstate;i++)
7216: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7217: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7218: /* We keep the original values on the extreme ages bage, fage and for
7219: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7220: we use a 5 terms etc. until the borders are no more concerned.
7221: */
7222: for (mob=3;mob <=mobilavrange;mob=mob+2){
7223: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7224: for (i=1; i<=nlstate;i++){
7225: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7226: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7227: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7228: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7229: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7230: }
7231: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7232: }
7233: }
7234: }/* end age */
7235: }/* end mob */
7236: }else
7237: return -1;
7238: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7239: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7240: if(invalidvarcomb[cptcod]){
7241: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7242: continue;
7243: }
1.219 brouard 7244:
1.222 brouard 7245: agemingood[cptcod]=fage-(mob-1)/2;
7246: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7247: sumnewm[cptcod]=0.;
7248: for (i=1; i<=nlstate;i++){
7249: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7250: }
7251: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7252: agemingood[cptcod]=age;
7253: }else{ /* bad */
7254: for (i=1; i<=nlstate;i++){
7255: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7256: } /* i */
7257: } /* end bad */
7258: }/* age */
7259: sum=0.;
7260: for (i=1; i<=nlstate;i++){
7261: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7262: }
7263: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7264: 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);
7265: /* for (i=1; i<=nlstate;i++){ */
7266: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7267: /* } /\* i *\/ */
7268: } /* end bad */
7269: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7270: /* From youngest, finding the oldest wrong */
7271: agemaxgood[cptcod]=bage+(mob-1)/2;
7272: for (age=bage+(mob-1)/2; age<=fage; age++){
7273: sumnewm[cptcod]=0.;
7274: for (i=1; i<=nlstate;i++){
7275: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7276: }
7277: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7278: agemaxgood[cptcod]=age;
7279: }else{ /* bad */
7280: for (i=1; i<=nlstate;i++){
7281: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7282: } /* i */
7283: } /* end bad */
7284: }/* age */
7285: sum=0.;
7286: for (i=1; i<=nlstate;i++){
7287: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7288: }
7289: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7290: 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);
7291: /* for (i=1; i<=nlstate;i++){ */
7292: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7293: /* } /\* i *\/ */
7294: } /* end bad */
7295:
7296: for (age=bage; age<=fage; age++){
1.235 brouard 7297: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7298: sumnewp[cptcod]=0.;
7299: sumnewm[cptcod]=0.;
7300: for (i=1; i<=nlstate;i++){
7301: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7302: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7303: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7304: }
7305: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7306: }
7307: /* printf("\n"); */
7308: /* } */
7309: /* brutal averaging */
7310: for (i=1; i<=nlstate;i++){
7311: for (age=1; age<=bage; age++){
7312: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7313: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7314: }
7315: for (age=fage; age<=AGESUP; age++){
7316: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7317: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7318: }
7319: } /* end i status */
7320: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7321: for (age=1; age<=AGESUP; age++){
7322: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7323: mobaverage[(int)age][i][cptcod]=0.;
7324: }
7325: }
7326: }/* end cptcod */
7327: free_vector(sumnewm,1, ncovcombmax);
7328: free_vector(sumnewp,1, ncovcombmax);
7329: free_vector(agemaxgood,1, ncovcombmax);
7330: free_vector(agemingood,1, ncovcombmax);
7331: return 0;
7332: }/* End movingaverage */
1.218 brouard 7333:
1.126 brouard 7334:
7335: /************** Forecasting ******************/
1.235 brouard 7336: 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 7337: /* proj1, year, month, day of starting projection
7338: agemin, agemax range of age
7339: dateprev1 dateprev2 range of dates during which prevalence is computed
7340: anproj2 year of en of projection (same day and month as proj1).
7341: */
1.235 brouard 7342: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7343: double agec; /* generic age */
7344: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7345: double *popeffectif,*popcount;
7346: double ***p3mat;
1.218 brouard 7347: /* double ***mobaverage; */
1.126 brouard 7348: char fileresf[FILENAMELENGTH];
7349:
7350: agelim=AGESUP;
1.211 brouard 7351: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7352: in each health status at the date of interview (if between dateprev1 and dateprev2).
7353: We still use firstpass and lastpass as another selection.
7354: */
1.214 brouard 7355: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7356: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7357:
1.201 brouard 7358: strcpy(fileresf,"F_");
7359: strcat(fileresf,fileresu);
1.126 brouard 7360: if((ficresf=fopen(fileresf,"w"))==NULL) {
7361: printf("Problem with forecast resultfile: %s\n", fileresf);
7362: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7363: }
1.235 brouard 7364: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7365: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7366:
1.225 brouard 7367: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7368:
7369:
7370: stepsize=(int) (stepm+YEARM-1)/YEARM;
7371: if (stepm<=12) stepsize=1;
7372: if(estepm < stepm){
7373: printf ("Problem %d lower than %d\n",estepm, stepm);
7374: }
7375: else hstepm=estepm;
7376:
7377: hstepm=hstepm/stepm;
7378: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7379: fractional in yp1 */
7380: anprojmean=yp;
7381: yp2=modf((yp1*12),&yp);
7382: mprojmean=yp;
7383: yp1=modf((yp2*30.5),&yp);
7384: jprojmean=yp;
7385: if(jprojmean==0) jprojmean=1;
7386: if(mprojmean==0) jprojmean=1;
7387:
1.227 brouard 7388: i1=pow(2,cptcoveff);
1.126 brouard 7389: if (cptcovn < 1){i1=1;}
7390:
7391: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7392:
7393: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7394:
1.126 brouard 7395: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7396: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7397: for(k=1; k<=i1;k++){
7398: if(TKresult[nres]!= k)
7399: continue;
1.227 brouard 7400: if(invalidvarcomb[k]){
7401: printf("\nCombination (%d) projection ignored because no cases \n",k);
7402: continue;
7403: }
7404: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7405: for(j=1;j<=cptcoveff;j++) {
7406: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7407: }
1.235 brouard 7408: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7409: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7410: }
1.227 brouard 7411: fprintf(ficresf," yearproj age");
7412: for(j=1; j<=nlstate+ndeath;j++){
7413: for(i=1; i<=nlstate;i++)
7414: fprintf(ficresf," p%d%d",i,j);
7415: fprintf(ficresf," wp.%d",j);
7416: }
7417: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7418: fprintf(ficresf,"\n");
7419: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7420: for (agec=fage; agec>=(ageminpar-1); agec--){
7421: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7422: nhstepm = nhstepm/hstepm;
7423: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7424: oldm=oldms;savm=savms;
1.235 brouard 7425: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7426:
7427: for (h=0; h<=nhstepm; h++){
7428: if (h*hstepm/YEARM*stepm ==yearp) {
7429: fprintf(ficresf,"\n");
7430: for(j=1;j<=cptcoveff;j++)
7431: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7432: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7433: }
7434: for(j=1; j<=nlstate+ndeath;j++) {
7435: ppij=0.;
7436: for(i=1; i<=nlstate;i++) {
7437: if (mobilav==1)
7438: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7439: else {
7440: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7441: }
7442: if (h*hstepm/YEARM*stepm== yearp) {
7443: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7444: }
7445: } /* end i */
7446: if (h*hstepm/YEARM*stepm==yearp) {
7447: fprintf(ficresf," %.3f", ppij);
7448: }
7449: }/* end j */
7450: } /* end h */
7451: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7452: } /* end agec */
7453: } /* end yearp */
7454: } /* end k */
1.219 brouard 7455:
1.126 brouard 7456: fclose(ficresf);
1.215 brouard 7457: printf("End of Computing forecasting \n");
7458: fprintf(ficlog,"End of Computing forecasting\n");
7459:
1.126 brouard 7460: }
7461:
1.218 brouard 7462: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7463: /* 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 7464: /* /\* back1, year, month, day of starting backection */
7465: /* agemin, agemax range of age */
7466: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7467: /* anback2 year of en of backection (same day and month as back1). */
7468: /* *\/ */
7469: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7470: /* double agec; /\* generic age *\/ */
7471: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7472: /* double *popeffectif,*popcount; */
7473: /* double ***p3mat; */
7474: /* /\* double ***mobaverage; *\/ */
7475: /* char fileresfb[FILENAMELENGTH]; */
7476:
7477: /* agelim=AGESUP; */
7478: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7479: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7480: /* We still use firstpass and lastpass as another selection. */
7481: /* *\/ */
7482: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7483: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7484: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7485:
7486: /* strcpy(fileresfb,"FB_"); */
7487: /* strcat(fileresfb,fileresu); */
7488: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7489: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7490: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7491: /* } */
7492: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7493: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7494:
1.225 brouard 7495: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7496:
7497: /* /\* if (mobilav!=0) { *\/ */
7498: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7499: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7500: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7501: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7502: /* /\* } *\/ */
7503: /* /\* } *\/ */
7504:
7505: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7506: /* if (stepm<=12) stepsize=1; */
7507: /* if(estepm < stepm){ */
7508: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7509: /* } */
7510: /* else hstepm=estepm; */
7511:
7512: /* hstepm=hstepm/stepm; */
7513: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7514: /* fractional in yp1 *\/ */
7515: /* anprojmean=yp; */
7516: /* yp2=modf((yp1*12),&yp); */
7517: /* mprojmean=yp; */
7518: /* yp1=modf((yp2*30.5),&yp); */
7519: /* jprojmean=yp; */
7520: /* if(jprojmean==0) jprojmean=1; */
7521: /* if(mprojmean==0) jprojmean=1; */
7522:
1.225 brouard 7523: /* i1=cptcoveff; */
1.218 brouard 7524: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7525:
1.218 brouard 7526: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7527:
1.218 brouard 7528: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7529:
7530: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7531: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7532: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7533: /* k=k+1; */
7534: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7535: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7536: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7537: /* } */
7538: /* fprintf(ficresfb," yearbproj age"); */
7539: /* for(j=1; j<=nlstate+ndeath;j++){ */
7540: /* for(i=1; i<=nlstate;i++) */
7541: /* fprintf(ficresfb," p%d%d",i,j); */
7542: /* fprintf(ficresfb," p.%d",j); */
7543: /* } */
7544: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7545: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7546: /* fprintf(ficresfb,"\n"); */
7547: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7548: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7549: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7550: /* nhstepm = nhstepm/hstepm; */
7551: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7552: /* oldm=oldms;savm=savms; */
7553: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7554: /* for (h=0; h<=nhstepm; h++){ */
7555: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7556: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7557: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7558: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7559: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7560: /* } */
7561: /* for(j=1; j<=nlstate+ndeath;j++) { */
7562: /* ppij=0.; */
7563: /* for(i=1; i<=nlstate;i++) { */
7564: /* if (mobilav==1) */
7565: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7566: /* else { */
7567: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7568: /* } */
7569: /* if (h*hstepm/YEARM*stepm== yearp) { */
7570: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7571: /* } */
7572: /* } /\* end i *\/ */
7573: /* if (h*hstepm/YEARM*stepm==yearp) { */
7574: /* fprintf(ficresfb," %.3f", ppij); */
7575: /* } */
7576: /* }/\* end j *\/ */
7577: /* } /\* end h *\/ */
7578: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7579: /* } /\* end agec *\/ */
7580: /* } /\* end yearp *\/ */
7581: /* } /\* end cptcod *\/ */
7582: /* } /\* end cptcov *\/ */
7583:
7584: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7585:
7586: /* fclose(ficresfb); */
7587: /* printf("End of Computing Back forecasting \n"); */
7588: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7589:
1.218 brouard 7590: /* } */
1.217 brouard 7591:
1.126 brouard 7592: /************** Forecasting *****not tested NB*************/
1.227 brouard 7593: /* 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 7594:
1.227 brouard 7595: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7596: /* int *popage; */
7597: /* double calagedatem, agelim, kk1, kk2; */
7598: /* double *popeffectif,*popcount; */
7599: /* double ***p3mat,***tabpop,***tabpopprev; */
7600: /* /\* double ***mobaverage; *\/ */
7601: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7602:
1.227 brouard 7603: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7604: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7605: /* agelim=AGESUP; */
7606: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7607:
1.227 brouard 7608: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7609:
7610:
1.227 brouard 7611: /* strcpy(filerespop,"POP_"); */
7612: /* strcat(filerespop,fileresu); */
7613: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7614: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7615: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7616: /* } */
7617: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7618: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7619:
1.227 brouard 7620: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7621:
1.227 brouard 7622: /* /\* if (mobilav!=0) { *\/ */
7623: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7624: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7625: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7626: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7627: /* /\* } *\/ */
7628: /* /\* } *\/ */
1.126 brouard 7629:
1.227 brouard 7630: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7631: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7632:
1.227 brouard 7633: /* agelim=AGESUP; */
1.126 brouard 7634:
1.227 brouard 7635: /* hstepm=1; */
7636: /* hstepm=hstepm/stepm; */
1.218 brouard 7637:
1.227 brouard 7638: /* if (popforecast==1) { */
7639: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7640: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7641: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7642: /* } */
7643: /* popage=ivector(0,AGESUP); */
7644: /* popeffectif=vector(0,AGESUP); */
7645: /* popcount=vector(0,AGESUP); */
1.126 brouard 7646:
1.227 brouard 7647: /* i=1; */
7648: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7649:
1.227 brouard 7650: /* imx=i; */
7651: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7652: /* } */
1.218 brouard 7653:
1.227 brouard 7654: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7655: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7656: /* k=k+1; */
7657: /* fprintf(ficrespop,"\n#******"); */
7658: /* for(j=1;j<=cptcoveff;j++) { */
7659: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7660: /* } */
7661: /* fprintf(ficrespop,"******\n"); */
7662: /* fprintf(ficrespop,"# Age"); */
7663: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7664: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7665:
1.227 brouard 7666: /* for (cpt=0; cpt<=0;cpt++) { */
7667: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7668:
1.227 brouard 7669: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7670: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7671: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7672:
1.227 brouard 7673: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7674: /* oldm=oldms;savm=savms; */
7675: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7676:
1.227 brouard 7677: /* for (h=0; h<=nhstepm; h++){ */
7678: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7679: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7680: /* } */
7681: /* for(j=1; j<=nlstate+ndeath;j++) { */
7682: /* kk1=0.;kk2=0; */
7683: /* for(i=1; i<=nlstate;i++) { */
7684: /* if (mobilav==1) */
7685: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7686: /* else { */
7687: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7688: /* } */
7689: /* } */
7690: /* if (h==(int)(calagedatem+12*cpt)){ */
7691: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7692: /* /\*fprintf(ficrespop," %.3f", kk1); */
7693: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7694: /* } */
7695: /* } */
7696: /* for(i=1; i<=nlstate;i++){ */
7697: /* kk1=0.; */
7698: /* for(j=1; j<=nlstate;j++){ */
7699: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7700: /* } */
7701: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7702: /* } */
1.218 brouard 7703:
1.227 brouard 7704: /* if (h==(int)(calagedatem+12*cpt)) */
7705: /* for(j=1; j<=nlstate;j++) */
7706: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7707: /* } */
7708: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7709: /* } */
7710: /* } */
1.218 brouard 7711:
1.227 brouard 7712: /* /\******\/ */
1.218 brouard 7713:
1.227 brouard 7714: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7715: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7716: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7717: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7718: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7719:
1.227 brouard 7720: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7721: /* oldm=oldms;savm=savms; */
7722: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7723: /* for (h=0; h<=nhstepm; h++){ */
7724: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7725: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7726: /* } */
7727: /* for(j=1; j<=nlstate+ndeath;j++) { */
7728: /* kk1=0.;kk2=0; */
7729: /* for(i=1; i<=nlstate;i++) { */
7730: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7731: /* } */
7732: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7733: /* } */
7734: /* } */
7735: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7736: /* } */
7737: /* } */
7738: /* } */
7739: /* } */
1.218 brouard 7740:
1.227 brouard 7741: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7742:
1.227 brouard 7743: /* if (popforecast==1) { */
7744: /* free_ivector(popage,0,AGESUP); */
7745: /* free_vector(popeffectif,0,AGESUP); */
7746: /* free_vector(popcount,0,AGESUP); */
7747: /* } */
7748: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7749: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7750: /* fclose(ficrespop); */
7751: /* } /\* End of popforecast *\/ */
1.218 brouard 7752:
1.126 brouard 7753: int fileappend(FILE *fichier, char *optionfich)
7754: {
7755: if((fichier=fopen(optionfich,"a"))==NULL) {
7756: printf("Problem with file: %s\n", optionfich);
7757: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7758: return (0);
7759: }
7760: fflush(fichier);
7761: return (1);
7762: }
7763:
7764:
7765: /**************** function prwizard **********************/
7766: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7767: {
7768:
7769: /* Wizard to print covariance matrix template */
7770:
1.164 brouard 7771: char ca[32], cb[32];
7772: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7773: int numlinepar;
7774:
7775: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7776: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7777: for(i=1; i <=nlstate; i++){
7778: jj=0;
7779: for(j=1; j <=nlstate+ndeath; j++){
7780: if(j==i) continue;
7781: jj++;
7782: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7783: printf("%1d%1d",i,j);
7784: fprintf(ficparo,"%1d%1d",i,j);
7785: for(k=1; k<=ncovmodel;k++){
7786: /* printf(" %lf",param[i][j][k]); */
7787: /* fprintf(ficparo," %lf",param[i][j][k]); */
7788: printf(" 0.");
7789: fprintf(ficparo," 0.");
7790: }
7791: printf("\n");
7792: fprintf(ficparo,"\n");
7793: }
7794: }
7795: printf("# Scales (for hessian or gradient estimation)\n");
7796: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7797: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7798: for(i=1; i <=nlstate; i++){
7799: jj=0;
7800: for(j=1; j <=nlstate+ndeath; j++){
7801: if(j==i) continue;
7802: jj++;
7803: fprintf(ficparo,"%1d%1d",i,j);
7804: printf("%1d%1d",i,j);
7805: fflush(stdout);
7806: for(k=1; k<=ncovmodel;k++){
7807: /* printf(" %le",delti3[i][j][k]); */
7808: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7809: printf(" 0.");
7810: fprintf(ficparo," 0.");
7811: }
7812: numlinepar++;
7813: printf("\n");
7814: fprintf(ficparo,"\n");
7815: }
7816: }
