Annotation of imach/src/imach.c, revision 1.254
1.254 ! brouard 1: /* $Id: imach.c,v 1.253 2016/12/15 11:59:41 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.254 ! brouard 4: Revision 1.253 2016/12/15 11:59:41 brouard
! 5: Summary: 0.99 in progress
! 6:
1.253 brouard 7: Revision 1.252 2016/09/15 21:15:37 brouard
8: *** empty log message ***
9:
1.252 brouard 10: Revision 1.251 2016/09/15 15:01:13 brouard
11: Summary: not working
12:
1.251 brouard 13: Revision 1.250 2016/09/08 16:07:27 brouard
14: Summary: continue
15:
1.250 brouard 16: Revision 1.249 2016/09/07 17:14:18 brouard
17: Summary: Starting values from frequencies
18:
1.249 brouard 19: Revision 1.248 2016/09/07 14:10:18 brouard
20: *** empty log message ***
21:
1.248 brouard 22: Revision 1.247 2016/09/02 11:11:21 brouard
23: *** empty log message ***
24:
1.247 brouard 25: Revision 1.246 2016/09/02 08:49:22 brouard
26: *** empty log message ***
27:
1.246 brouard 28: Revision 1.245 2016/09/02 07:25:01 brouard
29: *** empty log message ***
30:
1.245 brouard 31: Revision 1.244 2016/09/02 07:17:34 brouard
32: *** empty log message ***
33:
1.244 brouard 34: Revision 1.243 2016/09/02 06:45:35 brouard
35: *** empty log message ***
36:
1.243 brouard 37: Revision 1.242 2016/08/30 15:01:20 brouard
38: Summary: Fixing a lots
39:
1.242 brouard 40: Revision 1.241 2016/08/29 17:17:25 brouard
41: Summary: gnuplot problem in Back projection to fix
42:
1.241 brouard 43: Revision 1.240 2016/08/29 07:53:18 brouard
44: Summary: Better
45:
1.240 brouard 46: Revision 1.239 2016/08/26 15:51:03 brouard
47: Summary: Improvement in Powell output in order to copy and paste
48:
49: Author:
50:
1.239 brouard 51: Revision 1.238 2016/08/26 14:23:35 brouard
52: Summary: Starting tests of 0.99
53:
1.238 brouard 54: Revision 1.237 2016/08/26 09:20:19 brouard
55: Summary: to valgrind
56:
1.237 brouard 57: Revision 1.236 2016/08/25 10:50:18 brouard
58: *** empty log message ***
59:
1.236 brouard 60: Revision 1.235 2016/08/25 06:59:23 brouard
61: *** empty log message ***
62:
1.235 brouard 63: Revision 1.234 2016/08/23 16:51:20 brouard
64: *** empty log message ***
65:
1.234 brouard 66: Revision 1.233 2016/08/23 07:40:50 brouard
67: Summary: not working
68:
1.233 brouard 69: Revision 1.232 2016/08/22 14:20:21 brouard
70: Summary: not working
71:
1.232 brouard 72: Revision 1.231 2016/08/22 07:17:15 brouard
73: Summary: not working
74:
1.231 brouard 75: Revision 1.230 2016/08/22 06:55:53 brouard
76: Summary: Not working
77:
1.230 brouard 78: Revision 1.229 2016/07/23 09:45:53 brouard
79: Summary: Completing for func too
80:
1.229 brouard 81: Revision 1.228 2016/07/22 17:45:30 brouard
82: Summary: Fixing some arrays, still debugging
83:
1.227 brouard 84: Revision 1.226 2016/07/12 18:42:34 brouard
85: Summary: temp
86:
1.226 brouard 87: Revision 1.225 2016/07/12 08:40:03 brouard
88: Summary: saving but not running
89:
1.225 brouard 90: Revision 1.224 2016/07/01 13:16:01 brouard
91: Summary: Fixes
92:
1.224 brouard 93: Revision 1.223 2016/02/19 09:23:35 brouard
94: Summary: temporary
95:
1.223 brouard 96: Revision 1.222 2016/02/17 08:14:50 brouard
97: Summary: Probably last 0.98 stable version 0.98r6
98:
1.222 brouard 99: Revision 1.221 2016/02/15 23:35:36 brouard
100: Summary: minor bug
101:
1.220 brouard 102: Revision 1.219 2016/02/15 00:48:12 brouard
103: *** empty log message ***
104:
1.219 brouard 105: Revision 1.218 2016/02/12 11:29:23 brouard
106: Summary: 0.99 Back projections
107:
1.218 brouard 108: Revision 1.217 2015/12/23 17:18:31 brouard
109: Summary: Experimental backcast
110:
1.217 brouard 111: Revision 1.216 2015/12/18 17:32:11 brouard
112: Summary: 0.98r4 Warning and status=-2
113:
114: Version 0.98r4 is now:
115: - displaying an error when status is -1, date of interview unknown and date of death known;
116: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
117: Older changes concerning s=-2, dating from 2005 have been supersed.
118:
1.216 brouard 119: Revision 1.215 2015/12/16 08:52:24 brouard
120: Summary: 0.98r4 working
121:
1.215 brouard 122: Revision 1.214 2015/12/16 06:57:54 brouard
123: Summary: temporary not working
124:
1.214 brouard 125: Revision 1.213 2015/12/11 18:22:17 brouard
126: Summary: 0.98r4
127:
1.213 brouard 128: Revision 1.212 2015/11/21 12:47:24 brouard
129: Summary: minor typo
130:
1.212 brouard 131: Revision 1.211 2015/11/21 12:41:11 brouard
132: Summary: 0.98r3 with some graph of projected cross-sectional
133:
134: Author: Nicolas Brouard
135:
1.211 brouard 136: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 137: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 138: Summary: Adding ftolpl parameter
139: Author: N Brouard
140:
141: We had difficulties to get smoothed confidence intervals. It was due
142: to the period prevalence which wasn't computed accurately. The inner
143: parameter ftolpl is now an outer parameter of the .imach parameter
144: file after estepm. If ftolpl is small 1.e-4 and estepm too,
145: computation are long.
146:
1.209 brouard 147: Revision 1.208 2015/11/17 14:31:57 brouard
148: Summary: temporary
149:
1.208 brouard 150: Revision 1.207 2015/10/27 17:36:57 brouard
151: *** empty log message ***
152:
1.207 brouard 153: Revision 1.206 2015/10/24 07:14:11 brouard
154: *** empty log message ***
155:
1.206 brouard 156: Revision 1.205 2015/10/23 15:50:53 brouard
157: Summary: 0.98r3 some clarification for graphs on likelihood contributions
158:
1.205 brouard 159: Revision 1.204 2015/10/01 16:20:26 brouard
160: Summary: Some new graphs of contribution to likelihood
161:
1.204 brouard 162: Revision 1.203 2015/09/30 17:45:14 brouard
163: Summary: looking at better estimation of the hessian
164:
165: Also a better criteria for convergence to the period prevalence And
166: therefore adding the number of years needed to converge. (The
167: prevalence in any alive state shold sum to one
168:
1.203 brouard 169: Revision 1.202 2015/09/22 19:45:16 brouard
170: Summary: Adding some overall graph on contribution to likelihood. Might change
171:
1.202 brouard 172: Revision 1.201 2015/09/15 17:34:58 brouard
173: Summary: 0.98r0
174:
175: - Some new graphs like suvival functions
176: - Some bugs fixed like model=1+age+V2.
177:
1.201 brouard 178: Revision 1.200 2015/09/09 16:53:55 brouard
179: Summary: Big bug thanks to Flavia
180:
181: Even model=1+age+V2. did not work anymore
182:
1.200 brouard 183: Revision 1.199 2015/09/07 14:09:23 brouard
184: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
185:
1.199 brouard 186: Revision 1.198 2015/09/03 07:14:39 brouard
187: Summary: 0.98q5 Flavia
188:
1.198 brouard 189: Revision 1.197 2015/09/01 18:24:39 brouard
190: *** empty log message ***
191:
1.197 brouard 192: Revision 1.196 2015/08/18 23:17:52 brouard
193: Summary: 0.98q5
194:
1.196 brouard 195: Revision 1.195 2015/08/18 16:28:39 brouard
196: Summary: Adding a hack for testing purpose
197:
198: After reading the title, ftol and model lines, if the comment line has
199: a q, starting with #q, the answer at the end of the run is quit. It
200: permits to run test files in batch with ctest. The former workaround was
201: $ echo q | imach foo.imach
202:
1.195 brouard 203: Revision 1.194 2015/08/18 13:32:00 brouard
204: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
205:
1.194 brouard 206: Revision 1.193 2015/08/04 07:17:42 brouard
207: Summary: 0.98q4
208:
1.193 brouard 209: Revision 1.192 2015/07/16 16:49:02 brouard
210: Summary: Fixing some outputs
211:
1.192 brouard 212: Revision 1.191 2015/07/14 10:00:33 brouard
213: Summary: Some fixes
214:
1.191 brouard 215: Revision 1.190 2015/05/05 08:51:13 brouard
216: Summary: Adding digits in output parameters (7 digits instead of 6)
217:
218: Fix 1+age+.
219:
1.190 brouard 220: Revision 1.189 2015/04/30 14:45:16 brouard
221: Summary: 0.98q2
222:
1.189 brouard 223: Revision 1.188 2015/04/30 08:27:53 brouard
224: *** empty log message ***
225:
1.188 brouard 226: Revision 1.187 2015/04/29 09:11:15 brouard
227: *** empty log message ***
228:
1.187 brouard 229: Revision 1.186 2015/04/23 12:01:52 brouard
230: Summary: V1*age is working now, version 0.98q1
231:
232: Some codes had been disabled in order to simplify and Vn*age was
233: working in the optimization phase, ie, giving correct MLE parameters,
234: but, as usual, outputs were not correct and program core dumped.
235:
1.186 brouard 236: Revision 1.185 2015/03/11 13:26:42 brouard
237: Summary: Inclusion of compile and links command line for Intel Compiler
238:
1.185 brouard 239: Revision 1.184 2015/03/11 11:52:39 brouard
240: Summary: Back from Windows 8. Intel Compiler
241:
1.184 brouard 242: Revision 1.183 2015/03/10 20:34:32 brouard
243: Summary: 0.98q0, trying with directest, mnbrak fixed
244:
245: We use directest instead of original Powell test; probably no
246: incidence on the results, but better justifications;
247: We fixed Numerical Recipes mnbrak routine which was wrong and gave
248: wrong results.
249:
1.183 brouard 250: Revision 1.182 2015/02/12 08:19:57 brouard
251: Summary: Trying to keep directest which seems simpler and more general
252: Author: Nicolas Brouard
253:
1.182 brouard 254: Revision 1.181 2015/02/11 23:22:24 brouard
255: Summary: Comments on Powell added
256:
257: Author:
258:
1.181 brouard 259: Revision 1.180 2015/02/11 17:33:45 brouard
260: Summary: Finishing move from main to function (hpijx and prevalence_limit)
261:
1.180 brouard 262: Revision 1.179 2015/01/04 09:57:06 brouard
263: Summary: back to OS/X
264:
1.179 brouard 265: Revision 1.178 2015/01/04 09:35:48 brouard
266: *** empty log message ***
267:
1.178 brouard 268: Revision 1.177 2015/01/03 18:40:56 brouard
269: Summary: Still testing ilc32 on OSX
270:
1.177 brouard 271: Revision 1.176 2015/01/03 16:45:04 brouard
272: *** empty log message ***
273:
1.176 brouard 274: Revision 1.175 2015/01/03 16:33:42 brouard
275: *** empty log message ***
276:
1.175 brouard 277: Revision 1.174 2015/01/03 16:15:49 brouard
278: Summary: Still in cross-compilation
279:
1.174 brouard 280: Revision 1.173 2015/01/03 12:06:26 brouard
281: Summary: trying to detect cross-compilation
282:
1.173 brouard 283: Revision 1.172 2014/12/27 12:07:47 brouard
284: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
285:
1.172 brouard 286: Revision 1.171 2014/12/23 13:26:59 brouard
287: Summary: Back from Visual C
288:
289: Still problem with utsname.h on Windows
290:
1.171 brouard 291: Revision 1.170 2014/12/23 11:17:12 brouard
292: Summary: Cleaning some \%% back to %%
293:
294: The escape was mandatory for a specific compiler (which one?), but too many warnings.
295:
1.170 brouard 296: Revision 1.169 2014/12/22 23:08:31 brouard
297: Summary: 0.98p
298:
299: Outputs some informations on compiler used, OS etc. Testing on different platforms.
300:
1.169 brouard 301: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 302: Summary: update
1.169 brouard 303:
1.168 brouard 304: Revision 1.167 2014/12/22 13:50:56 brouard
305: Summary: Testing uname and compiler version and if compiled 32 or 64
306:
307: Testing on Linux 64
308:
1.167 brouard 309: Revision 1.166 2014/12/22 11:40:47 brouard
310: *** empty log message ***
311:
1.166 brouard 312: Revision 1.165 2014/12/16 11:20:36 brouard
313: Summary: After compiling on Visual C
314:
315: * imach.c (Module): Merging 1.61 to 1.162
316:
1.165 brouard 317: Revision 1.164 2014/12/16 10:52:11 brouard
318: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
319:
320: * imach.c (Module): Merging 1.61 to 1.162
321:
1.164 brouard 322: Revision 1.163 2014/12/16 10:30:11 brouard
323: * imach.c (Module): Merging 1.61 to 1.162
324:
1.163 brouard 325: Revision 1.162 2014/09/25 11:43:39 brouard
326: Summary: temporary backup 0.99!
327:
1.162 brouard 328: Revision 1.1 2014/09/16 11:06:58 brouard
329: Summary: With some code (wrong) for nlopt
330:
331: Author:
332:
333: Revision 1.161 2014/09/15 20:41:41 brouard
334: Summary: Problem with macro SQR on Intel compiler
335:
1.161 brouard 336: Revision 1.160 2014/09/02 09:24:05 brouard
337: *** empty log message ***
338:
1.160 brouard 339: Revision 1.159 2014/09/01 10:34:10 brouard
340: Summary: WIN32
341: Author: Brouard
342:
1.159 brouard 343: Revision 1.158 2014/08/27 17:11:51 brouard
344: *** empty log message ***
345:
1.158 brouard 346: Revision 1.157 2014/08/27 16:26:55 brouard
347: Summary: Preparing windows Visual studio version
348: Author: Brouard
349:
350: In order to compile on Visual studio, time.h is now correct and time_t
351: and tm struct should be used. difftime should be used but sometimes I
352: just make the differences in raw time format (time(&now).
353: Trying to suppress #ifdef LINUX
354: Add xdg-open for __linux in order to open default browser.
355:
1.157 brouard 356: Revision 1.156 2014/08/25 20:10:10 brouard
357: *** empty log message ***
358:
1.156 brouard 359: Revision 1.155 2014/08/25 18:32:34 brouard
360: Summary: New compile, minor changes
361: Author: Brouard
362:
1.155 brouard 363: Revision 1.154 2014/06/20 17:32:08 brouard
364: Summary: Outputs now all graphs of convergence to period prevalence
365:
1.154 brouard 366: Revision 1.153 2014/06/20 16:45:46 brouard
367: Summary: If 3 live state, convergence to period prevalence on same graph
368: Author: Brouard
369:
1.153 brouard 370: Revision 1.152 2014/06/18 17:54:09 brouard
371: Summary: open browser, use gnuplot on same dir than imach if not found in the path
372:
1.152 brouard 373: Revision 1.151 2014/06/18 16:43:30 brouard
374: *** empty log message ***
375:
1.151 brouard 376: Revision 1.150 2014/06/18 16:42:35 brouard
377: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
378: Author: brouard
379:
1.150 brouard 380: Revision 1.149 2014/06/18 15:51:14 brouard
381: Summary: Some fixes in parameter files errors
382: Author: Nicolas Brouard
383:
1.149 brouard 384: Revision 1.148 2014/06/17 17:38:48 brouard
385: Summary: Nothing new
386: Author: Brouard
387:
388: Just a new packaging for OS/X version 0.98nS
389:
1.148 brouard 390: Revision 1.147 2014/06/16 10:33:11 brouard
391: *** empty log message ***
392:
1.147 brouard 393: Revision 1.146 2014/06/16 10:20:28 brouard
394: Summary: Merge
395: Author: Brouard
396:
397: Merge, before building revised version.
398:
1.146 brouard 399: Revision 1.145 2014/06/10 21:23:15 brouard
400: Summary: Debugging with valgrind
401: Author: Nicolas Brouard
402:
403: Lot of changes in order to output the results with some covariates
404: After the Edimburgh REVES conference 2014, it seems mandatory to
405: improve the code.
406: No more memory valgrind error but a lot has to be done in order to
407: continue the work of splitting the code into subroutines.
408: Also, decodemodel has been improved. Tricode is still not
409: optimal. nbcode should be improved. Documentation has been added in
410: the source code.
411:
1.144 brouard 412: Revision 1.143 2014/01/26 09:45:38 brouard
413: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
414:
415: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
416: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
417:
1.143 brouard 418: Revision 1.142 2014/01/26 03:57:36 brouard
419: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
420:
421: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
422:
1.142 brouard 423: Revision 1.141 2014/01/26 02:42:01 brouard
424: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
425:
1.141 brouard 426: Revision 1.140 2011/09/02 10:37:54 brouard
427: Summary: times.h is ok with mingw32 now.
428:
1.140 brouard 429: Revision 1.139 2010/06/14 07:50:17 brouard
430: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
431: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
432:
1.139 brouard 433: Revision 1.138 2010/04/30 18:19:40 brouard
434: *** empty log message ***
435:
1.138 brouard 436: Revision 1.137 2010/04/29 18:11:38 brouard
437: (Module): Checking covariates for more complex models
438: than V1+V2. A lot of change to be done. Unstable.
439:
1.137 brouard 440: Revision 1.136 2010/04/26 20:30:53 brouard
441: (Module): merging some libgsl code. Fixing computation
442: of likelione (using inter/intrapolation if mle = 0) in order to
443: get same likelihood as if mle=1.
444: Some cleaning of code and comments added.
445:
1.136 brouard 446: Revision 1.135 2009/10/29 15:33:14 brouard
447: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
448:
1.135 brouard 449: Revision 1.134 2009/10/29 13:18:53 brouard
450: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
451:
1.134 brouard 452: Revision 1.133 2009/07/06 10:21:25 brouard
453: just nforces
454:
1.133 brouard 455: Revision 1.132 2009/07/06 08:22:05 brouard
456: Many tings
457:
1.132 brouard 458: Revision 1.131 2009/06/20 16:22:47 brouard
459: Some dimensions resccaled
460:
1.131 brouard 461: Revision 1.130 2009/05/26 06:44:34 brouard
462: (Module): Max Covariate is now set to 20 instead of 8. A
463: lot of cleaning with variables initialized to 0. Trying to make
464: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
465:
1.130 brouard 466: Revision 1.129 2007/08/31 13:49:27 lievre
467: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
468:
1.129 lievre 469: Revision 1.128 2006/06/30 13:02:05 brouard
470: (Module): Clarifications on computing e.j
471:
1.128 brouard 472: Revision 1.127 2006/04/28 18:11:50 brouard
473: (Module): Yes the sum of survivors was wrong since
474: imach-114 because nhstepm was no more computed in the age
475: loop. Now we define nhstepma in the age loop.
476: (Module): In order to speed up (in case of numerous covariates) we
477: compute health expectancies (without variances) in a first step
478: and then all the health expectancies with variances or standard
479: deviation (needs data from the Hessian matrices) which slows the
480: computation.
481: In the future we should be able to stop the program is only health
482: expectancies and graph are needed without standard deviations.
483:
1.127 brouard 484: Revision 1.126 2006/04/28 17:23:28 brouard
485: (Module): Yes the sum of survivors was wrong since
486: imach-114 because nhstepm was no more computed in the age
487: loop. Now we define nhstepma in the age loop.
488: Version 0.98h
489:
1.126 brouard 490: Revision 1.125 2006/04/04 15:20:31 lievre
491: Errors in calculation of health expectancies. Age was not initialized.
492: Forecasting file added.
493:
494: Revision 1.124 2006/03/22 17:13:53 lievre
495: Parameters are printed with %lf instead of %f (more numbers after the comma).
496: The log-likelihood is printed in the log file
497:
498: Revision 1.123 2006/03/20 10:52:43 brouard
499: * imach.c (Module): <title> changed, corresponds to .htm file
500: name. <head> headers where missing.
501:
502: * imach.c (Module): Weights can have a decimal point as for
503: English (a comma might work with a correct LC_NUMERIC environment,
504: otherwise the weight is truncated).
505: Modification of warning when the covariates values are not 0 or
506: 1.
507: Version 0.98g
508:
509: Revision 1.122 2006/03/20 09:45:41 brouard
510: (Module): Weights can have a decimal point as for
511: English (a comma might work with a correct LC_NUMERIC environment,
512: otherwise the weight is truncated).
513: Modification of warning when the covariates values are not 0 or
514: 1.
515: Version 0.98g
516:
517: Revision 1.121 2006/03/16 17:45:01 lievre
518: * imach.c (Module): Comments concerning covariates added
519:
520: * imach.c (Module): refinements in the computation of lli if
521: status=-2 in order to have more reliable computation if stepm is
522: not 1 month. Version 0.98f
523:
524: Revision 1.120 2006/03/16 15:10:38 lievre
525: (Module): refinements in the computation of lli if
526: status=-2 in order to have more reliable computation if stepm is
527: not 1 month. Version 0.98f
528:
529: Revision 1.119 2006/03/15 17:42:26 brouard
530: (Module): Bug if status = -2, the loglikelihood was
531: computed as likelihood omitting the logarithm. Version O.98e
532:
533: Revision 1.118 2006/03/14 18:20:07 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 1.117 2006/03/14 17:16:22 brouard
541: (Module): varevsij Comments added explaining the second
542: table of variances if popbased=1 .
543: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
544: (Module): Function pstamp added
545: (Module): Version 0.98d
546:
547: Revision 1.116 2006/03/06 10:29:27 brouard
548: (Module): Variance-covariance wrong links and
549: varian-covariance of ej. is needed (Saito).
550:
551: Revision 1.115 2006/02/27 12:17:45 brouard
552: (Module): One freematrix added in mlikeli! 0.98c
553:
554: Revision 1.114 2006/02/26 12:57:58 brouard
555: (Module): Some improvements in processing parameter
556: filename with strsep.
557:
558: Revision 1.113 2006/02/24 14:20:24 brouard
559: (Module): Memory leaks checks with valgrind and:
560: datafile was not closed, some imatrix were not freed and on matrix
561: allocation too.
562:
563: Revision 1.112 2006/01/30 09:55:26 brouard
564: (Module): Back to gnuplot.exe instead of wgnuplot.exe
565:
566: Revision 1.111 2006/01/25 20:38:18 brouard
567: (Module): Lots of cleaning and bugs added (Gompertz)
568: (Module): Comments can be added in data file. Missing date values
569: can be a simple dot '.'.
570:
571: Revision 1.110 2006/01/25 00:51:50 brouard
572: (Module): Lots of cleaning and bugs added (Gompertz)
573:
574: Revision 1.109 2006/01/24 19:37:15 brouard
575: (Module): Comments (lines starting with a #) are allowed in data.
576:
577: Revision 1.108 2006/01/19 18:05:42 lievre
578: Gnuplot problem appeared...
579: To be fixed
580:
581: Revision 1.107 2006/01/19 16:20:37 brouard
582: Test existence of gnuplot in imach path
583:
584: Revision 1.106 2006/01/19 13:24:36 brouard
585: Some cleaning and links added in html output
586:
587: Revision 1.105 2006/01/05 20:23:19 lievre
588: *** empty log message ***
589:
590: Revision 1.104 2005/09/30 16:11:43 lievre
591: (Module): sump fixed, loop imx fixed, and simplifications.
592: (Module): If the status is missing at the last wave but we know
593: that the person is alive, then we can code his/her status as -2
594: (instead of missing=-1 in earlier versions) and his/her
595: contributions to the likelihood is 1 - Prob of dying from last
596: health status (= 1-p13= p11+p12 in the easiest case of somebody in
597: the healthy state at last known wave). Version is 0.98
598:
599: Revision 1.103 2005/09/30 15:54:49 lievre
600: (Module): sump fixed, loop imx fixed, and simplifications.
601:
602: Revision 1.102 2004/09/15 17:31:30 brouard
603: Add the possibility to read data file including tab characters.
604:
605: Revision 1.101 2004/09/15 10:38:38 brouard
606: Fix on curr_time
607:
608: Revision 1.100 2004/07/12 18:29:06 brouard
609: Add version for Mac OS X. Just define UNIX in Makefile
610:
611: Revision 1.99 2004/06/05 08:57:40 brouard
612: *** empty log message ***
613:
614: Revision 1.98 2004/05/16 15:05:56 brouard
615: New version 0.97 . First attempt to estimate force of mortality
616: directly from the data i.e. without the need of knowing the health
617: state at each age, but using a Gompertz model: log u =a + b*age .
618: This is the basic analysis of mortality and should be done before any
619: other analysis, in order to test if the mortality estimated from the
620: cross-longitudinal survey is different from the mortality estimated
621: from other sources like vital statistic data.
622:
623: The same imach parameter file can be used but the option for mle should be -3.
624:
1.133 brouard 625: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 626: former routines in order to include the new code within the former code.
627:
628: The output is very simple: only an estimate of the intercept and of
629: the slope with 95% confident intervals.
630:
631: Current limitations:
632: A) Even if you enter covariates, i.e. with the
633: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
634: B) There is no computation of Life Expectancy nor Life Table.
635:
636: Revision 1.97 2004/02/20 13:25:42 lievre
637: Version 0.96d. Population forecasting command line is (temporarily)
638: suppressed.
639:
640: Revision 1.96 2003/07/15 15:38:55 brouard
641: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
642: rewritten within the same printf. Workaround: many printfs.
643:
644: Revision 1.95 2003/07/08 07:54:34 brouard
645: * imach.c (Repository):
646: (Repository): Using imachwizard code to output a more meaningful covariance
647: matrix (cov(a12,c31) instead of numbers.
648:
649: Revision 1.94 2003/06/27 13:00:02 brouard
650: Just cleaning
651:
652: Revision 1.93 2003/06/25 16:33:55 brouard
653: (Module): On windows (cygwin) function asctime_r doesn't
654: exist so I changed back to asctime which exists.
655: (Module): Version 0.96b
656:
657: Revision 1.92 2003/06/25 16:30:45 brouard
658: (Module): On windows (cygwin) function asctime_r doesn't
659: exist so I changed back to asctime which exists.
660:
661: Revision 1.91 2003/06/25 15:30:29 brouard
662: * imach.c (Repository): Duplicated warning errors corrected.
663: (Repository): Elapsed time after each iteration is now output. It
664: helps to forecast when convergence will be reached. Elapsed time
665: is stamped in powell. We created a new html file for the graphs
666: concerning matrix of covariance. It has extension -cov.htm.
667:
668: Revision 1.90 2003/06/24 12:34:15 brouard
669: (Module): Some bugs corrected for windows. Also, when
670: mle=-1 a template is output in file "or"mypar.txt with the design
671: of the covariance matrix to be input.
672:
673: Revision 1.89 2003/06/24 12:30:52 brouard
674: (Module): Some bugs corrected for windows. Also, when
675: mle=-1 a template is output in file "or"mypar.txt with the design
676: of the covariance matrix to be input.
677:
678: Revision 1.88 2003/06/23 17:54:56 brouard
679: * 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.
680:
681: Revision 1.87 2003/06/18 12:26:01 brouard
682: Version 0.96
683:
684: Revision 1.86 2003/06/17 20:04:08 brouard
685: (Module): Change position of html and gnuplot routines and added
686: routine fileappend.
687:
688: Revision 1.85 2003/06/17 13:12:43 brouard
689: * imach.c (Repository): Check when date of death was earlier that
690: current date of interview. It may happen when the death was just
691: prior to the death. In this case, dh was negative and likelihood
692: was wrong (infinity). We still send an "Error" but patch by
693: assuming that the date of death was just one stepm after the
694: interview.
695: (Repository): Because some people have very long ID (first column)
696: we changed int to long in num[] and we added a new lvector for
697: memory allocation. But we also truncated to 8 characters (left
698: truncation)
699: (Repository): No more line truncation errors.
700:
701: Revision 1.84 2003/06/13 21:44:43 brouard
702: * imach.c (Repository): Replace "freqsummary" at a correct
703: place. It differs from routine "prevalence" which may be called
704: many times. Probs is memory consuming and must be used with
705: parcimony.
706: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
707:
708: Revision 1.83 2003/06/10 13:39:11 lievre
709: *** empty log message ***
710:
711: Revision 1.82 2003/06/05 15:57:20 brouard
712: Add log in imach.c and fullversion number is now printed.
713:
714: */
715: /*
716: Interpolated Markov Chain
717:
718: Short summary of the programme:
719:
1.227 brouard 720: This program computes Healthy Life Expectancies or State-specific
721: (if states aren't health statuses) Expectancies from
722: cross-longitudinal data. Cross-longitudinal data consist in:
723:
724: -1- a first survey ("cross") where individuals from different ages
725: are interviewed on their health status or degree of disability (in
726: the case of a health survey which is our main interest)
727:
728: -2- at least a second wave of interviews ("longitudinal") which
729: measure each change (if any) in individual health status. Health
730: expectancies are computed from the time spent in each health state
731: according to a model. More health states you consider, more time is
732: necessary to reach the Maximum Likelihood of the parameters involved
733: in the model. The simplest model is the multinomial logistic model
734: where pij is the probability to be observed in state j at the second
735: wave conditional to be observed in state i at the first
736: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
737: etc , where 'age' is age and 'sex' is a covariate. If you want to
738: have a more complex model than "constant and age", you should modify
739: the program where the markup *Covariates have to be included here
740: again* invites you to do it. More covariates you add, slower the
1.126 brouard 741: convergence.