7817: printf("# Covariance matrix\n");
7818: /* # 121 Var(a12)\n\ */
7819: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7820: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7821: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7822: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7823: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7824: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7825: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7826: fflush(stdout);
7827: fprintf(ficparo,"# Covariance matrix\n");
7828: /* # 121 Var(a12)\n\ */
7829: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7830: /* # ...\n\ */
7831: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7832:
7833: for(itimes=1;itimes<=2;itimes++){
7834: jj=0;
7835: for(i=1; i <=nlstate; i++){
7836: for(j=1; j <=nlstate+ndeath; j++){
7837: if(j==i) continue;
7838: for(k=1; k<=ncovmodel;k++){
7839: jj++;
7840: ca[0]= k+'a'-1;ca[1]='\0';
7841: if(itimes==1){
7842: printf("#%1d%1d%d",i,j,k);
7843: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7844: }else{
7845: printf("%1d%1d%d",i,j,k);
7846: fprintf(ficparo,"%1d%1d%d",i,j,k);
7847: /* printf(" %.5le",matcov[i][j]); */
7848: }
7849: ll=0;
7850: for(li=1;li <=nlstate; li++){
7851: for(lj=1;lj <=nlstate+ndeath; lj++){
7852: if(lj==li) continue;
7853: for(lk=1;lk<=ncovmodel;lk++){
7854: ll++;
7855: if(ll<=jj){
7856: cb[0]= lk +'a'-1;cb[1]='\0';
7857: if(ll<jj){
7858: if(itimes==1){
7859: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7860: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7861: }else{
7862: printf(" 0.");
7863: fprintf(ficparo," 0.");
7864: }
7865: }else{
7866: if(itimes==1){
7867: printf(" Var(%s%1d%1d)",ca,i,j);
7868: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7869: }else{
7870: printf(" 0.");
7871: fprintf(ficparo," 0.");
7872: }
7873: }
7874: }
7875: } /* end lk */
7876: } /* end lj */
7877: } /* end li */
7878: printf("\n");
7879: fprintf(ficparo,"\n");
7880: numlinepar++;
7881: } /* end k*/
7882: } /*end j */
7883: } /* end i */
7884: } /* end itimes */
7885:
7886: } /* end of prwizard */
7887: /******************* Gompertz Likelihood ******************************/
7888: double gompertz(double x[])
7889: {
7890: double A,B,L=0.0,sump=0.,num=0.;
7891: int i,n=0; /* n is the size of the sample */
7892:
1.220 brouard 7893: for (i=1;i<=imx ; i++) {
1.126 brouard 7894: sump=sump+weight[i];
7895: /* sump=sump+1;*/
7896: num=num+1;
7897: }
7898:
7899:
7900: /* for (i=0; i<=imx; i++)
7901: 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]);*/
7902:
7903: for (i=1;i<=imx ; i++)
7904: {
7905: if (cens[i] == 1 && wav[i]>1)
7906: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7907:
7908: if (cens[i] == 0 && wav[i]>1)
7909: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7910: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7911:
7912: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7913: if (wav[i] > 1 ) { /* ??? */
7914: L=L+A*weight[i];
7915: /* 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]);*/
7916: }
7917: }
7918:
7919: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7920:
7921: return -2*L*num/sump;
7922: }
7923:
1.136 brouard 7924: #ifdef GSL
7925: /******************* Gompertz_f Likelihood ******************************/
7926: double gompertz_f(const gsl_vector *v, void *params)
7927: {
7928: double A,B,LL=0.0,sump=0.,num=0.;
7929: double *x= (double *) v->data;
7930: int i,n=0; /* n is the size of the sample */
7931:
7932: for (i=0;i<=imx-1 ; i++) {
7933: sump=sump+weight[i];
7934: /* sump=sump+1;*/
7935: num=num+1;
7936: }
7937:
7938:
7939: /* for (i=0; i<=imx; i++)
7940: 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]);*/
7941: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7942: for (i=1;i<=imx ; i++)
7943: {
7944: if (cens[i] == 1 && wav[i]>1)
7945: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7946:
7947: if (cens[i] == 0 && wav[i]>1)
7948: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7949: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7950:
7951: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7952: if (wav[i] > 1 ) { /* ??? */
7953: LL=LL+A*weight[i];
7954: /* 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]);*/
7955: }
7956: }
7957:
7958: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7959: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7960:
7961: return -2*LL*num/sump;
7962: }
7963: #endif
7964:
1.126 brouard 7965: /******************* Printing html file ***********/
1.201 brouard 7966: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7967: int lastpass, int stepm, int weightopt, char model[],\
7968: int imx, double p[],double **matcov,double agemortsup){
7969: int i,k;
7970:
7971: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7972: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7973: for (i=1;i<=2;i++)
7974: 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 7975: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7976: fprintf(fichtm,"</ul>");
7977:
7978: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7979:
7980: 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>");
7981:
7982: for (k=agegomp;k<(agemortsup-2);k++)
7983: 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]);
7984:
7985:
7986: fflush(fichtm);
7987: }
7988:
7989: /******************* Gnuplot file **************/
1.201 brouard 7990: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7991:
7992: char dirfileres[132],optfileres[132];
1.164 brouard 7993:
1.126 brouard 7994: int ng;
7995:
7996:
7997: /*#ifdef windows */
7998: fprintf(ficgp,"cd \"%s\" \n",pathc);
7999: /*#endif */
8000:
8001:
8002: strcpy(dirfileres,optionfilefiname);
8003: strcpy(optfileres,"vpl");
1.199 brouard 8004: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8005: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8006: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8007: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8008: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8009:
8010: }
8011:
1.136 brouard 8012: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8013: {
1.126 brouard 8014:
1.136 brouard 8015: /*-------- data file ----------*/
8016: FILE *fic;
8017: char dummy[]=" ";
1.240 brouard 8018: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8019: int lstra;
1.136 brouard 8020: int linei, month, year,iout;
8021: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8022: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8023: char *stratrunc;
1.223 brouard 8024:
1.240 brouard 8025: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8026: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8027:
1.240 brouard 8028: for(v=1; v <=ncovcol;v++){
8029: DummyV[v]=0;
8030: FixedV[v]=0;
8031: }
8032: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8033: DummyV[v]=1;
8034: FixedV[v]=0;
8035: }
8036: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8037: DummyV[v]=0;
8038: FixedV[v]=1;
8039: }
8040: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8041: DummyV[v]=1;
8042: FixedV[v]=1;
8043: }
8044: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8045: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8046: 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]);
8047: }
1.126 brouard 8048:
1.136 brouard 8049: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8050: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8051: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8052: }
1.126 brouard 8053:
1.136 brouard 8054: i=1;
8055: linei=0;
8056: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8057: linei=linei+1;
8058: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8059: if(line[j] == '\t')
8060: line[j] = ' ';
8061: }
8062: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8063: ;
8064: };
8065: line[j+1]=0; /* Trims blanks at end of line */
8066: if(line[0]=='#'){
8067: fprintf(ficlog,"Comment line\n%s\n",line);
8068: printf("Comment line\n%s\n",line);
8069: continue;
8070: }
8071: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8072: strcpy(line, linetmp);
1.223 brouard 8073:
8074: /* Loops on waves */
8075: for (j=maxwav;j>=1;j--){
8076: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8077: cutv(stra, strb, line, ' ');
8078: if(strb[0]=='.') { /* Missing value */
8079: lval=-1;
8080: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8081: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8082: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8083: 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);
8084: 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);
8085: return 1;
8086: }
8087: }else{
8088: errno=0;
8089: /* what_kind_of_number(strb); */
8090: dval=strtod(strb,&endptr);
8091: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8092: /* if(strb != endptr && *endptr == '\0') */
8093: /* dval=dlval; */
8094: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8095: if( strb[0]=='\0' || (*endptr != '\0')){
8096: 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);
8097: 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);
8098: return 1;
8099: }
8100: cotqvar[j][iv][i]=dval;
8101: cotvar[j][ntv+iv][i]=dval;
8102: }
8103: strcpy(line,stra);
1.223 brouard 8104: }/* end loop ntqv */
1.225 brouard 8105:
1.223 brouard 8106: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8107: cutv(stra, strb, line, ' ');
8108: if(strb[0]=='.') { /* Missing value */
8109: lval=-1;
8110: }else{
8111: errno=0;
8112: lval=strtol(strb,&endptr,10);
8113: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8114: if( strb[0]=='\0' || (*endptr != '\0')){
8115: 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);
8116: 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);
8117: return 1;
8118: }
8119: }
8120: if(lval <-1 || lval >1){
8121: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8122: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8123: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8124: For example, for multinomial values like 1, 2 and 3,\n \
8125: build V1=0 V2=0 for the reference value (1),\n \
8126: V1=1 V2=0 for (2) \n \
1.223 brouard 8127: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8128: output of IMaCh is often meaningless.\n \
1.223 brouard 8129: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8130: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8131: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8132: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8133: For example, for multinomial values like 1, 2 and 3,\n \
8134: build V1=0 V2=0 for the reference value (1),\n \
8135: V1=1 V2=0 for (2) \n \
1.223 brouard 8136: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8137: output of IMaCh is often meaningless.\n \
1.223 brouard 8138: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8139: return 1;
8140: }
8141: cotvar[j][iv][i]=(double)(lval);
8142: strcpy(line,stra);
1.223 brouard 8143: }/* end loop ntv */
1.225 brouard 8144:
1.223 brouard 8145: /* Statuses at wave */
1.137 brouard 8146: cutv(stra, strb, line, ' ');
1.223 brouard 8147: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8148: lval=-1;
1.136 brouard 8149: }else{
1.238 brouard 8150: errno=0;
8151: lval=strtol(strb,&endptr,10);
8152: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8153: if( strb[0]=='\0' || (*endptr != '\0')){
8154: 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);
8155: 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);
8156: return 1;
8157: }
1.136 brouard 8158: }
1.225 brouard 8159:
1.136 brouard 8160: s[j][i]=lval;
1.225 brouard 8161:
1.223 brouard 8162: /* Date of Interview */
1.136 brouard 8163: strcpy(line,stra);
8164: cutv(stra, strb,line,' ');
1.169 brouard 8165: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8166: }
1.169 brouard 8167: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8168: month=99;
8169: year=9999;
1.136 brouard 8170: }else{
1.225 brouard 8171: 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);
8172: 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);
8173: return 1;
1.136 brouard 8174: }
8175: anint[j][i]= (double) year;
8176: mint[j][i]= (double)month;
8177: strcpy(line,stra);
1.223 brouard 8178: } /* End loop on waves */
1.225 brouard 8179:
1.223 brouard 8180: /* Date of death */
1.136 brouard 8181: cutv(stra, strb,line,' ');
1.169 brouard 8182: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8183: }
1.169 brouard 8184: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8185: month=99;
8186: year=9999;
8187: }else{
1.141 brouard 8188: 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 8189: 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);
8190: return 1;
1.136 brouard 8191: }
8192: andc[i]=(double) year;
8193: moisdc[i]=(double) month;
8194: strcpy(line,stra);
8195:
1.223 brouard 8196: /* Date of birth */
1.136 brouard 8197: cutv(stra, strb,line,' ');
1.169 brouard 8198: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8199: }
1.169 brouard 8200: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8201: month=99;
8202: year=9999;
8203: }else{
1.141 brouard 8204: 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);
8205: 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 8206: return 1;
1.136 brouard 8207: }
8208: if (year==9999) {
1.141 brouard 8209: 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);
8210: 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 8211: return 1;
8212:
1.136 brouard 8213: }
8214: annais[i]=(double)(year);
8215: moisnais[i]=(double)(month);
8216: strcpy(line,stra);
1.225 brouard 8217:
1.223 brouard 8218: /* Sample weight */
1.136 brouard 8219: cutv(stra, strb,line,' ');
8220: errno=0;
8221: dval=strtod(strb,&endptr);
8222: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8223: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8224: 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 8225: fflush(ficlog);
8226: return 1;
8227: }
8228: weight[i]=dval;
8229: strcpy(line,stra);
1.225 brouard 8230:
1.223 brouard 8231: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8232: cutv(stra, strb, line, ' ');
8233: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8234: lval=-1;
1.223 brouard 8235: }else{
1.225 brouard 8236: errno=0;
8237: /* what_kind_of_number(strb); */
8238: dval=strtod(strb,&endptr);
8239: /* if(strb != endptr && *endptr == '\0') */
8240: /* dval=dlval; */
8241: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8242: if( strb[0]=='\0' || (*endptr != '\0')){
8243: 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);
8244: 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);
8245: return 1;
8246: }
8247: coqvar[iv][i]=dval;
1.226 brouard 8248: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8249: }
8250: strcpy(line,stra);
8251: }/* end loop nqv */
1.136 brouard 8252:
1.223 brouard 8253: /* Covariate values */
1.136 brouard 8254: for (j=ncovcol;j>=1;j--){
8255: cutv(stra, strb,line,' ');
1.223 brouard 8256: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8257: lval=-1;
1.136 brouard 8258: }else{
1.225 brouard 8259: errno=0;
8260: lval=strtol(strb,&endptr,10);
8261: if( strb[0]=='\0' || (*endptr != '\0')){
8262: 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);
8263: 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);
8264: return 1;
8265: }
1.136 brouard 8266: }
8267: if(lval <-1 || lval >1){
1.225 brouard 8268: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8269: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8270: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8271: For example, for multinomial values like 1, 2 and 3,\n \
8272: build V1=0 V2=0 for the reference value (1),\n \
8273: V1=1 V2=0 for (2) \n \
1.136 brouard 8274: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8275: output of IMaCh is often meaningless.\n \
1.136 brouard 8276: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8277: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8278: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8279: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8280: For example, for multinomial values like 1, 2 and 3,\n \
8281: build V1=0 V2=0 for the reference value (1),\n \
8282: V1=1 V2=0 for (2) \n \
1.136 brouard 8283: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8284: output of IMaCh is often meaningless.\n \
1.136 brouard 8285: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8286: return 1;
1.136 brouard 8287: }
8288: covar[j][i]=(double)(lval);
8289: strcpy(line,stra);
8290: }
8291: lstra=strlen(stra);
1.225 brouard 8292:
1.136 brouard 8293: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8294: stratrunc = &(stra[lstra-9]);
8295: num[i]=atol(stratrunc);
8296: }
8297: else
8298: num[i]=atol(stra);
8299: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8300: 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;}*/
8301:
8302: i=i+1;
8303: } /* End loop reading data */
1.225 brouard 8304:
1.136 brouard 8305: *imax=i-1; /* Number of individuals */
8306: fclose(fic);
1.225 brouard 8307:
1.136 brouard 8308: return (0);
1.164 brouard 8309: /* endread: */
1.225 brouard 8310: printf("Exiting readdata: ");
8311: fclose(fic);
8312: return (1);
1.223 brouard 8313: }
1.126 brouard 8314:
1.234 brouard 8315: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8316: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8317: while (*p2 == ' ')
1.234 brouard 8318: p2++;
8319: /* while ((*p1++ = *p2++) !=0) */
8320: /* ; */
8321: /* do */
8322: /* while (*p2 == ' ') */
8323: /* p2++; */
8324: /* while (*p1++ == *p2++); */
8325: *stri=p2;
1.145 brouard 8326: }
8327:
1.235 brouard 8328: int decoderesult ( char resultline[], int nres)
1.230 brouard 8329: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8330: {
1.235 brouard 8331: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8332: char resultsav[MAXLINE];
1.234 brouard 8333: int resultmodel[MAXLINE];
8334: int modelresult[MAXLINE];
1.230 brouard 8335: char stra[80], strb[80], strc[80], strd[80],stre[80];
8336:
1.234 brouard 8337: removefirstspace(&resultline);
1.233 brouard 8338: printf("decoderesult:%s\n",resultline);
1.230 brouard 8339:
8340: if (strstr(resultline,"v") !=0){
8341: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8342: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8343: return 1;
8344: }
8345: trimbb(resultsav, resultline);
8346: if (strlen(resultsav) >1){
8347: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8348: }
1.234 brouard 8349: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8350: 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);
8351: 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);
8352: }
8353: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8354: if(nbocc(resultsav,'=') >1){
8355: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8356: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8357: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8358: }else
8359: cutl(strc,strd,resultsav,'=');
1.230 brouard 8360: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8361:
1.230 brouard 8362: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8363: Tvarsel[k]=atoi(strc);
8364: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8365: /* cptcovsel++; */
8366: if (nbocc(stra,'=') >0)
8367: strcpy(resultsav,stra); /* and analyzes it */
8368: }
1.235 brouard 8369: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8370: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8371: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8372: match=0;
1.236 brouard 8373: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8374: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8375: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8376: match=1;
8377: break;
8378: }
8379: }
8380: if(match == 0){
8381: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8382: }
8383: }
8384: }
1.235 brouard 8385: /* Checking for missing or useless values in comparison of current model needs */
8386: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8387: match=0;
1.235 brouard 8388: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8389: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8390: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8391: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8392: ++match;
8393: }
8394: }
8395: }
8396: if(match == 0){
8397: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8398: }else if(match > 1){
8399: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8400: }
8401: }
1.235 brouard 8402:
1.234 brouard 8403: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8404: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8405: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8406: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8407: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8408: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8409: /* 1 0 0 0 */
8410: /* 2 1 0 0 */
8411: /* 3 0 1 0 */
8412: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8413: /* 5 0 0 1 */
8414: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8415: /* 7 0 1 1 */
8416: /* 8 1 1 1 */
1.237 brouard 8417: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8418: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8419: /* V5*age V5 known which value for nres? */
8420: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8421: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8422: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8423: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8424: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8425: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8426: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8427: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8428: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8429: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8430: k4++;;
8431: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8432: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8433: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8434: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8435: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8436: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8437: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8438: k4q++;;
8439: }
8440: }
1.234 brouard 8441:
1.235 brouard 8442: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8443: return (0);
8444: }
1.235 brouard 8445:
1.230 brouard 8446: int decodemodel( char model[], int lastobs)
8447: /**< This routine decodes the model and returns:
1.224 brouard 8448: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8449: * - nagesqr = 1 if age*age in the model, otherwise 0.