742:
743: The advantage of this computer programme, compared to a simple
744: multinomial logistic model, is clear when the delay between waves is not
745: identical for each individual. Also, if a individual missed an
746: intermediate interview, the information is lost, but taken into
747: account using an interpolation or extrapolation.
748:
749: hPijx is the probability to be observed in state i at age x+h
750: conditional to the observed state i at age x. The delay 'h' can be
751: split into an exact number (nh*stepm) of unobserved intermediate
752: states. This elementary transition (by month, quarter,
753: semester or year) is modelled as a multinomial logistic. The hPx
754: matrix is simply the matrix product of nh*stepm elementary matrices
755: and the contribution of each individual to the likelihood is simply
756: hPijx.
757:
758: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 759: of the life expectancies. It also computes the period (stable) prevalence.
760:
761: Back prevalence and projections:
1.227 brouard 762:
763: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
764: double agemaxpar, double ftolpl, int *ncvyearp, double
765: dateprev1,double dateprev2, int firstpass, int lastpass, int
766: mobilavproj)
767:
768: Computes the back prevalence limit for any combination of
769: covariate values k at any age between ageminpar and agemaxpar and
770: returns it in **bprlim. In the loops,
771:
772: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
773: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
774:
775: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 776: Computes for any combination of covariates k and any age between bage and fage
777: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
778: oldm=oldms;savm=savms;
1.227 brouard 779:
780: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 781: Computes the transition matrix starting at age 'age' over
782: 'nhstepm*hstepm*stepm' months (i.e. until
783: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 784: nhstepm*hstepm matrices.
785:
786: Returns p3mat[i][j][h] after calling
787: p3mat[i][j][h]=matprod2(newm,
788: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
789: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
790: oldm);
1.226 brouard 791:
792: Important routines
793:
794: - func (or funcone), computes logit (pij) distinguishing
795: o fixed variables (single or product dummies or quantitative);
796: o varying variables by:
797: (1) wave (single, product dummies, quantitative),
798: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
799: % fixed dummy (treated) or quantitative (not done because time-consuming);
800: % varying dummy (not done) or quantitative (not done);
801: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
802: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
803: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
804: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
805: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 806:
1.226 brouard 807:
808:
1.133 brouard 809: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
810: Institut national d'études démographiques, Paris.
1.126 brouard 811: This software have been partly granted by Euro-REVES, a concerted action
812: from the European Union.
813: It is copyrighted identically to a GNU software product, ie programme and
814: software can be distributed freely for non commercial use. Latest version
815: can be accessed at http://euroreves.ined.fr/imach .
816:
817: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
818: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
819:
820: **********************************************************************/
821: /*
822: main
823: read parameterfile
824: read datafile
825: concatwav
826: freqsummary
827: if (mle >= 1)
828: mlikeli
829: print results files
830: if mle==1
831: computes hessian
832: read end of parameter file: agemin, agemax, bage, fage, estepm
833: begin-prev-date,...
834: open gnuplot file
835: open html file
1.145 brouard 836: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
837: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
838: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
839: freexexit2 possible for memory heap.
840:
841: h Pij x | pij_nom ficrestpij
842: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
843: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
844: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
845:
846: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
847: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
848: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
849: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
850: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
851:
1.126 brouard 852: forecasting if prevfcast==1 prevforecast call prevalence()
853: health expectancies
854: Variance-covariance of DFLE
855: prevalence()
856: movingaverage()
857: varevsij()
858: if popbased==1 varevsij(,popbased)
859: total life expectancies
860: Variance of period (stable) prevalence
861: end
862: */
863:
1.187 brouard 864: /* #define DEBUG */
865: /* #define DEBUGBRENT */
1.203 brouard 866: /* #define DEBUGLINMIN */
867: /* #define DEBUGHESS */
868: #define DEBUGHESSIJ
1.224 brouard 869: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 870: #define POWELL /* Instead of NLOPT */
1.224 brouard 871: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 872: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
873: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 874:
875: #include <math.h>
876: #include <stdio.h>
877: #include <stdlib.h>
878: #include <string.h>
1.226 brouard 879: #include <ctype.h>
1.159 brouard 880:
881: #ifdef _WIN32
882: #include <io.h>
1.172 brouard 883: #include <windows.h>
884: #include <tchar.h>
1.159 brouard 885: #else
1.126 brouard 886: #include <unistd.h>
1.159 brouard 887: #endif
1.126 brouard 888:
889: #include <limits.h>
890: #include <sys/types.h>
1.171 brouard 891:
892: #if defined(__GNUC__)
893: #include <sys/utsname.h> /* Doesn't work on Windows */
894: #endif
895:
1.126 brouard 896: #include <sys/stat.h>
897: #include <errno.h>
1.159 brouard 898: /* extern int errno; */
1.126 brouard 899:
1.157 brouard 900: /* #ifdef LINUX */
901: /* #include <time.h> */
902: /* #include "timeval.h" */
903: /* #else */
904: /* #include <sys/time.h> */
905: /* #endif */
906:
1.126 brouard 907: #include <time.h>
908:
1.136 brouard 909: #ifdef GSL
910: #include <gsl/gsl_errno.h>
911: #include <gsl/gsl_multimin.h>
912: #endif
913:
1.167 brouard 914:
1.162 brouard 915: #ifdef NLOPT
916: #include <nlopt.h>
917: typedef struct {
918: double (* function)(double [] );
919: } myfunc_data ;
920: #endif
921:
1.126 brouard 922: /* #include <libintl.h> */
923: /* #define _(String) gettext (String) */
924:
1.251 brouard 925: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 926:
927: #define GNUPLOTPROGRAM "gnuplot"
928: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
929: #define FILENAMELENGTH 132
930:
931: #define GLOCK_ERROR_NOPATH -1 /* empty path */
932: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
933:
1.144 brouard 934: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
935: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 936:
937: #define NINTERVMAX 8
1.144 brouard 938: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
939: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
940: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 941: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 942: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
943: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 944: #define MAXN 20000
1.144 brouard 945: #define YEARM 12. /**< Number of months per year */
1.218 brouard 946: /* #define AGESUP 130 */
947: #define AGESUP 150
948: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 949: #define AGEBASE 40
1.194 brouard 950: #define AGEOVERFLOW 1.e20
1.164 brouard 951: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 952: #ifdef _WIN32
953: #define DIRSEPARATOR '\\'
954: #define CHARSEPARATOR "\\"
955: #define ODIRSEPARATOR '/'
956: #else
1.126 brouard 957: #define DIRSEPARATOR '/'
958: #define CHARSEPARATOR "/"
959: #define ODIRSEPARATOR '\\'
960: #endif
961:
1.254 ! brouard 962: /* $Id: imach.c,v 1.253 2016/12/15 11:59:41 brouard Exp $ */
1.126 brouard 963: /* $State: Exp $ */
1.196 brouard 964: #include "version.h"
965: char version[]=__IMACH_VERSION__;
1.224 brouard 966: 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";
1.254 ! brouard 967: char fullversion[]="$Revision: 1.253 $ $Date: 2016/12/15 11:59:41 $";
1.126 brouard 968: char strstart[80];
969: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 970: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 971: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 972: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
973: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
974: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 975: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
976: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 977: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
978: int cptcovprodnoage=0; /**< Number of covariate products without age */
979: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 980: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
981: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 982: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 983: int nsd=0; /**< Total number of single dummy variables (output) */
984: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 985: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 986: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 987: int ntveff=0; /**< ntveff number of effective time varying variables */
988: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 989: int cptcov=0; /* Working variable */
1.218 brouard 990: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 991: int npar=NPARMAX;
992: int nlstate=2; /* Number of live states */
993: int ndeath=1; /* Number of dead states */
1.130 brouard 994: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 995: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 996: int popbased=0;
997:
998: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 999: int maxwav=0; /* Maxim number of waves */
1000: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1001: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1002: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1003: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1004: int mle=1, weightopt=0;
1.126 brouard 1005: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1006: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1007: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1008: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1009: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1010: int selected(int kvar); /* Is covariate kvar selected for printing results */
1011:
1.130 brouard 1012: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1013: double **matprod2(); /* test */
1.126 brouard 1014: double **oldm, **newm, **savm; /* Working pointers to matrices */
1015: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1016: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1017:
1.136 brouard 1018: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1019: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1020: FILE *ficlog, *ficrespow;
1.130 brouard 1021: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1022: double fretone; /* Only one call to likelihood */
1.130 brouard 1023: long ipmx=0; /* Number of contributions */
1.126 brouard 1024: double sw; /* Sum of weights */
1025: char filerespow[FILENAMELENGTH];
1026: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1027: FILE *ficresilk;
1028: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1029: FILE *ficresprobmorprev;
1030: FILE *fichtm, *fichtmcov; /* Html File */
1031: FILE *ficreseij;
1032: char filerese[FILENAMELENGTH];
1033: FILE *ficresstdeij;
1034: char fileresstde[FILENAMELENGTH];
1035: FILE *ficrescveij;
1036: char filerescve[FILENAMELENGTH];
1037: FILE *ficresvij;
1038: char fileresv[FILENAMELENGTH];
1039: FILE *ficresvpl;
1040: char fileresvpl[FILENAMELENGTH];
1041: char title[MAXLINE];
1.234 brouard 1042: char model[MAXLINE]; /**< The model line */
1.217 brouard 1043: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1044: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1045: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1046: char command[FILENAMELENGTH];
1047: int outcmd=0;
1048:
1.217 brouard 1049: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1050: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1051: char filelog[FILENAMELENGTH]; /* Log file */
1052: char filerest[FILENAMELENGTH];
1053: char fileregp[FILENAMELENGTH];
1054: char popfile[FILENAMELENGTH];
1055:
1056: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1057:
1.157 brouard 1058: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1059: /* struct timezone tzp; */
1060: /* extern int gettimeofday(); */
1061: struct tm tml, *gmtime(), *localtime();
1062:
1063: extern time_t time();
1064:
1065: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1066: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1067: struct tm tm;
1068:
1.126 brouard 1069: char strcurr[80], strfor[80];
1070:
1071: char *endptr;
1072: long lval;
1073: double dval;
1074:
1075: #define NR_END 1
1076: #define FREE_ARG char*
1077: #define FTOL 1.0e-10
1078:
1079: #define NRANSI
1.240 brouard 1080: #define ITMAX 200
1081: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1082:
1083: #define TOL 2.0e-4
1084:
1085: #define CGOLD 0.3819660
1086: #define ZEPS 1.0e-10
1087: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1088:
1089: #define GOLD 1.618034
1090: #define GLIMIT 100.0
1091: #define TINY 1.0e-20
1092:
1093: static double maxarg1,maxarg2;
1094: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1095: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1096:
1097: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1098: #define rint(a) floor(a+0.5)
1.166 brouard 1099: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1100: #define mytinydouble 1.0e-16
1.166 brouard 1101: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1102: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1103: /* static double dsqrarg; */
1104: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1105: static double sqrarg;
1106: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1107: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1108: int agegomp= AGEGOMP;
1109:
1110: int imx;
1111: int stepm=1;
1112: /* Stepm, step in month: minimum step interpolation*/
1113:
1114: int estepm;
1115: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1116:
1117: int m,nb;
1118: long *num;
1.197 brouard 1119: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1120: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1121: covariate for which somebody answered excluding
1122: undefined. Usually 2: 0 and 1. */
1123: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1124: covariate for which somebody answered including
1125: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1126: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1127: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1128: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1129: double *ageexmed,*agecens;
1130: double dateintmean=0;
1131:
1132: double *weight;
1133: int **s; /* Status */
1.141 brouard 1134: double *agedc;
1.145 brouard 1135: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1136: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1137: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1138: double **coqvar; /* Fixed quantitative covariate iqv */
1139: double ***cotvar; /* Time varying covariate itv */
1140: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1141: double idx;
1142: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1143: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1144: /*k 1 2 3 4 5 6 7 8 9 */
1145: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1146: /* Tndvar[k] 1 2 3 4 5 */
1147: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1148: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1149: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1150: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1151: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1152: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1153: /* Tprod[i]=k 4 7 */
1154: /* Tage[i]=k 5 8 */
1155: /* */
1156: /* Type */
1157: /* V 1 2 3 4 5 */
1158: /* F F V V V */
1159: /* D Q D D Q */
1160: /* */
1161: int *TvarsD;
1162: int *TvarsDind;
1163: int *TvarsQ;
1164: int *TvarsQind;
1165:
1.235 brouard 1166: #define MAXRESULTLINES 10
1167: int nresult=0;
1168: int TKresult[MAXRESULTLINES];
1.237 brouard 1169: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1170: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1171: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1172: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1173: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1174: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1175:
1.234 brouard 1176: /* 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 1177: 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 */
1178: 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 */
1179: 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 */
1180: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1181: 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 */
1182: 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 1183: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1184: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1185: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1186: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1187: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1188: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1189: 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 */
1190: 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 */
1191:
1.230 brouard 1192: int *Tvarsel; /**< Selected covariates for output */
1193: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1194: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1195: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1196: 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 1197: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1198: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1199: int *Tage;
1.227 brouard 1200: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1201: 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 1202: 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*/
1203: 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 1204: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1205: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1206: int **Tvard;
1207: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1208: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1209: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1210: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1211: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1212: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1213: double *lsurv, *lpop, *tpop;
1214:
1.231 brouard 1215: #define FD 1; /* Fixed dummy covariate */
1216: #define FQ 2; /* Fixed quantitative covariate */
1217: #define FP 3; /* Fixed product covariate */
1218: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1219: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1220: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1221: #define VD 10; /* Varying dummy covariate */
1222: #define VQ 11; /* Varying quantitative covariate */
1223: #define VP 12; /* Varying product covariate */
1224: #define VPDD 13; /* Varying product dummy*dummy covariate */
1225: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1226: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1227: #define APFD 16; /* Age product * fixed dummy covariate */
1228: #define APFQ 17; /* Age product * fixed quantitative covariate */
1229: #define APVD 18; /* Age product * varying dummy covariate */
1230: #define APVQ 19; /* Age product * varying quantitative covariate */
1231:
1232: #define FTYPE 1; /* Fixed covariate */
1233: #define VTYPE 2; /* Varying covariate (loop in wave) */
1234: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1235:
1236: struct kmodel{
1237: int maintype; /* main type */
1238: int subtype; /* subtype */
1239: };
1240: struct kmodel modell[NCOVMAX];
1241:
1.143 brouard 1242: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1243: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1244:
1245: /**************** split *************************/
1246: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1247: {
1248: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1249: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1250: */
1251: char *ss; /* pointer */
1.186 brouard 1252: int l1=0, l2=0; /* length counters */
1.126 brouard 1253:
1254: l1 = strlen(path ); /* length of path */
1255: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1256: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1257: if ( ss == NULL ) { /* no directory, so determine current directory */
1258: strcpy( name, path ); /* we got the fullname name because no directory */
1259: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1260: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1261: /* get current working directory */
1262: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1263: #ifdef WIN32
1264: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1265: #else
1266: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1267: #endif
1.126 brouard 1268: return( GLOCK_ERROR_GETCWD );
1269: }
1270: /* got dirc from getcwd*/
1271: printf(" DIRC = %s \n",dirc);
1.205 brouard 1272: } else { /* strip directory from path */
1.126 brouard 1273: ss++; /* after this, the filename */
1274: l2 = strlen( ss ); /* length of filename */
1275: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1276: strcpy( name, ss ); /* save file name */
1277: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1278: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1279: printf(" DIRC2 = %s \n",dirc);
1280: }
1281: /* We add a separator at the end of dirc if not exists */
1282: l1 = strlen( dirc ); /* length of directory */
1283: if( dirc[l1-1] != DIRSEPARATOR ){
1284: dirc[l1] = DIRSEPARATOR;
1285: dirc[l1+1] = 0;
1286: printf(" DIRC3 = %s \n",dirc);
1287: }
1288: ss = strrchr( name, '.' ); /* find last / */
1289: if (ss >0){
1290: ss++;
1291: strcpy(ext,ss); /* save extension */
1292: l1= strlen( name);
1293: l2= strlen(ss)+1;
1294: strncpy( finame, name, l1-l2);
1295: finame[l1-l2]= 0;
1296: }
1297:
1298: return( 0 ); /* we're done */
1299: }
1300:
1301:
1302: /******************************************/
1303:
1304: void replace_back_to_slash(char *s, char*t)
1305: {
1306: int i;
1307: int lg=0;
1308: i=0;
1309: lg=strlen(t);
1310: for(i=0; i<= lg; i++) {
1311: (s[i] = t[i]);
1312: if (t[i]== '\\') s[i]='/';
1313: }
1314: }
1315:
1.132 brouard 1316: char *trimbb(char *out, char *in)
1.137 brouard 1317: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1318: char *s;
1319: s=out;
1320: while (*in != '\0'){
1.137 brouard 1321: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1322: in++;
1323: }
1324: *out++ = *in++;
1325: }
1326: *out='\0';
1327: return s;
1328: }
1329:
1.187 brouard 1330: /* char *substrchaine(char *out, char *in, char *chain) */
1331: /* { */
1332: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1333: /* char *s, *t; */
1334: /* t=in;s=out; */
1335: /* while ((*in != *chain) && (*in != '\0')){ */
1336: /* *out++ = *in++; */
1337: /* } */
1338:
1339: /* /\* *in matches *chain *\/ */
1340: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1341: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1342: /* } */
1343: /* in--; chain--; */
1344: /* while ( (*in != '\0')){ */
1345: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1346: /* *out++ = *in++; */
1347: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1348: /* } */
1349: /* *out='\0'; */
1350: /* out=s; */
1351: /* return out; */
1352: /* } */
1353: char *substrchaine(char *out, char *in, char *chain)
1354: {
1355: /* Substract chain 'chain' from 'in', return and output 'out' */
1356: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1357:
1358: char *strloc;
1359:
1360: strcpy (out, in);
1361: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1362: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1363: if(strloc != NULL){
1364: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1365: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1366: /* strcpy (strloc, strloc +strlen(chain));*/
1367: }
1368: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1369: return out;
1370: }
1371:
1372:
1.145 brouard 1373: char *cutl(char *blocc, char *alocc, char *in, char occ)
1374: {
1.187 brouard 1375: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1376: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1377: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1378: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1379: */
1.160 brouard 1380: char *s, *t;
1.145 brouard 1381: t=in;s=in;
1382: while ((*in != occ) && (*in != '\0')){
1383: *alocc++ = *in++;
1384: }
1385: if( *in == occ){
1386: *(alocc)='\0';
1387: s=++in;
1388: }
1389:
1390: if (s == t) {/* occ not found */
1391: *(alocc-(in-s))='\0';
1392: in=s;
1393: }
1394: while ( *in != '\0'){
1395: *blocc++ = *in++;
1396: }
1397:
1398: *blocc='\0';
1399: return t;
1400: }
1.137 brouard 1401: char *cutv(char *blocc, char *alocc, char *in, char occ)
1402: {
1.187 brouard 1403: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1404: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1405: gives blocc="abcdef2ghi" and alocc="j".
1406: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1407: */
1408: char *s, *t;
1409: t=in;s=in;
1410: while (*in != '\0'){
1411: while( *in == occ){
1412: *blocc++ = *in++;
1413: s=in;
1414: }
1415: *blocc++ = *in++;
1416: }
1417: if (s == t) /* occ not found */
1418: *(blocc-(in-s))='\0';
1419: else
1420: *(blocc-(in-s)-1)='\0';
1421: in=s;
1422: while ( *in != '\0'){
1423: *alocc++ = *in++;
1424: }
1425:
1426: *alocc='\0';
1427: return s;
1428: }
1429:
1.126 brouard 1430: int nbocc(char *s, char occ)
1431: {
1432: int i,j=0;
1433: int lg=20;
1434: i=0;
1435: lg=strlen(s);
1436: for(i=0; i<= lg; i++) {
1.234 brouard 1437: if (s[i] == occ ) j++;
1.126 brouard 1438: }
1439: return j;
1440: }
1441:
1.137 brouard 1442: /* void cutv(char *u,char *v, char*t, char occ) */
1443: /* { */
1444: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1445: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1446: /* gives u="abcdef2ghi" and v="j" *\/ */
1447: /* int i,lg,j,p=0; */
1448: /* i=0; */
1449: /* lg=strlen(t); */
1450: /* for(j=0; j<=lg-1; j++) { */
1451: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1452: /* } */
1.126 brouard 1453:
1.137 brouard 1454: /* for(j=0; j<p; j++) { */
1455: /* (u[j] = t[j]); */
1456: /* } */
1457: /* u[p]='\0'; */
1.126 brouard 1458:
1.137 brouard 1459: /* for(j=0; j<= lg; j++) { */
1460: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1461: /* } */
1462: /* } */
1.126 brouard 1463:
1.160 brouard 1464: #ifdef _WIN32
1465: char * strsep(char **pp, const char *delim)
1466: {
1467: char *p, *q;
1468:
1469: if ((p = *pp) == NULL)
1470: return 0;
1471: if ((q = strpbrk (p, delim)) != NULL)
1472: {
1473: *pp = q + 1;
1474: *q = '\0';
1475: }
1476: else
1477: *pp = 0;
1478: return p;
1479: }
1480: #endif
1481:
1.126 brouard 1482: /********************** nrerror ********************/
1483:
1484: void nrerror(char error_text[])
1485: {
1486: fprintf(stderr,"ERREUR ...\n");
1487: fprintf(stderr,"%s\n",error_text);
1488: exit(EXIT_FAILURE);
1489: }
1490: /*********************** vector *******************/
1491: double *vector(int nl, int nh)
1492: {
1493: double *v;
1494: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1495: if (!v) nrerror("allocation failure in vector");
1496: return v-nl+NR_END;
1497: }
1498:
1499: /************************ free vector ******************/
1500: void free_vector(double*v, int nl, int nh)
1501: {
1502: free((FREE_ARG)(v+nl-NR_END));
1503: }
1504:
1505: /************************ivector *******************************/
1506: int *ivector(long nl,long nh)
1507: {
1508: int *v;
1509: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1510: if (!v) nrerror("allocation failure in ivector");
1511: return v-nl+NR_END;
1512: }
1513:
1514: /******************free ivector **************************/
1515: void free_ivector(int *v, long nl, long nh)
1516: {
1517: free((FREE_ARG)(v+nl-NR_END));
1518: }
1519:
1520: /************************lvector *******************************/
1521: long *lvector(long nl,long nh)
1522: {
1523: long *v;
1524: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1525: if (!v) nrerror("allocation failure in ivector");
1526: return v-nl+NR_END;
1527: }
1528:
1529: /******************free lvector **************************/
1530: void free_lvector(long *v, long nl, long nh)
1531: {
1532: free((FREE_ARG)(v+nl-NR_END));
1533: }
1534:
1535: /******************* imatrix *******************************/
1536: int **imatrix(long nrl, long nrh, long ncl, long nch)
1537: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1538: {
1539: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1540: int **m;
1541:
1542: /* allocate pointers to rows */
1543: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1544: if (!m) nrerror("allocation failure 1 in matrix()");
1545: m += NR_END;
1546: m -= nrl;
1547:
1548:
1549: /* allocate rows and set pointers to them */
1550: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1551: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1552: m[nrl] += NR_END;
1553: m[nrl] -= ncl;
1554:
1555: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1556:
1557: /* return pointer to array of pointers to rows */
1558: return m;
1559: }
1560:
1561: /****************** free_imatrix *************************/
1562: void free_imatrix(m,nrl,nrh,ncl,nch)
1563: int **m;
1564: long nch,ncl,nrh,nrl;
1565: /* free an int matrix allocated by imatrix() */
1566: {
1567: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1568: free((FREE_ARG) (m+nrl-NR_END));
1569: }
1570:
1571: /******************* matrix *******************************/
1572: double **matrix(long nrl, long nrh, long ncl, long nch)
1573: {
1574: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1575: double **m;
1576:
1577: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1578: if (!m) nrerror("allocation failure 1 in matrix()");
1579: m += NR_END;
1580: m -= nrl;
1581:
1582: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1583: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1584: m[nrl] += NR_END;
1585: m[nrl] -= ncl;
1586:
1587: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1588: return m;
1.145 brouard 1589: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1590: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1591: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1592: */
1593: }
1594:
1595: /*************************free matrix ************************/
1596: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1597: {
1598: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1599: free((FREE_ARG)(m+nrl-NR_END));
1600: }
1601:
1602: /******************* ma3x *******************************/
1603: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1604: {
1605: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1606: double ***m;
1607:
1608: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1609: if (!m) nrerror("allocation failure 1 in matrix()");
1610: m += NR_END;
1611: m -= nrl;
1612:
1613: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1614: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1615: m[nrl] += NR_END;
1616: m[nrl] -= ncl;
1617:
1618: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1619:
1620: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1621: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1622: m[nrl][ncl] += NR_END;
1623: m[nrl][ncl] -= nll;
1624: for (j=ncl+1; j<=nch; j++)
1625: m[nrl][j]=m[nrl][j-1]+nlay;
1626:
1627: for (i=nrl+1; i<=nrh; i++) {
1628: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1629: for (j=ncl+1; j<=nch; j++)
1630: m[i][j]=m[i][j-1]+nlay;
1631: }
1632: return m;
1633: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1634: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1635: */
1636: }
1637:
1638: /*************************free ma3x ************************/
1639: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1640: {
1641: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1642: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1643: free((FREE_ARG)(m+nrl-NR_END));
1644: }
1645:
1646: /*************** function subdirf ***********/
1647: char *subdirf(char fileres[])
1648: {
1649: /* Caution optionfilefiname is hidden */
1650: strcpy(tmpout,optionfilefiname);
1651: strcat(tmpout,"/"); /* Add to the right */
1652: strcat(tmpout,fileres);
1653: return tmpout;
1654: }
1655:
1656: /*************** function subdirf2 ***********/
1657: char *subdirf2(char fileres[], char *preop)
1658: {
1659:
1660: /* Caution optionfilefiname is hidden */
1661: strcpy(tmpout,optionfilefiname);
1662: strcat(tmpout,"/");
1663: strcat(tmpout,preop);
1664: strcat(tmpout,fileres);
1665: return tmpout;
1666: }
1667:
1668: /*************** function subdirf3 ***********/
1669: char *subdirf3(char fileres[], char *preop, char *preop2)
1670: {
1671:
1672: /* Caution optionfilefiname is hidden */
1673: strcpy(tmpout,optionfilefiname);
1674: strcat(tmpout,"/");
1675: strcat(tmpout,preop);
1676: strcat(tmpout,preop2);
1677: strcat(tmpout,fileres);
1678: return tmpout;
1679: }
1.213 brouard 1680:
1681: /*************** function subdirfext ***********/
1682: char *subdirfext(char fileres[], char *preop, char *postop)
1683: {
1684:
1685: strcpy(tmpout,preop);
1686: strcat(tmpout,fileres);
1687: strcat(tmpout,postop);
1688: return tmpout;
1689: }
1.126 brouard 1690:
1.213 brouard 1691: /*************** function subdirfext3 ***********/
1692: char *subdirfext3(char fileres[], char *preop, char *postop)
1693: {
1694:
1695: /* Caution optionfilefiname is hidden */
1696: strcpy(tmpout,optionfilefiname);
1697: strcat(tmpout,"/");
1698: strcat(tmpout,preop);
1699: strcat(tmpout,fileres);
1700: strcat(tmpout,postop);
1701: return tmpout;
1702: }
1703:
1.162 brouard 1704: char *asc_diff_time(long time_sec, char ascdiff[])
1705: {
1706: long sec_left, days, hours, minutes;
1707: days = (time_sec) / (60*60*24);
1708: sec_left = (time_sec) % (60*60*24);
1709: hours = (sec_left) / (60*60) ;
1710: sec_left = (sec_left) %(60*60);
1711: minutes = (sec_left) /60;
1712: sec_left = (sec_left) % (60);
1713: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1714: return ascdiff;
1715: }
1716:
1.126 brouard 1717: /***************** f1dim *************************/
1718: extern int ncom;
1719: extern double *pcom,*xicom;
1720: extern double (*nrfunc)(double []);
1721:
1722: double f1dim(double x)
1723: {
1724: int j;
1725: double f;
1726: double *xt;
1727:
1728: xt=vector(1,ncom);
1729: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1730: f=(*nrfunc)(xt);
1731: free_vector(xt,1,ncom);
1732: return f;
1733: }
1734:
1735: /*****************brent *************************/
1736: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1737: {
1738: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1739: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1740: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1741: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1742: * returned function value.
1743: */
1.126 brouard 1744: int iter;
1745: double a,b,d,etemp;
1.159 brouard 1746: double fu=0,fv,fw,fx;
1.164 brouard 1747: double ftemp=0.;
1.126 brouard 1748: double p,q,r,tol1,tol2,u,v,w,x,xm;
1749: double e=0.0;
1750:
1751: a=(ax < cx ? ax : cx);
1752: b=(ax > cx ? ax : cx);
1753: x=w=v=bx;
1754: fw=fv=fx=(*f)(x);
1755: for (iter=1;iter<=ITMAX;iter++) {
1756: xm=0.5*(a+b);
1757: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1758: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1759: printf(".");fflush(stdout);
1760: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1761: #ifdef DEBUGBRENT
1.126 brouard 1762: 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);
1763: 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);
1764: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1765: #endif
1766: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1767: *xmin=x;
1768: return fx;
1769: }
1770: ftemp=fu;
1771: if (fabs(e) > tol1) {
1772: r=(x-w)*(fx-fv);
1773: q=(x-v)*(fx-fw);
1774: p=(x-v)*q-(x-w)*r;
1775: q=2.0*(q-r);
1776: if (q > 0.0) p = -p;
1777: q=fabs(q);
1778: etemp=e;
1779: e=d;
1780: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1781: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1782: else {
1.224 brouard 1783: d=p/q;
1784: u=x+d;
1785: if (u-a < tol2 || b-u < tol2)
1786: d=SIGN(tol1,xm-x);
1.126 brouard 1787: }
1788: } else {
1789: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1790: }
1791: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1792: fu=(*f)(u);
1793: if (fu <= fx) {
1794: if (u >= x) a=x; else b=x;
1795: SHFT(v,w,x,u)
1.183 brouard 1796: SHFT(fv,fw,fx,fu)
1797: } else {
1798: if (u < x) a=u; else b=u;
1799: if (fu <= fw || w == x) {
1.224 brouard 1800: v=w;
1801: w=u;
1802: fv=fw;
1803: fw=fu;
1.183 brouard 1804: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1805: v=u;
1806: fv=fu;
1.183 brouard 1807: }
1808: }
1.126 brouard 1809: }
1810: nrerror("Too many iterations in brent");
1811: *xmin=x;
1812: return fx;
1813: }
1814:
1815: /****************** mnbrak ***********************/
1816:
1817: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1818: double (*func)(double))
1.183 brouard 1819: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1820: the downhill direction (defined by the function as evaluated at the initial points) and returns
1821: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1822: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1823: */
1.126 brouard 1824: double ulim,u,r,q, dum;
1825: double fu;
1.187 brouard 1826:
1827: double scale=10.;
1828: int iterscale=0;
1829:
1830: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1831: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1832:
1833:
1834: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1835: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1836: /* *bx = *ax - (*ax - *bx)/scale; */
1837: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1838: /* } */
1839:
1.126 brouard 1840: if (*fb > *fa) {
1841: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1842: SHFT(dum,*fb,*fa,dum)
1843: }
1.126 brouard 1844: *cx=(*bx)+GOLD*(*bx-*ax);
1845: *fc=(*func)(*cx);
1.183 brouard 1846: #ifdef DEBUG
1.224 brouard 1847: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1848: 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 1849: #endif
1.224 brouard 1850: 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 1851: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1852: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1853: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1854: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1855: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1856: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1857: fu=(*func)(u);
1.163 brouard 1858: #ifdef DEBUG
1859: /* f(x)=A(x-u)**2+f(u) */
1860: double A, fparabu;
1861: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1862: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1863: 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);
1864: 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 1865: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1866: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1867: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1868: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1869: #endif
1.184 brouard 1870: #ifdef MNBRAKORIGINAL
1.183 brouard 1871: #else
1.191 brouard 1872: /* if (fu > *fc) { */
1873: /* #ifdef DEBUG */
1874: /* printf("mnbrak4 fu > fc \n"); */
1875: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1876: /* #endif */
1877: /* /\* 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 *\\/ *\/ */
1878: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1879: /* dum=u; /\* Shifting c and u *\/ */
1880: /* u = *cx; */
1881: /* *cx = dum; */
1882: /* dum = fu; */
1883: /* fu = *fc; */
1884: /* *fc =dum; */
1885: /* } else { /\* end *\/ */
1886: /* #ifdef DEBUG */
1887: /* printf("mnbrak3 fu < fc \n"); */
1888: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1889: /* #endif */
1890: /* dum=u; /\* Shifting c and u *\/ */
1891: /* u = *cx; */
1892: /* *cx = dum; */
1893: /* dum = fu; */
1894: /* fu = *fc; */
1895: /* *fc =dum; */
1896: /* } */
1.224 brouard 1897: #ifdef DEBUGMNBRAK
1898: double A, fparabu;
1899: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1900: fparabu= *fa - A*(*ax-u)*(*ax-u);
1901: 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);
1902: 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 1903: #endif
1.191 brouard 1904: dum=u; /* Shifting c and u */
1905: u = *cx;
1906: *cx = dum;
1907: dum = fu;
1908: fu = *fc;
1909: *fc =dum;
1.183 brouard 1910: #endif
1.162 brouard 1911: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1912: #ifdef DEBUG
1.224 brouard 1913: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1914: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1915: #endif
1.126 brouard 1916: fu=(*func)(u);
1917: if (fu < *fc) {
1.183 brouard 1918: #ifdef DEBUG
1.224 brouard 1919: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1920: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1921: #endif
1922: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1923: SHFT(*fb,*fc,fu,(*func)(u))
1924: #ifdef DEBUG
1925: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1926: #endif
1927: }
1.162 brouard 1928: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1929: #ifdef DEBUG
1.224 brouard 1930: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1931: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1932: #endif
1.126 brouard 1933: u=ulim;
1934: fu=(*func)(u);
1.183 brouard 1935: } else { /* u could be left to b (if r > q parabola has a maximum) */
1936: #ifdef DEBUG
1.224 brouard 1937: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1938: 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 1939: #endif
1.126 brouard 1940: u=(*cx)+GOLD*(*cx-*bx);
1941: fu=(*func)(u);
1.224 brouard 1942: #ifdef DEBUG
1943: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1944: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1945: #endif
1.183 brouard 1946: } /* end tests */
1.126 brouard 1947: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1948: SHFT(*fa,*fb,*fc,fu)
1949: #ifdef DEBUG
1.224 brouard 1950: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1951: 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 1952: #endif
1953: } /* 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 1954: }
1955:
1956: /*************** linmin ************************/
1.162 brouard 1957: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1958: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1959: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1960: the value of func at the returned location p . This is actually all accomplished by calling the
1961: routines mnbrak and brent .*/
1.126 brouard 1962: int ncom;
1963: double *pcom,*xicom;
1964: double (*nrfunc)(double []);
1965:
1.224 brouard 1966: #ifdef LINMINORIGINAL
1.126 brouard 1967: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1968: #else
1969: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1970: #endif
1.126 brouard 1971: {
1972: double brent(double ax, double bx, double cx,
1973: double (*f)(double), double tol, double *xmin);
1974: double f1dim(double x);
1975: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1976: double *fc, double (*func)(double));
1977: int j;
1978: double xx,xmin,bx,ax;
1979: double fx,fb,fa;
1.187 brouard 1980:
1.203 brouard 1981: #ifdef LINMINORIGINAL
1982: #else
1983: double scale=10., axs, xxs; /* Scale added for infinity */
1984: #endif
1985:
1.126 brouard 1986: ncom=n;
1987: pcom=vector(1,n);
1988: xicom=vector(1,n);
1989: nrfunc=func;
1990: for (j=1;j<=n;j++) {
1991: pcom[j]=p[j];
1.202 brouard 1992: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1993: }
1.187 brouard 1994:
1.203 brouard 1995: #ifdef LINMINORIGINAL
1996: xx=1.;
1997: #else
1998: axs=0.0;
1999: xxs=1.;
2000: do{
2001: xx= xxs;
2002: #endif
1.187 brouard 2003: ax=0.;
2004: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2005: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2006: /* 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)) */
2007: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2008: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2009: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2010: /* 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 2011: #ifdef LINMINORIGINAL
2012: #else
2013: if (fx != fx){
1.224 brouard 2014: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2015: printf("|");
2016: fprintf(ficlog,"|");
1.203 brouard 2017: #ifdef DEBUGLINMIN
1.224 brouard 2018: 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 2019: #endif
2020: }
1.224 brouard 2021: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2022: #endif
2023:
1.191 brouard 2024: #ifdef DEBUGLINMIN
2025: 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 2026: 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 2027: #endif
1.224 brouard 2028: #ifdef LINMINORIGINAL
2029: #else
2030: if(fb == fx){ /* Flat function in the direction */
2031: xmin=xx;
2032: *flat=1;
2033: }else{
2034: *flat=0;
2035: #endif
2036: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2037: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2038: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2039: /* fmin = f(p[j] + xmin * xi[j]) */
2040: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2041: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2042: #ifdef DEBUG
1.224 brouard 2043: 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);
2044: 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);
2045: #endif
2046: #ifdef LINMINORIGINAL
2047: #else
2048: }
1.126 brouard 2049: #endif
1.191 brouard 2050: #ifdef DEBUGLINMIN
2051: printf("linmin end ");
1.202 brouard 2052: fprintf(ficlog,"linmin end ");
1.191 brouard 2053: #endif
1.126 brouard 2054: for (j=1;j<=n;j++) {
1.203 brouard 2055: #ifdef LINMINORIGINAL
2056: xi[j] *= xmin;
2057: #else
2058: #ifdef DEBUGLINMIN
2059: if(xxs <1.0)
2060: printf(" before xi[%d]=%12.8f", j,xi[j]);
2061: #endif
2062: 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) */
2063: #ifdef DEBUGLINMIN
2064: if(xxs <1.0)
2065: 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 );
2066: #endif
2067: #endif
1.187 brouard 2068: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2069: }
1.191 brouard 2070: #ifdef DEBUGLINMIN
1.203 brouard 2071: printf("\n");
1.191 brouard 2072: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2073: 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 2074: for (j=1;j<=n;j++) {
1.202 brouard 2075: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2076: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2077: if(j % ncovmodel == 0){
1.191 brouard 2078: printf("\n");
1.202 brouard 2079: fprintf(ficlog,"\n");
2080: }
1.191 brouard 2081: }
1.203 brouard 2082: #else
1.191 brouard 2083: #endif
1.126 brouard 2084: free_vector(xicom,1,n);
2085: free_vector(pcom,1,n);
2086: }
2087:
2088:
2089: /*************** powell ************************/
1.162 brouard 2090: /*
2091: Minimization of a function func of n variables. Input consists of an initial starting point
2092: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2093: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2094: such that failure to decrease by more than this amount on one iteration signals doneness. On
2095: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2096: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2097: */
1.224 brouard 2098: #ifdef LINMINORIGINAL
2099: #else
2100: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2101: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2102: #endif
1.126 brouard 2103: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2104: double (*func)(double []))
2105: {
1.224 brouard 2106: #ifdef LINMINORIGINAL
2107: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2108: double (*func)(double []));
1.224 brouard 2109: #else
1.241 brouard 2110: void linmin(double p[], double xi[], int n, double *fret,
2111: double (*func)(double []),int *flat);
1.224 brouard 2112: #endif
1.239 brouard 2113: int i,ibig,j,jk,k;
1.126 brouard 2114: double del,t,*pt,*ptt,*xit;
1.181 brouard 2115: double directest;
1.126 brouard 2116: double fp,fptt;
2117: double *xits;
2118: int niterf, itmp;
1.224 brouard 2119: #ifdef LINMINORIGINAL
2120: #else
2121:
2122: flatdir=ivector(1,n);
2123: for (j=1;j<=n;j++) flatdir[j]=0;
2124: #endif
1.126 brouard 2125:
2126: pt=vector(1,n);
2127: ptt=vector(1,n);
2128: xit=vector(1,n);
2129: xits=vector(1,n);
2130: *fret=(*func)(p);
2131: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2132: rcurr_time = time(NULL);
1.126 brouard 2133: for (*iter=1;;++(*iter)) {
1.187 brouard 2134: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2135: ibig=0;
2136: del=0.0;
1.157 brouard 2137: rlast_time=rcurr_time;
2138: /* (void) gettimeofday(&curr_time,&tzp); */
2139: rcurr_time = time(NULL);
2140: curr_time = *localtime(&rcurr_time);
2141: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2142: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2143: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2144: for (i=1;i<=n;i++) {
1.126 brouard 2145: fprintf(ficrespow," %.12lf", p[i]);
2146: }
1.239 brouard 2147: fprintf(ficrespow,"\n");fflush(ficrespow);
2148: printf("\n#model= 1 + age ");
2149: fprintf(ficlog,"\n#model= 1 + age ");
2150: if(nagesqr==1){
1.241 brouard 2151: printf(" + age*age ");
2152: fprintf(ficlog," + age*age ");
1.239 brouard 2153: }
2154: for(j=1;j <=ncovmodel-2;j++){
2155: if(Typevar[j]==0) {
2156: printf(" + V%d ",Tvar[j]);
2157: fprintf(ficlog," + V%d ",Tvar[j]);
2158: }else if(Typevar[j]==1) {
2159: printf(" + V%d*age ",Tvar[j]);
2160: fprintf(ficlog," + V%d*age ",Tvar[j]);
2161: }else if(Typevar[j]==2) {
2162: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2163: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2164: }
2165: }
1.126 brouard 2166: printf("\n");
1.239 brouard 2167: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2168: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2169: fprintf(ficlog,"\n");
1.239 brouard 2170: for(i=1,jk=1; i <=nlstate; i++){
2171: for(k=1; k <=(nlstate+ndeath); k++){
2172: if (k != i) {
2173: printf("%d%d ",i,k);
2174: fprintf(ficlog,"%d%d ",i,k);
2175: for(j=1; j <=ncovmodel; j++){
2176: printf("%12.7f ",p[jk]);
2177: fprintf(ficlog,"%12.7f ",p[jk]);
2178: jk++;
2179: }
2180: printf("\n");
2181: fprintf(ficlog,"\n");
2182: }
2183: }
2184: }
1.241 brouard 2185: if(*iter <=3 && *iter >1){
1.157 brouard 2186: tml = *localtime(&rcurr_time);
2187: strcpy(strcurr,asctime(&tml));
2188: rforecast_time=rcurr_time;
1.126 brouard 2189: itmp = strlen(strcurr);
2190: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2191: strcurr[itmp-1]='\0';
1.162 brouard 2192: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2193: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2194: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2195: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2196: forecast_time = *localtime(&rforecast_time);
2197: strcpy(strfor,asctime(&forecast_time));
2198: itmp = strlen(strfor);
2199: if(strfor[itmp-1]=='\n')
2200: strfor[itmp-1]='\0';
2201: 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);
2202: 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 2203: }
2204: }
1.187 brouard 2205: for (i=1;i<=n;i++) { /* For each direction i */
2206: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2207: fptt=(*fret);
2208: #ifdef DEBUG
1.203 brouard 2209: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2210: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2211: #endif
1.203 brouard 2212: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2213: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2214: #ifdef LINMINORIGINAL
1.188 brouard 2215: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2216: #else
2217: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2218: flatdir[i]=flat; /* Function is vanishing in that direction i */
2219: #endif
2220: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2221: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2222: /* because that direction will be replaced unless the gain del is small */
2223: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2224: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2225: /* with the new direction. */
2226: del=fabs(fptt-(*fret));
2227: ibig=i;
1.126 brouard 2228: }
2229: #ifdef DEBUG
2230: printf("%d %.12e",i,(*fret));
2231: fprintf(ficlog,"%d %.12e",i,(*fret));
2232: for (j=1;j<=n;j++) {
1.224 brouard 2233: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2234: printf(" x(%d)=%.12e",j,xit[j]);
2235: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2236: }
2237: for(j=1;j<=n;j++) {
1.225 brouard 2238: printf(" p(%d)=%.12e",j,p[j]);
2239: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2240: }
2241: printf("\n");
2242: fprintf(ficlog,"\n");
2243: #endif
1.187 brouard 2244: } /* end loop on each direction i */
2245: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2246: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2247: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2248: for(j=1;j<=n;j++) {
1.225 brouard 2249: if(flatdir[j] >0){
2250: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2251: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2252: }
2253: /* printf("\n"); */
2254: /* fprintf(ficlog,"\n"); */
2255: }
1.243 brouard 2256: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2257: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2258: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2259: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2260: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2261: /* decreased of more than 3.84 */
2262: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2263: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2264: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2265:
1.188 brouard 2266: /* Starting the program with initial values given by a former maximization will simply change */
2267: /* the scales of the directions and the directions, because the are reset to canonical directions */
2268: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2269: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2270: #ifdef DEBUG
2271: int k[2],l;
2272: k[0]=1;
2273: k[1]=-1;
2274: printf("Max: %.12e",(*func)(p));
2275: fprintf(ficlog,"Max: %.12e",(*func)(p));
2276: for (j=1;j<=n;j++) {
2277: printf(" %.12e",p[j]);
2278: fprintf(ficlog," %.12e",p[j]);
2279: }
2280: printf("\n");
2281: fprintf(ficlog,"\n");
2282: for(l=0;l<=1;l++) {
2283: for (j=1;j<=n;j++) {
2284: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2285: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2286: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2287: }
2288: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2289: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2290: }
2291: #endif
2292:
1.224 brouard 2293: #ifdef LINMINORIGINAL
2294: #else
2295: free_ivector(flatdir,1,n);
2296: #endif
1.126 brouard 2297: free_vector(xit,1,n);
2298: free_vector(xits,1,n);
2299: free_vector(ptt,1,n);
2300: free_vector(pt,1,n);
2301: return;
1.192 brouard 2302: } /* enough precision */
1.240 brouard 2303: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2304: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2305: ptt[j]=2.0*p[j]-pt[j];
2306: xit[j]=p[j]-pt[j];
2307: pt[j]=p[j];
2308: }
1.181 brouard 2309: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2310: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2311: if (*iter <=4) {
1.225 brouard 2312: #else
2313: #endif
1.224 brouard 2314: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2315: #else
1.161 brouard 2316: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2317: #endif
1.162 brouard 2318: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2319: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2320: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2321: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2322: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2323: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2324: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2325: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2326: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2327: /* Even if f3 <f1, directest can be negative and t >0 */
2328: /* mu² and del² are equal when f3=f1 */
2329: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2330: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2331: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2332: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2333: #ifdef NRCORIGINAL
2334: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2335: #else
2336: 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 2337: t= t- del*SQR(fp-fptt);
1.183 brouard 2338: #endif
1.202 brouard 2339: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2340: #ifdef DEBUG
1.181 brouard 2341: 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);
2342: 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 2343: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2344: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2345: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2346: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2347: 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);
2348: 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);
2349: #endif
1.183 brouard 2350: #ifdef POWELLORIGINAL
2351: if (t < 0.0) { /* Then we use it for new direction */
2352: #else
1.182 brouard 2353: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2354: 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 2355: 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 2356: 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 2357: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2358: }
1.181 brouard 2359: if (directest < 0.0) { /* Then we use it for new direction */
2360: #endif
1.191 brouard 2361: #ifdef DEBUGLINMIN
1.234 brouard 2362: printf("Before linmin in direction P%d-P0\n",n);
2363: for (j=1;j<=n;j++) {
2364: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2365: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2366: if(j % ncovmodel == 0){
2367: printf("\n");
2368: fprintf(ficlog,"\n");
2369: }
2370: }
1.224 brouard 2371: #endif
2372: #ifdef LINMINORIGINAL
1.234 brouard 2373: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2374: #else
1.234 brouard 2375: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2376: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2377: #endif
1.234 brouard 2378:
1.191 brouard 2379: #ifdef DEBUGLINMIN
1.234 brouard 2380: for (j=1;j<=n;j++) {
2381: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2382: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2383: if(j % ncovmodel == 0){
2384: printf("\n");
2385: fprintf(ficlog,"\n");
2386: }
2387: }
1.224 brouard 2388: #endif
1.234 brouard 2389: for (j=1;j<=n;j++) {
2390: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2391: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2392: }
1.224 brouard 2393: #ifdef LINMINORIGINAL
2394: #else
1.234 brouard 2395: for (j=1, flatd=0;j<=n;j++) {
2396: if(flatdir[j]>0)
2397: flatd++;
2398: }
2399: if(flatd >0){
2400: printf("%d flat directions\n",flatd);
2401: fprintf(ficlog,"%d flat directions\n",flatd);
2402: for (j=1;j<=n;j++) {
2403: if(flatdir[j]>0){
2404: printf("%d ",j);
2405: fprintf(ficlog,"%d ",j);
2406: }
2407: }
2408: printf("\n");
2409: fprintf(ficlog,"\n");
2410: }
1.191 brouard 2411: #endif
1.234 brouard 2412: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2413: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2414:
1.126 brouard 2415: #ifdef DEBUG
1.234 brouard 2416: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2417: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2418: for(j=1;j<=n;j++){
2419: printf(" %lf",xit[j]);
2420: fprintf(ficlog," %lf",xit[j]);
2421: }
2422: printf("\n");
2423: fprintf(ficlog,"\n");
1.126 brouard 2424: #endif
1.192 brouard 2425: } /* end of t or directest negative */
1.224 brouard 2426: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2427: #else
1.234 brouard 2428: } /* end if (fptt < fp) */
1.192 brouard 2429: #endif
1.225 brouard 2430: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2431: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2432: #else
1.224 brouard 2433: #endif
1.234 brouard 2434: } /* loop iteration */
1.126 brouard 2435: }
1.234 brouard 2436:
1.126 brouard 2437: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2438:
1.235 brouard 2439: 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 2440: {
1.235 brouard 2441: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2442: (and selected quantitative values in nres)
2443: by left multiplying the unit
1.234 brouard 2444: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2445: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2446: /* Wx is row vector: population in state 1, population in state 2, population dead */
2447: /* or prevalence in state 1, prevalence in state 2, 0 */
2448: /* newm is the matrix after multiplications, its rows are identical at a factor */
2449: /* Initial matrix pimij */
2450: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2451: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2452: /* 0, 0 , 1} */
2453: /*
2454: * and after some iteration: */
2455: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2456: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2457: /* 0, 0 , 1} */
2458: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2459: /* {0.51571254859325999, 0.4842874514067399, */
2460: /* 0.51326036147820708, 0.48673963852179264} */
2461: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2462:
1.126 brouard 2463: int i, ii,j,k;
1.209 brouard 2464: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2465: /* double **matprod2(); */ /* test */
1.218 brouard 2466: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2467: double **newm;
1.209 brouard 2468: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2469: int ncvloop=0;
1.169 brouard 2470:
1.209 brouard 2471: min=vector(1,nlstate);
2472: max=vector(1,nlstate);
2473: meandiff=vector(1,nlstate);
2474:
1.218 brouard 2475: /* Starting with matrix unity */
1.126 brouard 2476: for (ii=1;ii<=nlstate+ndeath;ii++)
2477: for (j=1;j<=nlstate+ndeath;j++){
2478: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2479: }
1.169 brouard 2480:
2481: cov[1]=1.;
2482:
2483: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2484: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2485: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2486: ncvloop++;
1.126 brouard 2487: newm=savm;
2488: /* Covariates have to be included here again */
1.138 brouard 2489: cov[2]=agefin;
1.187 brouard 2490: if(nagesqr==1)
2491: cov[3]= agefin*agefin;;
1.234 brouard 2492: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2493: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2494: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2495: /* 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 2496: }
2497: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2498: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2499: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2500: /* 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 2501: }
1.237 brouard 2502: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2503: if(Dummy[Tvar[Tage[k]]]){
2504: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2505: } else{
1.235 brouard 2506: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2507: }
1.235 brouard 2508: /* 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 2509: }
1.237 brouard 2510: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2511: /* 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 2512: if(Dummy[Tvard[k][1]==0]){
2513: if(Dummy[Tvard[k][2]==0]){
2514: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2515: }else{
2516: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2517: }
2518: }else{
2519: if(Dummy[Tvard[k][2]==0]){
2520: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2521: }else{
2522: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2523: }
2524: }
1.234 brouard 2525: }
1.138 brouard 2526: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2527: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2528: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2529: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2530: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2531: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2532: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2533:
1.126 brouard 2534: savm=oldm;
2535: oldm=newm;
1.209 brouard 2536:
2537: for(j=1; j<=nlstate; j++){
2538: max[j]=0.;
2539: min[j]=1.;
2540: }
2541: for(i=1;i<=nlstate;i++){
2542: sumnew=0;
2543: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2544: for(j=1; j<=nlstate; j++){
2545: prlim[i][j]= newm[i][j]/(1-sumnew);
2546: max[j]=FMAX(max[j],prlim[i][j]);
2547: min[j]=FMIN(min[j],prlim[i][j]);
2548: }
2549: }
2550:
1.126 brouard 2551: maxmax=0.;
1.209 brouard 2552: for(j=1; j<=nlstate; j++){
2553: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2554: maxmax=FMAX(maxmax,meandiff[j]);
2555: /* 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 2556: } /* j loop */
1.203 brouard 2557: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2558: /* 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 2559: if(maxmax < ftolpl){
1.209 brouard 2560: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2561: free_vector(min,1,nlstate);
2562: free_vector(max,1,nlstate);
2563: free_vector(meandiff,1,nlstate);
1.126 brouard 2564: return prlim;
2565: }
1.169 brouard 2566: } /* age loop */
1.208 brouard 2567: /* After some age loop it doesn't converge */
1.209 brouard 2568: 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 2569: 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 2570: /* 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); */
2571: free_vector(min,1,nlstate);
2572: free_vector(max,1,nlstate);
2573: free_vector(meandiff,1,nlstate);
1.208 brouard 2574:
1.169 brouard 2575: return prlim; /* should not reach here */
1.126 brouard 2576: }
2577:
1.217 brouard 2578:
2579: /**** Back Prevalence limit (stable or period prevalence) ****************/
2580:
1.218 brouard 2581: /* 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) */
2582: /* 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 2583: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2584: {
1.218 brouard 2585: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2586: matrix by transitions matrix until convergence is reached with precision ftolpl */
2587: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2588: /* Wx is row vector: population in state 1, population in state 2, population dead */
2589: /* or prevalence in state 1, prevalence in state 2, 0 */
2590: /* newm is the matrix after multiplications, its rows are identical at a factor */
2591: /* Initial matrix pimij */
2592: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2593: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2594: /* 0, 0 , 1} */
2595: /*
2596: * and after some iteration: */
2597: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2598: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2599: /* 0, 0 , 1} */
2600: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2601: /* {0.51571254859325999, 0.4842874514067399, */
2602: /* 0.51326036147820708, 0.48673963852179264} */
2603: /* If we start from prlim again, prlim tends to a constant matrix */
2604:
2605: int i, ii,j,k;
1.247 brouard 2606: int first=0;
1.217 brouard 2607: double *min, *max, *meandiff, maxmax,sumnew=0.;
2608: /* double **matprod2(); */ /* test */
2609: double **out, cov[NCOVMAX+1], **bmij();
2610: double **newm;
1.218 brouard 2611: double **dnewm, **doldm, **dsavm; /* for use */
2612: double **oldm, **savm; /* for use */
2613:
1.217 brouard 2614: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2615: int ncvloop=0;
2616:
2617: min=vector(1,nlstate);
2618: max=vector(1,nlstate);
2619: meandiff=vector(1,nlstate);
2620:
1.218 brouard 2621: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2622: oldm=oldms; savm=savms;
2623:
2624: /* Starting with matrix unity */
2625: for (ii=1;ii<=nlstate+ndeath;ii++)
2626: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2627: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2628: }
2629:
2630: cov[1]=1.;
2631:
2632: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2633: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2634: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2635: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2636: ncvloop++;
1.218 brouard 2637: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2638: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2639: /* Covariates have to be included here again */
2640: cov[2]=agefin;
2641: if(nagesqr==1)
2642: cov[3]= agefin*agefin;;
1.242 brouard 2643: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2644: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2645: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2646: /* 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)); */
2647: }
2648: /* for (k=1; k<=cptcovn;k++) { */
2649: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2650: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2651: /* /\* 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])]); *\/ */
2652: /* } */
2653: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2654: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2655: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2656: /* 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]); */
2657: }
2658: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2659: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2660: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2661: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2662: for (k=1; k<=cptcovage;k++){ /* For product with age */
2663: if(Dummy[Tvar[Tage[k]]]){
2664: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2665: } else{
2666: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2667: }
2668: /* 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]); */
2669: }
2670: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2671: /* 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]); */
2672: if(Dummy[Tvard[k][1]==0]){
2673: if(Dummy[Tvard[k][2]==0]){
2674: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2675: }else{
2676: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2677: }
2678: }else{
2679: if(Dummy[Tvard[k][2]==0]){
2680: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2681: }else{
2682: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2683: }
2684: }
1.217 brouard 2685: }
2686:
2687: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2688: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2689: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2690: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2691: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2692: /* ij should be linked to the correct index of cov */
2693: /* age and covariate values ij are in 'cov', but we need to pass
2694: * ij for the observed prevalence at age and status and covariate
2695: * number: prevacurrent[(int)agefin][ii][ij]
2696: */
2697: /* 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 *\/ */
2698: /* 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 *\/ */
2699: 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 2700: savm=oldm;
2701: oldm=newm;
2702: for(j=1; j<=nlstate; j++){
2703: max[j]=0.;
2704: min[j]=1.;
2705: }
2706: for(j=1; j<=nlstate; j++){
2707: for(i=1;i<=nlstate;i++){
1.234 brouard 2708: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2709: bprlim[i][j]= newm[i][j];
2710: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2711: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2712: }
2713: }
1.218 brouard 2714:
1.217 brouard 2715: maxmax=0.;
2716: for(i=1; i<=nlstate; i++){
2717: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2718: maxmax=FMAX(maxmax,meandiff[i]);
2719: /* 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); */
2720: } /* j loop */
2721: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2722: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2723: if(maxmax < ftolpl){
1.220 brouard 2724: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2725: free_vector(min,1,nlstate);
2726: free_vector(max,1,nlstate);
2727: free_vector(meandiff,1,nlstate);
2728: return bprlim;
2729: }
2730: } /* age loop */
2731: /* After some age loop it doesn't converge */
1.247 brouard 2732: if(first){
2733: first=1;
2734: 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\
2735: 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);
2736: }
2737: 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 2738: 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);
2739: /* 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); */
2740: free_vector(min,1,nlstate);
2741: free_vector(max,1,nlstate);
2742: free_vector(meandiff,1,nlstate);
2743:
2744: return bprlim; /* should not reach here */
2745: }
2746:
1.126 brouard 2747: /*************** transition probabilities ***************/
2748:
2749: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2750: {
1.138 brouard 2751: /* According to parameters values stored in x and the covariate's values stored in cov,
2752: computes the probability to be observed in state j being in state i by appying the
2753: model to the ncovmodel covariates (including constant and age).