8450: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8451: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8452: * - cptcovage number of covariates with age*products =2
8453: * - cptcovs number of simple covariates
8454: * - 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
8455: * which is a new column after the 9 (ncovcol) variables.
8456: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8457: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8458: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8459: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8460: */
1.136 brouard 8461: {
1.238 brouard 8462: int i, j, k, ks, v;
1.227 brouard 8463: int j1, k1, k2, k3, k4;
1.136 brouard 8464: char modelsav[80];
1.145 brouard 8465: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8466: char *strpt;
1.136 brouard 8467:
1.145 brouard 8468: /*removespace(model);*/
1.136 brouard 8469: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8470: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8471: if (strstr(model,"AGE") !=0){
1.192 brouard 8472: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8473: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8474: return 1;
8475: }
1.141 brouard 8476: if (strstr(model,"v") !=0){
8477: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8478: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8479: return 1;
8480: }
1.187 brouard 8481: strcpy(modelsav,model);
8482: if ((strpt=strstr(model,"age*age")) !=0){
8483: printf(" strpt=%s, model=%s\n",strpt, model);
8484: if(strpt != model){
1.234 brouard 8485: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8486: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8487: corresponding column of parameters.\n",model);
1.234 brouard 8488: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8489: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8490: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8491: return 1;
1.225 brouard 8492: }
1.187 brouard 8493: nagesqr=1;
8494: if (strstr(model,"+age*age") !=0)
1.234 brouard 8495: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8496: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8497: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8498: else
1.234 brouard 8499: substrchaine(modelsav, model, "age*age");
1.187 brouard 8500: }else
8501: nagesqr=0;
8502: if (strlen(modelsav) >1){
8503: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8504: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8505: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8506: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8507: * cst, age and age*age
8508: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8509: /* including age products which are counted in cptcovage.
8510: * but the covariates which are products must be treated
8511: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8512: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8513: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8514:
8515:
1.187 brouard 8516: /* Design
8517: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8518: * < ncovcol=8 >
8519: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8520: * k= 1 2 3 4 5 6 7 8
8521: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8522: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8523: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8524: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8525: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8526: * Tage[++cptcovage]=k
8527: * if products, new covar are created after ncovcol with k1
8528: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8529: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8530: * 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
8531: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8532: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8533: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8534: * < ncovcol=8 >
8535: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8536: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8537: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8538: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8539: * p Tprod[1]@2={ 6, 5}
8540: *p Tvard[1][1]@4= {7, 8, 5, 6}
8541: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8542: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8543: *How to reorganize?
8544: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8545: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8546: * {2, 1, 4, 8, 5, 6, 3, 7}
8547: * Struct []
8548: */
1.225 brouard 8549:
1.187 brouard 8550: /* This loop fills the array Tvar from the string 'model'.*/
8551: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8552: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8553: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8554: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8555: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8556: /* k=1 Tvar[1]=2 (from V2) */
8557: /* k=5 Tvar[5] */
8558: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8559: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8560: /* } */
1.198 brouard 8561: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8562: /*
8563: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8564: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8565: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8566: }
1.187 brouard 8567: cptcovage=0;
8568: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8569: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8570: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8571: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8572: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8573: /*scanf("%d",i);*/
8574: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8575: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8576: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8577: /* covar is not filled and then is empty */
8578: cptcovprod--;
8579: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8580: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8581: Typevar[k]=1; /* 1 for age product */
8582: cptcovage++; /* Sums the number of covariates which include age as a product */
8583: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8584: /*printf("stre=%s ", stre);*/
8585: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8586: cptcovprod--;
8587: cutl(stre,strb,strc,'V');
8588: Tvar[k]=atoi(stre);
8589: Typevar[k]=1; /* 1 for age product */
8590: cptcovage++;
8591: Tage[cptcovage]=k;
8592: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8593: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8594: cptcovn++;
8595: cptcovprodnoage++;k1++;
8596: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8597: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8598: because this model-covariate is a construction we invent a new column
8599: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8600: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8601: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8602: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8603: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8604: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8605: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8606: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8607: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8608: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8609: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8610: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8611: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8612: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8613: for (i=1; i<=lastobs;i++){
8614: /* Computes the new covariate which is a product of
8615: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8616: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8617: }
8618: } /* End age is not in the model */
8619: } /* End if model includes a product */
8620: else { /* no more sum */
8621: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8622: /* scanf("%d",i);*/
8623: cutl(strd,strc,strb,'V');
8624: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8625: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8626: Tvar[k]=atoi(strd);
8627: Typevar[k]=0; /* 0 for simple covariates */
8628: }
8629: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8630: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8631: scanf("%d",i);*/
1.187 brouard 8632: } /* end of loop + on total covariates */
8633: } /* end if strlen(modelsave == 0) age*age might exist */
8634: } /* end if strlen(model == 0) */
1.136 brouard 8635:
8636: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8637: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8638:
1.136 brouard 8639: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8640: printf("cptcovprod=%d ", cptcovprod);
8641: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8642: scanf("%d ",i);*/
8643:
8644:
1.230 brouard 8645: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8646: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8647: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8648: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8649: k = 1 2 3 4 5 6 7 8 9
8650: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8651: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8652: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8653: Dummy[k] 1 0 0 0 3 1 1 2 3
8654: Tmodelind[combination of covar]=k;
1.225 brouard 8655: */
8656: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8657: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8658: /* 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 8659: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8660: printf("Model=%s\n\
8661: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8662: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8663: 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);
8664: fprintf(ficlog,"Model=%s\n\
8665: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8666: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8667: 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 8668: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8669: 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 */
8670: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8671: Fixed[k]= 0;
8672: Dummy[k]= 0;
1.225 brouard 8673: ncoveff++;
1.232 brouard 8674: ncovf++;
1.234 brouard 8675: nsd++;
8676: modell[k].maintype= FTYPE;
8677: TvarsD[nsd]=Tvar[k];
8678: TvarsDind[nsd]=k;
8679: TvarF[ncovf]=Tvar[k];
8680: TvarFind[ncovf]=k;
8681: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8682: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8683: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8684: Fixed[k]= 0;
8685: Dummy[k]= 0;
8686: ncoveff++;
8687: ncovf++;
8688: modell[k].maintype= FTYPE;
8689: TvarF[ncovf]=Tvar[k];
8690: TvarFind[ncovf]=k;
1.230 brouard 8691: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8692: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8693: }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 8694: Fixed[k]= 0;
8695: Dummy[k]= 1;
1.230 brouard 8696: nqfveff++;
1.234 brouard 8697: modell[k].maintype= FTYPE;
8698: modell[k].subtype= FQ;
8699: nsq++;
8700: TvarsQ[nsq]=Tvar[k];
8701: TvarsQind[nsq]=k;
1.232 brouard 8702: ncovf++;
1.234 brouard 8703: TvarF[ncovf]=Tvar[k];
8704: TvarFind[ncovf]=k;
1.231 brouard 8705: 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 8706: 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 8707: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8708: Fixed[k]= 1;
8709: Dummy[k]= 0;
1.225 brouard 8710: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8711: modell[k].maintype= VTYPE;
8712: modell[k].subtype= VD;
8713: nsd++;
8714: TvarsD[nsd]=Tvar[k];
8715: TvarsDind[nsd]=k;
8716: ncovv++; /* Only simple time varying variables */
8717: TvarV[ncovv]=Tvar[k];
1.242 brouard 8718: 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 8719: 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 */
8720: 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 8721: 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);
8722: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8723: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8724: Fixed[k]= 1;
8725: Dummy[k]= 1;
8726: nqtveff++;
8727: modell[k].maintype= VTYPE;
8728: modell[k].subtype= VQ;
8729: ncovv++; /* Only simple time varying variables */
8730: nsq++;
8731: TvarsQ[nsq]=Tvar[k];
8732: TvarsQind[nsq]=k;
8733: TvarV[ncovv]=Tvar[k];
1.242 brouard 8734: 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 8735: 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 */
8736: 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 8737: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8738: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8739: 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 8740: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8741: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8742: ncova++;
8743: TvarA[ncova]=Tvar[k];
8744: TvarAind[ncova]=k;
1.231 brouard 8745: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8746: Fixed[k]= 2;
8747: Dummy[k]= 2;
8748: modell[k].maintype= ATYPE;
8749: modell[k].subtype= APFD;
8750: /* ncoveff++; */
1.227 brouard 8751: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8752: Fixed[k]= 2;
8753: Dummy[k]= 3;
8754: modell[k].maintype= ATYPE;
8755: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8756: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8757: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8758: Fixed[k]= 3;
8759: Dummy[k]= 2;
8760: modell[k].maintype= ATYPE;
8761: modell[k].subtype= APVD; /* Product age * varying dummy */
8762: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8763: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8764: Fixed[k]= 3;
8765: Dummy[k]= 3;
8766: modell[k].maintype= ATYPE;
8767: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8768: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8769: }
8770: }else if (Typevar[k] == 2) { /* product without age */
8771: k1=Tposprod[k];
8772: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8773: if(Tvard[k1][2] <=ncovcol){
8774: Fixed[k]= 1;
8775: Dummy[k]= 0;
8776: modell[k].maintype= FTYPE;
8777: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8778: ncovf++; /* Fixed variables without age */
8779: TvarF[ncovf]=Tvar[k];
8780: TvarFind[ncovf]=k;
8781: }else if(Tvard[k1][2] <=ncovcol+nqv){
8782: Fixed[k]= 0; /* or 2 ?*/
8783: Dummy[k]= 1;
8784: modell[k].maintype= FTYPE;
8785: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8786: ncovf++; /* Varying variables without age */
8787: TvarF[ncovf]=Tvar[k];
8788: TvarFind[ncovf]=k;
8789: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8790: Fixed[k]= 1;
8791: Dummy[k]= 0;
8792: modell[k].maintype= VTYPE;
8793: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8794: ncovv++; /* Varying variables without age */
8795: TvarV[ncovv]=Tvar[k];
8796: TvarVind[ncovv]=k;
8797: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8798: Fixed[k]= 1;
8799: Dummy[k]= 1;
8800: modell[k].maintype= VTYPE;
8801: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8802: ncovv++; /* Varying variables without age */
8803: TvarV[ncovv]=Tvar[k];
8804: TvarVind[ncovv]=k;
8805: }
1.227 brouard 8806: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8807: if(Tvard[k1][2] <=ncovcol){
8808: Fixed[k]= 0; /* or 2 ?*/
8809: Dummy[k]= 1;
8810: modell[k].maintype= FTYPE;
8811: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8812: ncovf++; /* Fixed variables without age */
8813: TvarF[ncovf]=Tvar[k];
8814: TvarFind[ncovf]=k;
8815: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8816: Fixed[k]= 1;
8817: Dummy[k]= 1;
8818: modell[k].maintype= VTYPE;
8819: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8820: ncovv++; /* Varying variables without age */
8821: TvarV[ncovv]=Tvar[k];
8822: TvarVind[ncovv]=k;
8823: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8824: Fixed[k]= 1;
8825: Dummy[k]= 1;
8826: modell[k].maintype= VTYPE;
8827: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8828: ncovv++; /* Varying variables without age */
8829: TvarV[ncovv]=Tvar[k];
8830: TvarVind[ncovv]=k;
8831: ncovv++; /* Varying variables without age */
8832: TvarV[ncovv]=Tvar[k];
8833: TvarVind[ncovv]=k;
8834: }
1.227 brouard 8835: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8836: if(Tvard[k1][2] <=ncovcol){
8837: Fixed[k]= 1;
8838: Dummy[k]= 1;
8839: modell[k].maintype= VTYPE;
8840: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8841: ncovv++; /* Varying variables without age */
8842: TvarV[ncovv]=Tvar[k];
8843: TvarVind[ncovv]=k;
8844: }else if(Tvard[k1][2] <=ncovcol+nqv){
8845: Fixed[k]= 1;
8846: Dummy[k]= 1;
8847: modell[k].maintype= VTYPE;
8848: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8849: ncovv++; /* Varying variables without age */
8850: TvarV[ncovv]=Tvar[k];
8851: TvarVind[ncovv]=k;
8852: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8853: Fixed[k]= 1;
8854: Dummy[k]= 0;
8855: modell[k].maintype= VTYPE;
8856: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8857: ncovv++; /* Varying variables without age */
8858: TvarV[ncovv]=Tvar[k];
8859: TvarVind[ncovv]=k;
8860: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8861: Fixed[k]= 1;
8862: Dummy[k]= 1;
8863: modell[k].maintype= VTYPE;
8864: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8865: ncovv++; /* Varying variables without age */
8866: TvarV[ncovv]=Tvar[k];
8867: TvarVind[ncovv]=k;
8868: }
1.227 brouard 8869: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8870: if(Tvard[k1][2] <=ncovcol){
8871: Fixed[k]= 1;
8872: Dummy[k]= 1;
8873: modell[k].maintype= VTYPE;
8874: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8875: ncovv++; /* Varying variables without age */
8876: TvarV[ncovv]=Tvar[k];
8877: TvarVind[ncovv]=k;
8878: }else if(Tvard[k1][2] <=ncovcol+nqv){
8879: Fixed[k]= 1;
8880: Dummy[k]= 1;
8881: modell[k].maintype= VTYPE;
8882: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8883: ncovv++; /* Varying variables without age */
8884: TvarV[ncovv]=Tvar[k];
8885: TvarVind[ncovv]=k;
8886: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8887: Fixed[k]= 1;
8888: Dummy[k]= 1;
8889: modell[k].maintype= VTYPE;
8890: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8891: ncovv++; /* Varying variables without age */
8892: TvarV[ncovv]=Tvar[k];
8893: TvarVind[ncovv]=k;
8894: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8895: Fixed[k]= 1;
8896: Dummy[k]= 1;
8897: modell[k].maintype= VTYPE;
8898: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8899: ncovv++; /* Varying variables without age */
8900: TvarV[ncovv]=Tvar[k];
8901: TvarVind[ncovv]=k;
8902: }
1.227 brouard 8903: }else{
1.240 brouard 8904: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8905: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8906: } /*end k1*/
1.225 brouard 8907: }else{
1.226 brouard 8908: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8909: 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 8910: }