2754: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2755: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2756: ncth covariate in the global vector x is given by the formula:
2757: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2758: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2759: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2760: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2761: Outputs ps[i][j] the probability to be observed in j being in j according to
2762: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2763: */
2764: double s1, lnpijopii;
1.126 brouard 2765: /*double t34;*/
1.164 brouard 2766: int i,j, nc, ii, jj;
1.126 brouard 2767:
1.223 brouard 2768: for(i=1; i<= nlstate; i++){
2769: for(j=1; j<i;j++){
2770: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2771: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2772: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2773: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2774: }
2775: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2776: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2777: }
2778: for(j=i+1; j<=nlstate+ndeath;j++){
2779: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2780: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2781: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2782: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2783: }
2784: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2785: }
2786: }
1.218 brouard 2787:
1.223 brouard 2788: for(i=1; i<= nlstate; i++){
2789: s1=0;
2790: for(j=1; j<i; j++){
2791: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2792: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2793: }
2794: for(j=i+1; j<=nlstate+ndeath; j++){
2795: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2796: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2797: }
2798: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2799: ps[i][i]=1./(s1+1.);
2800: /* Computing other pijs */
2801: for(j=1; j<i; j++)
2802: ps[i][j]= exp(ps[i][j])*ps[i][i];
2803: for(j=i+1; j<=nlstate+ndeath; j++)
2804: ps[i][j]= exp(ps[i][j])*ps[i][i];
2805: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2806: } /* end i */
1.218 brouard 2807:
1.223 brouard 2808: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2809: for(jj=1; jj<= nlstate+ndeath; jj++){
2810: ps[ii][jj]=0;
2811: ps[ii][ii]=1;
2812: }
2813: }
1.218 brouard 2814:
2815:
1.223 brouard 2816: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2817: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2818: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2819: /* } */
2820: /* printf("\n "); */
2821: /* } */
2822: /* printf("\n ");printf("%lf ",cov[2]);*/
2823: /*
2824: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2825: goto end;*/
1.223 brouard 2826: return ps;
1.126 brouard 2827: }
2828:
1.218 brouard 2829: /*************** backward transition probabilities ***************/
2830:
2831: /* 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 ) */
2832: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2833: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2834: {
1.222 brouard 2835: /* Computes the backward probability at age agefin and covariate ij
2836: * and returns in **ps as well as **bmij.
2837: */
1.218 brouard 2838: int i, ii, j,k;
1.222 brouard 2839:
2840: double **out, **pmij();
2841: double sumnew=0.;
1.218 brouard 2842: double agefin;
1.222 brouard 2843:
2844: double **dnewm, **dsavm, **doldm;
2845: double **bbmij;
2846:
1.218 brouard 2847: doldm=ddoldms; /* global pointers */
1.222 brouard 2848: dnewm=ddnewms;
2849: dsavm=ddsavms;
2850:
2851: agefin=cov[2];
2852: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2853: the observed prevalence (with this covariate ij) */
2854: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2855: /* We do have the matrix Px in savm and we need pij */
2856: for (j=1;j<=nlstate+ndeath;j++){
2857: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2858: for (ii=1;ii<=nlstate;ii++){
2859: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2860: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2861: for (ii=1;ii<=nlstate+ndeath;ii++){
2862: if(sumnew >= 1.e-10){
2863: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2864: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2865: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2866: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2867: /* }else */
2868: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2869: }else{
1.242 brouard 2870: ;
2871: /* 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 2872: }
2873: } /*End ii */
2874: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2875: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2876: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2877: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2878: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2879: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2880: /* left Product of this matrix by diag matrix of prevalences (savm) */
2881: for (j=1;j<=nlstate+ndeath;j++){
2882: for (ii=1;ii<=nlstate+ndeath;ii++){
2883: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2884: }
2885: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2886: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2887: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2888: /* end bmij */
2889: return ps;
1.218 brouard 2890: }
1.217 brouard 2891: /*************** transition probabilities ***************/
2892:
1.218 brouard 2893: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2894: {
2895: /* According to parameters values stored in x and the covariate's values stored in cov,
2896: computes the probability to be observed in state j being in state i by appying the
2897: model to the ncovmodel covariates (including constant and age).
2898: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2899: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2900: ncth covariate in the global vector x is given by the formula:
2901: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2902: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2903: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2904: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2905: Outputs ps[i][j] the probability to be observed in j being in j according to
2906: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2907: */
2908: double s1, lnpijopii;
2909: /*double t34;*/
2910: int i,j, nc, ii, jj;
2911:
1.234 brouard 2912: for(i=1; i<= nlstate; i++){
2913: for(j=1; j<i;j++){
2914: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2915: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2916: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2917: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2918: }
2919: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2920: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2921: }
2922: for(j=i+1; j<=nlstate+ndeath;j++){
2923: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2924: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2925: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2926: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2927: }
2928: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2929: }
2930: }
2931:
2932: for(i=1; i<= nlstate; i++){
2933: s1=0;
2934: for(j=1; j<i; j++){
2935: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2936: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2937: }
2938: for(j=i+1; j<=nlstate+ndeath; j++){
2939: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2940: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2941: }
2942: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2943: ps[i][i]=1./(s1+1.);
2944: /* Computing other pijs */
2945: for(j=1; j<i; j++)
2946: ps[i][j]= exp(ps[i][j])*ps[i][i];
2947: for(j=i+1; j<=nlstate+ndeath; j++)
2948: ps[i][j]= exp(ps[i][j])*ps[i][i];
2949: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2950: } /* end i */
2951:
2952: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2953: for(jj=1; jj<= nlstate+ndeath; jj++){
2954: ps[ii][jj]=0;
2955: ps[ii][ii]=1;
2956: }
2957: }
2958: /* Added for backcast */ /* Transposed matrix too */
2959: for(jj=1; jj<= nlstate+ndeath; jj++){
2960: s1=0.;
2961: for(ii=1; ii<= nlstate+ndeath; ii++){
2962: s1+=ps[ii][jj];
2963: }
2964: for(ii=1; ii<= nlstate; ii++){
2965: ps[ii][jj]=ps[ii][jj]/s1;
2966: }
2967: }
2968: /* Transposition */
2969: for(jj=1; jj<= nlstate+ndeath; jj++){
2970: for(ii=jj; ii<= nlstate+ndeath; ii++){
2971: s1=ps[ii][jj];
2972: ps[ii][jj]=ps[jj][ii];
2973: ps[jj][ii]=s1;
2974: }
2975: }
2976: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2977: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2978: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2979: /* } */
2980: /* printf("\n "); */
2981: /* } */
2982: /* printf("\n ");printf("%lf ",cov[2]);*/
2983: /*
2984: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2985: goto end;*/
2986: return ps;
1.217 brouard 2987: }
2988:
2989:
1.126 brouard 2990: /**************** Product of 2 matrices ******************/
2991:
1.145 brouard 2992: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2993: {
2994: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2995: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2996: /* in, b, out are matrice of pointers which should have been initialized
2997: before: only the contents of out is modified. The function returns
2998: a pointer to pointers identical to out */
1.145 brouard 2999: int i, j, k;
1.126 brouard 3000: for(i=nrl; i<= nrh; i++)
1.145 brouard 3001: for(k=ncolol; k<=ncoloh; k++){
3002: out[i][k]=0.;
3003: for(j=ncl; j<=nch; j++)
3004: out[i][k] +=in[i][j]*b[j][k];
3005: }
1.126 brouard 3006: return out;
3007: }
3008:
3009:
3010: /************* Higher Matrix Product ***************/
3011:
1.235 brouard 3012: 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 3013: {
1.218 brouard 3014: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3015: 'nhstepm*hstepm*stepm' months (i.e. until
3016: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3017: nhstepm*hstepm matrices.
3018: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3019: (typically every 2 years instead of every month which is too big
3020: for the memory).
3021: Model is determined by parameters x and covariates have to be
3022: included manually here.
3023:
3024: */
3025:
3026: int i, j, d, h, k;
1.131 brouard 3027: double **out, cov[NCOVMAX+1];
1.126 brouard 3028: double **newm;
1.187 brouard 3029: double agexact;
1.214 brouard 3030: double agebegin, ageend;
1.126 brouard 3031:
3032: /* Hstepm could be zero and should return the unit matrix */
3033: for (i=1;i<=nlstate+ndeath;i++)
3034: for (j=1;j<=nlstate+ndeath;j++){
3035: oldm[i][j]=(i==j ? 1.0 : 0.0);
3036: po[i][j][0]=(i==j ? 1.0 : 0.0);
3037: }
3038: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3039: for(h=1; h <=nhstepm; h++){
3040: for(d=1; d <=hstepm; d++){
3041: newm=savm;
3042: /* Covariates have to be included here again */
3043: cov[1]=1.;
1.214 brouard 3044: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3045: cov[2]=agexact;
3046: if(nagesqr==1)
1.227 brouard 3047: cov[3]= agexact*agexact;
1.235 brouard 3048: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3049: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3050: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3051: /* 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)); */
3052: }
3053: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3054: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3055: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3056: /* 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]); */
3057: }
3058: for (k=1; k<=cptcovage;k++){
3059: if(Dummy[Tvar[Tage[k]]]){
3060: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3061: } else{
3062: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3063: }
3064: /* 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]); */
3065: }
3066: for (k=1; k<=cptcovprod;k++){ /* */
3067: /* 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]); */
3068: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3069: }
3070: /* for (k=1; k<=cptcovn;k++) */
3071: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3072: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3073: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3074: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3075: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3076:
3077:
1.126 brouard 3078: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3079: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3080: /* right multiplication of oldm by the current matrix */
1.126 brouard 3081: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3082: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3083: /* if((int)age == 70){ */
3084: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3085: /* for(i=1; i<=nlstate+ndeath; i++) { */
3086: /* printf("%d pmmij ",i); */
3087: /* for(j=1;j<=nlstate+ndeath;j++) { */
3088: /* printf("%f ",pmmij[i][j]); */
3089: /* } */
3090: /* printf(" oldm "); */
3091: /* for(j=1;j<=nlstate+ndeath;j++) { */
3092: /* printf("%f ",oldm[i][j]); */
3093: /* } */
3094: /* printf("\n"); */
3095: /* } */
3096: /* } */
1.126 brouard 3097: savm=oldm;
3098: oldm=newm;
3099: }
3100: for(i=1; i<=nlstate+ndeath; i++)
3101: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3102: po[i][j][h]=newm[i][j];
3103: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3104: }
1.128 brouard 3105: /*printf("h=%d ",h);*/
1.126 brouard 3106: } /* end h */
1.218 brouard 3107: /* printf("\n H=%d \n",h); */
1.126 brouard 3108: return po;
3109: }
3110:
1.217 brouard 3111: /************* Higher Back Matrix Product ***************/
1.218 brouard 3112: /* 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 3113: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3114: {
1.218 brouard 3115: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3116: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3117: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3118: nhstepm*hstepm matrices.
3119: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3120: (typically every 2 years instead of every month which is too big
1.217 brouard 3121: for the memory).
1.218 brouard 3122: Model is determined by parameters x and covariates have to be
3123: included manually here.
1.217 brouard 3124:
1.222 brouard 3125: */
1.217 brouard 3126:
3127: int i, j, d, h, k;
3128: double **out, cov[NCOVMAX+1];
3129: double **newm;
3130: double agexact;
3131: double agebegin, ageend;
1.222 brouard 3132: double **oldm, **savm;
1.217 brouard 3133:
1.222 brouard 3134: oldm=oldms;savm=savms;
1.217 brouard 3135: /* Hstepm could be zero and should return the unit matrix */
3136: for (i=1;i<=nlstate+ndeath;i++)
3137: for (j=1;j<=nlstate+ndeath;j++){
3138: oldm[i][j]=(i==j ? 1.0 : 0.0);
3139: po[i][j][0]=(i==j ? 1.0 : 0.0);
3140: }
3141: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3142: for(h=1; h <=nhstepm; h++){
3143: for(d=1; d <=hstepm; d++){
3144: newm=savm;
3145: /* Covariates have to be included here again */
3146: cov[1]=1.;
3147: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3148: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3149: cov[2]=agexact;
3150: if(nagesqr==1)
1.222 brouard 3151: cov[3]= agexact*agexact;
1.218 brouard 3152: for (k=1; k<=cptcovn;k++)
1.222 brouard 3153: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3154: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3155: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3156: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3157: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3158: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3159: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3160: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3161: /* 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 3162:
3163:
1.217 brouard 3164: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3165: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3166: /* Careful transposed matrix */
1.222 brouard 3167: /* age is in cov[2] */
1.218 brouard 3168: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3169: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3170: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3171: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3172: /* if((int)age == 70){ */
3173: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3174: /* for(i=1; i<=nlstate+ndeath; i++) { */
3175: /* printf("%d pmmij ",i); */
3176: /* for(j=1;j<=nlstate+ndeath;j++) { */
3177: /* printf("%f ",pmmij[i][j]); */
3178: /* } */
3179: /* printf(" oldm "); */
3180: /* for(j=1;j<=nlstate+ndeath;j++) { */
3181: /* printf("%f ",oldm[i][j]); */
3182: /* } */
3183: /* printf("\n"); */
3184: /* } */
3185: /* } */
3186: savm=oldm;
3187: oldm=newm;
3188: }
3189: for(i=1; i<=nlstate+ndeath; i++)
3190: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3191: po[i][j][h]=newm[i][j];
3192: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3193: }
3194: /*printf("h=%d ",h);*/
3195: } /* end h */
1.222 brouard 3196: /* printf("\n H=%d \n",h); */
1.217 brouard 3197: return po;
3198: }
3199:
3200:
1.162 brouard 3201: #ifdef NLOPT
3202: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3203: double fret;
3204: double *xt;
3205: int j;
3206: myfunc_data *d2 = (myfunc_data *) pd;
3207: /* xt = (p1-1); */
3208: xt=vector(1,n);
3209: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3210:
3211: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3212: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3213: printf("Function = %.12lf ",fret);
3214: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3215: printf("\n");
3216: free_vector(xt,1,n);
3217: return fret;
3218: }
3219: #endif
1.126 brouard 3220:
3221: /*************** log-likelihood *************/
3222: double func( double *x)
3223: {
1.226 brouard 3224: int i, ii, j, k, mi, d, kk;
3225: int ioffset=0;
3226: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3227: double **out;
3228: double lli; /* Individual log likelihood */
3229: int s1, s2;
1.228 brouard 3230: 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 3231: double bbh, survp;
3232: long ipmx;
3233: double agexact;
3234: /*extern weight */
3235: /* We are differentiating ll according to initial status */
3236: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3237: /*for(i=1;i<imx;i++)
3238: printf(" %d\n",s[4][i]);
3239: */
1.162 brouard 3240:
1.226 brouard 3241: ++countcallfunc;
1.162 brouard 3242:
1.226 brouard 3243: cov[1]=1.;
1.126 brouard 3244:
1.226 brouard 3245: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3246: ioffset=0;
1.226 brouard 3247: if(mle==1){
3248: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3249: /* Computes the values of the ncovmodel covariates of the model
3250: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3251: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3252: to be observed in j being in i according to the model.
3253: */
1.243 brouard 3254: ioffset=2+nagesqr ;
1.233 brouard 3255: /* Fixed */
1.234 brouard 3256: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3257: 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)*/
3258: }
1.226 brouard 3259: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3260: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3261: has been calculated etc */
3262: /* For an individual i, wav[i] gives the number of effective waves */
3263: /* We compute the contribution to Likelihood of each effective transition
3264: mw[mi][i] is real wave of the mi th effectve wave */
3265: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3266: s2=s[mw[mi+1][i]][i];
3267: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3268: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3269: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3270: */
3271: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3272: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3273: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3274: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3275: }
3276: for (ii=1;ii<=nlstate+ndeath;ii++)
3277: for (j=1;j<=nlstate+ndeath;j++){
3278: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3279: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3280: }
3281: for(d=0; d<dh[mi][i]; d++){
3282: newm=savm;
3283: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3284: cov[2]=agexact;
3285: if(nagesqr==1)
3286: cov[3]= agexact*agexact; /* Should be changed here */
3287: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3288: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3289: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3290: else
3291: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3292: }
3293: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3294: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3295: savm=oldm;
3296: oldm=newm;
3297: } /* end mult */
3298:
3299: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3300: /* But now since version 0.9 we anticipate for bias at large stepm.
3301: * If stepm is larger than one month (smallest stepm) and if the exact delay
3302: * (in months) between two waves is not a multiple of stepm, we rounded to
3303: * the nearest (and in case of equal distance, to the lowest) interval but now
3304: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3305: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3306: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3307: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3308: * -stepm/2 to stepm/2 .
3309: * For stepm=1 the results are the same as for previous versions of Imach.
3310: * For stepm > 1 the results are less biased than in previous versions.
3311: */
1.234 brouard 3312: s1=s[mw[mi][i]][i];
3313: s2=s[mw[mi+1][i]][i];
3314: bbh=(double)bh[mi][i]/(double)stepm;
3315: /* bias bh is positive if real duration
3316: * is higher than the multiple of stepm and negative otherwise.
3317: */
3318: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3319: if( s2 > nlstate){
3320: /* i.e. if s2 is a death state and if the date of death is known
3321: then the contribution to the likelihood is the probability to
3322: die between last step unit time and current step unit time,
3323: which is also equal to probability to die before dh
3324: minus probability to die before dh-stepm .
3325: In version up to 0.92 likelihood was computed
3326: as if date of death was unknown. Death was treated as any other
3327: health state: the date of the interview describes the actual state
3328: and not the date of a change in health state. The former idea was
3329: to consider that at each interview the state was recorded
3330: (healthy, disable or death) and IMaCh was corrected; but when we
3331: introduced the exact date of death then we should have modified
3332: the contribution of an exact death to the likelihood. This new
3333: contribution is smaller and very dependent of the step unit
3334: stepm. It is no more the probability to die between last interview
3335: and month of death but the probability to survive from last
3336: interview up to one month before death multiplied by the
3337: probability to die within a month. Thanks to Chris
3338: Jackson for correcting this bug. Former versions increased
3339: mortality artificially. The bad side is that we add another loop
3340: which slows down the processing. The difference can be up to 10%
3341: lower mortality.
3342: */
3343: /* If, at the beginning of the maximization mostly, the
3344: cumulative probability or probability to be dead is
3345: constant (ie = 1) over time d, the difference is equal to
3346: 0. out[s1][3] = savm[s1][3]: probability, being at state
3347: s1 at precedent wave, to be dead a month before current
3348: wave is equal to probability, being at state s1 at
3349: precedent wave, to be dead at mont of the current
3350: wave. Then the observed probability (that this person died)
3351: is null according to current estimated parameter. In fact,
3352: it should be very low but not zero otherwise the log go to
3353: infinity.
3354: */
1.183 brouard 3355: /* #ifdef INFINITYORIGINAL */
3356: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3357: /* #else */
3358: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3359: /* lli=log(mytinydouble); */
3360: /* else */
3361: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3362: /* #endif */
1.226 brouard 3363: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3364:
1.226 brouard 3365: } else if ( s2==-1 ) { /* alive */
3366: for (j=1,survp=0. ; j<=nlstate; j++)
3367: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3368: /*survp += out[s1][j]; */
3369: lli= log(survp);
3370: }
3371: else if (s2==-4) {
3372: for (j=3,survp=0. ; j<=nlstate; j++)
3373: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3374: lli= log(survp);
3375: }
3376: else if (s2==-5) {
3377: for (j=1,survp=0. ; j<=2; j++)
3378: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3379: lli= log(survp);
3380: }
3381: else{
3382: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3383: /* 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 */
3384: }
3385: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3386: /*if(lli ==000.0)*/
3387: /*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); */
3388: ipmx +=1;
3389: sw += weight[i];
3390: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3391: /* if (lli < log(mytinydouble)){ */
3392: /* 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); */
3393: /* 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]); */
3394: /* } */
3395: } /* end of wave */
3396: } /* end of individual */
3397: } else if(mle==2){
3398: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3399: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3400: for(mi=1; mi<= wav[i]-1; mi++){
3401: for (ii=1;ii<=nlstate+ndeath;ii++)
3402: for (j=1;j<=nlstate+ndeath;j++){
3403: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3404: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3405: }
3406: for(d=0; d<=dh[mi][i]; d++){
3407: newm=savm;
3408: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3409: cov[2]=agexact;
3410: if(nagesqr==1)
3411: cov[3]= agexact*agexact;
3412: for (kk=1; kk<=cptcovage;kk++) {
3413: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3414: }
3415: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3416: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3417: savm=oldm;
3418: oldm=newm;
3419: } /* end mult */
3420:
3421: s1=s[mw[mi][i]][i];
3422: s2=s[mw[mi+1][i]][i];
3423: bbh=(double)bh[mi][i]/(double)stepm;
3424: 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 */
3425: ipmx +=1;
3426: sw += weight[i];
3427: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3428: } /* end of wave */
3429: } /* end of individual */
3430: } else if(mle==3){ /* exponential inter-extrapolation */
3431: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3432: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3433: for(mi=1; mi<= wav[i]-1; mi++){
3434: for (ii=1;ii<=nlstate+ndeath;ii++)
3435: for (j=1;j<=nlstate+ndeath;j++){
3436: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3437: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3438: }
3439: for(d=0; d<dh[mi][i]; d++){
3440: newm=savm;
3441: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3442: cov[2]=agexact;
3443: if(nagesqr==1)
3444: cov[3]= agexact*agexact;
3445: for (kk=1; kk<=cptcovage;kk++) {
3446: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3447: }
3448: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3449: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3450: savm=oldm;
3451: oldm=newm;
3452: } /* end mult */
3453:
3454: s1=s[mw[mi][i]][i];
3455: s2=s[mw[mi+1][i]][i];
3456: bbh=(double)bh[mi][i]/(double)stepm;
3457: 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 */
3458: ipmx +=1;
3459: sw += weight[i];
3460: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3461: } /* end of wave */
3462: } /* end of individual */
3463: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3464: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3465: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3466: for(mi=1; mi<= wav[i]-1; mi++){
3467: for (ii=1;ii<=nlstate+ndeath;ii++)
3468: for (j=1;j<=nlstate+ndeath;j++){
3469: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3470: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3471: }
3472: for(d=0; d<dh[mi][i]; d++){
3473: newm=savm;
3474: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3475: cov[2]=agexact;
3476: if(nagesqr==1)
3477: cov[3]= agexact*agexact;
3478: for (kk=1; kk<=cptcovage;kk++) {
3479: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3480: }
1.126 brouard 3481:
1.226 brouard 3482: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3483: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3484: savm=oldm;
3485: oldm=newm;
3486: } /* end mult */
3487:
3488: s1=s[mw[mi][i]][i];
3489: s2=s[mw[mi+1][i]][i];
3490: if( s2 > nlstate){
3491: lli=log(out[s1][s2] - savm[s1][s2]);
3492: } else if ( s2==-1 ) { /* alive */
3493: for (j=1,survp=0. ; j<=nlstate; j++)
3494: survp += out[s1][j];
3495: lli= log(survp);
3496: }else{
3497: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3498: }
3499: ipmx +=1;
3500: sw += weight[i];
3501: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3502: /* 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 3503: } /* end of wave */
3504: } /* end of individual */
3505: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3506: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3507: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3508: for(mi=1; mi<= wav[i]-1; mi++){
3509: for (ii=1;ii<=nlstate+ndeath;ii++)
3510: for (j=1;j<=nlstate+ndeath;j++){
3511: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3512: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3513: }
3514: for(d=0; d<dh[mi][i]; d++){
3515: newm=savm;
3516: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3517: cov[2]=agexact;
3518: if(nagesqr==1)
3519: cov[3]= agexact*agexact;
3520: for (kk=1; kk<=cptcovage;kk++) {
3521: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3522: }
1.126 brouard 3523:
1.226 brouard 3524: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3525: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3526: savm=oldm;
3527: oldm=newm;
3528: } /* end mult */
3529:
3530: s1=s[mw[mi][i]][i];
3531: s2=s[mw[mi+1][i]][i];
3532: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3533: ipmx +=1;
3534: sw += weight[i];
3535: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3536: /*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]);*/
3537: } /* end of wave */
3538: } /* end of individual */
3539: } /* End of if */
3540: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3541: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3542: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3543: return -l;
1.126 brouard 3544: }
3545:
3546: /*************** log-likelihood *************/
3547: double funcone( double *x)
3548: {
1.228 brouard 3549: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3550: int i, ii, j, k, mi, d, kk;
1.228 brouard 3551: int ioffset=0;
1.131 brouard 3552: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3553: double **out;
3554: double lli; /* Individual log likelihood */
3555: double llt;
3556: int s1, s2;
1.228 brouard 3557: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3558:
1.126 brouard 3559: double bbh, survp;
1.187 brouard 3560: double agexact;
1.214 brouard 3561: double agebegin, ageend;
1.126 brouard 3562: /*extern weight */
3563: /* We are differentiating ll according to initial status */
3564: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3565: /*for(i=1;i<imx;i++)
3566: printf(" %d\n",s[4][i]);
3567: */
3568: cov[1]=1.;
3569:
3570: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3571: ioffset=0;
3572: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3573: /* ioffset=2+nagesqr+cptcovage; */
3574: ioffset=2+nagesqr;
1.232 brouard 3575: /* Fixed */
1.224 brouard 3576: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3577: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3578: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3579: 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)*/
3580: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3581: /* cov[2+6]=covar[Tvar[6]][i]; */
3582: /* cov[2+6]=covar[2][i]; V2 */
3583: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3584: /* cov[2+7]=covar[Tvar[7]][i]; */
3585: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3586: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3587: /* cov[2+9]=covar[Tvar[9]][i]; */
3588: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3589: }
1.232 brouard 3590: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3591: /* 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?)*\/ */
3592: /* } */
1.231 brouard 3593: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3594: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3595: /* } */
1.225 brouard 3596:
1.233 brouard 3597:
3598: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3599: /* Wave varying (but not age varying) */
3600: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3601: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3602: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3603: }
1.232 brouard 3604: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3605: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3606: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3607: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3608: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3609: /* 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 3610: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3611: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3612: /* /\* 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]); *\/ */
3613: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3614: /* } */
1.126 brouard 3615: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3616: for (j=1;j<=nlstate+ndeath;j++){
3617: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3618: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3619: }
1.214 brouard 3620:
3621: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3622: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3623: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3624: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3625: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3626: and mw[mi+1][i]. dh depends on stepm.*/
3627: newm=savm;
1.247 brouard 3628: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3629: cov[2]=agexact;
3630: if(nagesqr==1)
3631: cov[3]= agexact*agexact;
3632: for (kk=1; kk<=cptcovage;kk++) {
3633: if(!FixedV[Tvar[Tage[kk]]])
3634: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3635: else
3636: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3637: }
3638: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3639: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3640: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3641: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3642: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3643: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3644: savm=oldm;
3645: oldm=newm;
1.126 brouard 3646: } /* end mult */
3647:
3648: s1=s[mw[mi][i]][i];
3649: s2=s[mw[mi+1][i]][i];
1.217 brouard 3650: /* if(s2==-1){ */
3651: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3652: /* /\* exit(1); *\/ */
3653: /* } */
1.126 brouard 3654: bbh=(double)bh[mi][i]/(double)stepm;
3655: /* bias is positive if real duration
3656: * is higher than the multiple of stepm and negative otherwise.
3657: */
3658: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3659: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3660: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3661: for (j=1,survp=0. ; j<=nlstate; j++)
3662: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3663: lli= log(survp);
1.126 brouard 3664: }else if (mle==1){
1.242 brouard 3665: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3666: } else if(mle==2){
1.242 brouard 3667: 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 3668: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3669: 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 3670: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3671: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3672: } else{ /* mle=0 back to 1 */
1.242 brouard 3673: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3674: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3675: } /* End of if */
3676: ipmx +=1;
3677: sw += weight[i];
3678: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3679: /*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 3680: if(globpr){
1.246 brouard 3681: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3682: %11.6f %11.6f %11.6f ", \
1.242 brouard 3683: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3684: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3685: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3686: llt +=ll[k]*gipmx/gsw;
3687: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3688: }
3689: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3690: }
1.232 brouard 3691: } /* end of wave */
3692: } /* end of individual */
3693: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3694: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3695: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3696: if(globpr==0){ /* First time we count the contributions and weights */
3697: gipmx=ipmx;
3698: gsw=sw;
3699: }
3700: return -l;
1.126 brouard 3701: }
3702:
3703:
3704: /*************** function likelione ***********/
3705: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3706: {
3707: /* This routine should help understanding what is done with
3708: the selection of individuals/waves and
3709: to check the exact contribution to the likelihood.
3710: Plotting could be done.
3711: */
3712: int k;
3713:
3714: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3715: strcpy(fileresilk,"ILK_");
1.202 brouard 3716: strcat(fileresilk,fileresu);
1.126 brouard 3717: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3718: printf("Problem with resultfile: %s\n", fileresilk);
3719: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3720: }
1.214 brouard 3721: 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");
3722: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3723: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3724: for(k=1; k<=nlstate; k++)
3725: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3726: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3727: }
3728:
3729: *fretone=(*funcone)(p);
3730: if(*globpri !=0){
3731: fclose(ficresilk);
1.205 brouard 3732: if (mle ==0)
3733: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3734: else if(mle >=1)
3735: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3736: 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 3737:
1.208 brouard 3738:
3739: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3740: 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 3741: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3742: }
1.207 brouard 3743: 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 3744: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3745: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3746: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3747: fflush(fichtm);
1.205 brouard 3748: }
1.126 brouard 3749: return;
3750: }
3751:
3752:
3753: /*********** Maximum Likelihood Estimation ***************/
3754:
3755: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3756: {
1.165 brouard 3757: int i,j, iter=0;
1.126 brouard 3758: double **xi;
3759: double fret;
3760: double fretone; /* Only one call to likelihood */
3761: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3762:
3763: #ifdef NLOPT
3764: int creturn;
3765: nlopt_opt opt;
3766: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3767: double *lb;
3768: double minf; /* the minimum objective value, upon return */
3769: double * p1; /* Shifted parameters from 0 instead of 1 */
3770: myfunc_data dinst, *d = &dinst;
3771: #endif
3772:
3773:
1.126 brouard 3774: xi=matrix(1,npar,1,npar);
3775: for (i=1;i<=npar;i++)
3776: for (j=1;j<=npar;j++)
3777: xi[i][j]=(i==j ? 1.0 : 0.0);
3778: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3779: strcpy(filerespow,"POW_");
1.126 brouard 3780: strcat(filerespow,fileres);
3781: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3782: printf("Problem with resultfile: %s\n", filerespow);
3783: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3784: }
3785: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3786: for (i=1;i<=nlstate;i++)
3787: for(j=1;j<=nlstate+ndeath;j++)
3788: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3789: fprintf(ficrespow,"\n");
1.162 brouard 3790: #ifdef POWELL
1.126 brouard 3791: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3792: #endif
1.126 brouard 3793:
1.162 brouard 3794: #ifdef NLOPT
3795: #ifdef NEWUOA
3796: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3797: #else
3798: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3799: #endif
3800: lb=vector(0,npar-1);
3801: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3802: nlopt_set_lower_bounds(opt, lb);
3803: nlopt_set_initial_step1(opt, 0.1);
3804:
3805: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3806: d->function = func;
3807: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3808: nlopt_set_min_objective(opt, myfunc, d);
3809: nlopt_set_xtol_rel(opt, ftol);
3810: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3811: printf("nlopt failed! %d\n",creturn);
3812: }
3813: else {
3814: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3815: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3816: iter=1; /* not equal */
3817: }
3818: nlopt_destroy(opt);
3819: #endif
1.126 brouard 3820: free_matrix(xi,1,npar,1,npar);
3821: fclose(ficrespow);
1.203 brouard 3822: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3823: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3824: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3825:
3826: }
3827:
3828: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3829: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3830: {
3831: double **a,**y,*x,pd;
1.203 brouard 3832: /* double **hess; */
1.164 brouard 3833: int i, j;
1.126 brouard 3834: int *indx;
3835:
3836: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3837: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3838: void lubksb(double **a, int npar, int *indx, double b[]) ;
3839: void ludcmp(double **a, int npar, int *indx, double *d) ;
3840: double gompertz(double p[]);
1.203 brouard 3841: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3842:
3843: printf("\nCalculation of the hessian matrix. Wait...\n");
3844: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3845: for (i=1;i<=npar;i++){
1.203 brouard 3846: printf("%d-",i);fflush(stdout);
3847: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3848:
3849: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3850:
3851: /* printf(" %f ",p[i]);
3852: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3853: }
3854:
3855: for (i=1;i<=npar;i++) {
3856: for (j=1;j<=npar;j++) {
3857: if (j>i) {
1.203 brouard 3858: printf(".%d-%d",i,j);fflush(stdout);
3859: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3860: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3861:
3862: hess[j][i]=hess[i][j];
3863: /*printf(" %lf ",hess[i][j]);*/
3864: }
3865: }
3866: }
3867: printf("\n");
3868: fprintf(ficlog,"\n");
3869:
3870: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3871: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3872:
3873: a=matrix(1,npar,1,npar);
3874: y=matrix(1,npar,1,npar);
3875: x=vector(1,npar);
3876: indx=ivector(1,npar);
3877: for (i=1;i<=npar;i++)
3878: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3879: ludcmp(a,npar,indx,&pd);
3880:
3881: for (j=1;j<=npar;j++) {
3882: for (i=1;i<=npar;i++) x[i]=0;
3883: x[j]=1;
3884: lubksb(a,npar,indx,x);
3885: for (i=1;i<=npar;i++){
3886: matcov[i][j]=x[i];
3887: }
3888: }
3889:
3890: printf("\n#Hessian matrix#\n");
3891: fprintf(ficlog,"\n#Hessian matrix#\n");
3892: for (i=1;i<=npar;i++) {
3893: for (j=1;j<=npar;j++) {
1.203 brouard 3894: printf("%.6e ",hess[i][j]);
3895: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3896: }
3897: printf("\n");
3898: fprintf(ficlog,"\n");
3899: }
3900:
1.203 brouard 3901: /* printf("\n#Covariance matrix#\n"); */
3902: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3903: /* for (i=1;i<=npar;i++) { */
3904: /* for (j=1;j<=npar;j++) { */
3905: /* printf("%.6e ",matcov[i][j]); */
3906: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3907: /* } */
3908: /* printf("\n"); */
3909: /* fprintf(ficlog,"\n"); */
3910: /* } */
3911:
1.126 brouard 3912: /* Recompute Inverse */
1.203 brouard 3913: /* for (i=1;i<=npar;i++) */
3914: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3915: /* ludcmp(a,npar,indx,&pd); */
3916:
3917: /* printf("\n#Hessian matrix recomputed#\n"); */
3918:
3919: /* for (j=1;j<=npar;j++) { */
3920: /* for (i=1;i<=npar;i++) x[i]=0; */
3921: /* x[j]=1; */
3922: /* lubksb(a,npar,indx,x); */
3923: /* for (i=1;i<=npar;i++){ */
3924: /* y[i][j]=x[i]; */
3925: /* printf("%.3e ",y[i][j]); */
3926: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3927: /* } */
3928: /* printf("\n"); */
3929: /* fprintf(ficlog,"\n"); */
3930: /* } */
3931:
3932: /* Verifying the inverse matrix */
3933: #ifdef DEBUGHESS
3934: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3935:
1.203 brouard 3936: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3937: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3938:
3939: for (j=1;j<=npar;j++) {
3940: for (i=1;i<=npar;i++){
1.203 brouard 3941: printf("%.2f ",y[i][j]);
3942: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3943: }
3944: printf("\n");
3945: fprintf(ficlog,"\n");
3946: }
1.203 brouard 3947: #endif
1.126 brouard 3948:
3949: free_matrix(a,1,npar,1,npar);
3950: free_matrix(y,1,npar,1,npar);
3951: free_vector(x,1,npar);
3952: free_ivector(indx,1,npar);
1.203 brouard 3953: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3954:
3955:
3956: }
3957:
3958: /*************** hessian matrix ****************/
3959: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3960: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3961: int i;
3962: int l=1, lmax=20;
1.203 brouard 3963: double k1,k2, res, fx;
1.132 brouard 3964: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3965: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3966: int k=0,kmax=10;
3967: double l1;
3968:
3969: fx=func(x);
3970: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3971: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3972: l1=pow(10,l);
3973: delts=delt;
3974: for(k=1 ; k <kmax; k=k+1){
3975: delt = delta*(l1*k);
3976: p2[theta]=x[theta] +delt;
1.145 brouard 3977: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3978: p2[theta]=x[theta]-delt;
3979: k2=func(p2)-fx;
3980: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3981: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3982:
1.203 brouard 3983: #ifdef DEBUGHESSII
1.126 brouard 3984: 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);
3985: 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);
3986: #endif
3987: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3988: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3989: k=kmax;
3990: }
3991: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3992: k=kmax; l=lmax*10;
1.126 brouard 3993: }
3994: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3995: delts=delt;
3996: }
1.203 brouard 3997: } /* End loop k */
1.126 brouard 3998: }
3999: delti[theta]=delts;
4000: return res;
4001:
4002: }
4003:
1.203 brouard 4004: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4005: {
4006: int i;
1.164 brouard 4007: int l=1, lmax=20;
1.126 brouard 4008: double k1,k2,k3,k4,res,fx;
1.132 brouard 4009: double p2[MAXPARM+1];
1.203 brouard 4010: int k, kmax=1;
4011: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4012:
4013: int firstime=0;
1.203 brouard 4014:
1.126 brouard 4015: fx=func(x);
1.203 brouard 4016: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4017: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4018: p2[thetai]=x[thetai]+delti[thetai]*k;
4019: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4020: k1=func(p2)-fx;
4021:
1.203 brouard 4022: p2[thetai]=x[thetai]+delti[thetai]*k;
4023: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4024: k2=func(p2)-fx;
4025:
1.203 brouard 4026: p2[thetai]=x[thetai]-delti[thetai]*k;
4027: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4028: k3=func(p2)-fx;
4029:
1.203 brouard 4030: p2[thetai]=x[thetai]-delti[thetai]*k;
4031: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4032: k4=func(p2)-fx;
1.203 brouard 4033: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4034: if(k1*k2*k3*k4 <0.){
1.208 brouard 4035: firstime=1;
1.203 brouard 4036: kmax=kmax+10;
1.208 brouard 4037: }
4038: if(kmax >=10 || firstime ==1){
1.246 brouard 4039: 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);
4040: 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 4041: 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);
4042: 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);
4043: }
4044: #ifdef DEBUGHESSIJ
4045: v1=hess[thetai][thetai];
4046: v2=hess[thetaj][thetaj];
4047: cv12=res;
4048: /* Computing eigen value of Hessian matrix */
4049: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4050: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4051: if ((lc2 <0) || (lc1 <0) ){
4052: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4053: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4054: 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);
4055: 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);
4056: }
1.126 brouard 4057: #endif
4058: }
4059: return res;
4060: }
4061:
1.203 brouard 4062: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4063: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4064: /* { */
4065: /* int i; */
4066: /* int l=1, lmax=20; */
4067: /* double k1,k2,k3,k4,res,fx; */
4068: /* double p2[MAXPARM+1]; */
4069: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4070: /* int k=0,kmax=10; */
4071: /* double l1; */
4072:
4073: /* fx=func(x); */
4074: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4075: /* l1=pow(10,l); */
4076: /* delts=delt; */
4077: /* for(k=1 ; k <kmax; k=k+1){ */
4078: /* delt = delti*(l1*k); */
4079: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4080: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4081: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4082: /* k1=func(p2)-fx; */
4083:
4084: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4085: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4086: /* k2=func(p2)-fx; */
4087:
4088: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4089: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4090: /* k3=func(p2)-fx; */
4091:
4092: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4093: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4094: /* k4=func(p2)-fx; */
4095: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4096: /* #ifdef DEBUGHESSIJ */
4097: /* 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); */
4098: /* 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); */
4099: /* #endif */
4100: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4101: /* k=kmax; */
4102: /* } */
4103: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4104: /* k=kmax; l=lmax*10; */
4105: /* } */
4106: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4107: /* delts=delt; */
4108: /* } */
4109: /* } /\* End loop k *\/ */
4110: /* } */
4111: /* delti[theta]=delts; */
4112: /* return res; */
4113: /* } */
4114:
4115:
1.126 brouard 4116: /************** Inverse of matrix **************/
4117: void ludcmp(double **a, int n, int *indx, double *d)
4118: {
4119: int i,imax,j,k;
4120: double big,dum,sum,temp;
4121: double *vv;
4122:
4123: vv=vector(1,n);
4124: *d=1.0;
4125: for (i=1;i<=n;i++) {
4126: big=0.0;
4127: for (j=1;j<=n;j++)
4128: if ((temp=fabs(a[i][j])) > big) big=temp;
4129: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4130: vv[i]=1.0/big;
4131: }
4132: for (j=1;j<=n;j++) {
4133: for (i=1;i<j;i++) {
4134: sum=a[i][j];
4135: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4136: a[i][j]=sum;
4137: }
4138: big=0.0;
4139: for (i=j;i<=n;i++) {
4140: sum=a[i][j];
4141: for (k=1;k<j;k++)
4142: sum -= a[i][k]*a[k][j];
4143: a[i][j]=sum;
4144: if ( (dum=vv[i]*fabs(sum)) >= big) {
4145: big=dum;
4146: imax=i;
4147: }
4148: }
4149: if (j != imax) {
4150: for (k=1;k<=n;k++) {
4151: dum=a[imax][k];
4152: a[imax][k]=a[j][k];
4153: a[j][k]=dum;
4154: }
4155: *d = -(*d);
4156: vv[imax]=vv[j];
4157: }
4158: indx[j]=imax;
4159: if (a[j][j] == 0.0) a[j][j]=TINY;
4160: if (j != n) {
4161: dum=1.0/(a[j][j]);
4162: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4163: }
4164: }
4165: free_vector(vv,1,n); /* Doesn't work */
4166: ;
4167: }
4168:
4169: void lubksb(double **a, int n, int *indx, double b[])
4170: {
4171: int i,ii=0,ip,j;
4172: double sum;
4173:
4174: for (i=1;i<=n;i++) {
4175: ip=indx[i];
4176: sum=b[ip];
4177: b[ip]=b[i];
4178: if (ii)
4179: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4180: else if (sum) ii=i;
4181: b[i]=sum;
4182: }
4183: for (i=n;i>=1;i--) {
4184: sum=b[i];
4185: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4186: b[i]=sum/a[i][i];
4187: }
4188: }
4189:
4190: void pstamp(FILE *fichier)
4191: {
1.196 brouard 4192: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4193: }
4194:
1.253 brouard 4195: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4196:
4197: /* y=a+bx regression */
4198: double sumx = 0.0; /* sum of x */
4199: double sumx2 = 0.0; /* sum of x**2 */
4200: double sumxy = 0.0; /* sum of x * y */
4201: double sumy = 0.0; /* sum of y */
4202: double sumy2 = 0.0; /* sum of y**2 */
4203: double sume2; /* sum of square or residuals */
4204: double yhat;
4205:
4206: double denom=0;
4207: int i;
4208: int ne=*no;
4209:
4210: for ( i=ifi, ne=0;i<=ila;i++) {
4211: if(!isfinite(x[i]) || !isfinite(y[i])){
4212: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4213: continue;
4214: }
4215: ne=ne+1;
4216: sumx += x[i];
4217: sumx2 += x[i]*x[i];
4218: sumxy += x[i] * y[i];
4219: sumy += y[i];
4220: sumy2 += y[i]*y[i];
4221: denom = (ne * sumx2 - sumx*sumx);
4222: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4223: }
4224:
4225: denom = (ne * sumx2 - sumx*sumx);
4226: if (denom == 0) {
4227: // vertical, slope m is infinity
4228: *b = INFINITY;
4229: *a = 0;
4230: if (r) *r = 0;
4231: return 1;
4232: }
4233:
4234: *b = (ne * sumxy - sumx * sumy) / denom;
4235: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4236: if (r!=NULL) {
4237: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4238: sqrt((sumx2 - sumx*sumx/ne) *
4239: (sumy2 - sumy*sumy/ne));
4240: }
4241: *no=ne;
4242: for ( i=ifi, ne=0;i<=ila;i++) {
4243: if(!isfinite(x[i]) || !isfinite(y[i])){
4244: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4245: continue;
4246: }
4247: ne=ne+1;
4248: yhat = y[i] - *a -*b* x[i];
4249: sume2 += yhat * yhat ;
4250:
4251: denom = (ne * sumx2 - sumx*sumx);
4252: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4253: }
4254: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4255: *sa= *sb * sqrt(sumx2/ne);
4256:
4257: return 0;
4258: }
4259:
1.126 brouard 4260: /************ Frequencies ********************/
1.251 brouard 4261: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4262: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4263: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4264: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4265:
1.253 brouard 4266: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4267: int iind=0, iage=0;
4268: int mi; /* Effective wave */
4269: int first;
4270: double ***freq; /* Frequencies */
1.253 brouard 4271: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4272: int no;
1.226 brouard 4273: double *meanq;
4274: double **meanqt;
4275: double *pp, **prop, *posprop, *pospropt;
4276: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4277: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4278: double agebegin, ageend;
4279:
4280: pp=vector(1,nlstate);
1.251 brouard 4281: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4282: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4283: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4284: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4285: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4286: meanqt=matrix(1,lastpass,1,nqtveff);
4287: strcpy(fileresp,"P_");
4288: strcat(fileresp,fileresu);
4289: /*strcat(fileresphtm,fileresu);*/
4290: if((ficresp=fopen(fileresp,"w"))==NULL) {
4291: printf("Problem with prevalence resultfile: %s\n", fileresp);
4292: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4293: exit(0);
4294: }
1.240 brouard 4295:
1.226 brouard 4296: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4297: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4298: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4299: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4300: fflush(ficlog);
4301: exit(70);
4302: }
4303: else{
4304: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4305: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4306: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4307: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4308: }
1.237 brouard 4309: 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 4310:
1.226 brouard 4311: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4312: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4313: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4314: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4315: fflush(ficlog);
4316: exit(70);
1.240 brouard 4317: } else{
1.226 brouard 4318: 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 4319: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4320: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4321: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4322: }
1.240 brouard 4323: 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);
4324:
1.253 brouard 4325: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4326: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4327: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4328: j1=0;
1.126 brouard 4329:
1.227 brouard 4330: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4331: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4332: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4333:
4334:
1.226 brouard 4335: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4336: reference=low_education V1=0,V2=0
4337: med_educ V1=1 V2=0,
4338: high_educ V1=0 V2=1
4339: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4340: */
1.249 brouard 4341: dateintsum=0;
4342: k2cpt=0;
4343:
1.253 brouard 4344: if(cptcoveff == 0 )
4345: nl=1; /* Constant model only */
4346: else
4347: nl=2;
4348: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4349: if(nj==1)
4350: j=0; /* First pass for the constant */
4351: else
4352: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4353: first=1;
4354: 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 */
4355: posproptt=0.;
4356: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4357: scanf("%d", i);*/
4358: for (i=-5; i<=nlstate+ndeath; i++)
4359: for (jk=-5; jk<=nlstate+ndeath; jk++)
4360: for(m=iagemin; m <= iagemax+3; m++)
4361: freq[i][jk][m]=0;
4362:
4363: for (i=1; i<=nlstate; i++) {
1.240 brouard 4364: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4365: prop[i][m]=0;
4366: posprop[i]=0;
4367: pospropt[i]=0;
4368: }
4369: /* for (z1=1; z1<= nqfveff; z1++) { */
4370: /* meanq[z1]+=0.; */
4371: /* for(m=1;m<=lastpass;m++){ */
4372: /* meanqt[m][z1]=0.; */
4373: /* } */
4374: /* } */
4375:
4376: /* dateintsum=0; */
4377: /* k2cpt=0; */
4378:
4379: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4380: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4381: bool=1;
4382: if(j !=0){
4383: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4384: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4385: /* for (z1=1; z1<= nqfveff; z1++) { */
4386: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4387: /* } */
4388: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4389: /* if(Tvaraff[z1] ==-20){ */
4390: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4391: /* }else if(Tvaraff[z1] ==-10){ */
4392: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4393: /* }else */
4394: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4395: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4396: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4397: /* 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",
4398: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4399: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4400: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4401: } /* Onlyf fixed */
4402: } /* end z1 */
4403: } /* cptcovn > 0 */
4404: } /* end any */
4405: }/* end j==0 */
4406: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4407: /* for(m=firstpass; m<=lastpass; m++){ */
4408: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4409: m=mw[mi][iind];
4410: if(j!=0){
4411: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4412: for (z1=1; z1<=cptcoveff; z1++) {
4413: if( Fixed[Tmodelind[z1]]==1){
4414: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4415: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4416: value is -1, we don't select. It differs from the
4417: constant and age model which counts them. */
4418: bool=0; /* not selected */
4419: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4420: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4421: bool=0;
4422: }
4423: }
4424: }
4425: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4426: } /* end j==0 */
4427: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4428: if(bool==1){
4429: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4430: and mw[mi+1][iind]. dh depends on stepm. */
4431: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4432: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4433: if(m >=firstpass && m <=lastpass){
4434: k2=anint[m][iind]+(mint[m][iind]/12.);
4435: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4436: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4437: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4438: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4439: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4440: if (m<lastpass) {
4441: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4442: /* 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]); */
4443: if(s[m][iind]==-1)
4444: 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.));
4445: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4446: /* if((int)agev[m][iind] == 55) */
4447: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4448: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4449: 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 4450: }
1.251 brouard 4451: } /* end if between passes */
4452: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4453: dateintsum=dateintsum+k2; /* on all covariates ?*/
4454: k2cpt++;
4455: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4456: }
1.251 brouard 4457: }else{
4458: bool=1;
4459: }/* end bool 2 */
4460: } /* end m */
4461: } /* end bool */
4462: } /* end iind = 1 to imx */
4463: /* prop[s][age] is feeded for any initial and valid live state as well as
4464: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4465:
4466:
4467: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4468: pstamp(ficresp);
4469: if (cptcoveff>0 && j!=0){
4470: printf( "\n#********** Variable ");
4471: fprintf(ficresp, "\n#********** Variable ");
4472: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4473: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4474: fprintf(ficlog, "\n#********** Variable ");
4475: for (z1=1; z1<=cptcoveff; z1++){
4476: if(!FixedV[Tvaraff[z1]]){
4477: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4478: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4479: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4480: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4481: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4482: }else{
1.251 brouard 4483: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4484: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4485: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4486: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4487: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4488: }
4489: }
4490: printf( "**********\n#");
4491: fprintf(ficresp, "**********\n#");
4492: fprintf(ficresphtm, "**********</h3>\n");
4493: fprintf(ficresphtmfr, "**********</h3>\n");
4494: fprintf(ficlog, "**********\n");
4495: }
4496: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4497: for(i=1; i<=nlstate;i++) {
4498: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4499: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4500: }
4501: fprintf(ficresp, "\n");
4502: fprintf(ficresphtm, "\n");
4503:
4504: /* Header of frequency table by age */
4505: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4506: fprintf(ficresphtmfr,"<th>Age</th> ");
4507: for(jk=-1; jk <=nlstate+ndeath; jk++){
4508: for(m=-1; m <=nlstate+ndeath; m++){
4509: if(jk!=0 && m!=0)
4510: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4511: }
1.226 brouard 4512: }
1.251 brouard 4513: fprintf(ficresphtmfr, "\n");
4514:
4515: /* For each age */
4516: for(iage=iagemin; iage <= iagemax+3; iage++){
4517: fprintf(ficresphtm,"<tr>");
4518: if(iage==iagemax+1){
4519: fprintf(ficlog,"1");
4520: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4521: }else if(iage==iagemax+2){
4522: fprintf(ficlog,"0");
4523: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4524: }else if(iage==iagemax+3){
4525: fprintf(ficlog,"Total");
4526: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4527: }else{
1.240 brouard 4528: if(first==1){
1.251 brouard 4529: first=0;
4530: printf("See log file for details...\n");
4531: }
4532: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4533: fprintf(ficlog,"Age %d", iage);
4534: }
4535: for(jk=1; jk <=nlstate ; jk++){
4536: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4537: pp[jk] += freq[jk][m][iage];
4538: }
4539: for(jk=1; jk <=nlstate ; jk++){
4540: for(m=-1, pos=0; m <=0 ; m++)
4541: pos += freq[jk][m][iage];
4542: if(pp[jk]>=1.e-10){
4543: if(first==1){
4544: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4545: }
4546: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4547: }else{
4548: if(first==1)
4549: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4550: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4551: }
4552: }
4553:
1.251 brouard 4554: for(jk=1; jk <=nlstate ; jk++){
4555: /* posprop[jk]=0; */
4556: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4557: pp[jk] += freq[jk][m][iage];
4558: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4559:
4560: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4561: pos += pp[jk]; /* pos is the total number of transitions until this age */
4562: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4563: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4564: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4565: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4566: }
1.251 brouard 4567: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4568: if(pos>=1.e-5){
1.251 brouard 4569: if(first==1)
4570: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4571: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4572: }else{
4573: if(first==1)
4574: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4575: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4576: }
4577: if( iage <= iagemax){
4578: if(pos>=1.e-5){
4579: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4580: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4581: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4582: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4583: }
4584: else{
4585: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4586: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4587: }
1.240 brouard 4588: }
1.251 brouard 4589: pospropt[jk] +=posprop[jk];
4590: } /* end loop jk */
4591: /* pospropt=0.; */
4592: for(jk=-1; jk <=nlstate+ndeath; jk++){
4593: for(m=-1; m <=nlstate+ndeath; m++){
4594: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4595: if(first==1){
4596: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4597: }
1.253 brouard 4598: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4599: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4600: }
4601: if(jk!=0 && m!=0)
4602: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4603: }
1.251 brouard 4604: } /* end loop jk */
4605: posproptt=0.;
4606: for(jk=1; jk <=nlstate; jk++){
4607: posproptt += pospropt[jk];
4608: }
4609: fprintf(ficresphtmfr,"</tr>\n ");
4610: if(iage <= iagemax){
4611: fprintf(ficresp,"\n");
4612: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4613: }
1.251 brouard 4614: if(first==1)
4615: printf("Others in log...\n");
4616: fprintf(ficlog,"\n");
4617: } /* end loop age iage */
4618: fprintf(ficresphtm,"<tr><th>Tot</th>");
4619: for(jk=1; jk <=nlstate ; jk++){
4620: if(posproptt < 1.e-5){
4621: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4622: }else{
4623: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4624: }
1.226 brouard 4625: }
1.251 brouard 4626: fprintf(ficresphtm,"</tr>\n");
4627: fprintf(ficresphtm,"</table>\n");
4628: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4629: if(posproptt < 1.e-5){
1.251 brouard 4630: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4631: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4632: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4633: invalidvarcomb[j1]=1;
1.226 brouard 4634: }else{
1.251 brouard 4635: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4636: invalidvarcomb[j1]=0;
1.226 brouard 4637: }
1.251 brouard 4638: fprintf(ficresphtmfr,"</table>\n");
4639: fprintf(ficlog,"\n");
4640: if(j!=0){
4641: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4642: for(i=1,jk=1; i <=nlstate; i++){
4643: for(k=1; k <=(nlstate+ndeath); k++){
4644: if (k != i) {
4645: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4646: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4647: if(j1==1){ /* All dummy covariates to zero */
4648: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4649: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4650: printf("%d%d ",i,k);
4651: fprintf(ficlog,"%d%d ",i,k);
4652: 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]));
4653: 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]));
4654: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4655: }
1.253 brouard 4656: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4657: for(iage=iagemin; iage <= iagemax+3; iage++){
4658: x[iage]= (double)iage;
4659: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4660: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4661: }
4662: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4663: pstart[jk]=b;
4664: pstart[jk-1]=a;
1.252 brouard 4665: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
4666: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
4667: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.251 brouard 4668: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4669: printf("%d%d ",i,k);
4670: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4671: 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]));
4672: }else{ /* Other cases, like quantitative fixed or varying covariates */
4673: ;
4674: }
4675: /* printf("%12.7f )", param[i][jj][k]); */
4676: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4677: jk++;
4678: } /* end jj */
4679: } /* end k!= i */
4680: } /* end k */
4681: } /* end i, jk */
4682: } /* end j !=0 */
4683: } /* end selected combination of covariate j1 */
4684: if(j==0){ /* We can estimate starting values from the occurences in each case */
4685: printf("#Freqsummary: Starting values for the constants:\n");
4686: fprintf(ficlog,"\n");
4687: for(i=1,jk=1; i <=nlstate; i++){
4688: for(k=1; k <=(nlstate+ndeath); k++){
4689: if (k != i) {
4690: printf("%d%d ",i,k);
4691: fprintf(ficlog,"%d%d ",i,k);
4692: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4693: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4694: if(jj==1){ /* Age has to be done */
4695: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4696: 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]));
4697: 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]));
4698: }
4699: /* printf("%12.7f )", param[i][jj][k]); */
4700: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4701: jk++;
1.250 brouard 4702: }
1.251 brouard 4703: printf("\n");
4704: fprintf(ficlog,"\n");
1.250 brouard 4705: }
4706: }
4707: }
1.251 brouard 4708: printf("#Freqsummary\n");
4709: fprintf(ficlog,"\n");
4710: for(jk=-1; jk <=nlstate+ndeath; jk++){
4711: for(m=-1; m <=nlstate+ndeath; m++){
4712: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4713: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4714: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4715: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4716: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4717: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4718: /* } */
4719: }
4720: } /* end loop jk */
4721:
4722: printf("\n");
4723: fprintf(ficlog,"\n");
4724: } /* end j=0 */
1.249 brouard 4725: } /* end j */
1.252 brouard 4726:
1.253 brouard 4727: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4728: for(i=1, jk=1; i <=nlstate; i++){
4729: for(j=1; j <=nlstate+ndeath; j++){
4730: if(j!=i){
4731: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4732: printf("%1d%1d",i,j);
4733: fprintf(ficparo,"%1d%1d",i,j);
4734: for(k=1; k<=ncovmodel;k++){
4735: /* printf(" %lf",param[i][j][k]); */
4736: /* fprintf(ficparo," %lf",param[i][j][k]); */
4737: p[jk]=pstart[jk];
4738: printf(" %f ",pstart[jk]);
4739: fprintf(ficparo," %f ",pstart[jk]);
4740: jk++;
4741: }
4742: printf("\n");
4743: fprintf(ficparo,"\n");
4744: }
4745: }
4746: }
4747: } /* end mle=-2 */
1.226 brouard 4748: dateintmean=dateintsum/k2cpt;
1.240 brouard 4749:
1.226 brouard 4750: fclose(ficresp);
4751: fclose(ficresphtm);
4752: fclose(ficresphtmfr);
4753: free_vector(meanq,1,nqfveff);
4754: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4755: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4756: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4757: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4758: free_vector(pospropt,1,nlstate);
4759: free_vector(posprop,1,nlstate);
1.251 brouard 4760: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4761: free_vector(pp,1,nlstate);
4762: /* End of freqsummary */
4763: }
1.126 brouard 4764:
4765: /************ Prevalence ********************/
1.227 brouard 4766: 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)
4767: {
4768: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4769: in each health status at the date of interview (if between dateprev1 and dateprev2).
4770: We still use firstpass and lastpass as another selection.