1.227 brouard 8911: 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 8912: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8913: 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]);
8914: }
8915: /* Searching for doublons in the model */
8916: for(k1=1; k1<= cptcovt;k1++){
8917: for(k2=1; k2 <k1;k2++){
8918: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8919: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8920: if(Tvar[k1]==Tvar[k2]){
8921: 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]]);
8922: 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);
8923: return(1);
8924: }
8925: }else if (Typevar[k1] ==2){
8926: k3=Tposprod[k1];
8927: k4=Tposprod[k2];
8928: 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])) ){
8929: 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]]);
8930: 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);
8931: return(1);
8932: }
8933: }
1.227 brouard 8934: }
8935: }
1.225 brouard 8936: }
8937: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8938: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8939: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8940: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8941: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8942: /*endread:*/
1.225 brouard 8943: printf("Exiting decodemodel: ");
8944: return (1);
1.136 brouard 8945: }
8946:
1.169 brouard 8947: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 8948: {/* Check ages at death */
1.136 brouard 8949: int i, m;
1.218 brouard 8950: int firstone=0;
8951:
1.136 brouard 8952: for (i=1; i<=imx; i++) {
8953: for(m=2; (m<= maxwav); m++) {
8954: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8955: anint[m][i]=9999;
1.216 brouard 8956: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8957: s[m][i]=-1;
1.136 brouard 8958: }
8959: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8960: *nberr = *nberr + 1;
1.218 brouard 8961: if(firstone == 0){
8962: firstone=1;
8963: 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);
8964: }
8965: 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 8966: s[m][i]=-1;
8967: }
8968: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8969: (*nberr)++;
1.136 brouard 8970: 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]);
8971: 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]);
8972: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8973: }
8974: }
8975: }
8976:
8977: for (i=1; i<=imx; i++) {
8978: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8979: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8980: 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 8981: if (s[m][i] >= nlstate+1) {
1.169 brouard 8982: if(agedc[i]>0){
8983: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8984: agev[m][i]=agedc[i];
1.214 brouard 8985: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8986: }else {
1.136 brouard 8987: if ((int)andc[i]!=9999){
8988: nbwarn++;
8989: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8990: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8991: agev[m][i]=-1;
8992: }
8993: }
1.169 brouard 8994: } /* agedc > 0 */
1.214 brouard 8995: } /* end if */
1.136 brouard 8996: else if(s[m][i] !=9){ /* Standard case, age in fractional
8997: years but with the precision of a month */
8998: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8999: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9000: agev[m][i]=1;
9001: else if(agev[m][i] < *agemin){
9002: *agemin=agev[m][i];
9003: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9004: }
9005: else if(agev[m][i] >*agemax){
9006: *agemax=agev[m][i];
1.156 brouard 9007: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9008: }
9009: /*agev[m][i]=anint[m][i]-annais[i];*/
9010: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9011: } /* en if 9*/
1.136 brouard 9012: else { /* =9 */
1.214 brouard 9013: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9014: agev[m][i]=1;
9015: s[m][i]=-1;
9016: }
9017: }
1.214 brouard 9018: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9019: agev[m][i]=1;
1.214 brouard 9020: else{
9021: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9022: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9023: agev[m][i]=0;
9024: }
9025: } /* End for lastpass */
9026: }
1.136 brouard 9027:
9028: for (i=1; i<=imx; i++) {
9029: for(m=firstpass; (m<=lastpass); m++){
9030: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9031: (*nberr)++;
1.136 brouard 9032: 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);
9033: 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);
9034: return 1;
9035: }
9036: }
9037: }
9038:
9039: /*for (i=1; i<=imx; i++){
9040: for (m=firstpass; (m<lastpass); m++){
9041: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9042: }
9043:
9044: }*/
9045:
9046:
1.139 brouard 9047: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9048: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9049:
9050: return (0);
1.164 brouard 9051: /* endread:*/
1.136 brouard 9052: printf("Exiting calandcheckages: ");
9053: return (1);
9054: }
9055:
1.172 brouard 9056: #if defined(_MSC_VER)
9057: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9058: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9059: //#include "stdafx.h"
9060: //#include <stdio.h>
9061: //#include <tchar.h>
9062: //#include <windows.h>
9063: //#include <iostream>
9064: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9065:
9066: LPFN_ISWOW64PROCESS fnIsWow64Process;
9067:
9068: BOOL IsWow64()
9069: {
9070: BOOL bIsWow64 = FALSE;
9071:
9072: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9073: // (HANDLE, PBOOL);
9074:
9075: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9076:
9077: HMODULE module = GetModuleHandle(_T("kernel32"));
9078: const char funcName[] = "IsWow64Process";
9079: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9080: GetProcAddress(module, funcName);
9081:
9082: if (NULL != fnIsWow64Process)
9083: {
9084: if (!fnIsWow64Process(GetCurrentProcess(),
9085: &bIsWow64))
9086: //throw std::exception("Unknown error");
9087: printf("Unknown error\n");
9088: }
9089: return bIsWow64 != FALSE;
9090: }
9091: #endif
1.177 brouard 9092:
1.191 brouard 9093: void syscompilerinfo(int logged)
1.167 brouard 9094: {
9095: /* #include "syscompilerinfo.h"*/
1.185 brouard 9096: /* command line Intel compiler 32bit windows, XP compatible:*/
9097: /* /GS /W3 /Gy
9098: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9099: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9100: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9101: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9102: */
9103: /* 64 bits */
1.185 brouard 9104: /*
9105: /GS /W3 /Gy
9106: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9107: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9108: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9109: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9110: /* Optimization are useless and O3 is slower than O2 */
9111: /*
9112: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9113: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9114: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9115: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9116: */
1.186 brouard 9117: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9118: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9119: /PDB:"visual studio
9120: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9121: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9122: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9123: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9124: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9125: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9126: uiAccess='false'"
9127: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9128: /NOLOGO /TLBID:1
9129: */
1.177 brouard 9130: #if defined __INTEL_COMPILER
1.178 brouard 9131: #if defined(__GNUC__)
9132: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9133: #endif
1.177 brouard 9134: #elif defined(__GNUC__)
1.179 brouard 9135: #ifndef __APPLE__
1.174 brouard 9136: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9137: #endif
1.177 brouard 9138: struct utsname sysInfo;
1.178 brouard 9139: int cross = CROSS;
9140: if (cross){
9141: printf("Cross-");
1.191 brouard 9142: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9143: }
1.174 brouard 9144: #endif
9145:
1.171 brouard 9146: #include <stdint.h>
1.178 brouard 9147:
1.191 brouard 9148: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9149: #if defined(__clang__)
1.191 brouard 9150: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9151: #endif
9152: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9153: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9154: #endif
9155: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9156: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9157: #endif
9158: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9159: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9160: #endif
9161: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9162: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9163: #endif
9164: #if defined(_MSC_VER)
1.191 brouard 9165: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9166: #endif
9167: #if defined(__PGI)
1.191 brouard 9168: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9169: #endif
9170: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9171: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9172: #endif
1.191 brouard 9173: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9174:
1.167 brouard 9175: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9176: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9177: // Windows (x64 and x86)
1.191 brouard 9178: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9179: #elif __unix__ // all unices, not all compilers
9180: // Unix
1.191 brouard 9181: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9182: #elif __linux__
9183: // linux
1.191 brouard 9184: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9185: #elif __APPLE__
1.174 brouard 9186: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9187: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9188: #endif
9189:
9190: /* __MINGW32__ */
9191: /* __CYGWIN__ */
9192: /* __MINGW64__ */
9193: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9194: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9195: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9196: /* _WIN64 // Defined for applications for Win64. */
9197: /* _M_X64 // Defined for compilations that target x64 processors. */
9198: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9199:
1.167 brouard 9200: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9201: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9202: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9203: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9204: #else
1.191 brouard 9205: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9206: #endif
9207:
1.169 brouard 9208: #if defined(__GNUC__)
9209: # if defined(__GNUC_PATCHLEVEL__)
9210: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9211: + __GNUC_MINOR__ * 100 \
9212: + __GNUC_PATCHLEVEL__)
9213: # else
9214: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9215: + __GNUC_MINOR__ * 100)
9216: # endif
1.174 brouard 9217: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9218: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9219:
9220: if (uname(&sysInfo) != -1) {
9221: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9222: 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 9223: }
9224: else
9225: perror("uname() error");
1.179 brouard 9226: //#ifndef __INTEL_COMPILER
9227: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9228: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9229: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9230: #endif
1.169 brouard 9231: #endif
1.172 brouard 9232:
9233: // void main()
9234: // {
1.169 brouard 9235: #if defined(_MSC_VER)
1.174 brouard 9236: if (IsWow64()){
1.191 brouard 9237: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9238: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9239: }
9240: else{
1.191 brouard 9241: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9242: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9243: }
1.172 brouard 9244: // printf("\nPress Enter to continue...");
9245: // getchar();
9246: // }
9247:
1.169 brouard 9248: #endif
9249:
1.167 brouard 9250:
1.219 brouard 9251: }
1.136 brouard 9252:
1.219 brouard 9253: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9254: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9255: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9256: /* double ftolpl = 1.e-10; */
1.180 brouard 9257: double age, agebase, agelim;
1.203 brouard 9258: double tot;
1.180 brouard 9259:
1.202 brouard 9260: strcpy(filerespl,"PL_");
9261: strcat(filerespl,fileresu);
9262: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9263: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9264: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9265: }
1.227 brouard 9266: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9267: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9268: pstamp(ficrespl);
1.203 brouard 9269: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9270: fprintf(ficrespl,"#Age ");
9271: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9272: fprintf(ficrespl,"\n");
1.180 brouard 9273:
1.219 brouard 9274: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9275:
1.219 brouard 9276: agebase=ageminpar;
9277: agelim=agemaxpar;
1.180 brouard 9278:
1.227 brouard 9279: /* i1=pow(2,ncoveff); */
1.234 brouard 9280: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9281: if (cptcovn < 1){i1=1;}
1.180 brouard 9282:
1.238 brouard 9283: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9284: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9285: if(TKresult[nres]!= k)
9286: continue;
1.235 brouard 9287:
1.238 brouard 9288: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9289: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9290: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9291: /* k=k+1; */
9292: /* to clean */
9293: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9294: fprintf(ficrespl,"#******");
9295: printf("#******");
9296: fprintf(ficlog,"#******");
9297: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9298: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9299: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9300: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9301: }
9302: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9303: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9304: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9305: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9306: }
9307: fprintf(ficrespl,"******\n");
9308: printf("******\n");
9309: fprintf(ficlog,"******\n");
9310: if(invalidvarcomb[k]){
9311: printf("\nCombination (%d) ignored because no case \n",k);
9312: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9313: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9314: continue;
9315: }
1.219 brouard 9316:
1.238 brouard 9317: fprintf(ficrespl,"#Age ");
9318: for(j=1;j<=cptcoveff;j++) {
9319: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9320: }
9321: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9322: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9323:
1.238 brouard 9324: for (age=agebase; age<=agelim; age++){
9325: /* for (age=agebase; age<=agebase; age++){ */
9326: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9327: fprintf(ficrespl,"%.0f ",age );
9328: for(j=1;j<=cptcoveff;j++)
9329: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9330: tot=0.;
9331: for(i=1; i<=nlstate;i++){
9332: tot += prlim[i][i];
9333: fprintf(ficrespl," %.5f", prlim[i][i]);
9334: }
9335: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9336: } /* Age */
9337: /* was end of cptcod */
9338: } /* cptcov */
9339: } /* nres */
1.219 brouard 9340: return 0;
1.180 brouard 9341: }
9342:
1.218 brouard 9343: 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){
9344: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9345:
9346: /* Computes the back prevalence limit for any combination of covariate values
9347: * at any age between ageminpar and agemaxpar
9348: */
1.235 brouard 9349: int i, j, k, i1, nres=0 ;
1.217 brouard 9350: /* double ftolpl = 1.e-10; */
9351: double age, agebase, agelim;
9352: double tot;
1.218 brouard 9353: /* double ***mobaverage; */
9354: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9355:
9356: strcpy(fileresplb,"PLB_");
9357: strcat(fileresplb,fileresu);
9358: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9359: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9360: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9361: }
9362: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9363: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9364: pstamp(ficresplb);
9365: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9366: fprintf(ficresplb,"#Age ");
9367: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9368: fprintf(ficresplb,"\n");
9369:
1.218 brouard 9370:
9371: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9372:
9373: agebase=ageminpar;
9374: agelim=agemaxpar;
9375:
9376:
1.227 brouard 9377: i1=pow(2,cptcoveff);
1.218 brouard 9378: if (cptcovn < 1){i1=1;}
1.227 brouard 9379:
1.238 brouard 9380: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9381: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9382: if(TKresult[nres]!= k)
9383: continue;
9384: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9385: fprintf(ficresplb,"#******");
9386: printf("#******");
9387: fprintf(ficlog,"#******");
9388: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9389: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9390: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9391: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9392: }
9393: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9394: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9395: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9396: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9397: }
9398: fprintf(ficresplb,"******\n");
9399: printf("******\n");
9400: fprintf(ficlog,"******\n");
9401: if(invalidvarcomb[k]){
9402: printf("\nCombination (%d) ignored because no cases \n",k);
9403: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9404: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9405: continue;
9406: }
1.218 brouard 9407:
1.238 brouard 9408: fprintf(ficresplb,"#Age ");
9409: for(j=1;j<=cptcoveff;j++) {
9410: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9411: }
9412: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9413: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9414:
9415:
1.238 brouard 9416: for (age=agebase; age<=agelim; age++){
9417: /* for (age=agebase; age<=agebase; age++){ */
9418: if(mobilavproj > 0){
9419: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9420: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9421: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9422: }else if (mobilavproj == 0){