4771: */
1.126 brouard 4772:
1.227 brouard 4773: int i, m, jk, j1, bool, z1,j, iv;
4774: int mi; /* Effective wave */
4775: int iage;
4776: double agebegin, ageend;
4777:
4778: double **prop;
4779: double posprop;
4780: double y2; /* in fractional years */
4781: int iagemin, iagemax;
4782: int first; /** to stop verbosity which is redirected to log file */
4783:
4784: iagemin= (int) agemin;
4785: iagemax= (int) agemax;
4786: /*pp=vector(1,nlstate);*/
1.251 brouard 4787: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4788: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4789: j1=0;
1.222 brouard 4790:
1.227 brouard 4791: /*j=cptcoveff;*/
4792: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4793:
1.227 brouard 4794: first=1;
4795: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4796: for (i=1; i<=nlstate; i++)
1.251 brouard 4797: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4798: prop[i][iage]=0.0;
4799: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4800: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4801: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4802:
4803: for (i=1; i<=imx; i++) { /* Each individual */
4804: bool=1;
4805: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4806: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4807: m=mw[mi][i];
4808: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4809: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4810: for (z1=1; z1<=cptcoveff; z1++){
4811: if( Fixed[Tmodelind[z1]]==1){
4812: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4813: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4814: bool=0;
4815: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4816: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4817: bool=0;
4818: }
4819: }
4820: if(bool==1){ /* Otherwise we skip that wave/person */
4821: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4822: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4823: if(m >=firstpass && m <=lastpass){
4824: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4825: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4826: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4827: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4828: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4829: 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);
4830: exit(1);
4831: }
4832: if (s[m][i]>0 && s[m][i]<=nlstate) {
4833: /*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]]);*/
4834: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4835: prop[s[m][i]][iagemax+3] += weight[i];
4836: } /* end valid statuses */
4837: } /* end selection of dates */
4838: } /* end selection of waves */
4839: } /* end bool */
4840: } /* end wave */
4841: } /* end individual */
4842: for(i=iagemin; i <= iagemax+3; i++){
4843: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4844: posprop += prop[jk][i];
4845: }
4846:
4847: for(jk=1; jk <=nlstate ; jk++){
4848: if( i <= iagemax){
4849: if(posprop>=1.e-5){
4850: probs[i][jk][j1]= prop[jk][i]/posprop;
4851: } else{
4852: if(first==1){
4853: first=0;
4854: 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]);
4855: }
4856: }
4857: }
4858: }/* end jk */
4859: }/* end i */
1.222 brouard 4860: /*} *//* end i1 */
1.227 brouard 4861: } /* end j1 */
1.222 brouard 4862:
1.227 brouard 4863: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4864: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4865: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4866: } /* End of prevalence */
1.126 brouard 4867:
4868: /************* Waves Concatenation ***************/
4869:
4870: 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)
4871: {
4872: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4873: Death is a valid wave (if date is known).
4874: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4875: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4876: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4877: */
1.126 brouard 4878:
1.224 brouard 4879: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4880: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4881: double sum=0., jmean=0.;*/
1.224 brouard 4882: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4883: int j, k=0,jk, ju, jl;
4884: double sum=0.;
4885: first=0;
1.214 brouard 4886: firstwo=0;
1.217 brouard 4887: firsthree=0;
1.218 brouard 4888: firstfour=0;
1.164 brouard 4889: jmin=100000;
1.126 brouard 4890: jmax=-1;
4891: jmean=0.;
1.224 brouard 4892:
4893: /* Treating live states */
1.214 brouard 4894: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4895: mi=0; /* First valid wave */
1.227 brouard 4896: mli=0; /* Last valid wave */
1.126 brouard 4897: m=firstpass;
1.214 brouard 4898: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4899: 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 */
4900: mli=m-1;/* mw[++mi][i]=m-1; */
4901: }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 */
4902: mw[++mi][i]=m;
4903: mli=m;
1.224 brouard 4904: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4905: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4906: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4907: }
1.227 brouard 4908: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4909: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4910: break;
1.224 brouard 4911: #else
1.227 brouard 4912: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4913: if(firsthree == 0){
4914: 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);
4915: firsthree=1;
4916: }
4917: 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);
4918: mw[++mi][i]=m;
4919: mli=m;
4920: }
4921: if(s[m][i]==-2){ /* Vital status is really unknown */
4922: nbwarn++;
4923: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4924: 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);
4925: 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);
4926: }
4927: break;
4928: }
4929: break;
1.224 brouard 4930: #endif
1.227 brouard 4931: }/* End m >= lastpass */
1.126 brouard 4932: }/* end while */
1.224 brouard 4933:
1.227 brouard 4934: /* 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 4935: /* After last pass */
1.224 brouard 4936: /* Treating death states */
1.214 brouard 4937: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4938: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4939: /* } */
1.126 brouard 4940: mi++; /* Death is another wave */
4941: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4942: /* Only death is a correct wave */
1.126 brouard 4943: mw[mi][i]=m;
1.224 brouard 4944: }
4945: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4946: 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 4947: /* m++; */
4948: /* mi++; */
4949: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4950: /* mw[mi][i]=m; */
1.218 brouard 4951: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4952: 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 */
4953: nbwarn++;
4954: if(firstfiv==0){
4955: 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 );
4956: firstfiv=1;
4957: }else{
4958: 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 );
4959: }
4960: }else{ /* Death occured afer last wave potential bias */
4961: nberr++;
4962: if(firstwo==0){
4963: 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 );
4964: firstwo=1;
4965: }
4966: 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 );
4967: }
1.218 brouard 4968: }else{ /* end date of interview is known */
1.227 brouard 4969: /* death is known but not confirmed by death status at any wave */
4970: if(firstfour==0){
4971: 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 );
4972: firstfour=1;
4973: }
4974: 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 4975: }
1.224 brouard 4976: } /* end if date of death is known */
4977: #endif
4978: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4979: /* wav[i]=mw[mi][i]; */
1.126 brouard 4980: if(mi==0){
4981: nbwarn++;
4982: if(first==0){
1.227 brouard 4983: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4984: first=1;
1.126 brouard 4985: }
4986: if(first==1){
1.227 brouard 4987: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4988: }
4989: } /* end mi==0 */
4990: } /* End individuals */
1.214 brouard 4991: /* wav and mw are no more changed */
1.223 brouard 4992:
1.214 brouard 4993:
1.126 brouard 4994: for(i=1; i<=imx; i++){
4995: for(mi=1; mi<wav[i];mi++){
4996: if (stepm <=0)
1.227 brouard 4997: dh[mi][i]=1;
1.126 brouard 4998: else{
1.227 brouard 4999: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
5000: if (agedc[i] < 2*AGESUP) {
5001: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5002: if(j==0) j=1; /* Survives at least one month after exam */
5003: else if(j<0){
5004: nberr++;
5005: 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]);
5006: j=1; /* Temporary Dangerous patch */
5007: 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);
5008: 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]);
5009: 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);
5010: }
5011: k=k+1;
5012: if (j >= jmax){
5013: jmax=j;
5014: ijmax=i;
5015: }
5016: if (j <= jmin){
5017: jmin=j;
5018: ijmin=i;
5019: }
5020: sum=sum+j;
5021: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5022: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5023: }
5024: }
5025: else{
5026: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5027: /* 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 5028:
1.227 brouard 5029: k=k+1;
5030: if (j >= jmax) {
5031: jmax=j;
5032: ijmax=i;
5033: }
5034: else if (j <= jmin){
5035: jmin=j;
5036: ijmin=i;
5037: }
5038: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5039: /*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]);*/
5040: if(j<0){
5041: nberr++;
5042: 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]);
5043: 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]);
5044: }
5045: sum=sum+j;
5046: }
5047: jk= j/stepm;
5048: jl= j -jk*stepm;
5049: ju= j -(jk+1)*stepm;
5050: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5051: if(jl==0){
5052: dh[mi][i]=jk;
5053: bh[mi][i]=0;
5054: }else{ /* We want a negative bias in order to only have interpolation ie
5055: * to avoid the price of an extra matrix product in likelihood */
5056: dh[mi][i]=jk+1;
5057: bh[mi][i]=ju;
5058: }
5059: }else{
5060: if(jl <= -ju){
5061: dh[mi][i]=jk;
5062: bh[mi][i]=jl; /* bias is positive if real duration
5063: * is higher than the multiple of stepm and negative otherwise.
5064: */
5065: }
5066: else{
5067: dh[mi][i]=jk+1;
5068: bh[mi][i]=ju;
5069: }
5070: if(dh[mi][i]==0){
5071: dh[mi][i]=1; /* At least one step */
5072: bh[mi][i]=ju; /* At least one step */
5073: /* 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);*/
5074: }
5075: } /* end if mle */
1.126 brouard 5076: }
5077: } /* end wave */
5078: }
5079: jmean=sum/k;
5080: 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 5081: 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 5082: }
1.126 brouard 5083:
5084: /*********** Tricode ****************************/
1.220 brouard 5085: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5086: {
5087: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5088: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5089: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5090: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5091: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5092: */
1.130 brouard 5093:
1.242 brouard 5094: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5095: int modmaxcovj=0; /* Modality max of covariates j */
5096: int cptcode=0; /* Modality max of covariates j */
5097: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5098:
5099:
1.242 brouard 5100: /* cptcoveff=0; */
5101: /* *cptcov=0; */
1.126 brouard 5102:
1.242 brouard 5103: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5104:
1.242 brouard 5105: /* Loop on covariates without age and products and no quantitative variable */
5106: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5107: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5108: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5109: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5110: switch(Fixed[k]) {
5111: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5112: 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*/
5113: ij=(int)(covar[Tvar[k]][i]);
5114: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5115: * If product of Vn*Vm, still boolean *:
5116: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5117: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5118: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5119: modality of the nth covariate of individual i. */
5120: if (ij > modmaxcovj)
5121: modmaxcovj=ij;
5122: else if (ij < modmincovj)
5123: modmincovj=ij;
5124: if ((ij < -1) && (ij > NCOVMAX)){
5125: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5126: exit(1);
5127: }else
5128: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5129: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5130: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5131: /* getting the maximum value of the modality of the covariate
5132: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5133: female ies 1, then modmaxcovj=1.
5134: */
5135: } /* end for loop on individuals i */
5136: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5137: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5138: cptcode=modmaxcovj;
5139: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5140: /*for (i=0; i<=cptcode; i++) {*/
5141: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5142: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5143: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5144: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5145: if( j != -1){
5146: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5147: covariate for which somebody answered excluding
5148: undefined. Usually 2: 0 and 1. */
5149: }
5150: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5151: covariate for which somebody answered including
5152: undefined. Usually 3: -1, 0 and 1. */
5153: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5154: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5155: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5156:
1.242 brouard 5157: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5158: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5159: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5160: /* modmincovj=3; modmaxcovj = 7; */
5161: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5162: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5163: /* defining two dummy variables: variables V1_1 and V1_2.*/
5164: /* nbcode[Tvar[j]][ij]=k; */
5165: /* nbcode[Tvar[j]][1]=0; */
5166: /* nbcode[Tvar[j]][2]=1; */
5167: /* nbcode[Tvar[j]][3]=2; */
5168: /* To be continued (not working yet). */
5169: ij=0; /* ij is similar to i but can jump over null modalities */
5170: 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*/
5171: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5172: break;
5173: }
5174: ij++;
5175: 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*/
5176: cptcode = ij; /* New max modality for covar j */
5177: } /* end of loop on modality i=-1 to 1 or more */
5178: break;
5179: case 1: /* Testing on varying covariate, could be simple and
5180: * should look at waves or product of fixed *
5181: * varying. No time to test -1, assuming 0 and 1 only */
5182: ij=0;
5183: for(i=0; i<=1;i++){
5184: nbcode[Tvar[k]][++ij]=i;
5185: }
5186: break;
5187: default:
5188: break;
5189: } /* end switch */
5190: } /* end dummy test */
5191:
5192: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5193: /* /\*recode from 0 *\/ */
5194: /* k is a modality. If we have model=V1+V1*sex */
5195: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5196: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5197: /* } */
5198: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5199: /* if (ij > ncodemax[j]) { */
5200: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5201: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5202: /* break; */
5203: /* } */
5204: /* } /\* end of loop on modality k *\/ */
5205: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5206:
5207: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5208: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5209: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5210: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5211: 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 */
5212: 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 */
5213: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5214: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5215:
5216: ij=0;
5217: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5218: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5219: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5220: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5221: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5222: /* If product not in single variable we don't print results */
5223: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5224: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5225: 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*/
5226: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5227: 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 */
5228: if(Fixed[k]!=0)
5229: anyvaryingduminmodel=1;
5230: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5231: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5232: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5233: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5234: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5235: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5236: }
5237: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5238: /* ij--; */
5239: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5240: *cptcov=ij; /*Number of total real effective covariates: effective
5241: * because they can be excluded from the model and real
5242: * if in the model but excluded because missing values, but how to get k from ij?*/
5243: for(j=ij+1; j<= cptcovt; j++){
5244: Tvaraff[j]=0;
5245: Tmodelind[j]=0;
5246: }
5247: for(j=ntveff+1; j<= cptcovt; j++){
5248: TmodelInvind[j]=0;
5249: }
5250: /* To be sorted */
5251: ;
5252: }
1.126 brouard 5253:
1.145 brouard 5254:
1.126 brouard 5255: /*********** Health Expectancies ****************/
5256:
1.235 brouard 5257: 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 5258:
5259: {
5260: /* Health expectancies, no variances */
1.164 brouard 5261: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5262: int nhstepma, nstepma; /* Decreasing with age */
5263: double age, agelim, hf;
5264: double ***p3mat;
5265: double eip;
5266:
1.238 brouard 5267: /* pstamp(ficreseij); */
1.126 brouard 5268: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5269: fprintf(ficreseij,"# Age");
5270: for(i=1; i<=nlstate;i++){
5271: for(j=1; j<=nlstate;j++){
5272: fprintf(ficreseij," e%1d%1d ",i,j);
5273: }
5274: fprintf(ficreseij," e%1d. ",i);
5275: }
5276: fprintf(ficreseij,"\n");
5277:
5278:
5279: if(estepm < stepm){
5280: printf ("Problem %d lower than %d\n",estepm, stepm);
5281: }
5282: else hstepm=estepm;
5283: /* We compute the life expectancy from trapezoids spaced every estepm months
5284: * This is mainly to measure the difference between two models: for example
5285: * if stepm=24 months pijx are given only every 2 years and by summing them
5286: * we are calculating an estimate of the Life Expectancy assuming a linear
5287: * progression in between and thus overestimating or underestimating according
5288: * to the curvature of the survival function. If, for the same date, we
5289: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5290: * to compare the new estimate of Life expectancy with the same linear
5291: * hypothesis. A more precise result, taking into account a more precise
5292: * curvature will be obtained if estepm is as small as stepm. */
5293:
5294: /* For example we decided to compute the life expectancy with the smallest unit */
5295: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5296: nhstepm is the number of hstepm from age to agelim
5297: nstepm is the number of stepm from age to agelin.
5298: Look at hpijx to understand the reason of that which relies in memory size
5299: and note for a fixed period like estepm months */
5300: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5301: survival function given by stepm (the optimization length). Unfortunately it
5302: means that if the survival funtion is printed only each two years of age and if
5303: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5304: results. So we changed our mind and took the option of the best precision.
5305: */
5306: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5307:
5308: agelim=AGESUP;
5309: /* If stepm=6 months */
5310: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5311: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5312:
5313: /* nhstepm age range expressed in number of stepm */
5314: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5315: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5316: /* if (stepm >= YEARM) hstepm=1;*/
5317: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5318: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5319:
5320: for (age=bage; age<=fage; age ++){
5321: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5322: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5323: /* if (stepm >= YEARM) hstepm=1;*/
5324: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5325:
5326: /* If stepm=6 months */
5327: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5328: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5329:
1.235 brouard 5330: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5331:
5332: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5333:
5334: printf("%d|",(int)age);fflush(stdout);
5335: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5336:
5337: /* Computing expectancies */
5338: for(i=1; i<=nlstate;i++)
5339: for(j=1; j<=nlstate;j++)
5340: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5341: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5342:
5343: /* 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]);*/
5344:
5345: }
5346:
5347: fprintf(ficreseij,"%3.0f",age );
5348: for(i=1; i<=nlstate;i++){
5349: eip=0;
5350: for(j=1; j<=nlstate;j++){
5351: eip +=eij[i][j][(int)age];
5352: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5353: }
5354: fprintf(ficreseij,"%9.4f", eip );
5355: }
5356: fprintf(ficreseij,"\n");
5357:
5358: }
5359: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5360: printf("\n");
5361: fprintf(ficlog,"\n");
5362:
5363: }
5364:
1.235 brouard 5365: 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 5366:
5367: {
5368: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5369: to initial status i, ei. .
1.126 brouard 5370: */
5371: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5372: int nhstepma, nstepma; /* Decreasing with age */
5373: double age, agelim, hf;
5374: double ***p3matp, ***p3matm, ***varhe;
5375: double **dnewm,**doldm;
5376: double *xp, *xm;
5377: double **gp, **gm;
5378: double ***gradg, ***trgradg;
5379: int theta;
5380:
5381: double eip, vip;
5382:
5383: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5384: xp=vector(1,npar);
5385: xm=vector(1,npar);
5386: dnewm=matrix(1,nlstate*nlstate,1,npar);
5387: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5388:
5389: pstamp(ficresstdeij);
5390: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5391: fprintf(ficresstdeij,"# Age");
5392: for(i=1; i<=nlstate;i++){
5393: for(j=1; j<=nlstate;j++)
5394: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5395: fprintf(ficresstdeij," e%1d. ",i);
5396: }
5397: fprintf(ficresstdeij,"\n");
5398:
5399: pstamp(ficrescveij);
5400: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5401: fprintf(ficrescveij,"# Age");
5402: for(i=1; i<=nlstate;i++)
5403: for(j=1; j<=nlstate;j++){
5404: cptj= (j-1)*nlstate+i;
5405: for(i2=1; i2<=nlstate;i2++)
5406: for(j2=1; j2<=nlstate;j2++){
5407: cptj2= (j2-1)*nlstate+i2;
5408: if(cptj2 <= cptj)
5409: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5410: }
5411: }
5412: fprintf(ficrescveij,"\n");
5413:
5414: if(estepm < stepm){
5415: printf ("Problem %d lower than %d\n",estepm, stepm);
5416: }
5417: else hstepm=estepm;
5418: /* We compute the life expectancy from trapezoids spaced every estepm months
5419: * This is mainly to measure the difference between two models: for example
5420: * if stepm=24 months pijx are given only every 2 years and by summing them
5421: * we are calculating an estimate of the Life Expectancy assuming a linear
5422: * progression in between and thus overestimating or underestimating according
5423: * to the curvature of the survival function. If, for the same date, we
5424: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5425: * to compare the new estimate of Life expectancy with the same linear
5426: * hypothesis. A more precise result, taking into account a more precise
5427: * curvature will be obtained if estepm is as small as stepm. */
5428:
5429: /* For example we decided to compute the life expectancy with the smallest unit */
5430: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5431: nhstepm is the number of hstepm from age to agelim
5432: nstepm is the number of stepm from age to agelin.
5433: Look at hpijx to understand the reason of that which relies in memory size
5434: and note for a fixed period like estepm months */
5435: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5436: survival function given by stepm (the optimization length). Unfortunately it
5437: means that if the survival funtion is printed only each two years of age and if
5438: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5439: results. So we changed our mind and took the option of the best precision.
5440: */
5441: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5442:
5443: /* If stepm=6 months */
5444: /* nhstepm age range expressed in number of stepm */
5445: agelim=AGESUP;
5446: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5447: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5448: /* if (stepm >= YEARM) hstepm=1;*/
5449: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5450:
5451: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5452: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5453: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5454: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5455: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5456: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5457:
5458: for (age=bage; age<=fage; age ++){
5459: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5460: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5461: /* if (stepm >= YEARM) hstepm=1;*/
5462: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5463:
1.126 brouard 5464: /* If stepm=6 months */
5465: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5466: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5467:
5468: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5469:
1.126 brouard 5470: /* Computing Variances of health expectancies */
5471: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5472: decrease memory allocation */
5473: for(theta=1; theta <=npar; theta++){
5474: for(i=1; i<=npar; i++){
1.222 brouard 5475: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5476: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5477: }
1.235 brouard 5478: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5479: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5480:
1.126 brouard 5481: for(j=1; j<= nlstate; j++){
1.222 brouard 5482: for(i=1; i<=nlstate; i++){
5483: for(h=0; h<=nhstepm-1; h++){
5484: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5485: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5486: }
5487: }
1.126 brouard 5488: }
1.218 brouard 5489:
1.126 brouard 5490: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5491: for(h=0; h<=nhstepm-1; h++){
5492: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5493: }
1.126 brouard 5494: }/* End theta */
5495:
5496:
5497: for(h=0; h<=nhstepm-1; h++)
5498: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5499: for(theta=1; theta <=npar; theta++)
5500: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5501:
1.218 brouard 5502:
1.222 brouard 5503: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5504: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5505: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5506:
1.222 brouard 5507: printf("%d|",(int)age);fflush(stdout);
5508: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5509: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5510: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5511: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5512: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5513: for(ij=1;ij<=nlstate*nlstate;ij++)
5514: for(ji=1;ji<=nlstate*nlstate;ji++)
5515: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5516: }
5517: }
1.218 brouard 5518:
1.126 brouard 5519: /* Computing expectancies */
1.235 brouard 5520: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5521: for(i=1; i<=nlstate;i++)
5522: for(j=1; j<=nlstate;j++)
1.222 brouard 5523: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5524: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5525:
1.222 brouard 5526: /* 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 5527:
1.222 brouard 5528: }
1.218 brouard 5529:
1.126 brouard 5530: fprintf(ficresstdeij,"%3.0f",age );
5531: for(i=1; i<=nlstate;i++){
5532: eip=0.;
5533: vip=0.;
5534: for(j=1; j<=nlstate;j++){
1.222 brouard 5535: eip += eij[i][j][(int)age];
5536: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5537: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5538: 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 5539: }
5540: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5541: }
5542: fprintf(ficresstdeij,"\n");
1.218 brouard 5543:
1.126 brouard 5544: fprintf(ficrescveij,"%3.0f",age );
5545: for(i=1; i<=nlstate;i++)
5546: for(j=1; j<=nlstate;j++){
1.222 brouard 5547: cptj= (j-1)*nlstate+i;
5548: for(i2=1; i2<=nlstate;i2++)
5549: for(j2=1; j2<=nlstate;j2++){
5550: cptj2= (j2-1)*nlstate+i2;
5551: if(cptj2 <= cptj)
5552: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5553: }
1.126 brouard 5554: }
5555: fprintf(ficrescveij,"\n");
1.218 brouard 5556:
1.126 brouard 5557: }
5558: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5559: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5560: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5561: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5562: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5563: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5564: printf("\n");
5565: fprintf(ficlog,"\n");
1.218 brouard 5566:
1.126 brouard 5567: free_vector(xm,1,npar);
5568: free_vector(xp,1,npar);
5569: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5570: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5571: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5572: }
1.218 brouard 5573:
1.126 brouard 5574: /************ Variance ******************/
1.235 brouard 5575: 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 5576: {
5577: /* Variance of health expectancies */
5578: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5579: /* double **newm;*/
5580: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5581:
5582: /* int movingaverage(); */
5583: double **dnewm,**doldm;
5584: double **dnewmp,**doldmp;
5585: int i, j, nhstepm, hstepm, h, nstepm ;
5586: int k;
5587: double *xp;
5588: double **gp, **gm; /* for var eij */
5589: double ***gradg, ***trgradg; /*for var eij */
5590: double **gradgp, **trgradgp; /* for var p point j */
5591: double *gpp, *gmp; /* for var p point j */
5592: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5593: double ***p3mat;
5594: double age,agelim, hf;
5595: /* double ***mobaverage; */
5596: int theta;
5597: char digit[4];
5598: char digitp[25];
5599:
5600: char fileresprobmorprev[FILENAMELENGTH];
5601:
5602: if(popbased==1){
5603: if(mobilav!=0)
5604: strcpy(digitp,"-POPULBASED-MOBILAV_");
5605: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5606: }
5607: else
5608: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5609:
1.218 brouard 5610: /* if (mobilav!=0) { */
5611: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5612: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5613: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5614: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5615: /* } */
5616: /* } */
5617:
5618: strcpy(fileresprobmorprev,"PRMORPREV-");
5619: sprintf(digit,"%-d",ij);
5620: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5621: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5622: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5623: strcat(fileresprobmorprev,fileresu);
5624: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5625: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5626: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5627: }
5628: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5629: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5630: pstamp(ficresprobmorprev);
5631: 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 5632: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5633: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5634: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5635: }
5636: for(j=1;j<=cptcoveff;j++)
5637: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5638: fprintf(ficresprobmorprev,"\n");
5639:
1.218 brouard 5640: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5641: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5642: fprintf(ficresprobmorprev," p.%-d SE",j);
5643: for(i=1; i<=nlstate;i++)
5644: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5645: }
5646: fprintf(ficresprobmorprev,"\n");
5647:
5648: fprintf(ficgp,"\n# Routine varevsij");
5649: fprintf(ficgp,"\nunset title \n");
5650: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5651: 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");
5652: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5653: /* } */
5654: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5655: pstamp(ficresvij);
5656: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5657: if(popbased==1)
5658: 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);
5659: else
5660: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5661: fprintf(ficresvij,"# Age");
5662: for(i=1; i<=nlstate;i++)
5663: for(j=1; j<=nlstate;j++)
5664: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5665: fprintf(ficresvij,"\n");
5666:
5667: xp=vector(1,npar);
5668: dnewm=matrix(1,nlstate,1,npar);
5669: doldm=matrix(1,nlstate,1,nlstate);
5670: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5671: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5672:
5673: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5674: gpp=vector(nlstate+1,nlstate+ndeath);
5675: gmp=vector(nlstate+1,nlstate+ndeath);
5676: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5677:
1.218 brouard 5678: if(estepm < stepm){
5679: printf ("Problem %d lower than %d\n",estepm, stepm);
5680: }
5681: else hstepm=estepm;
5682: /* For example we decided to compute the life expectancy with the smallest unit */
5683: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5684: nhstepm is the number of hstepm from age to agelim
5685: nstepm is the number of stepm from age to agelim.
5686: Look at function hpijx to understand why because of memory size limitations,
5687: we decided (b) to get a life expectancy respecting the most precise curvature of the
5688: survival function given by stepm (the optimization length). Unfortunately it
5689: means that if the survival funtion is printed every two years of age and if
5690: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5691: results. So we changed our mind and took the option of the best precision.
5692: */
5693: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5694: agelim = AGESUP;
5695: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5696: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5697: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5698: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5699: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5700: gp=matrix(0,nhstepm,1,nlstate);
5701: gm=matrix(0,nhstepm,1,nlstate);
5702:
5703:
5704: for(theta=1; theta <=npar; theta++){
5705: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5706: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5707: }
5708:
1.242 brouard 5709: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5710:
5711: if (popbased==1) {
5712: if(mobilav ==0){
5713: for(i=1; i<=nlstate;i++)
5714: prlim[i][i]=probs[(int)age][i][ij];
5715: }else{ /* mobilav */
5716: for(i=1; i<=nlstate;i++)
5717: prlim[i][i]=mobaverage[(int)age][i][ij];
5718: }
5719: }
5720:
1.235 brouard 5721: 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 5722: for(j=1; j<= nlstate; j++){
5723: for(h=0; h<=nhstepm; h++){
5724: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5725: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5726: }
5727: }
5728: /* Next for computing probability of death (h=1 means
5729: computed over hstepm matrices product = hstepm*stepm months)
5730: as a weighted average of prlim.
5731: */
5732: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5733: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5734: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5735: }
5736: /* end probability of death */
5737:
5738: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5739: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5740:
1.242 brouard 5741: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5742:
5743: if (popbased==1) {
5744: if(mobilav ==0){
5745: for(i=1; i<=nlstate;i++)
5746: prlim[i][i]=probs[(int)age][i][ij];
5747: }else{ /* mobilav */
5748: for(i=1; i<=nlstate;i++)
5749: prlim[i][i]=mobaverage[(int)age][i][ij];
5750: }
5751: }
5752:
1.235 brouard 5753: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5754:
5755: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5756: for(h=0; h<=nhstepm; h++){
5757: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5758: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5759: }
5760: }
5761: /* This for computing probability of death (h=1 means
5762: computed over hstepm matrices product = hstepm*stepm months)
5763: as a weighted average of prlim.
5764: */
5765: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5766: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5767: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5768: }
5769: /* end probability of death */
5770:
5771: for(j=1; j<= nlstate; j++) /* vareij */
5772: for(h=0; h<=nhstepm; h++){
5773: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5774: }
5775:
5776: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5777: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5778: }
5779:
5780: } /* End theta */
5781:
5782: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5783:
5784: for(h=0; h<=nhstepm; h++) /* veij */
5785: for(j=1; j<=nlstate;j++)
5786: for(theta=1; theta <=npar; theta++)
5787: trgradg[h][j][theta]=gradg[h][theta][j];
5788:
5789: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5790: for(theta=1; theta <=npar; theta++)
5791: trgradgp[j][theta]=gradgp[theta][j];
5792:
5793:
5794: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5795: for(i=1;i<=nlstate;i++)
5796: for(j=1;j<=nlstate;j++)
5797: vareij[i][j][(int)age] =0.;
5798:
5799: for(h=0;h<=nhstepm;h++){
5800: for(k=0;k<=nhstepm;k++){
5801: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5802: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5803: for(i=1;i<=nlstate;i++)
5804: for(j=1;j<=nlstate;j++)
5805: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5806: }
5807: }
5808:
5809: /* pptj */
5810: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5811: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5812: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5813: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5814: varppt[j][i]=doldmp[j][i];
5815: /* end ppptj */
5816: /* x centered again */
5817:
1.242 brouard 5818: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5819:
5820: if (popbased==1) {
5821: if(mobilav ==0){
5822: for(i=1; i<=nlstate;i++)
5823: prlim[i][i]=probs[(int)age][i][ij];
5824: }else{ /* mobilav */
5825: for(i=1; i<=nlstate;i++)
5826: prlim[i][i]=mobaverage[(int)age][i][ij];
5827: }
5828: }
5829:
5830: /* This for computing probability of death (h=1 means
5831: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5832: as a weighted average of prlim.