9423: 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);
9424: 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);
9425: exit(1);
9426: }else{
9427: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9428: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9429: }
9430: fprintf(ficresplb,"%.0f ",age );
9431: for(j=1;j<=cptcoveff;j++)
9432: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9433: tot=0.;
9434: for(i=1; i<=nlstate;i++){
9435: tot += bprlim[i][i];
9436: fprintf(ficresplb," %.5f", bprlim[i][i]);
9437: }
9438: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9439: } /* Age */
9440: /* was end of cptcod */
9441: } /* end of any combination */
9442: } /* end of nres */
1.218 brouard 9443: /* hBijx(p, bage, fage); */
9444: /* fclose(ficrespijb); */
9445:
9446: return 0;
1.217 brouard 9447: }
1.218 brouard 9448:
1.180 brouard 9449: int hPijx(double *p, int bage, int fage){
9450: /*------------- h Pij x at various ages ------------*/
9451:
9452: int stepsize;
9453: int agelim;
9454: int hstepm;
9455: int nhstepm;
1.235 brouard 9456: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9457:
9458: double agedeb;
9459: double ***p3mat;
9460:
1.201 brouard 9461: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9462: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9463: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9464: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9465: }
9466: printf("Computing pij: result on file '%s' \n", filerespij);
9467: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9468:
9469: stepsize=(int) (stepm+YEARM-1)/YEARM;
9470: /*if (stepm<=24) stepsize=2;*/
9471:
9472: agelim=AGESUP;
9473: hstepm=stepsize*YEARM; /* Every year of age */
9474: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9475:
1.180 brouard 9476: /* hstepm=1; aff par mois*/
9477: pstamp(ficrespij);
9478: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9479: i1= pow(2,cptcoveff);
1.218 brouard 9480: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9481: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9482: /* k=k+1; */
1.235 brouard 9483: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9484: for(k=1; k<=i1;k++){
9485: if(TKresult[nres]!= k)
9486: continue;
1.183 brouard 9487: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9488: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9489: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9490: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9491: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9492: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9493: }
1.183 brouard 9494: fprintf(ficrespij,"******\n");
9495:
9496: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9497: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9498: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9499:
9500: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9501:
1.183 brouard 9502: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9503: oldm=oldms;savm=savms;
1.235 brouard 9504: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9505: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9506: for(i=1; i<=nlstate;i++)
9507: for(j=1; j<=nlstate+ndeath;j++)
9508: fprintf(ficrespij," %1d-%1d",i,j);
9509: fprintf(ficrespij,"\n");
9510: for (h=0; h<=nhstepm; h++){
9511: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9512: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9513: for(i=1; i<=nlstate;i++)
9514: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9515: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9516: fprintf(ficrespij,"\n");
9517: }
1.183 brouard 9518: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9519: fprintf(ficrespij,"\n");
9520: }
1.180 brouard 9521: /*}*/
9522: }
1.218 brouard 9523: return 0;
1.180 brouard 9524: }
1.218 brouard 9525:
9526: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9527: /*------------- h Bij x at various ages ------------*/
9528:
9529: int stepsize;
1.218 brouard 9530: /* int agelim; */
9531: int ageminl;
1.217 brouard 9532: int hstepm;
9533: int nhstepm;
1.238 brouard 9534: int h, i, i1, j, k, nres;
1.218 brouard 9535:
1.217 brouard 9536: double agedeb;
9537: double ***p3mat;
1.218 brouard 9538:
9539: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9540: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9541: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9542: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9543: }
9544: printf("Computing pij back: result on file '%s' \n", filerespijb);
9545: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9546:
9547: stepsize=(int) (stepm+YEARM-1)/YEARM;
9548: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9549:
1.218 brouard 9550: /* agelim=AGESUP; */
9551: ageminl=30;
9552: hstepm=stepsize*YEARM; /* Every year of age */
9553: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9554:
9555: /* hstepm=1; aff par mois*/
9556: pstamp(ficrespijb);
9557: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9558: i1= pow(2,cptcoveff);
1.218 brouard 9559: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9560: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9561: /* k=k+1; */
1.238 brouard 9562: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9563: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9564: if(TKresult[nres]!= k)
9565: continue;
9566: fprintf(ficrespijb,"\n#****** ");
9567: for(j=1;j<=cptcoveff;j++)
9568: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9569: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9570: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9571: }
9572: fprintf(ficrespijb,"******\n");
9573: if(invalidvarcomb[k]){
9574: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9575: continue;
9576: }
9577:
9578: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9579: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9580: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9581: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9582: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9583:
9584: /* nhstepm=nhstepm*YEARM; aff par mois*/
9585:
9586: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9587: /* oldm=oldms;savm=savms; */
9588: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9589: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9590: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9591: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9592: for(i=1; i<=nlstate;i++)
9593: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9594: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9595: fprintf(ficrespijb,"\n");
1.238 brouard 9596: for (h=0; h<=nhstepm; h++){
9597: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9598: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9599: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9600: for(i=1; i<=nlstate;i++)
9601: for(j=1; j<=nlstate+ndeath;j++)
9602: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9603: fprintf(ficrespijb,"\n");
9604: }
9605: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9606: fprintf(ficrespijb,"\n");
9607: } /* end age deb */
9608: } /* end combination */
9609: } /* end nres */
1.218 brouard 9610: return 0;
9611: } /* hBijx */
1.217 brouard 9612:
1.180 brouard 9613:
1.136 brouard 9614: /***********************************************/
9615: /**************** Main Program *****************/
9616: /***********************************************/
9617:
9618: int main(int argc, char *argv[])
9619: {
9620: #ifdef GSL
9621: const gsl_multimin_fminimizer_type *T;
9622: size_t iteri = 0, it;
9623: int rval = GSL_CONTINUE;
9624: int status = GSL_SUCCESS;
9625: double ssval;
9626: #endif
9627: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9628: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9629: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9630: int jj, ll, li, lj, lk;
1.136 brouard 9631: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9632: int num_filled;
1.136 brouard 9633: int itimes;
9634: int NDIM=2;
9635: int vpopbased=0;
1.235 brouard 9636: int nres=0;
1.136 brouard 9637:
1.164 brouard 9638: char ca[32], cb[32];
1.136 brouard 9639: /* FILE *fichtm; *//* Html File */
9640: /* FILE *ficgp;*/ /*Gnuplot File */
9641: struct stat info;
1.191 brouard 9642: double agedeb=0.;
1.194 brouard 9643:
9644: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9645: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9646:
1.165 brouard 9647: double fret;
1.191 brouard 9648: double dum=0.; /* Dummy variable */
1.136 brouard 9649: double ***p3mat;
1.218 brouard 9650: /* double ***mobaverage; */
1.164 brouard 9651:
9652: char line[MAXLINE];
1.197 brouard 9653: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9654:
1.234 brouard 9655: char modeltemp[MAXLINE];
1.230 brouard 9656: char resultline[MAXLINE];
9657:
1.136 brouard 9658: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9659: char *tok, *val; /* pathtot */
1.136 brouard 9660: int firstobs=1, lastobs=10;
1.195 brouard 9661: int c, h , cpt, c2;
1.191 brouard 9662: int jl=0;
9663: int i1, j1, jk, stepsize=0;
1.194 brouard 9664: int count=0;
9665:
1.164 brouard 9666: int *tab;
1.136 brouard 9667: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9668: int backcast=0;
1.136 brouard 9669: int mobilav=0,popforecast=0;
1.191 brouard 9670: int hstepm=0, nhstepm=0;
1.136 brouard 9671: int agemortsup;
9672: float sumlpop=0.;
9673: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9674: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9675:
1.191 brouard 9676: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9677: double ftolpl=FTOL;
9678: double **prlim;
1.217 brouard 9679: double **bprlim;
1.136 brouard 9680: double ***param; /* Matrix of parameters */
1.251 ! brouard 9681: double ***paramstart; /* Matrix of starting parameter values */
! 9682: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9683: double **matcov; /* Matrix of covariance */
1.203 brouard 9684: double **hess; /* Hessian matrix */
1.136 brouard 9685: double ***delti3; /* Scale */
9686: double *delti; /* Scale */
9687: double ***eij, ***vareij;
9688: double **varpl; /* Variances of prevalence limits by age */
9689: double *epj, vepp;
1.164 brouard 9690:
1.136 brouard 9691: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9692: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9693:
1.136 brouard 9694: double **ximort;
1.145 brouard 9695: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9696: int *dcwave;
9697:
1.164 brouard 9698: char z[1]="c";
1.136 brouard 9699:
9700: /*char *strt;*/
9701: char strtend[80];
1.126 brouard 9702:
1.164 brouard 9703:
1.126 brouard 9704: /* setlocale (LC_ALL, ""); */
9705: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9706: /* textdomain (PACKAGE); */
9707: /* setlocale (LC_CTYPE, ""); */
9708: /* setlocale (LC_MESSAGES, ""); */
9709:
9710: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9711: rstart_time = time(NULL);
9712: /* (void) gettimeofday(&start_time,&tzp);*/
9713: start_time = *localtime(&rstart_time);
1.126 brouard 9714: curr_time=start_time;
1.157 brouard 9715: /*tml = *localtime(&start_time.tm_sec);*/
9716: /* strcpy(strstart,asctime(&tml)); */
9717: strcpy(strstart,asctime(&start_time));
1.126 brouard 9718:
9719: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9720: /* tp.tm_sec = tp.tm_sec +86400; */
9721: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9722: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9723: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9724: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9725: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9726: /* strt=asctime(&tmg); */
9727: /* printf("Time(after) =%s",strstart); */
9728: /* (void) time (&time_value);
9729: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9730: * tm = *localtime(&time_value);
9731: * strstart=asctime(&tm);
9732: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9733: */
9734:
9735: nberr=0; /* Number of errors and warnings */
9736: nbwarn=0;
1.184 brouard 9737: #ifdef WIN32
9738: _getcwd(pathcd, size);
9739: #else
1.126 brouard 9740: getcwd(pathcd, size);
1.184 brouard 9741: #endif
1.191 brouard 9742: syscompilerinfo(0);
1.196 brouard 9743: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9744: if(argc <=1){
9745: printf("\nEnter the parameter file name: ");
1.205 brouard 9746: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9747: printf("ERROR Empty parameter file name\n");
9748: goto end;
9749: }
1.126 brouard 9750: i=strlen(pathr);
9751: if(pathr[i-1]=='\n')
9752: pathr[i-1]='\0';
1.156 brouard 9753: i=strlen(pathr);
1.205 brouard 9754: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9755: pathr[i-1]='\0';
1.205 brouard 9756: }
9757: i=strlen(pathr);
9758: if( i==0 ){
9759: printf("ERROR Empty parameter file name\n");
9760: goto end;
9761: }
9762: for (tok = pathr; tok != NULL; ){
1.126 brouard 9763: printf("Pathr |%s|\n",pathr);
9764: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9765: printf("val= |%s| pathr=%s\n",val,pathr);
9766: strcpy (pathtot, val);
9767: if(pathr[0] == '\0') break; /* Dirty */
9768: }
9769: }
9770: else{
9771: strcpy(pathtot,argv[1]);
9772: }
9773: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9774: /*cygwin_split_path(pathtot,path,optionfile);
9775: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9776: /* cutv(path,optionfile,pathtot,'\\');*/
9777:
9778: /* Split argv[0], imach program to get pathimach */
9779: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9780: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9781: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9782: /* strcpy(pathimach,argv[0]); */
9783: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9784: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9785: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9786: #ifdef WIN32
9787: _chdir(path); /* Can be a relative path */
9788: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9789: #else
1.126 brouard 9790: chdir(path); /* Can be a relative path */
1.184 brouard 9791: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9792: #endif
9793: printf("Current directory %s!\n",pathcd);
1.126 brouard 9794: strcpy(command,"mkdir ");
9795: strcat(command,optionfilefiname);
9796: if((outcmd=system(command)) != 0){
1.169 brouard 9797: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9798: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9799: /* fclose(ficlog); */
9800: /* exit(1); */
9801: }
9802: /* if((imk=mkdir(optionfilefiname))<0){ */
9803: /* perror("mkdir"); */
9804: /* } */
9805:
9806: /*-------- arguments in the command line --------*/
9807:
1.186 brouard 9808: /* Main Log file */
1.126 brouard 9809: strcat(filelog, optionfilefiname);
9810: strcat(filelog,".log"); /* */
9811: if((ficlog=fopen(filelog,"w"))==NULL) {
9812: printf("Problem with logfile %s\n",filelog);
9813: goto end;
9814: }
9815: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9816: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9817: fprintf(ficlog,"\nEnter the parameter file name: \n");
9818: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9819: path=%s \n\
9820: optionfile=%s\n\
9821: optionfilext=%s\n\
1.156 brouard 9822: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9823:
1.197 brouard 9824: syscompilerinfo(1);
1.167 brouard 9825:
1.126 brouard 9826: printf("Local time (at start):%s",strstart);
9827: fprintf(ficlog,"Local time (at start): %s",strstart);
9828: fflush(ficlog);
9829: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9830: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9831:
9832: /* */
9833: strcpy(fileres,"r");
9834: strcat(fileres, optionfilefiname);
1.201 brouard 9835: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9836: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9837: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9838:
1.186 brouard 9839: /* Main ---------arguments file --------*/
1.126 brouard 9840:
9841: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9842: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9843: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9844: fflush(ficlog);
1.149 brouard 9845: /* goto end; */
9846: exit(70);
1.126 brouard 9847: }
9848:
9849:
9850:
9851: strcpy(filereso,"o");
1.201 brouard 9852: strcat(filereso,fileresu);
1.126 brouard 9853: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9854: printf("Problem with Output resultfile: %s\n", filereso);
9855: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9856: fflush(ficlog);
9857: goto end;
9858: }
9859:
9860: /* Reads comments: lines beginning with '#' */
9861: numlinepar=0;
1.197 brouard 9862:
9863: /* First parameter line */
9864: while(fgets(line, MAXLINE, ficpar)) {
9865: /* If line starts with a # it is a comment */
9866: if (line[0] == '#') {
9867: numlinepar++;
9868: fputs(line,stdout);
9869: fputs(line,ficparo);
9870: fputs(line,ficlog);
9871: continue;
9872: }else
9873: break;
9874: }
9875: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9876: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9877: if (num_filled != 5) {
9878: printf("Should be 5 parameters\n");
9879: }
1.126 brouard 9880: numlinepar++;
1.197 brouard 9881: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9882: }
9883: /* Second parameter line */
9884: while(fgets(line, MAXLINE, ficpar)) {
9885: /* If line starts with a # it is a comment */
9886: if (line[0] == '#') {
9887: numlinepar++;
9888: fputs(line,stdout);
9889: fputs(line,ficparo);
9890: fputs(line,ficlog);
9891: continue;
9892: }else
9893: break;
9894: }
1.223 brouard 9895: 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", \
9896: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9897: if (num_filled != 11) {
9898: 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 9899: printf("but line=%s\n",line);
1.197 brouard 9900: }
1.223 brouard 9901: 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 9902: }
1.203 brouard 9903: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9904: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9905: /* Third parameter line */
9906: while(fgets(line, MAXLINE, ficpar)) {
9907: /* If line starts with a # it is a comment */
9908: if (line[0] == '#') {
9909: numlinepar++;
9910: fputs(line,stdout);
9911: fputs(line,ficparo);
9912: fputs(line,ficlog);
9913: continue;
9914: }else
9915: break;
9916: }
1.201 brouard 9917: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9918: if (num_filled == 0)
9919: model[0]='\0';
9920: else if (num_filled != 1){
1.197 brouard 9921: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9922: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9923: model[0]='\0';
9924: goto end;
9925: }
9926: else{
9927: if (model[0]=='+'){
9928: for(i=1; i<=strlen(model);i++)
9929: modeltemp[i-1]=model[i];
1.201 brouard 9930: strcpy(model,modeltemp);
1.197 brouard 9931: }
9932: }
1.199 brouard 9933: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9934: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9935: }
9936: /* 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); */
9937: /* numlinepar=numlinepar+3; /\* In general *\/ */
9938: /* 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 9939: 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);
9940: 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 9941: fflush(ficlog);
1.190 brouard 9942: /* if(model[0]=='#'|| model[0]== '\0'){ */
9943: if(model[0]=='#'){
1.187 brouard 9944: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9945: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9946: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9947: if(mle != -1){
9948: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9949: exit(1);
9950: }
9951: }
1.126 brouard 9952: while((c=getc(ficpar))=='#' && c!= EOF){
9953: ungetc(c,ficpar);
9954: fgets(line, MAXLINE, ficpar);
9955: numlinepar++;
1.195 brouard 9956: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9957: z[0]=line[1];
9958: }
9959: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9960: fputs(line, stdout);
9961: //puts(line);
1.126 brouard 9962: fputs(line,ficparo);
9963: fputs(line,ficlog);
9964: }
9965: ungetc(c,ficpar);
9966:
9967:
1.145 brouard 9968: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9969: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9970: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9971: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9972: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9973: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9974: v1+v2*age+v2*v3 makes cptcovn = 3
9975: */
9976: if (strlen(model)>1)
1.187 brouard 9977: 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 9978: else
1.187 brouard 9979: ncovmodel=2; /* Constant and age */
1.133 brouard 9980: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9981: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9982: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9983: 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);
9984: 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);
9985: fflush(stdout);
9986: fclose (ficlog);
9987: goto end;
9988: }
1.126 brouard 9989: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9990: delti=delti3[1][1];
9991: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9992: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 9993: /* We could also provide initial parameters values giving by simple logistic regression
9994: * only one way, that is without matrix product. We will have nlstate maximizations */
9995: /* for(i=1;i<nlstate;i++){ */
9996: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
9997: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
9998: /* } */
1.126 brouard 9999: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10000: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10001: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10002: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10003: fclose (ficparo);
10004: fclose (ficlog);
10005: goto end;
10006: exit(0);
1.248 brouard 10007: } else if(mle==-2) { /* Guessing from means */
10008: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
10009: printf(" You chose mle=-2, look at file %s for a template of covariance matrix \n",filereso);
10010: fprintf(ficlog," You chose mle=-2, look at file %s for a template of covariance matrix \n",filereso);
10011:
1.220 brouard 10012: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10013: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10014: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10015: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10016: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10017: matcov=matrix(1,npar,1,npar);
1.203 brouard 10018: hess=matrix(1,npar,1,npar);
1.220 brouard 10019: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10020: /* Read guessed parameters */
1.126 brouard 10021: /* Reads comments: lines beginning with '#' */
10022: while((c=getc(ficpar))=='#' && c!= EOF){
10023: ungetc(c,ficpar);
10024: fgets(line, MAXLINE, ficpar);
10025: numlinepar++;
1.141 brouard 10026: fputs(line,stdout);
1.126 brouard 10027: fputs(line,ficparo);
10028: fputs(line,ficlog);
10029: }
10030: ungetc(c,ficpar);
10031:
10032: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 ! brouard 10033: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10034: for(i=1; i <=nlstate; i++){
1.234 brouard 10035: j=0;
1.126 brouard 10036: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10037: if(jj==i) continue;
10038: j++;
10039: fscanf(ficpar,"%1d%1d",&i1,&j1);
10040: if ((i1 != i) || (j1 != jj)){
10041: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10042: It might be a problem of design; if ncovcol and the model are correct\n \
10043: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10044: exit(1);
10045: }
10046: fprintf(ficparo,"%1d%1d",i1,j1);
10047: if(mle==1)
10048: printf("%1d%1d",i,jj);
10049: fprintf(ficlog,"%1d%1d",i,jj);
10050: for(k=1; k<=ncovmodel;k++){
10051: fscanf(ficpar," %lf",¶m[i][j][k]);
10052: if(mle==1){
10053: printf(" %lf",param[i][j][k]);
10054: fprintf(ficlog," %lf",param[i][j][k]);
10055: }
10056: else
10057: fprintf(ficlog," %lf",param[i][j][k]);
10058: fprintf(ficparo," %lf",param[i][j][k]);
10059: }
10060: fscanf(ficpar,"\n");
10061: numlinepar++;
10062: if(mle==1)
10063: printf("\n");
10064: fprintf(ficlog,"\n");
10065: fprintf(ficparo,"\n");
1.126 brouard 10066: }
10067: }
10068: fflush(ficlog);
1.234 brouard 10069:
1.251 ! brouard 10070: /* Reads parameters values */
1.126 brouard 10071: p=param[1][1];
1.251 ! brouard 10072: pstart=paramstart[1][1];
1.126 brouard 10073:
10074: /* Reads comments: lines beginning with '#' */
10075: while((c=getc(ficpar))=='#' && c!= EOF){
10076: ungetc(c,ficpar);
10077: fgets(line, MAXLINE, ficpar);
10078: numlinepar++;
1.141 brouard 10079: fputs(line,stdout);
1.126 brouard 10080: fputs(line,ficparo);
10081: fputs(line,ficlog);
10082: }
10083: ungetc(c,ficpar);
10084:
10085: for(i=1; i <=nlstate; i++){
10086: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10087: fscanf(ficpar,"%1d%1d",&i1,&j1);
10088: if ( (i1-i) * (j1-j) != 0){
10089: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10090: exit(1);
10091: }
10092: printf("%1d%1d",i,j);
10093: fprintf(ficparo,"%1d%1d",i1,j1);
10094: fprintf(ficlog,"%1d%1d",i1,j1);
10095: for(k=1; k<=ncovmodel;k++){
10096: fscanf(ficpar,"%le",&delti3[i][j][k]);
10097: printf(" %le",delti3[i][j][k]);
10098: fprintf(ficparo," %le",delti3[i][j][k]);
10099: fprintf(ficlog," %le",delti3[i][j][k]);
10100: }
10101: fscanf(ficpar,"\n");
10102: numlinepar++;
10103: printf("\n");
10104: fprintf(ficparo,"\n");
10105: fprintf(ficlog,"\n");
1.126 brouard 10106: }
10107: }
10108: fflush(ficlog);
1.234 brouard 10109:
1.145 brouard 10110: /* Reads covariance matrix */
1.126 brouard 10111: delti=delti3[1][1];
1.220 brouard 10112:
10113:
1.126 brouard 10114: /* 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 10115:
1.126 brouard 10116: /* Reads comments: lines beginning with '#' */
10117: while((c=getc(ficpar))=='#' && c!= EOF){
10118: ungetc(c,ficpar);
10119: fgets(line, MAXLINE, ficpar);
10120: numlinepar++;
1.141 brouard 10121: fputs(line,stdout);
1.126 brouard 10122: fputs(line,ficparo);
10123: fputs(line,ficlog);
10124: }
10125: ungetc(c,ficpar);
1.220 brouard 10126:
1.126 brouard 10127: matcov=matrix(1,npar,1,npar);
1.203 brouard 10128: hess=matrix(1,npar,1,npar);
1.131 brouard 10129: for(i=1; i <=npar; i++)
10130: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10131:
1.194 brouard 10132: /* Scans npar lines */
1.126 brouard 10133: for(i=1; i <=npar; i++){
1.226 brouard 10134: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10135: if(count != 3){
1.226 brouard 10136: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10137: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10138: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10139: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10140: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10141: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10142: exit(1);
1.220 brouard 10143: }else{
1.226 brouard 10144: if(mle==1)
10145: printf("%1d%1d%d",i1,j1,jk);
10146: }
10147: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10148: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10149: for(j=1; j <=i; j++){
1.226 brouard 10150: fscanf(ficpar," %le",&matcov[i][j]);
10151: if(mle==1){
10152: printf(" %.5le",matcov[i][j]);
10153: }
10154: fprintf(ficlog," %.5le",matcov[i][j]);
10155: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10156: }
10157: fscanf(ficpar,"\n");
10158: numlinepar++;
10159: if(mle==1)
1.220 brouard 10160: printf("\n");
1.126 brouard 10161: fprintf(ficlog,"\n");
10162: fprintf(ficparo,"\n");
10163: }
1.194 brouard 10164: /* End of read covariance matrix npar lines */
1.126 brouard 10165: for(i=1; i <=npar; i++)
10166: for(j=i+1;j<=npar;j++)
1.226 brouard 10167: matcov[i][j]=matcov[j][i];
1.126 brouard 10168:
10169: if(mle==1)
10170: printf("\n");
10171: fprintf(ficlog,"\n");
10172:
10173: fflush(ficlog);
10174:
10175: /*-------- Rewriting parameter file ----------*/
10176: strcpy(rfileres,"r"); /* "Rparameterfile */
10177: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10178: strcat(rfileres,"."); /* */
10179: strcat(rfileres,optionfilext); /* Other files have txt extension */
10180: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10181: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10182: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10183: }
10184: fprintf(ficres,"#%s\n",version);
10185: } /* End of mle != -3 */
1.218 brouard 10186:
1.186 brouard 10187: /* Main data
10188: */
1.126 brouard 10189: n= lastobs;
10190: num=lvector(1,n);
10191: moisnais=vector(1,n);
10192: annais=vector(1,n);
10193: moisdc=vector(1,n);
10194: andc=vector(1,n);
1.220 brouard 10195: weight=vector(1,n);
1.126 brouard 10196: agedc=vector(1,n);
10197: cod=ivector(1,n);
1.220 brouard 10198: for(i=1;i<=n;i++){
1.234 brouard 10199: num[i]=0;
10200: moisnais[i]=0;
10201: annais[i]=0;
10202: moisdc[i]=0;
10203: andc[i]=0;
10204: agedc[i]=0;
10205: cod[i]=0;
10206: weight[i]=1.0; /* Equal weights, 1 by default */
10207: }
1.126 brouard 10208: mint=matrix(1,maxwav,1,n);
10209: anint=matrix(1,maxwav,1,n);
1.131 brouard 10210: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10211: tab=ivector(1,NCOVMAX);
1.144 brouard 10212: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10213: 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 10214:
1.136 brouard 10215: /* Reads data from file datafile */
10216: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10217: goto end;
10218:
10219: /* Calculation of the number of parameters from char model */
1.234 brouard 10220: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10221: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10222: k=3 V4 Tvar[k=3]= 4 (from V4)
10223: k=2 V1 Tvar[k=2]= 1 (from V1)
10224: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10225: */
10226:
10227: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10228: TvarsDind=ivector(1,NCOVMAX); /* */
10229: TvarsD=ivector(1,NCOVMAX); /* */
10230: TvarsQind=ivector(1,NCOVMAX); /* */
10231: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10232: TvarF=ivector(1,NCOVMAX); /* */
10233: TvarFind=ivector(1,NCOVMAX); /* */
10234: TvarV=ivector(1,NCOVMAX); /* */
10235: TvarVind=ivector(1,NCOVMAX); /* */
10236: TvarA=ivector(1,NCOVMAX); /* */
10237: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10238: TvarFD=ivector(1,NCOVMAX); /* */
10239: TvarFDind=ivector(1,NCOVMAX); /* */
10240: TvarFQ=ivector(1,NCOVMAX); /* */
10241: TvarFQind=ivector(1,NCOVMAX); /* */
10242: TvarVD=ivector(1,NCOVMAX); /* */
10243: TvarVDind=ivector(1,NCOVMAX); /* */
10244: TvarVQ=ivector(1,NCOVMAX); /* */
10245: TvarVQind=ivector(1,NCOVMAX); /* */
10246:
1.230 brouard 10247: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10248: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10249: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10250: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10251: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10252: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10253: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10254: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10255: */
10256: /* For model-covariate k tells which data-covariate to use but
10257: because this model-covariate is a construction we invent a new column
10258: ncovcol + k1
10259: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10260: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10261: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10262: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10263: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10264: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10265: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10266: */
1.145 brouard 10267: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10268: 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 10269: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10270: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10271: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10272: 4 covariates (3 plus signs)
10273: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10274: */
1.230 brouard 10275: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10276: * individual dummy, fixed or varying:
10277: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10278: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10279: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10280: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10281: * Tmodelind[1]@9={9,0,3,2,}*/
10282: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10283: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10284: * individual quantitative, fixed or varying:
10285: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10286: * 3, 1, 0, 0, 0, 0, 0, 0},
10287: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10288: /* Main decodemodel */
10289:
1.187 brouard 10290:
1.223 brouard 10291: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10292: goto end;
10293:
1.137 brouard 10294: if((double)(lastobs-imx)/(double)imx > 1.10){
10295: nbwarn++;
10296: 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);
10297: 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);
10298: }
1.136 brouard 10299: /* if(mle==1){*/
1.137 brouard 10300: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10301: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10302: }
10303:
10304: /*-calculation of age at interview from date of interview and age at death -*/
10305: agev=matrix(1,maxwav,1,imx);
10306:
10307: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10308: goto end;
10309:
1.126 brouard 10310:
1.136 brouard 10311: agegomp=(int)agemin;
10312: free_vector(moisnais,1,n);
10313: free_vector(annais,1,n);
1.126 brouard 10314: /* free_matrix(mint,1,maxwav,1,n);
10315: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10316: /* free_vector(moisdc,1,n); */
10317: /* free_vector(andc,1,n); */
1.145 brouard 10318: /* */
10319:
1.126 brouard 10320: wav=ivector(1,imx);
1.214 brouard 10321: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10322: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10323: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10324: 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.*/
10325: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10326: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10327:
10328: /* Concatenates waves */
1.214 brouard 10329: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10330: Death is a valid wave (if date is known).
10331: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10332: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10333: and mw[mi+1][i]. dh depends on stepm.
10334: */
10335:
1.126 brouard 10336: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10337: /* Concatenates waves */
1.145 brouard 10338:
1.215 brouard 10339: free_vector(moisdc,1,n);
10340: free_vector(andc,1,n);
10341:
1.126 brouard 10342: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10343: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10344: ncodemax[1]=1;
1.145 brouard 10345: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10346: cptcoveff=0;
1.220 brouard 10347: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10348: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10349: }
10350:
10351: ncovcombmax=pow(2,cptcoveff);
10352: invalidvarcomb=ivector(1, ncovcombmax);
10353: for(i=1;i<ncovcombmax;i++)
10354: invalidvarcomb[i]=0;
10355:
1.211 brouard 10356: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10357: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10358: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10359:
1.200 brouard 10360: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10361: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10362: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10363: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10364: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10365: * (currently 0 or 1) in the data.
10366: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10367: * corresponding modality (h,j).
10368: */
10369:
1.145 brouard 10370: h=0;
10371: /*if (cptcovn > 0) */
1.126 brouard 10372: m=pow(2,cptcoveff);
10373:
1.144 brouard 10374: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10375: * For k=4 covariates, h goes from 1 to m=2**k
10376: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10377: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10378: * h\k 1 2 3 4
1.143 brouard 10379: *______________________________
10380: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10381: * 2 2 1 1 1
10382: * 3 i=2 1 2 1 1
10383: * 4 2 2 1 1
10384: * 5 i=3 1 i=2 1 2 1
10385: * 6 2 1 2 1
10386: * 7 i=4 1 2 2 1
10387: * 8 2 2 2 1
1.197 brouard 10388: * 9 i=5 1 i=3 1 i=2 1 2
10389: * 10 2 1 1 2
10390: * 11 i=6 1 2 1 2
10391: * 12 2 2 1 2
10392: * 13 i=7 1 i=4 1 2 2
10393: * 14 2 1 2 2
10394: * 15 i=8 1 2 2 2
10395: * 16 2 2 2 2
1.143 brouard 10396: */
1.212 brouard 10397: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10398: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10399: * and the value of each covariate?
10400: * V1=1, V2=1, V3=2, V4=1 ?
10401: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10402: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10403: * In order to get the real value in the data, we use nbcode
10404: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10405: * We are keeping this crazy system in order to be able (in the future?)
10406: * to have more than 2 values (0 or 1) for a covariate.
10407: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10408: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10409: * bbbbbbbb
10410: * 76543210
10411: * h-1 00000101 (6-1=5)
1.219 brouard 10412: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10413: * &
10414: * 1 00000001 (1)
1.219 brouard 10415: * 00000000 = 1 & ((h-1) >> (k-1))
10416: * +1= 00000001 =1
1.211 brouard 10417: *
10418: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10419: * h' 1101 =2^3+2^2+0x2^1+2^0
10420: * >>k' 11
10421: * & 00000001
10422: * = 00000001
10423: * +1 = 00000010=2 = codtabm(14,3)
10424: * Reverse h=6 and m=16?