5833: */
1.235 brouard 5834: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5835: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5836: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5837: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5838: }
5839: /* end probability of death */
5840:
5841: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5842: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5843: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5844: for(i=1; i<=nlstate;i++){
5845: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5846: }
5847: }
5848: fprintf(ficresprobmorprev,"\n");
5849:
5850: fprintf(ficresvij,"%.0f ",age );
5851: for(i=1; i<=nlstate;i++)
5852: for(j=1; j<=nlstate;j++){
5853: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5854: }
5855: fprintf(ficresvij,"\n");
5856: free_matrix(gp,0,nhstepm,1,nlstate);
5857: free_matrix(gm,0,nhstepm,1,nlstate);
5858: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5859: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5860: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5861: } /* End age */
5862: free_vector(gpp,nlstate+1,nlstate+ndeath);
5863: free_vector(gmp,nlstate+1,nlstate+ndeath);
5864: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5865: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5866: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5867: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5868: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5869: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5870: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5871: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5872: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5873: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5874: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5875: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5876: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5877: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5878: 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);
5879: /* 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 5880: */
1.218 brouard 5881: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5882: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5883:
1.218 brouard 5884: free_vector(xp,1,npar);
5885: free_matrix(doldm,1,nlstate,1,nlstate);
5886: free_matrix(dnewm,1,nlstate,1,npar);
5887: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5888: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5889: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5890: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5891: fclose(ficresprobmorprev);
5892: fflush(ficgp);
5893: fflush(fichtm);
5894: } /* end varevsij */
1.126 brouard 5895:
5896: /************ Variance of prevlim ******************/
1.235 brouard 5897: 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 5898: {
1.205 brouard 5899: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5900: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5901:
1.126 brouard 5902: double **dnewm,**doldm;
5903: int i, j, nhstepm, hstepm;
5904: double *xp;
5905: double *gp, *gm;
5906: double **gradg, **trgradg;
1.208 brouard 5907: double **mgm, **mgp;
1.126 brouard 5908: double age,agelim;
5909: int theta;
5910:
5911: pstamp(ficresvpl);
5912: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5913: fprintf(ficresvpl,"# Age ");
5914: if(nresult >=1)
5915: fprintf(ficresvpl," Result# ");
1.126 brouard 5916: for(i=1; i<=nlstate;i++)
5917: fprintf(ficresvpl," %1d-%1d",i,i);
5918: fprintf(ficresvpl,"\n");
5919:
5920: xp=vector(1,npar);
5921: dnewm=matrix(1,nlstate,1,npar);
5922: doldm=matrix(1,nlstate,1,nlstate);
5923:
5924: hstepm=1*YEARM; /* Every year of age */
5925: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5926: agelim = AGESUP;
5927: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5928: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5929: if (stepm >= YEARM) hstepm=1;
5930: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5931: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5932: mgp=matrix(1,npar,1,nlstate);
5933: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5934: gp=vector(1,nlstate);
5935: gm=vector(1,nlstate);
5936:
5937: for(theta=1; theta <=npar; theta++){
5938: for(i=1; i<=npar; i++){ /* Computes gradient */
5939: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5940: }
1.209 brouard 5941: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5942: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5943: else
1.235 brouard 5944: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5945: for(i=1;i<=nlstate;i++){
1.126 brouard 5946: gp[i] = prlim[i][i];
1.208 brouard 5947: mgp[theta][i] = prlim[i][i];
5948: }
1.126 brouard 5949: for(i=1; i<=npar; i++) /* Computes gradient */
5950: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5951: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5952: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5953: else
1.235 brouard 5954: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5955: for(i=1;i<=nlstate;i++){
1.126 brouard 5956: gm[i] = prlim[i][i];
1.208 brouard 5957: mgm[theta][i] = prlim[i][i];
5958: }
1.126 brouard 5959: for(i=1;i<=nlstate;i++)
5960: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5961: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5962: } /* End theta */
5963:
5964: trgradg =matrix(1,nlstate,1,npar);
5965:
5966: for(j=1; j<=nlstate;j++)
5967: for(theta=1; theta <=npar; theta++)
5968: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5969: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5970: /* printf("\nmgm mgp %d ",(int)age); */
5971: /* for(j=1; j<=nlstate;j++){ */
5972: /* printf(" %d ",j); */
5973: /* for(theta=1; theta <=npar; theta++) */
5974: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5975: /* printf("\n "); */
5976: /* } */
5977: /* } */
5978: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5979: /* printf("\n gradg %d ",(int)age); */
5980: /* for(j=1; j<=nlstate;j++){ */
5981: /* printf("%d ",j); */
5982: /* for(theta=1; theta <=npar; theta++) */
5983: /* printf("%d %lf ",theta,gradg[theta][j]); */
5984: /* printf("\n "); */
5985: /* } */
5986: /* } */
1.126 brouard 5987:
5988: for(i=1;i<=nlstate;i++)
5989: varpl[i][(int)age] =0.;
1.209 brouard 5990: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5991: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5992: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5993: }else{
1.126 brouard 5994: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5995: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5996: }
1.126 brouard 5997: for(i=1;i<=nlstate;i++)
5998: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5999:
6000: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6001: if(nresult >=1)
6002: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6003: for(i=1; i<=nlstate;i++)
6004: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6005: fprintf(ficresvpl,"\n");
6006: free_vector(gp,1,nlstate);
6007: free_vector(gm,1,nlstate);
1.208 brouard 6008: free_matrix(mgm,1,npar,1,nlstate);
6009: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6010: free_matrix(gradg,1,npar,1,nlstate);
6011: free_matrix(trgradg,1,nlstate,1,npar);
6012: } /* End age */
6013:
6014: free_vector(xp,1,npar);
6015: free_matrix(doldm,1,nlstate,1,npar);
6016: free_matrix(dnewm,1,nlstate,1,nlstate);
6017:
6018: }
6019:
6020: /************ Variance of one-step probabilities ******************/
6021: 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 6022: {
6023: int i, j=0, k1, l1, tj;
6024: int k2, l2, j1, z1;
6025: int k=0, l;
6026: int first=1, first1, first2;
6027: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6028: double **dnewm,**doldm;
6029: double *xp;
6030: double *gp, *gm;
6031: double **gradg, **trgradg;
6032: double **mu;
6033: double age, cov[NCOVMAX+1];
6034: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6035: int theta;
6036: char fileresprob[FILENAMELENGTH];
6037: char fileresprobcov[FILENAMELENGTH];
6038: char fileresprobcor[FILENAMELENGTH];
6039: double ***varpij;
6040:
6041: strcpy(fileresprob,"PROB_");
6042: strcat(fileresprob,fileres);
6043: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6044: printf("Problem with resultfile: %s\n", fileresprob);
6045: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6046: }
6047: strcpy(fileresprobcov,"PROBCOV_");
6048: strcat(fileresprobcov,fileresu);
6049: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6050: printf("Problem with resultfile: %s\n", fileresprobcov);
6051: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6052: }
6053: strcpy(fileresprobcor,"PROBCOR_");
6054: strcat(fileresprobcor,fileresu);
6055: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6056: printf("Problem with resultfile: %s\n", fileresprobcor);
6057: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6058: }
6059: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6060: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6061: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6062: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6063: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6064: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6065: pstamp(ficresprob);
6066: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6067: fprintf(ficresprob,"# Age");
6068: pstamp(ficresprobcov);
6069: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6070: fprintf(ficresprobcov,"# Age");
6071: pstamp(ficresprobcor);
6072: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6073: fprintf(ficresprobcor,"# Age");
1.126 brouard 6074:
6075:
1.222 brouard 6076: for(i=1; i<=nlstate;i++)
6077: for(j=1; j<=(nlstate+ndeath);j++){
6078: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6079: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6080: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6081: }
6082: /* fprintf(ficresprob,"\n");
6083: fprintf(ficresprobcov,"\n");
6084: fprintf(ficresprobcor,"\n");
6085: */
6086: xp=vector(1,npar);
6087: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6088: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6089: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6090: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6091: first=1;
6092: fprintf(ficgp,"\n# Routine varprob");
6093: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6094: fprintf(fichtm,"\n");
6095:
6096: 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);
6097: 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);
6098: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6099: and drawn. It helps understanding how is the covariance between two incidences.\
6100: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6101: 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 6102: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6103: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6104: standard deviations wide on each axis. <br>\
6105: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6106: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6107: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6108:
1.222 brouard 6109: cov[1]=1;
6110: /* tj=cptcoveff; */
1.225 brouard 6111: tj = (int) pow(2,cptcoveff);
1.222 brouard 6112: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6113: j1=0;
1.224 brouard 6114: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6115: if (cptcovn>0) {
6116: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6117: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6118: fprintf(ficresprob, "**********\n#\n");
6119: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6120: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6121: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6122:
1.222 brouard 6123: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6124: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6125: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6126:
6127:
1.222 brouard 6128: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6129: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6130: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6131:
1.222 brouard 6132: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6133: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6134: fprintf(ficresprobcor, "**********\n#");
6135: if(invalidvarcomb[j1]){
6136: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6137: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6138: continue;
6139: }
6140: }
6141: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6142: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6143: gp=vector(1,(nlstate)*(nlstate+ndeath));
6144: gm=vector(1,(nlstate)*(nlstate+ndeath));
6145: for (age=bage; age<=fage; age ++){
6146: cov[2]=age;
6147: if(nagesqr==1)
6148: cov[3]= age*age;
6149: for (k=1; k<=cptcovn;k++) {
6150: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6151: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6152: * 1 1 1 1 1
6153: * 2 2 1 1 1
6154: * 3 1 2 1 1
6155: */
6156: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6157: }
6158: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6159: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6160: for (k=1; k<=cptcovprod;k++)
6161: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6162:
6163:
1.222 brouard 6164: for(theta=1; theta <=npar; theta++){
6165: for(i=1; i<=npar; i++)
6166: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6167:
1.222 brouard 6168: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6169:
1.222 brouard 6170: k=0;
6171: for(i=1; i<= (nlstate); i++){
6172: for(j=1; j<=(nlstate+ndeath);j++){
6173: k=k+1;
6174: gp[k]=pmmij[i][j];
6175: }
6176: }
1.220 brouard 6177:
1.222 brouard 6178: for(i=1; i<=npar; i++)
6179: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6180:
1.222 brouard 6181: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6182: k=0;
6183: for(i=1; i<=(nlstate); i++){
6184: for(j=1; j<=(nlstate+ndeath);j++){
6185: k=k+1;
6186: gm[k]=pmmij[i][j];
6187: }
6188: }
1.220 brouard 6189:
1.222 brouard 6190: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6191: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6192: }
1.126 brouard 6193:
1.222 brouard 6194: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6195: for(theta=1; theta <=npar; theta++)
6196: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6197:
1.222 brouard 6198: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6199: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6200:
1.222 brouard 6201: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6202:
1.222 brouard 6203: k=0;
6204: for(i=1; i<=(nlstate); i++){
6205: for(j=1; j<=(nlstate+ndeath);j++){
6206: k=k+1;
6207: mu[k][(int) age]=pmmij[i][j];
6208: }
6209: }
6210: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6211: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6212: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6213:
1.222 brouard 6214: /*printf("\n%d ",(int)age);
6215: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6216: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6217: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6218: }*/
1.220 brouard 6219:
1.222 brouard 6220: fprintf(ficresprob,"\n%d ",(int)age);
6221: fprintf(ficresprobcov,"\n%d ",(int)age);
6222: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6223:
1.222 brouard 6224: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6225: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6226: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6227: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6228: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6229: }
6230: i=0;
6231: for (k=1; k<=(nlstate);k++){
6232: for (l=1; l<=(nlstate+ndeath);l++){
6233: i++;
6234: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6235: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6236: for (j=1; j<=i;j++){
6237: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6238: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6239: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6240: }
6241: }
6242: }/* end of loop for state */
6243: } /* end of loop for age */
6244: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6245: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6246: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6247: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6248:
6249: /* Confidence intervalle of pij */
6250: /*
6251: fprintf(ficgp,"\nunset parametric;unset label");
6252: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6253: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6254: 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);
6255: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6256: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6257: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6258: */
6259:
6260: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6261: first1=1;first2=2;
6262: for (k2=1; k2<=(nlstate);k2++){
6263: for (l2=1; l2<=(nlstate+ndeath);l2++){
6264: if(l2==k2) continue;
6265: j=(k2-1)*(nlstate+ndeath)+l2;
6266: for (k1=1; k1<=(nlstate);k1++){
6267: for (l1=1; l1<=(nlstate+ndeath);l1++){
6268: if(l1==k1) continue;
6269: i=(k1-1)*(nlstate+ndeath)+l1;
6270: if(i<=j) continue;
6271: for (age=bage; age<=fage; age ++){
6272: if ((int)age %5==0){
6273: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6274: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6275: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6276: mu1=mu[i][(int) age]/stepm*YEARM ;
6277: mu2=mu[j][(int) age]/stepm*YEARM;
6278: c12=cv12/sqrt(v1*v2);
6279: /* Computing eigen value of matrix of covariance */
6280: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6281: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6282: if ((lc2 <0) || (lc1 <0) ){
6283: if(first2==1){
6284: first1=0;
6285: 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);
6286: }
6287: 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);
6288: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6289: /* lc2=fabs(lc2); */
6290: }
1.220 brouard 6291:
1.222 brouard 6292: /* Eigen vectors */
6293: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6294: /*v21=sqrt(1.-v11*v11); *//* error */
6295: v21=(lc1-v1)/cv12*v11;
6296: v12=-v21;
6297: v22=v11;
6298: tnalp=v21/v11;
6299: if(first1==1){
6300: first1=0;
6301: 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);
6302: }
6303: 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);
6304: /*printf(fignu*/
6305: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6306: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6307: if(first==1){
6308: first=0;
6309: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6310: fprintf(ficgp,"\nset parametric;unset label");
6311: 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);
6312: fprintf(ficgp,"\nset ter svg size 640, 480");
6313: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6314: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6315: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6316: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6317: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6318: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6319: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6320: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6321: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6322: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6323: 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", \
6324: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6325: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6326: }else{
6327: first=0;
6328: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6329: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6330: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6331: 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", \
6332: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6333: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6334: }/* if first */
6335: } /* age mod 5 */
6336: } /* end loop age */
6337: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6338: first=1;
6339: } /*l12 */
6340: } /* k12 */
6341: } /*l1 */
6342: }/* k1 */
6343: } /* loop on combination of covariates j1 */
6344: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6345: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6346: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6347: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6348: free_vector(xp,1,npar);
6349: fclose(ficresprob);
6350: fclose(ficresprobcov);
6351: fclose(ficresprobcor);
6352: fflush(ficgp);
6353: fflush(fichtmcov);
6354: }
1.126 brouard 6355:
6356:
6357: /******************* Printing html file ***********/
1.201 brouard 6358: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6359: int lastpass, int stepm, int weightopt, char model[],\
6360: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6361: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6362: double jprev1, double mprev1,double anprev1, double dateprev1, \
6363: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6364: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6365:
6366: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6367: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6368: </ul>");
1.237 brouard 6369: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6370: </ul>", model);
1.214 brouard 6371: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6372: 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",
6373: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6374: 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 6375: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6376: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6377: fprintf(fichtm,"\
6378: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6379: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6380: fprintf(fichtm,"\
1.217 brouard 6381: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6382: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6383: fprintf(fichtm,"\
1.126 brouard 6384: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6385: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6386: fprintf(fichtm,"\
1.217 brouard 6387: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6388: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6389: fprintf(fichtm,"\
1.211 brouard 6390: - (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 6391: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6392: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6393: if(prevfcast==1){
6394: fprintf(fichtm,"\
6395: - Prevalence projections by age and states: \
1.201 brouard 6396: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6397: }
1.126 brouard 6398:
1.222 brouard 6399: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6400:
1.225 brouard 6401: m=pow(2,cptcoveff);
1.222 brouard 6402: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6403:
1.222 brouard 6404: jj1=0;
1.237 brouard 6405:
6406: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6407: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6408: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6409: continue;
1.220 brouard 6410:
1.222 brouard 6411: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6412: jj1++;
6413: if (cptcovn > 0) {
6414: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6415: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6416: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6417: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6418: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6419: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6420: }
1.237 brouard 6421: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6422: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6423: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6424: }
6425:
1.230 brouard 6426: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6427: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6428: if(invalidvarcomb[k1]){
6429: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6430: printf("\nCombination (%d) ignored because no cases \n",k1);
6431: continue;
6432: }
6433: }
6434: /* aij, bij */
1.241 brouard 6435: 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> \
6436: <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 6437: /* Pij */
1.241 brouard 6438: 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> \
6439: <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 6440: /* Quasi-incidences */
6441: 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 6442: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6443: 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 6444: 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> \
6445: <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 6446: /* Survival functions (period) in state j */
6447: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6448: 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> \
6449: <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 6450: }
6451: /* State specific survival functions (period) */
6452: for(cpt=1; cpt<=nlstate;cpt++){
6453: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6454: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6455: <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 6456: }
6457: /* Period (stable) prevalence in each health state */
6458: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6459: 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> \
6460: <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 6461: }
6462: if(backcast==1){
6463: /* Period (stable) back prevalence in each health state */
6464: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6465: 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> \
6466: <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 6467: }
1.217 brouard 6468: }
1.222 brouard 6469: if(prevfcast==1){
6470: /* Projection of prevalence up to period (stable) prevalence in each health state */
6471: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6472: 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> \
6473: <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 6474: }
6475: }
1.220 brouard 6476:
1.222 brouard 6477: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6478: 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> \
6479: <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 6480: }
6481: /* } /\* end i1 *\/ */
6482: }/* End k1 */
6483: fprintf(fichtm,"</ul>");
1.126 brouard 6484:
1.222 brouard 6485: fprintf(fichtm,"\
1.126 brouard 6486: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6487: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6488: - 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 6489: But because parameters are usually highly correlated (a higher incidence of disability \
6490: and a higher incidence of recovery can give very close observed transition) it might \
6491: be very useful to look not only at linear confidence intervals estimated from the \
6492: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6493: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6494: covariance matrix of the one-step probabilities. \
6495: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6496:
1.222 brouard 6497: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6498: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6499: fprintf(fichtm,"\
1.126 brouard 6500: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6501: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6502:
1.222 brouard 6503: fprintf(fichtm,"\
1.126 brouard 6504: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6505: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6506: fprintf(fichtm,"\
1.126 brouard 6507: - 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): \
6508: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6509: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6510: fprintf(fichtm,"\
1.126 brouard 6511: - (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): \
6512: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6513: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6514: fprintf(fichtm,"\
1.128 brouard 6515: - 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 6516: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6517: fprintf(fichtm,"\
1.128 brouard 6518: - 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 6519: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6520: fprintf(fichtm,"\
1.126 brouard 6521: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6522: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6523:
6524: /* if(popforecast==1) fprintf(fichtm,"\n */
6525: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6526: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6527: /* <br>",fileres,fileres,fileres,fileres); */
6528: /* else */
6529: /* 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 6530: fflush(fichtm);
6531: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6532:
1.225 brouard 6533: m=pow(2,cptcoveff);
1.222 brouard 6534: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6535:
1.222 brouard 6536: jj1=0;
1.237 brouard 6537:
1.241 brouard 6538: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6539: for(k1=1; k1<=m;k1++){
1.253 brouard 6540: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6541: continue;
1.222 brouard 6542: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6543: jj1++;
1.126 brouard 6544: if (cptcovn > 0) {
6545: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6546: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6547: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6548: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6549: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6550: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6551: }
6552:
1.126 brouard 6553: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6554:
1.222 brouard 6555: if(invalidvarcomb[k1]){
6556: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6557: continue;
6558: }
1.126 brouard 6559: }
6560: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6561: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6562: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6563: <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 6564: }
6565: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6566: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6567: true period expectancies (those weighted with period prevalences are also\
6568: drawn in addition to the population based expectancies computed using\
1.241 brouard 6569: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6570: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6571: /* } /\* end i1 *\/ */
6572: }/* End k1 */
1.241 brouard 6573: }/* End nres */
1.222 brouard 6574: fprintf(fichtm,"</ul>");
6575: fflush(fichtm);
1.126 brouard 6576: }
6577:
6578: /******************* Gnuplot file **************/
1.223 brouard 6579: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6580:
6581: char dirfileres[132],optfileres[132];
1.223 brouard 6582: char gplotcondition[132];
1.237 brouard 6583: 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 6584: int lv=0, vlv=0, kl=0;
1.130 brouard 6585: int ng=0;
1.201 brouard 6586: int vpopbased;
1.223 brouard 6587: int ioffset; /* variable offset for columns */
1.235 brouard 6588: int nres=0; /* Index of resultline */
1.219 brouard 6589:
1.126 brouard 6590: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6591: /* printf("Problem with file %s",optionfilegnuplot); */
6592: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6593: /* } */
6594:
6595: /*#ifdef windows */
6596: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6597: /*#endif */
1.225 brouard 6598: m=pow(2,cptcoveff);
1.126 brouard 6599:
1.202 brouard 6600: /* Contribution to likelihood */
6601: /* Plot the probability implied in the likelihood */
1.223 brouard 6602: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6603: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6604: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6605: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6606: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6607: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6608: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6609: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6610: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6611: 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));
6612: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6613: 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));
6614: for (i=1; i<= nlstate ; i ++) {
6615: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6616: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6617: 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);
6618: for (j=2; j<= nlstate+ndeath ; j ++) {
6619: 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);
6620: }
6621: fprintf(ficgp,";\nset out; unset ylabel;\n");
6622: }
6623: /* 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 */
6624: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6625: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6626: fprintf(ficgp,"\nset out;unset log\n");
6627: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6628:
1.126 brouard 6629: strcpy(dirfileres,optionfilefiname);
6630: strcpy(optfileres,"vpl");
1.223 brouard 6631: /* 1eme*/
1.238 brouard 6632: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6633: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6634: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6635: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6636: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6637: continue;
6638: /* We are interested in selected combination by the resultline */
1.246 brouard 6639: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6640: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6641: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6642: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6643: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6644: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6645: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6646: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6647: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6648: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6649: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6650: }
6651: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6652: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6653: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6654: }
1.246 brouard 6655: /* printf("\n#\n"); */
1.238 brouard 6656: fprintf(ficgp,"\n#\n");
6657: if(invalidvarcomb[k1]){
6658: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6659: continue;
6660: }
1.235 brouard 6661:
1.241 brouard 6662: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6663: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6664: 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 6665:
1.238 brouard 6666: for (i=1; i<= nlstate ; i ++) {
6667: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6668: else fprintf(ficgp," %%*lf (%%*lf)");
6669: }
1.242 brouard 6670: 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 6671: for (i=1; i<= nlstate ; i ++) {
6672: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6673: else fprintf(ficgp," %%*lf (%%*lf)");
6674: }
1.242 brouard 6675: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6676: for (i=1; i<= nlstate ; i ++) {
6677: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6678: else fprintf(ficgp," %%*lf (%%*lf)");
6679: }
6680: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6681: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6682: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6683: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6684: if(cptcoveff ==0){
1.245 brouard 6685: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6686: }else{
6687: kl=0;
6688: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6689: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6690: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6691: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6692: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6693: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6694: kl++;
1.238 brouard 6695: /* 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 *\/ */
6696: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6697: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6698: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6699: if(k==cptcoveff){
1.245 brouard 6700: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6701: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6702: }else{
6703: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6704: kl++;
6705: }
6706: } /* end covariate */
6707: } /* end if no covariate */
6708: } /* end if backcast */
6709: fprintf(ficgp,"\nset out \n");
6710: } /* nres */
1.201 brouard 6711: } /* k1 */
6712: } /* cpt */
1.235 brouard 6713:
6714:
1.126 brouard 6715: /*2 eme*/
1.238 brouard 6716: for (k1=1; k1<= m ; k1 ++){
6717: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6718: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6719: continue;
6720: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6721: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6722: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6723: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6724: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6725: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6726: vlv= nbcode[Tvaraff[k]][lv];
6727: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6728: }
1.237 brouard 6729: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6730: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6731: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6732: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6733: }
1.211 brouard 6734: fprintf(ficgp,"\n#\n");
1.223 brouard 6735: if(invalidvarcomb[k1]){
6736: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6737: continue;
6738: }
1.219 brouard 6739:
1.241 brouard 6740: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6741: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6742: if(vpopbased==0)
6743: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6744: else
6745: fprintf(ficgp,"\nreplot ");
6746: for (i=1; i<= nlstate+1 ; i ++) {
6747: k=2*i;
6748: 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);
6749: for (j=1; j<= nlstate+1 ; j ++) {
6750: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6751: else fprintf(ficgp," %%*lf (%%*lf)");
6752: }
6753: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6754: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6755: 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);
6756: for (j=1; j<= nlstate+1 ; j ++) {
6757: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6758: else fprintf(ficgp," %%*lf (%%*lf)");
6759: }
6760: fprintf(ficgp,"\" t\"\" w l lt 0,");
6761: 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);
6762: for (j=1; j<= nlstate+1 ; j ++) {
6763: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6764: else fprintf(ficgp," %%*lf (%%*lf)");
6765: }
6766: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6767: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6768: } /* state */
6769: } /* vpopbased */
1.244 brouard 6770: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6771: } /* end nres */
6772: } /* k1 end 2 eme*/
6773:
6774:
6775: /*3eme*/
6776: for (k1=1; k1<= m ; k1 ++){
6777: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6778: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6779: continue;
6780:
6781: for (cpt=1; cpt<= nlstate ; cpt ++) {
6782: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6783: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6784: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6785: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6786: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6787: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6788: vlv= nbcode[Tvaraff[k]][lv];
6789: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6790: }
6791: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6792: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6793: }
6794: fprintf(ficgp,"\n#\n");
6795: if(invalidvarcomb[k1]){
6796: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6797: continue;
6798: }
6799:
6800: /* k=2+nlstate*(2*cpt-2); */
6801: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6802: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6803: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6804: 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 6805: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6806: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6807: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6808: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6809: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6810: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6811:
1.238 brouard 6812: */
6813: for (i=1; i< nlstate ; i ++) {
6814: 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);
6815: /* 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 6816:
1.238 brouard 6817: }
6818: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6819: }
6820: } /* end nres */
6821: } /* end kl 3eme */
1.126 brouard 6822:
1.223 brouard 6823: /* 4eme */
1.201 brouard 6824: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6825: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6826: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6827: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6828: continue;
1.238 brouard 6829: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6830: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6831: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6832: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6833: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6834: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6835: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6836: vlv= nbcode[Tvaraff[k]][lv];
6837: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6838: }
6839: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6840: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6841: }
6842: fprintf(ficgp,"\n#\n");
6843: if(invalidvarcomb[k1]){
6844: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6845: continue;
1.223 brouard 6846: }
1.238 brouard 6847:
1.241 brouard 6848: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6849: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6850: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6851: k=3;
6852: for (i=1; i<= nlstate ; i ++){
6853: if(i==1){
6854: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6855: }else{
6856: fprintf(ficgp,", '' ");
6857: }
6858: l=(nlstate+ndeath)*(i-1)+1;
6859: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6860: for (j=2; j<= nlstate+ndeath ; j ++)
6861: fprintf(ficgp,"+$%d",k+l+j-1);
6862: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6863: } /* nlstate */
6864: fprintf(ficgp,"\nset out\n");
6865: } /* end cpt state*/
6866: } /* end nres */
6867: } /* end covariate k1 */
6868:
1.220 brouard 6869: /* 5eme */
1.201 brouard 6870: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6871: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6872: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6873: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6874: continue;