10425: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10426: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10427: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10428: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10429: * V3=decodtabm(14,3,2**4)=2
10430: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10431: *(h-1) >> (j-1) 0011 =13 >> 2
10432: * &1 000000001
10433: * = 000000001
10434: * +1= 000000010 =2
10435: * 2211
10436: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10437: * V3=2
1.220 brouard 10438: * codtabm and decodtabm are identical
1.211 brouard 10439: */
10440:
1.145 brouard 10441:
10442: free_ivector(Ndum,-1,NCOVMAX);
10443:
10444:
1.126 brouard 10445:
1.186 brouard 10446: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10447: strcpy(optionfilegnuplot,optionfilefiname);
10448: if(mle==-3)
1.201 brouard 10449: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10450: strcat(optionfilegnuplot,".gp");
10451:
10452: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10453: printf("Problem with file %s",optionfilegnuplot);
10454: }
10455: else{
1.204 brouard 10456: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10457: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10458: //fprintf(ficgp,"set missing 'NaNq'\n");
10459: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10460: }
10461: /* fclose(ficgp);*/
1.186 brouard 10462:
10463:
10464: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10465:
10466: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10467: if(mle==-3)
1.201 brouard 10468: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10469: strcat(optionfilehtm,".htm");
10470: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10471: printf("Problem with %s \n",optionfilehtm);
10472: exit(0);
1.126 brouard 10473: }
10474:
10475: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10476: strcat(optionfilehtmcov,"-cov.htm");
10477: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10478: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10479: }
10480: else{
10481: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10482: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10483: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10484: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10485: }
10486:
1.213 brouard 10487: 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 10488: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10489: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10490: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10491: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10492: \n\
10493: <hr size=\"2\" color=\"#EC5E5E\">\
10494: <ul><li><h4>Parameter files</h4>\n\
10495: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10496: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10497: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10498: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10499: - Date and time at start: %s</ul>\n",\
10500: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10501: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10502: fileres,fileres,\
10503: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10504: fflush(fichtm);
10505:
10506: strcpy(pathr,path);
10507: strcat(pathr,optionfilefiname);
1.184 brouard 10508: #ifdef WIN32
10509: _chdir(optionfilefiname); /* Move to directory named optionfile */
10510: #else
1.126 brouard 10511: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10512: #endif
10513:
1.126 brouard 10514:
1.220 brouard 10515: /* Calculates basic frequencies. Computes observed prevalence at single age
10516: and for any valid combination of covariates
1.126 brouard 10517: and prints on file fileres'p'. */
1.251 ! brouard 10518: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10519: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10520:
10521: fprintf(fichtm,"\n");
10522: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10523: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10524: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10525: imx,agemin,agemax,jmin,jmax,jmean);
10526: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10527: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10528: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10529: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10530: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10531:
1.126 brouard 10532: /* For Powell, parameters are in a vector p[] starting at p[1]
10533: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10534: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10535:
10536: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10537: /* For mortality only */
1.126 brouard 10538: if (mle==-3){
1.136 brouard 10539: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10540: for(i=1;i<=NDIM;i++)
10541: for(j=1;j<=NDIM;j++)
10542: ximort[i][j]=0.;
1.186 brouard 10543: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10544: cens=ivector(1,n);
10545: ageexmed=vector(1,n);
10546: agecens=vector(1,n);
10547: dcwave=ivector(1,n);
1.223 brouard 10548:
1.126 brouard 10549: for (i=1; i<=imx; i++){
10550: dcwave[i]=-1;
10551: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10552: if (s[m][i]>nlstate) {
10553: dcwave[i]=m;
10554: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10555: break;
10556: }
1.126 brouard 10557: }
1.226 brouard 10558:
1.126 brouard 10559: for (i=1; i<=imx; i++) {
10560: if (wav[i]>0){
1.226 brouard 10561: ageexmed[i]=agev[mw[1][i]][i];
10562: j=wav[i];
10563: agecens[i]=1.;
10564:
10565: if (ageexmed[i]> 1 && wav[i] > 0){
10566: agecens[i]=agev[mw[j][i]][i];
10567: cens[i]= 1;
10568: }else if (ageexmed[i]< 1)
10569: cens[i]= -1;
10570: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10571: cens[i]=0 ;
1.126 brouard 10572: }
10573: else cens[i]=-1;
10574: }
10575:
10576: for (i=1;i<=NDIM;i++) {
10577: for (j=1;j<=NDIM;j++)
1.226 brouard 10578: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10579: }
10580:
1.145 brouard 10581: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10582: /*printf("%lf %lf", p[1], p[2]);*/
10583:
10584:
1.136 brouard 10585: #ifdef GSL
10586: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10587: #else
1.126 brouard 10588: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10589: #endif
1.201 brouard 10590: strcpy(filerespow,"POW-MORT_");
10591: strcat(filerespow,fileresu);
1.126 brouard 10592: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10593: printf("Problem with resultfile: %s\n", filerespow);
10594: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10595: }
1.136 brouard 10596: #ifdef GSL
10597: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10598: #else
1.126 brouard 10599: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10600: #endif
1.126 brouard 10601: /* for (i=1;i<=nlstate;i++)
10602: for(j=1;j<=nlstate+ndeath;j++)
10603: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10604: */
10605: fprintf(ficrespow,"\n");
1.136 brouard 10606: #ifdef GSL
10607: /* gsl starts here */
10608: T = gsl_multimin_fminimizer_nmsimplex;
10609: gsl_multimin_fminimizer *sfm = NULL;
10610: gsl_vector *ss, *x;
10611: gsl_multimin_function minex_func;
10612:
10613: /* Initial vertex size vector */
10614: ss = gsl_vector_alloc (NDIM);
10615:
10616: if (ss == NULL){
10617: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10618: }
10619: /* Set all step sizes to 1 */
10620: gsl_vector_set_all (ss, 0.001);
10621:
10622: /* Starting point */
1.126 brouard 10623:
1.136 brouard 10624: x = gsl_vector_alloc (NDIM);
10625:
10626: if (x == NULL){
10627: gsl_vector_free(ss);
10628: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10629: }
10630:
10631: /* Initialize method and iterate */
10632: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10633: /* gsl_vector_set(x, 0, 0.0268); */
10634: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10635: gsl_vector_set(x, 0, p[1]);
10636: gsl_vector_set(x, 1, p[2]);
10637:
10638: minex_func.f = &gompertz_f;
10639: minex_func.n = NDIM;
10640: minex_func.params = (void *)&p; /* ??? */
10641:
10642: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10643: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10644:
10645: printf("Iterations beginning .....\n\n");
10646: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10647:
10648: iteri=0;
10649: while (rval == GSL_CONTINUE){
10650: iteri++;
10651: status = gsl_multimin_fminimizer_iterate(sfm);
10652:
10653: if (status) printf("error: %s\n", gsl_strerror (status));
10654: fflush(0);
10655:
10656: if (status)
10657: break;
10658:
10659: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10660: ssval = gsl_multimin_fminimizer_size (sfm);
10661:
10662: if (rval == GSL_SUCCESS)
10663: printf ("converged to a local maximum at\n");
10664:
10665: printf("%5d ", iteri);
10666: for (it = 0; it < NDIM; it++){
10667: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10668: }
10669: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10670: }
10671:
10672: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10673:
10674: gsl_vector_free(x); /* initial values */
10675: gsl_vector_free(ss); /* inital step size */
10676: for (it=0; it<NDIM; it++){
10677: p[it+1]=gsl_vector_get(sfm->x,it);
10678: fprintf(ficrespow," %.12lf", p[it]);
10679: }
10680: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10681: #endif
10682: #ifdef POWELL
10683: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10684: #endif
1.126 brouard 10685: fclose(ficrespow);
10686:
1.203 brouard 10687: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10688:
10689: for(i=1; i <=NDIM; i++)
10690: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10691: matcov[i][j]=matcov[j][i];
1.126 brouard 10692:
10693: printf("\nCovariance matrix\n ");
1.203 brouard 10694: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10695: for(i=1; i <=NDIM; i++) {
10696: for(j=1;j<=NDIM;j++){
1.220 brouard 10697: printf("%f ",matcov[i][j]);
10698: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10699: }
1.203 brouard 10700: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10701: }
10702:
10703: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10704: for (i=1;i<=NDIM;i++) {
1.126 brouard 10705: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10706: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10707: }
1.126 brouard 10708: lsurv=vector(1,AGESUP);
10709: lpop=vector(1,AGESUP);
10710: tpop=vector(1,AGESUP);
10711: lsurv[agegomp]=100000;
10712:
10713: for (k=agegomp;k<=AGESUP;k++) {
10714: agemortsup=k;
10715: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10716: }
10717:
10718: for (k=agegomp;k<agemortsup;k++)
10719: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10720:
10721: for (k=agegomp;k<agemortsup;k++){
10722: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10723: sumlpop=sumlpop+lpop[k];
10724: }
10725:
10726: tpop[agegomp]=sumlpop;
10727: for (k=agegomp;k<(agemortsup-3);k++){
10728: /* tpop[k+1]=2;*/
10729: tpop[k+1]=tpop[k]-lpop[k];
10730: }
10731:
10732:
10733: printf("\nAge lx qx dx Lx Tx e(x)\n");
10734: for (k=agegomp;k<(agemortsup-2);k++)
10735: 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]);
10736:
10737:
10738: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10739: ageminpar=50;
10740: agemaxpar=100;
1.194 brouard 10741: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10742: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10743: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10744: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10745: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10746: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10747: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10748: }else{
10749: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10750: 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 10751: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10752: }
1.201 brouard 10753: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10754: stepm, weightopt,\
10755: model,imx,p,matcov,agemortsup);
10756:
10757: free_vector(lsurv,1,AGESUP);
10758: free_vector(lpop,1,AGESUP);
10759: free_vector(tpop,1,AGESUP);
1.220 brouard 10760: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10761: free_ivector(cens,1,n);
10762: free_vector(agecens,1,n);
10763: free_ivector(dcwave,1,n);
1.220 brouard 10764: #ifdef GSL
1.136 brouard 10765: #endif
1.186 brouard 10766: } /* Endof if mle==-3 mortality only */
1.205 brouard 10767: /* Standard */
10768: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10769: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10770: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10771: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10772: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10773: for (k=1; k<=npar;k++)
10774: printf(" %d %8.5f",k,p[k]);
10775: printf("\n");
1.205 brouard 10776: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10777: /* mlikeli uses func not funcone */
1.247 brouard 10778: /* for(i=1;i<nlstate;i++){ */
10779: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10780: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10781: /* } */
1.205 brouard 10782: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10783: }
10784: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10785: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10786: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10787: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10788: }
10789: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10790: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10791: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10792: for (k=1; k<=npar;k++)
10793: printf(" %d %8.5f",k,p[k]);
10794: printf("\n");
10795:
10796: /*--------- results files --------------*/
1.224 brouard 10797: 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 10798:
10799:
10800: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10801: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10802: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10803: for(i=1,jk=1; i <=nlstate; i++){
10804: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10805: if (k != i) {
10806: printf("%d%d ",i,k);
10807: fprintf(ficlog,"%d%d ",i,k);
10808: fprintf(ficres,"%1d%1d ",i,k);
10809: for(j=1; j <=ncovmodel; j++){
10810: printf("%12.7f ",p[jk]);
10811: fprintf(ficlog,"%12.7f ",p[jk]);
10812: fprintf(ficres,"%12.7f ",p[jk]);
10813: jk++;
10814: }
10815: printf("\n");
10816: fprintf(ficlog,"\n");
10817: fprintf(ficres,"\n");
10818: }
1.126 brouard 10819: }
10820: }
1.203 brouard 10821: if(mle != 0){
10822: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10823: ftolhess=ftol; /* Usually correct */
1.203 brouard 10824: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10825: 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");
10826: 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");
10827: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10828: for(k=1; k <=(nlstate+ndeath); k++){
10829: if (k != i) {
10830: printf("%d%d ",i,k);
10831: fprintf(ficlog,"%d%d ",i,k);
10832: for(j=1; j <=ncovmodel; j++){
10833: 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]));
10834: 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]));
10835: jk++;
10836: }
10837: printf("\n");
10838: fprintf(ficlog,"\n");
10839: }
10840: }
1.193 brouard 10841: }
1.203 brouard 10842: } /* end of hesscov and Wald tests */
1.225 brouard 10843:
1.203 brouard 10844: /* */
1.126 brouard 10845: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10846: printf("# Scales (for hessian or gradient estimation)\n");
10847: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10848: for(i=1,jk=1; i <=nlstate; i++){
10849: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10850: if (j!=i) {
10851: fprintf(ficres,"%1d%1d",i,j);
10852: printf("%1d%1d",i,j);
10853: fprintf(ficlog,"%1d%1d",i,j);
10854: for(k=1; k<=ncovmodel;k++){
10855: printf(" %.5e",delti[jk]);
10856: fprintf(ficlog," %.5e",delti[jk]);
10857: fprintf(ficres," %.5e",delti[jk]);
10858: jk++;
10859: }
10860: printf("\n");
10861: fprintf(ficlog,"\n");
10862: fprintf(ficres,"\n");
10863: }
1.126 brouard 10864: }
10865: }
10866:
10867: 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 10868: if(mle >= 1) /* To big for the screen */
1.126 brouard 10869: 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");
10870: 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");
10871: /* # 121 Var(a12)\n\ */
10872: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10873: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10874: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10875: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10876: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10877: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10878: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10879:
10880:
10881: /* Just to have a covariance matrix which will be more understandable
10882: even is we still don't want to manage dictionary of variables
10883: */
10884: for(itimes=1;itimes<=2;itimes++){
10885: jj=0;
10886: for(i=1; i <=nlstate; i++){
1.225 brouard 10887: for(j=1; j <=nlstate+ndeath; j++){
10888: if(j==i) continue;
10889: for(k=1; k<=ncovmodel;k++){
10890: jj++;
10891: ca[0]= k+'a'-1;ca[1]='\0';
10892: if(itimes==1){
10893: if(mle>=1)
10894: printf("#%1d%1d%d",i,j,k);
10895: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10896: fprintf(ficres,"#%1d%1d%d",i,j,k);
10897: }else{
10898: if(mle>=1)
10899: printf("%1d%1d%d",i,j,k);
10900: fprintf(ficlog,"%1d%1d%d",i,j,k);
10901: fprintf(ficres,"%1d%1d%d",i,j,k);
10902: }
10903: ll=0;
10904: for(li=1;li <=nlstate; li++){
10905: for(lj=1;lj <=nlstate+ndeath; lj++){
10906: if(lj==li) continue;
10907: for(lk=1;lk<=ncovmodel;lk++){
10908: ll++;
10909: if(ll<=jj){
10910: cb[0]= lk +'a'-1;cb[1]='\0';
10911: if(ll<jj){
10912: if(itimes==1){
10913: if(mle>=1)
10914: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10915: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10916: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10917: }else{
10918: if(mle>=1)
10919: printf(" %.5e",matcov[jj][ll]);
10920: fprintf(ficlog," %.5e",matcov[jj][ll]);
10921: fprintf(ficres," %.5e",matcov[jj][ll]);
10922: }
10923: }else{
10924: if(itimes==1){
10925: if(mle>=1)
10926: printf(" Var(%s%1d%1d)",ca,i,j);
10927: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10928: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10929: }else{
10930: if(mle>=1)
10931: printf(" %.7e",matcov[jj][ll]);
10932: fprintf(ficlog," %.7e",matcov[jj][ll]);
10933: fprintf(ficres," %.7e",matcov[jj][ll]);
10934: }
10935: }
10936: }
10937: } /* end lk */
10938: } /* end lj */
10939: } /* end li */
10940: if(mle>=1)
10941: printf("\n");
10942: fprintf(ficlog,"\n");
10943: fprintf(ficres,"\n");
10944: numlinepar++;
10945: } /* end k*/
10946: } /*end j */
1.126 brouard 10947: } /* end i */
10948: } /* end itimes */
10949:
10950: fflush(ficlog);
10951: fflush(ficres);
1.225 brouard 10952: while(fgets(line, MAXLINE, ficpar)) {
10953: /* If line starts with a # it is a comment */
10954: if (line[0] == '#') {
10955: numlinepar++;
10956: fputs(line,stdout);
10957: fputs(line,ficparo);
10958: fputs(line,ficlog);
10959: continue;
10960: }else
10961: break;
10962: }
10963:
1.209 brouard 10964: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10965: /* ungetc(c,ficpar); */
10966: /* fgets(line, MAXLINE, ficpar); */
10967: /* fputs(line,stdout); */
10968: /* fputs(line,ficparo); */
10969: /* } */
10970: /* ungetc(c,ficpar); */
1.126 brouard 10971:
10972: estepm=0;
1.209 brouard 10973: 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 10974:
10975: if (num_filled != 6) {
10976: 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);
10977: 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);
10978: goto end;
10979: }
10980: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10981: }
10982: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10983: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10984:
1.209 brouard 10985: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10986: if (estepm==0 || estepm < stepm) estepm=stepm;
10987: if (fage <= 2) {
10988: bage = ageminpar;
10989: fage = agemaxpar;
10990: }
10991:
10992: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10993: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10994: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10995:
1.186 brouard 10996: /* Other stuffs, more or less useful */
1.126 brouard 10997: while((c=getc(ficpar))=='#' && c!= EOF){
10998: ungetc(c,ficpar);
10999: fgets(line, MAXLINE, ficpar);
1.141 brouard 11000: fputs(line,stdout);
1.126 brouard 11001: fputs(line,ficparo);
11002: }
11003: ungetc(c,ficpar);
11004:
11005: 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);
11006: 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);
11007: 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);
11008: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11009: 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);
11010:
11011: while((c=getc(ficpar))=='#' && c!= EOF){
11012: ungetc(c,ficpar);
11013: fgets(line, MAXLINE, ficpar);
1.141 brouard 11014: fputs(line,stdout);
1.126 brouard 11015: fputs(line,ficparo);
11016: }
11017: ungetc(c,ficpar);
11018:
11019:
11020: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11021: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11022:
11023: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 11024: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 11025: fprintf(ficparo,"pop_based=%d\n",popbased);
11026: fprintf(ficres,"pop_based=%d\n",popbased);
11027:
11028: while((c=getc(ficpar))=='#' && c!= EOF){
11029: ungetc(c,ficpar);
11030: fgets(line, MAXLINE, ficpar);
1.141 brouard 11031: fputs(line,stdout);
1.238 brouard 11032: fputs(line,ficres);
1.126 brouard 11033: fputs(line,ficparo);
11034: }
11035: ungetc(c,ficpar);
11036:
11037: 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);
11038: 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);
11039: 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);
11040: 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);
11041: 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);
11042: /* day and month of proj2 are not used but only year anproj2.*/
11043:
1.217 brouard 11044: while((c=getc(ficpar))=='#' && c!= EOF){
11045: ungetc(c,ficpar);
11046: fgets(line, MAXLINE, ficpar);
11047: fputs(line,stdout);
11048: fputs(line,ficparo);
1.238 brouard 11049: fputs(line,ficres);
1.217 brouard 11050: }
11051: ungetc(c,ficpar);
11052:
11053: 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 11054: 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);
11055: 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);
11056: 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 11057: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 11058:
1.230 brouard 11059: /* Results */
1.235 brouard 11060: nresult=0;
1.230 brouard 11061: while(fgets(line, MAXLINE, ficpar)) {
11062: /* If line starts with a # it is a comment */
11063: if (line[0] == '#') {
11064: numlinepar++;
11065: fputs(line,stdout);
11066: fputs(line,ficparo);
11067: fputs(line,ficlog);
1.238 brouard 11068: fputs(line,ficres);
1.230 brouard 11069: continue;
11070: }else
11071: break;
11072: }
1.240 brouard 11073: if (!feof(ficpar))
1.230 brouard 11074: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11075: if (num_filled == 0){
1.230 brouard 11076: resultline[0]='\0';
1.240 brouard 11077: break;
11078: } else if (num_filled != 1){
1.230 brouard 11079: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
11080: }
1.235 brouard 11081: nresult++; /* Sum of resultlines */
11082: printf("Result %d: result=%s\n",nresult, resultline);
11083: if(nresult > MAXRESULTLINES){
11084: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11085: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11086: goto end;
11087: }
11088: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11089: fprintf(ficparo,"result: %s\n",resultline);
11090: fprintf(ficres,"result: %s\n",resultline);
11091: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11092: while(fgets(line, MAXLINE, ficpar)) {
11093: /* If line starts with a # it is a comment */
11094: if (line[0] == '#') {
11095: numlinepar++;
11096: fputs(line,stdout);
11097: fputs(line,ficparo);
1.238 brouard 11098: fputs(line,ficres);
1.230 brouard 11099: fputs(line,ficlog);
11100: continue;
11101: }else
11102: break;
11103: }
11104: if (feof(ficpar))
11105: break;
11106: else{ /* Processess output results for this combination of covariate values */
11107: }
1.240 brouard 11108: } /* end while */
1.230 brouard 11109:
11110:
1.126 brouard 11111:
1.230 brouard 11112: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11113: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11114:
11115: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11116: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11117: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11118: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11119: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11120: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11121: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11122: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11123: }else{
1.218 brouard 11124: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11125: }
11126: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11127: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11128: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11129:
1.225 brouard 11130: /*------------ free_vector -------------*/
11131: /* chdir(path); */
1.220 brouard 11132:
1.215 brouard 11133: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11134: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11135: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11136: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11137: free_lvector(num,1,n);
11138: free_vector(agedc,1,n);
11139: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11140: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11141: fclose(ficparo);
11142: fclose(ficres);
1.220 brouard 11143:
11144:
1.186 brouard 11145: /* Other results (useful)*/
1.220 brouard 11146:
11147:
1.126 brouard 11148: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11149: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11150: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11151: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11152: fclose(ficrespl);
11153:
11154: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11155: /*#include "hpijx.h"*/
11156: hPijx(p, bage, fage);
1.145 brouard 11157: fclose(ficrespij);
1.227 brouard 11158:
1.220 brouard 11159: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11160: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11161: k=1;
1.126 brouard 11162: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11163:
1.219 brouard 11164: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11165: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11166: for(i=1;i<=AGESUP;i++)
1.219 brouard 11167: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11168: for(k=1;k<=ncovcombmax;k++)
11169: probs[i][j][k]=0.;
1.219 brouard 11170: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11171: if (mobilav!=0 ||mobilavproj !=0 ) {
11172: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11173: for(i=1;i<=AGESUP;i++)
11174: for(j=1;j<=nlstate;j++)
11175: for(k=1;k<=ncovcombmax;k++)
11176: mobaverages[i][j][k]=0.;
1.219 brouard 11177: mobaverage=mobaverages;
11178: if (mobilav!=0) {
1.235 brouard 11179: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11180: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11181: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11182: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11183: }
1.219 brouard 11184: }
11185: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11186: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11187: else if (mobilavproj !=0) {
1.235 brouard 11188: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11189: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11190: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11191: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11192: }
1.219 brouard 11193: }
11194: }/* end if moving average */
1.227 brouard 11195:
1.126 brouard 11196: /*---------- Forecasting ------------------*/
11197: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11198: if(prevfcast==1){
11199: /* if(stepm ==1){*/
1.225 brouard 11200: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11201: }
1.217 brouard 11202: if(backcast==1){
1.219 brouard 11203: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11204: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11205: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11206:
11207: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11208:
11209: bprlim=matrix(1,nlstate,1,nlstate);
11210: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11211: fclose(ficresplb);
11212:
1.222 brouard 11213: hBijx(p, bage, fage, mobaverage);
11214: fclose(ficrespijb);
1.219 brouard 11215: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11216:
11217: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11218: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11219: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11220: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11221: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11222: }
1.217 brouard 11223:
1.186 brouard 11224:
11225: /* ------ Other prevalence ratios------------ */
1.126 brouard 11226:
1.215 brouard 11227: free_ivector(wav,1,imx);
11228: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11229: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11230: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11231:
11232:
1.127 brouard 11233: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11234:
1.201 brouard 11235: strcpy(filerese,"E_");
11236: strcat(filerese,fileresu);
1.126 brouard 11237: if((ficreseij=fopen(filerese,"w"))==NULL) {
11238: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11239: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11240: }
1.208 brouard 11241: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11242: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11243:
11244: pstamp(ficreseij);
1.219 brouard 11245:
1.235 brouard 11246: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11247: if (cptcovn < 1){i1=1;}
11248:
11249: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11250: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11251: if(TKresult[nres]!= k)
11252: continue;
1.219 brouard 11253: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11254: printf("\n#****** ");
1.225 brouard 11255: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11256: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11257: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11258: }
11259: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11260: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11261: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11262: }
11263: fprintf(ficreseij,"******\n");
1.235 brouard 11264: printf("******\n");
1.219 brouard 11265:
11266: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11267: oldm=oldms;savm=savms;
1.235 brouard 11268: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11269:
1.219 brouard 11270: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11271: }
11272: fclose(ficreseij);
1.208 brouard 11273: printf("done evsij\n");fflush(stdout);
11274: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11275:
1.227 brouard 11276: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11277:
11278:
1.201 brouard 11279: strcpy(filerest,"T_");
11280: strcat(filerest,fileresu);
1.127 brouard 11281: if((ficrest=fopen(filerest,"w"))==NULL) {
11282: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11283: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11284: }
1.208 brouard 11285: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11286: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11287:
1.126 brouard 11288:
1.201 brouard 11289: strcpy(fileresstde,"STDE_");
11290: strcat(fileresstde,fileresu);
1.126 brouard 11291: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11292: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11293: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11294: }
1.227 brouard 11295: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11296: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11297:
1.201 brouard 11298: strcpy(filerescve,"CVE_");
11299: strcat(filerescve,fileresu);
1.126 brouard 11300: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11301: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11302: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11303: }
1.227 brouard 11304: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11305: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11306:
1.201 brouard 11307: strcpy(fileresv,"V_");
11308: strcat(fileresv,fileresu);
1.126 brouard 11309: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11310: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11311: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11312: }
1.227 brouard 11313: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11314: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11315:
1.145 brouard 11316: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11317: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11318:
1.235 brouard 11319: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11320: if (cptcovn < 1){i1=1;}
11321:
11322: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11323: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11324: if(TKresult[nres]!= k)
11325: continue;
1.242 brouard 11326: printf("\n#****** Result for:");
11327: fprintf(ficrest,"\n#****** Result for:");
11328: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11329: for(j=1;j<=cptcoveff;j++){
11330: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11331: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11332: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11333: }
1.235 brouard 11334: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11335: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11336: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11337: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11338: }
1.208 brouard 11339: fprintf(ficrest,"******\n");
1.227 brouard 11340: fprintf(ficlog,"******\n");
11341: printf("******\n");
1.208 brouard 11342:
11343: fprintf(ficresstdeij,"\n#****** ");
11344: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11345: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11346: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11347: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11348: }
1.235 brouard 11349: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11350: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11351: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11352: }
1.208 brouard 11353: fprintf(ficresstdeij,"******\n");
11354: fprintf(ficrescveij,"******\n");
11355:
11356: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11357: /* pstamp(ficresvij); */
1.225 brouard 11358: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11359: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11360: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11361: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11362: }
1.208 brouard 11363: fprintf(ficresvij,"******\n");
11364:
11365: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11366: oldm=oldms;savm=savms;
1.235 brouard 11367: printf(" cvevsij ");
11368: fprintf(ficlog, " cvevsij ");
11369: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11370: printf(" end cvevsij \n ");
11371: fprintf(ficlog, " end cvevsij \n ");
11372:
11373: /*
11374: */
11375: /* goto endfree; */
11376:
11377: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11378: pstamp(ficrest);
11379:
11380:
11381: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11382: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11383: cptcod= 0; /* To be deleted */
11384: printf("varevsij vpopbased=%d \n",vpopbased);
11385: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11386: 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 11387: 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 ");
11388: if(vpopbased==1)
11389: 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);
11390: else
11391: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11392: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11393: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11394: fprintf(ficrest,"\n");
11395: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11396: epj=vector(1,nlstate+1);
11397: printf("Computing age specific period (stable) prevalences in each health state \n");
11398: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11399: for(age=bage; age <=fage ;age++){
1.235 brouard 11400: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11401: if (vpopbased==1) {
11402: if(mobilav ==0){
11403: for(i=1; i<=nlstate;i++)
11404: prlim[i][i]=probs[(int)age][i][k];
11405: }else{ /* mobilav */
11406: for(i=1; i<=nlstate;i++)
11407: prlim[i][i]=mobaverage[(int)age][i][k];
11408: }
11409: }
1.219 brouard 11410:
1.227 brouard 11411: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11412: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11413: /* printf(" age %4.0f ",age); */
11414: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11415: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11416: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11417: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11418: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11419: }
11420: epj[nlstate+1] +=epj[j];
11421: }
11422: /* printf(" age %4.0f \n",age); */
1.219 brouard 11423:
1.227 brouard 11424: for(i=1, vepp=0.;i <=nlstate;i++)
11425: for(j=1;j <=nlstate;j++)
11426: vepp += vareij[i][j][(int)age];
11427: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11428: for(j=1;j <=nlstate;j++){
11429: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11430: }
11431: fprintf(ficrest,"\n");
11432: }
1.208 brouard 11433: } /* End vpopbased */
11434: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11435: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11436: free_vector(epj,1,nlstate+1);
1.235 brouard 11437: printf("done selection\n");fflush(stdout);
11438: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11439:
1.145 brouard 11440: /*}*/
1.235 brouard 11441: } /* End k selection */
1.227 brouard 11442:
11443: printf("done State-specific expectancies\n");fflush(stdout);
11444: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11445:
1.126 brouard 11446: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11447:
1.201 brouard 11448: strcpy(fileresvpl,"VPL_");
11449: strcat(fileresvpl,fileresu);
1.126 brouard 11450: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11451: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11452: exit(0);
11453: }
1.208 brouard 11454: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11455: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11456:
1.145 brouard 11457: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11458: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11459:
1.235 brouard 11460: i1=pow(2,cptcoveff);
11461: if (cptcovn < 1){i1=1;}
11462:
11463: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11464: for(k=1; k<=i1;k++){
11465: if(TKresult[nres]!= k)
11466: continue;
1.227 brouard 11467: fprintf(ficresvpl,"\n#****** ");
11468: printf("\n#****** ");
11469: fprintf(ficlog,"\n#****** ");
11470: for(j=1;j<=cptcoveff;j++) {
11471: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11472: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11473: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11474: }
1.235 brouard 11475: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11476: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11477: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11478: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11479: }
1.227 brouard 11480: fprintf(ficresvpl,"******\n");
11481: printf("******\n");
11482: fprintf(ficlog,"******\n");
11483:
11484: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11485: oldm=oldms;savm=savms;
1.235 brouard 11486: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11487: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11488: /*}*/
1.126 brouard 11489: }
1.227 brouard 11490:
1.126 brouard 11491: fclose(ficresvpl);
1.208 brouard 11492: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11493: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11494:
11495: free_vector(weight,1,n);
11496: free_imatrix(Tvard,1,NCOVMAX,1,2);
11497: free_imatrix(s,1,maxwav+1,1,n);
11498: free_matrix(anint,1,maxwav,1,n);
11499: free_matrix(mint,1,maxwav,1,n);
11500: free_ivector(cod,1,n);
11501: free_ivector(tab,1,NCOVMAX);
11502: fclose(ficresstdeij);
11503: fclose(ficrescveij);
11504: fclose(ficresvij);
11505: fclose(ficrest);
11506: fclose(ficpar);
11507:
11508:
1.126 brouard 11509: /*---------- End : free ----------------*/
1.219 brouard 11510: if (mobilav!=0 ||mobilavproj !=0)
11511: 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 11512: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11513: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11514: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11515: } /* mle==-3 arrives here for freeing */
1.227 brouard 11516: /* endfree:*/
11517: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11518: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11519: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11520: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11521: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11522: free_matrix(coqvar,1,maxwav,1,n);
11523: free_matrix(covar,0,NCOVMAX,1,n);
11524: free_matrix(matcov,1,npar,1,npar);
11525: free_matrix(hess,1,npar,1,npar);
11526: /*free_vector(delti,1,npar);*/
11527: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11528: free_matrix(agev,1,maxwav,1,imx);
11529: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11530:
11531: free_ivector(ncodemax,1,NCOVMAX);
11532: free_ivector(ncodemaxwundef,1,NCOVMAX);
11533: free_ivector(Dummy,-1,NCOVMAX);
11534: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11535: free_ivector(DummyV,1,NCOVMAX);
11536: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11537: free_ivector(Typevar,-1,NCOVMAX);
11538: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11539: free_ivector(TvarsQ,1,NCOVMAX);
11540: free_ivector(TvarsQind,1,NCOVMAX);
11541: free_ivector(TvarsD,1,NCOVMAX);
11542: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11543: free_ivector(TvarFD,1,NCOVMAX);
11544: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11545: free_ivector(TvarF,1,NCOVMAX);
11546: free_ivector(TvarFind,1,NCOVMAX);
11547: free_ivector(TvarV,1,NCOVMAX);
11548: free_ivector(TvarVind,1,NCOVMAX);
11549: free_ivector(TvarA,1,NCOVMAX);
11550: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11551: free_ivector(TvarFQ,1,NCOVMAX);
11552: free_ivector(TvarFQind,1,NCOVMAX);
11553: free_ivector(TvarVD,1,NCOVMAX);
11554: free_ivector(TvarVDind,1,NCOVMAX);
11555: free_ivector(TvarVQ,1,NCOVMAX);
11556: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11557: free_ivector(Tvarsel,1,NCOVMAX);
11558: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11559: free_ivector(Tposprod,1,NCOVMAX);
11560: free_ivector(Tprod,1,NCOVMAX);
11561: free_ivector(Tvaraff,1,NCOVMAX);
11562: free_ivector(invalidvarcomb,1,ncovcombmax);
11563: free_ivector(Tage,1,NCOVMAX);
11564: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11565: free_ivector(TmodelInvind,1,NCOVMAX);
11566: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11567:
11568: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11569: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11570: fflush(fichtm);
11571: fflush(ficgp);
11572:
1.227 brouard 11573:
1.126 brouard 11574: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11575: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11576: 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 11577: }else{
11578: printf("End of Imach\n");
11579: fprintf(ficlog,"End of Imach\n");
11580: }
11581: printf("See log file on %s\n",filelog);
11582: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11583: /*(void) gettimeofday(&end_time,&tzp);*/
11584: rend_time = time(NULL);
11585: end_time = *localtime(&rend_time);
11586: /* tml = *localtime(&end_time.tm_sec); */
11587: strcpy(strtend,asctime(&end_time));
1.126 brouard 11588: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11589: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11590: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11591:
1.157 brouard 11592: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11593: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11594: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11595: /* printf("Total time was %d uSec.\n", total_usecs);*/
11596: /* if(fileappend(fichtm,optionfilehtm)){ */
11597: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11598: fclose(fichtm);
11599: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11600: fclose(fichtmcov);
11601: fclose(ficgp);
11602: fclose(ficlog);
11603: /*------ End -----------*/
1.227 brouard 11604:
11605:
11606: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11607: #ifdef WIN32
1.227 brouard 11608: if (_chdir(pathcd) != 0)
11609: printf("Can't move to directory %s!\n",path);
11610: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11611: #else
1.227 brouard 11612: if(chdir(pathcd) != 0)
11613: printf("Can't move to directory %s!\n", path);
11614: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11615: #endif
1.126 brouard 11616: printf("Current directory %s!\n",pathcd);
11617: /*strcat(plotcmd,CHARSEPARATOR);*/
11618: sprintf(plotcmd,"gnuplot");
1.157 brouard 11619: #ifdef _WIN32
1.126 brouard 11620: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11621: #endif
11622: if(!stat(plotcmd,&info)){
1.158 brouard 11623: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11624: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11625: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11626: }else
11627: strcpy(pplotcmd,plotcmd);
1.157 brouard 11628: #ifdef __unix
1.126 brouard 11629: strcpy(plotcmd,GNUPLOTPROGRAM);
11630: if(!stat(plotcmd,&info)){
1.158 brouard 11631: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11632: }else
11633: strcpy(pplotcmd,plotcmd);
11634: #endif
11635: }else
11636: strcpy(pplotcmd,plotcmd);
11637:
11638: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11639: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11640:
1.126 brouard 11641: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11642: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11643: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11644: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11645: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11646: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11647: }
1.158 brouard 11648: printf(" Successful, please wait...");
1.126 brouard 11649: while (z[0] != 'q') {
11650: /* chdir(path); */
1.154 brouard 11651: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11652: scanf("%s",z);
11653: /* if (z[0] == 'c') system("./imach"); */
11654: if (z[0] == 'e') {
1.158 brouard 11655: #ifdef __APPLE__
1.152 brouard 11656: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11657: #elif __linux
11658: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11659: #else
1.152 brouard 11660: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11661: #endif
11662: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11663: system(pplotcmd);
1.126 brouard 11664: }
11665: else if (z[0] == 'g') system(plotcmd);
11666: else if (z[0] == 'q') exit(0);
11667: }
1.227 brouard 11668: end:
1.126 brouard 11669: while (z[0] != 'q') {
1.195 brouard 11670: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11671: scanf("%s",z);
11672: }
11673: }
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