1.238 brouard 6875: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6876: 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);
6877: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6878: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6879: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6880: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6881: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6882: vlv= nbcode[Tvaraff[k]][lv];
6883: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6884: }
6885: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6886: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6887: }
6888: fprintf(ficgp,"\n#\n");
6889: if(invalidvarcomb[k1]){
6890: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6891: continue;
6892: }
1.227 brouard 6893:
1.241 brouard 6894: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6895: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6896: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6897: k=3;
6898: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6899: if(j==1)
6900: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6901: else
6902: fprintf(ficgp,", '' ");
6903: l=(nlstate+ndeath)*(cpt-1) +j;
6904: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6905: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6906: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6907: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6908: } /* nlstate */
6909: fprintf(ficgp,", '' ");
6910: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6911: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6912: l=(nlstate+ndeath)*(cpt-1) +j;
6913: if(j < nlstate)
6914: fprintf(ficgp,"$%d +",k+l);
6915: else
6916: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6917: }
6918: fprintf(ficgp,"\nset out\n");
6919: } /* end cpt state*/
6920: } /* end covariate */
6921: } /* end nres */
1.227 brouard 6922:
1.220 brouard 6923: /* 6eme */
1.202 brouard 6924: /* CV preval stable (period) for each covariate */
1.237 brouard 6925: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6926: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6927: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6928: continue;
1.153 brouard 6929: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6930:
1.211 brouard 6931: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6932: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6933: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6934: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6935: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6936: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6937: vlv= nbcode[Tvaraff[k]][lv];
6938: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6939: }
1.237 brouard 6940: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6941: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6942: }
1.211 brouard 6943: fprintf(ficgp,"\n#\n");
1.223 brouard 6944: if(invalidvarcomb[k1]){
1.227 brouard 6945: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6946: continue;
1.223 brouard 6947: }
1.227 brouard 6948:
1.241 brouard 6949: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6950: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6951: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6952: k=3; /* Offset */
1.153 brouard 6953: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6954: if(i==1)
6955: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6956: else
6957: fprintf(ficgp,", '' ");
6958: l=(nlstate+ndeath)*(i-1)+1;
6959: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6960: for (j=2; j<= nlstate ; j ++)
6961: fprintf(ficgp,"+$%d",k+l+j-1);
6962: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6963: } /* nlstate */
1.201 brouard 6964: fprintf(ficgp,"\nset out\n");
1.153 brouard 6965: } /* end cpt state*/
6966: } /* end covariate */
1.227 brouard 6967:
6968:
1.220 brouard 6969: /* 7eme */
1.218 brouard 6970: if(backcast == 1){
1.217 brouard 6971: /* CV back preval stable (period) for each covariate */
1.237 brouard 6972: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6973: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6974: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6975: continue;
1.218 brouard 6976: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6977: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6978: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6979: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6980: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6981: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6982: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6983: vlv= nbcode[Tvaraff[k]][lv];
6984: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6985: }
1.237 brouard 6986: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6987: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6988: }
1.227 brouard 6989: fprintf(ficgp,"\n#\n");
6990: if(invalidvarcomb[k1]){
6991: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6992: continue;
6993: }
6994:
1.241 brouard 6995: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6996: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6997: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6998: k=3; /* Offset */
6999: for (i=1; i<= nlstate ; i ++){
7000: if(i==1)
7001: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7002: else
7003: fprintf(ficgp,", '' ");
7004: /* l=(nlstate+ndeath)*(i-1)+1; */
7005: l=(nlstate+ndeath)*(cpt-1)+1;
7006: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7007: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
7008: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
7009: /* for (j=2; j<= nlstate ; j ++) */
7010: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7011: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7012: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7013: } /* nlstate */
7014: fprintf(ficgp,"\nset out\n");
1.218 brouard 7015: } /* end cpt state*/
7016: } /* end covariate */
7017: } /* End if backcast */
7018:
1.223 brouard 7019: /* 8eme */
1.218 brouard 7020: if(prevfcast==1){
7021: /* Projection from cross-sectional to stable (period) for each covariate */
7022:
1.237 brouard 7023: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7024: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7025: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7026: continue;
1.211 brouard 7027: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7028: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7029: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7030: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7031: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7032: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7033: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7034: vlv= nbcode[Tvaraff[k]][lv];
7035: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7036: }
1.237 brouard 7037: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7038: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7039: }
1.227 brouard 7040: fprintf(ficgp,"\n#\n");
7041: if(invalidvarcomb[k1]){
7042: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7043: continue;
7044: }
7045:
7046: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7047: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7048: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7049: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7050: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7051: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7052: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7053: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7054: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7055: if(i==1){
7056: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7057: }else{
7058: fprintf(ficgp,",\\\n '' ");
7059: }
7060: if(cptcoveff ==0){ /* No covariate */
7061: ioffset=2; /* Age is in 2 */
7062: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7063: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7064: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7065: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7066: fprintf(ficgp," u %d:(", ioffset);
7067: if(i==nlstate+1)
7068: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7069: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7070: else
7071: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7072: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7073: }else{ /* more than 2 covariates */
7074: if(cptcoveff ==1){
7075: ioffset=4; /* Age is in 4 */
7076: }else{
7077: ioffset=6; /* Age is in 6 */
7078: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7079: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7080: }
7081: fprintf(ficgp," u %d:(",ioffset);
7082: kl=0;
7083: strcpy(gplotcondition,"(");
7084: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7085: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7086: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7087: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7088: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7089: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7090: kl++;
7091: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7092: kl++;
7093: if(k <cptcoveff && cptcoveff>1)
7094: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7095: }
7096: strcpy(gplotcondition+strlen(gplotcondition),")");
7097: /* 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 *\/ */
7098: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7099: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7100: /* '' 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*/
7101: if(i==nlstate+1){
7102: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7103: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7104: }else{
7105: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7106: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7107: }
7108: } /* end if covariate */
7109: } /* nlstate */
7110: fprintf(ficgp,"\nset out\n");
1.223 brouard 7111: } /* end cpt state*/
7112: } /* end covariate */
7113: } /* End if prevfcast */
1.227 brouard 7114:
7115:
1.238 brouard 7116: /* 9eme writing MLE parameters */
7117: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7118: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7119: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7120: for(k=1; k <=(nlstate+ndeath); k++){
7121: if (k != i) {
1.227 brouard 7122: fprintf(ficgp,"# current state %d\n",k);
7123: for(j=1; j <=ncovmodel; j++){
7124: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7125: jk++;
7126: }
7127: fprintf(ficgp,"\n");
1.126 brouard 7128: }
7129: }
1.223 brouard 7130: }
1.187 brouard 7131: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7132:
1.145 brouard 7133: /*goto avoid;*/
1.238 brouard 7134: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7135: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7136: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7137: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7138: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7139: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7140: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7141: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7142: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7143: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7144: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7145: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7146: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7147: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7148: fprintf(ficgp,"#\n");
1.223 brouard 7149: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7150: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7151: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7152: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7153: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7154: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7155: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7156: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7157: continue;
7158: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7159: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7160: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7161: }
7162: fprintf(ficgp,"\n#\n");
1.241 brouard 7163: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7164: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7165: if (ng==1){
7166: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7167: fprintf(ficgp,"\nunset log y");
7168: }else if (ng==2){
7169: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7170: fprintf(ficgp,"\nset log y");
7171: }else if (ng==3){
7172: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7173: fprintf(ficgp,"\nset log y");
7174: }else
7175: fprintf(ficgp,"\nunset title ");
7176: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7177: i=1;
7178: for(k2=1; k2<=nlstate; k2++) {
7179: k3=i;
7180: for(k=1; k<=(nlstate+ndeath); k++) {
7181: if (k != k2){
7182: switch( ng) {
7183: case 1:
7184: if(nagesqr==0)
7185: fprintf(ficgp," p%d+p%d*x",i,i+1);
7186: else /* nagesqr =1 */
7187: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7188: break;
7189: case 2: /* ng=2 */
7190: if(nagesqr==0)
7191: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7192: else /* nagesqr =1 */
7193: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7194: break;
7195: case 3:
7196: if(nagesqr==0)
7197: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7198: else /* nagesqr =1 */
7199: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7200: break;
7201: }
7202: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7203: ijp=1; /* product no age */
7204: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7205: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7206: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7207: if(j==Tage[ij]) { /* Product by age */
7208: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7209: if(DummyV[j]==0){
1.237 brouard 7210: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7211: }else{ /* quantitative */
7212: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7213: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7214: }
7215: ij++;
7216: }
7217: }else if(j==Tprod[ijp]) { /* */
7218: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7219: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7220: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7221: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7222: /* 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)]); */
7223: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7224: }else{ /* Vn is dummy and Vm is quanti */
7225: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7226: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7227: }
7228: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7229: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7230: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7231: }else{ /* Both quanti */
7232: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7233: }
7234: }
1.238 brouard 7235: ijp++;
1.237 brouard 7236: }
7237: } else{ /* simple covariate */
7238: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7239: if(Dummy[j]==0){
7240: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7241: }else{ /* quantitative */
7242: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7243: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7244: }
1.237 brouard 7245: } /* end simple */
7246: } /* end j */
1.223 brouard 7247: }else{
7248: i=i-ncovmodel;
7249: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7250: fprintf(ficgp," (1.");
7251: }
1.227 brouard 7252:
1.223 brouard 7253: if(ng != 1){
7254: fprintf(ficgp,")/(1");
1.227 brouard 7255:
1.223 brouard 7256: for(k1=1; k1 <=nlstate; k1++){
7257: if(nagesqr==0)
7258: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7259: else /* nagesqr =1 */
7260: 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 7261:
1.223 brouard 7262: ij=1;
7263: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7264: if((j-2)==Tage[ij]) { /* Bug valgrind */
7265: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7266: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7267: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7268: ij++;
7269: }
7270: }
7271: else
1.225 brouard 7272: 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 7273: }
7274: fprintf(ficgp,")");
7275: }
7276: fprintf(ficgp,")");
7277: if(ng ==2)
7278: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7279: else /* ng= 3 */
7280: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7281: }else{ /* end ng <> 1 */
7282: if( k !=k2) /* logit p11 is hard to draw */
7283: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7284: }
7285: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7286: fprintf(ficgp,",");
7287: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7288: fprintf(ficgp,",");
7289: i=i+ncovmodel;
7290: } /* end k */
7291: } /* end k2 */
7292: fprintf(ficgp,"\n set out\n");
7293: } /* end jk */
7294: } /* end ng */
7295: /* avoid: */
7296: fflush(ficgp);
1.126 brouard 7297: } /* end gnuplot */
7298:
7299:
7300: /*************** Moving average **************/
1.219 brouard 7301: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7302: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7303:
1.222 brouard 7304: int i, cpt, cptcod;
7305: int modcovmax =1;
7306: int mobilavrange, mob;
7307: int iage=0;
7308:
7309: double sum=0.;
7310: double age;
7311: double *sumnewp, *sumnewm;
7312: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7313:
7314:
1.225 brouard 7315: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7316: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7317:
7318: sumnewp = vector(1,ncovcombmax);
7319: sumnewm = vector(1,ncovcombmax);
7320: agemingood = vector(1,ncovcombmax);
7321: agemaxgood = vector(1,ncovcombmax);
7322:
7323: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7324: sumnewm[cptcod]=0.;
7325: sumnewp[cptcod]=0.;
7326: agemingood[cptcod]=0;
7327: agemaxgood[cptcod]=0;
7328: }
7329: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7330:
7331: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7332: if(mobilav==1) mobilavrange=5; /* default */
7333: else mobilavrange=mobilav;
7334: for (age=bage; age<=fage; age++)
7335: for (i=1; i<=nlstate;i++)
7336: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7337: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7338: /* We keep the original values on the extreme ages bage, fage and for
7339: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7340: we use a 5 terms etc. until the borders are no more concerned.
7341: */
7342: for (mob=3;mob <=mobilavrange;mob=mob+2){
7343: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7344: for (i=1; i<=nlstate;i++){
7345: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7346: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7347: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7348: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7349: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7350: }
7351: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7352: }
7353: }
7354: }/* end age */
7355: }/* end mob */
7356: }else
7357: return -1;
7358: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7359: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7360: if(invalidvarcomb[cptcod]){
7361: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7362: continue;
7363: }
1.219 brouard 7364:
1.222 brouard 7365: agemingood[cptcod]=fage-(mob-1)/2;
7366: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7367: sumnewm[cptcod]=0.;
7368: for (i=1; i<=nlstate;i++){
7369: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7370: }
7371: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7372: agemingood[cptcod]=age;
7373: }else{ /* bad */
7374: for (i=1; i<=nlstate;i++){
7375: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7376: } /* i */
7377: } /* end bad */
7378: }/* age */
7379: sum=0.;
7380: for (i=1; i<=nlstate;i++){
7381: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7382: }
7383: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7384: 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);
7385: /* for (i=1; i<=nlstate;i++){ */
7386: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7387: /* } /\* i *\/ */
7388: } /* end bad */
7389: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7390: /* From youngest, finding the oldest wrong */
7391: agemaxgood[cptcod]=bage+(mob-1)/2;
7392: for (age=bage+(mob-1)/2; age<=fage; age++){
7393: sumnewm[cptcod]=0.;
7394: for (i=1; i<=nlstate;i++){
7395: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7396: }
7397: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7398: agemaxgood[cptcod]=age;
7399: }else{ /* bad */
7400: for (i=1; i<=nlstate;i++){
7401: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7402: } /* i */
7403: } /* end bad */
7404: }/* age */
7405: sum=0.;
7406: for (i=1; i<=nlstate;i++){
7407: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7408: }
7409: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7410: 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);
7411: /* for (i=1; i<=nlstate;i++){ */
7412: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7413: /* } /\* i *\/ */
7414: } /* end bad */
7415:
7416: for (age=bage; age<=fage; age++){
1.235 brouard 7417: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7418: sumnewp[cptcod]=0.;
7419: sumnewm[cptcod]=0.;
7420: for (i=1; i<=nlstate;i++){
7421: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7422: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7423: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7424: }
7425: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7426: }
7427: /* printf("\n"); */
7428: /* } */
7429: /* brutal averaging */
7430: for (i=1; i<=nlstate;i++){
7431: for (age=1; age<=bage; age++){
7432: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7433: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7434: }
7435: for (age=fage; age<=AGESUP; age++){
7436: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7437: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7438: }
7439: } /* end i status */
7440: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7441: for (age=1; age<=AGESUP; age++){
7442: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7443: mobaverage[(int)age][i][cptcod]=0.;
7444: }
7445: }
7446: }/* end cptcod */
7447: free_vector(sumnewm,1, ncovcombmax);
7448: free_vector(sumnewp,1, ncovcombmax);
7449: free_vector(agemaxgood,1, ncovcombmax);
7450: free_vector(agemingood,1, ncovcombmax);
7451: return 0;
7452: }/* End movingaverage */
1.218 brouard 7453:
1.126 brouard 7454:
7455: /************** Forecasting ******************/
1.235 brouard 7456: 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 7457: /* proj1, year, month, day of starting projection
7458: agemin, agemax range of age
7459: dateprev1 dateprev2 range of dates during which prevalence is computed
7460: anproj2 year of en of projection (same day and month as proj1).
7461: */
1.235 brouard 7462: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7463: double agec; /* generic age */
7464: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7465: double *popeffectif,*popcount;
7466: double ***p3mat;
1.218 brouard 7467: /* double ***mobaverage; */
1.126 brouard 7468: char fileresf[FILENAMELENGTH];
7469:
7470: agelim=AGESUP;
1.211 brouard 7471: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7472: in each health status at the date of interview (if between dateprev1 and dateprev2).
7473: We still use firstpass and lastpass as another selection.
7474: */
1.214 brouard 7475: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7476: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7477:
1.201 brouard 7478: strcpy(fileresf,"F_");
7479: strcat(fileresf,fileresu);
1.126 brouard 7480: if((ficresf=fopen(fileresf,"w"))==NULL) {
7481: printf("Problem with forecast resultfile: %s\n", fileresf);
7482: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7483: }
1.235 brouard 7484: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7485: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7486:
1.225 brouard 7487: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7488:
7489:
7490: stepsize=(int) (stepm+YEARM-1)/YEARM;
7491: if (stepm<=12) stepsize=1;
7492: if(estepm < stepm){
7493: printf ("Problem %d lower than %d\n",estepm, stepm);
7494: }
7495: else hstepm=estepm;
7496:
7497: hstepm=hstepm/stepm;
7498: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7499: fractional in yp1 */
7500: anprojmean=yp;
7501: yp2=modf((yp1*12),&yp);
7502: mprojmean=yp;
7503: yp1=modf((yp2*30.5),&yp);
7504: jprojmean=yp;
7505: if(jprojmean==0) jprojmean=1;
7506: if(mprojmean==0) jprojmean=1;
7507:
1.227 brouard 7508: i1=pow(2,cptcoveff);
1.126 brouard 7509: if (cptcovn < 1){i1=1;}
7510:
7511: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7512:
7513: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7514:
1.126 brouard 7515: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7516: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7517: for(k=1; k<=i1;k++){
1.253 brouard 7518: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7519: continue;
1.227 brouard 7520: if(invalidvarcomb[k]){
7521: printf("\nCombination (%d) projection ignored because no cases \n",k);
7522: continue;
7523: }
7524: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7525: for(j=1;j<=cptcoveff;j++) {
7526: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7527: }
1.235 brouard 7528: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7529: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7530: }
1.227 brouard 7531: fprintf(ficresf," yearproj age");
7532: for(j=1; j<=nlstate+ndeath;j++){
7533: for(i=1; i<=nlstate;i++)
7534: fprintf(ficresf," p%d%d",i,j);
7535: fprintf(ficresf," wp.%d",j);
7536: }
7537: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7538: fprintf(ficresf,"\n");
7539: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7540: for (agec=fage; agec>=(ageminpar-1); agec--){
7541: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7542: nhstepm = nhstepm/hstepm;
7543: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7544: oldm=oldms;savm=savms;
1.235 brouard 7545: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7546:
7547: for (h=0; h<=nhstepm; h++){
7548: if (h*hstepm/YEARM*stepm ==yearp) {
7549: fprintf(ficresf,"\n");
7550: for(j=1;j<=cptcoveff;j++)
7551: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7552: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7553: }
7554: for(j=1; j<=nlstate+ndeath;j++) {
7555: ppij=0.;
7556: for(i=1; i<=nlstate;i++) {
7557: if (mobilav==1)
7558: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7559: else {
7560: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7561: }
7562: if (h*hstepm/YEARM*stepm== yearp) {
7563: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7564: }
7565: } /* end i */
7566: if (h*hstepm/YEARM*stepm==yearp) {
7567: fprintf(ficresf," %.3f", ppij);
7568: }
7569: }/* end j */
7570: } /* end h */
7571: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7572: } /* end agec */
7573: } /* end yearp */
7574: } /* end k */
1.219 brouard 7575:
1.126 brouard 7576: fclose(ficresf);
1.215 brouard 7577: printf("End of Computing forecasting \n");
7578: fprintf(ficlog,"End of Computing forecasting\n");
7579:
1.126 brouard 7580: }
7581:
1.218 brouard 7582: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7583: /* 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 7584: /* /\* back1, year, month, day of starting backection */
7585: /* agemin, agemax range of age */
7586: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7587: /* anback2 year of en of backection (same day and month as back1). */
7588: /* *\/ */
7589: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7590: /* double agec; /\* generic age *\/ */
7591: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7592: /* double *popeffectif,*popcount; */
7593: /* double ***p3mat; */
7594: /* /\* double ***mobaverage; *\/ */
7595: /* char fileresfb[FILENAMELENGTH]; */
7596:
7597: /* agelim=AGESUP; */
7598: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7599: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7600: /* We still use firstpass and lastpass as another selection. */
7601: /* *\/ */
7602: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7603: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7604: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7605:
7606: /* strcpy(fileresfb,"FB_"); */
7607: /* strcat(fileresfb,fileresu); */
7608: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7609: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7610: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7611: /* } */
7612: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7613: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7614:
1.225 brouard 7615: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7616:
7617: /* /\* if (mobilav!=0) { *\/ */
7618: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7619: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7620: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7621: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7622: /* /\* } *\/ */
7623: /* /\* } *\/ */
7624:
7625: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7626: /* if (stepm<=12) stepsize=1; */
7627: /* if(estepm < stepm){ */
7628: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7629: /* } */
7630: /* else hstepm=estepm; */
7631:
7632: /* hstepm=hstepm/stepm; */
7633: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7634: /* fractional in yp1 *\/ */
7635: /* anprojmean=yp; */
7636: /* yp2=modf((yp1*12),&yp); */
7637: /* mprojmean=yp; */
7638: /* yp1=modf((yp2*30.5),&yp); */
7639: /* jprojmean=yp; */
7640: /* if(jprojmean==0) jprojmean=1; */
7641: /* if(mprojmean==0) jprojmean=1; */
7642:
1.225 brouard 7643: /* i1=cptcoveff; */
1.218 brouard 7644: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7645:
1.218 brouard 7646: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7647:
1.218 brouard 7648: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7649:
7650: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7651: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7652: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7653: /* k=k+1; */
7654: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7655: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7656: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7657: /* } */
7658: /* fprintf(ficresfb," yearbproj age"); */
7659: /* for(j=1; j<=nlstate+ndeath;j++){ */
7660: /* for(i=1; i<=nlstate;i++) */
7661: /* fprintf(ficresfb," p%d%d",i,j); */
7662: /* fprintf(ficresfb," p.%d",j); */
7663: /* } */
7664: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7665: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7666: /* fprintf(ficresfb,"\n"); */
7667: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7668: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7669: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7670: /* nhstepm = nhstepm/hstepm; */
7671: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7672: /* oldm=oldms;savm=savms; */
7673: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7674: /* for (h=0; h<=nhstepm; h++){ */
7675: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7676: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7677: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7678: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7679: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7680: /* } */
7681: /* for(j=1; j<=nlstate+ndeath;j++) { */
7682: /* ppij=0.; */
7683: /* for(i=1; i<=nlstate;i++) { */
7684: /* if (mobilav==1) */
7685: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7686: /* else { */
7687: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7688: /* } */
7689: /* if (h*hstepm/YEARM*stepm== yearp) { */
7690: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7691: /* } */
7692: /* } /\* end i *\/ */
7693: /* if (h*hstepm/YEARM*stepm==yearp) { */
7694: /* fprintf(ficresfb," %.3f", ppij); */
7695: /* } */
7696: /* }/\* end j *\/ */
7697: /* } /\* end h *\/ */
7698: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7699: /* } /\* end agec *\/ */
7700: /* } /\* end yearp *\/ */
7701: /* } /\* end cptcod *\/ */
7702: /* } /\* end cptcov *\/ */
7703:
7704: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7705:
7706: /* fclose(ficresfb); */
7707: /* printf("End of Computing Back forecasting \n"); */
7708: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7709:
1.218 brouard 7710: /* } */
1.217 brouard 7711:
1.126 brouard 7712: /************** Forecasting *****not tested NB*************/
1.227 brouard 7713: /* 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 7714:
1.227 brouard 7715: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7716: /* int *popage; */
7717: /* double calagedatem, agelim, kk1, kk2; */
7718: /* double *popeffectif,*popcount; */
7719: /* double ***p3mat,***tabpop,***tabpopprev; */
7720: /* /\* double ***mobaverage; *\/ */
7721: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7722:
1.227 brouard 7723: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7724: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7725: /* agelim=AGESUP; */
7726: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7727:
1.227 brouard 7728: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7729:
7730:
1.227 brouard 7731: /* strcpy(filerespop,"POP_"); */
7732: /* strcat(filerespop,fileresu); */
7733: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7734: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7735: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7736: /* } */
7737: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7738: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7739:
1.227 brouard 7740: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7741:
1.227 brouard 7742: /* /\* if (mobilav!=0) { *\/ */
7743: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7744: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7745: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7746: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7747: /* /\* } *\/ */
7748: /* /\* } *\/ */
1.126 brouard 7749:
1.227 brouard 7750: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7751: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7752:
1.227 brouard 7753: /* agelim=AGESUP; */
1.126 brouard 7754:
1.227 brouard 7755: /* hstepm=1; */
7756: /* hstepm=hstepm/stepm; */
1.218 brouard 7757:
1.227 brouard 7758: /* if (popforecast==1) { */
7759: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7760: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7761: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7762: /* } */
7763: /* popage=ivector(0,AGESUP); */
7764: /* popeffectif=vector(0,AGESUP); */
7765: /* popcount=vector(0,AGESUP); */
1.126 brouard 7766:
1.227 brouard 7767: /* i=1; */
7768: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7769:
1.227 brouard 7770: /* imx=i; */
7771: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7772: /* } */
1.218 brouard 7773:
1.227 brouard 7774: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7775: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7776: /* k=k+1; */
7777: /* fprintf(ficrespop,"\n#******"); */
7778: /* for(j=1;j<=cptcoveff;j++) { */
7779: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7780: /* } */
7781: /* fprintf(ficrespop,"******\n"); */
7782: /* fprintf(ficrespop,"# Age"); */
7783: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7784: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7785:
1.227 brouard 7786: /* for (cpt=0; cpt<=0;cpt++) { */
7787: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7788:
1.227 brouard 7789: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7790: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7791: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7792:
1.227 brouard 7793: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7794: /* oldm=oldms;savm=savms; */
7795: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7796:
1.227 brouard 7797: /* for (h=0; h<=nhstepm; h++){ */
7798: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7799: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7800: /* } */
7801: /* for(j=1; j<=nlstate+ndeath;j++) { */
7802: /* kk1=0.;kk2=0; */
7803: /* for(i=1; i<=nlstate;i++) { */
7804: /* if (mobilav==1) */
7805: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7806: /* else { */
7807: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7808: /* } */
7809: /* } */
7810: /* if (h==(int)(calagedatem+12*cpt)){ */
7811: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7812: /* /\*fprintf(ficrespop," %.3f", kk1); */
7813: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7814: /* } */
7815: /* } */
7816: /* for(i=1; i<=nlstate;i++){ */
7817: /* kk1=0.; */
7818: /* for(j=1; j<=nlstate;j++){ */
7819: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7820: /* } */
7821: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7822: /* } */
1.218 brouard 7823:
1.227 brouard 7824: /* if (h==(int)(calagedatem+12*cpt)) */
7825: /* for(j=1; j<=nlstate;j++) */
7826: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7827: /* } */
7828: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7829: /* } */
7830: /* } */
1.218 brouard 7831:
1.227 brouard 7832: /* /\******\/ */
1.218 brouard 7833:
1.227 brouard 7834: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7835: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7836: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7837: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7838: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7839:
1.227 brouard 7840: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7841: /* oldm=oldms;savm=savms; */
7842: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7843: /* for (h=0; h<=nhstepm; h++){ */
7844: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7845: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7846: /* } */
7847: /* for(j=1; j<=nlstate+ndeath;j++) { */
7848: /* kk1=0.;kk2=0; */
7849: /* for(i=1; i<=nlstate;i++) { */
7850: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7851: /* } */
7852: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7853: /* } */
7854: /* } */
7855: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7856: /* } */
7857: /* } */
7858: /* } */
7859: /* } */
1.218 brouard 7860:
1.227 brouard 7861: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7862:
1.227 brouard 7863: /* if (popforecast==1) { */
7864: /* free_ivector(popage,0,AGESUP); */
7865: /* free_vector(popeffectif,0,AGESUP); */
7866: /* free_vector(popcount,0,AGESUP); */
7867: /* } */
7868: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7869: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7870: /* fclose(ficrespop); */
7871: /* } /\* End of popforecast *\/ */
1.218 brouard 7872:
1.126 brouard 7873: int fileappend(FILE *fichier, char *optionfich)
7874: {
7875: if((fichier=fopen(optionfich,"a"))==NULL) {
7876: printf("Problem with file: %s\n", optionfich);
7877: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7878: return (0);
7879: }
7880: fflush(fichier);
7881: return (1);
7882: }
7883:
7884:
7885: /**************** function prwizard **********************/
7886: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7887: {
7888:
7889: /* Wizard to print covariance matrix template */
7890:
1.164 brouard 7891: char ca[32], cb[32];
7892: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7893: int numlinepar;
7894:
7895: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7896: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7897: for(i=1; i <=nlstate; i++){
7898: jj=0;
7899: for(j=1; j <=nlstate+ndeath; j++){
7900: if(j==i) continue;
7901: jj++;
7902: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7903: printf("%1d%1d",i,j);
7904: fprintf(ficparo,"%1d%1d",i,j);
7905: for(k=1; k<=ncovmodel;k++){
7906: /* printf(" %lf",param[i][j][k]); */
7907: /* fprintf(ficparo," %lf",param[i][j][k]); */
7908: printf(" 0.");
7909: fprintf(ficparo," 0.");
7910: }
7911: printf("\n");
7912: fprintf(ficparo,"\n");
7913: }
7914: }
7915: printf("# Scales (for hessian or gradient estimation)\n");
7916: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7917: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7918: for(i=1; i <=nlstate; i++){
7919: jj=0;
7920: for(j=1; j <=nlstate+ndeath; j++){
7921: if(j==i) continue;
7922: jj++;
7923: fprintf(ficparo,"%1d%1d",i,j);
7924: printf("%1d%1d",i,j);
7925: fflush(stdout);
7926: for(k=1; k<=ncovmodel;k++){
7927: /* printf(" %le",delti3[i][j][k]); */
7928: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7929: printf(" 0.");
7930: fprintf(ficparo," 0.");
7931: }
7932: numlinepar++;
7933: printf("\n");
7934: fprintf(ficparo,"\n");
7935: }
7936: }
7937: printf("# Covariance matrix\n");
7938: /* # 121 Var(a12)\n\ */
7939: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7940: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7941: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7942: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7943: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7944: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7945: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7946: fflush(stdout);
7947: fprintf(ficparo,"# Covariance matrix\n");
7948: /* # 121 Var(a12)\n\ */
7949: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7950: /* # ...\n\ */
7951: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7952:
7953: for(itimes=1;itimes<=2;itimes++){
7954: jj=0;
7955: for(i=1; i <=nlstate; i++){
7956: for(j=1; j <=nlstate+ndeath; j++){
7957: if(j==i) continue;
7958: for(k=1; k<=ncovmodel;k++){
7959: jj++;
7960: ca[0]= k+'a'-1;ca[1]='\0';
7961: if(itimes==1){
7962: printf("#%1d%1d%d",i,j,k);
7963: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7964: }else{
7965: printf("%1d%1d%d",i,j,k);
7966: fprintf(ficparo,"%1d%1d%d",i,j,k);
7967: /* printf(" %.5le",matcov[i][j]); */
7968: }
7969: ll=0;
7970: for(li=1;li <=nlstate; li++){
7971: for(lj=1;lj <=nlstate+ndeath; lj++){
7972: if(lj==li) continue;
7973: for(lk=1;lk<=ncovmodel;lk++){
7974: ll++;
7975: if(ll<=jj){
7976: cb[0]= lk +'a'-1;cb[1]='\0';
7977: if(ll<jj){
7978: if(itimes==1){
7979: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7980: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7981: }else{
7982: printf(" 0.");
7983: fprintf(ficparo," 0.");
7984: }
7985: }else{
7986: if(itimes==1){
7987: printf(" Var(%s%1d%1d)",ca,i,j);
7988: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7989: }else{
7990: printf(" 0.");
7991: fprintf(ficparo," 0.");
7992: }
7993: }
7994: }
7995: } /* end lk */
7996: } /* end lj */
7997: } /* end li */
7998: printf("\n");
7999: fprintf(ficparo,"\n");
8000: numlinepar++;
8001: } /* end k*/
8002: } /*end j */
8003: } /* end i */
8004: } /* end itimes */
8005:
8006: } /* end of prwizard */
8007: /******************* Gompertz Likelihood ******************************/
8008: double gompertz(double x[])
8009: {
8010: double A,B,L=0.0,sump=0.,num=0.;
8011: int i,n=0; /* n is the size of the sample */
8012:
1.220 brouard 8013: for (i=1;i<=imx ; i++) {
1.126 brouard 8014: sump=sump+weight[i];
8015: /* sump=sump+1;*/
8016: num=num+1;
8017: }
8018:
8019:
8020: /* for (i=0; i<=imx; i++)
8021: 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]);*/
8022:
8023: for (i=1;i<=imx ; i++)
8024: {
8025: if (cens[i] == 1 && wav[i]>1)
8026: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8027:
8028: if (cens[i] == 0 && wav[i]>1)
8029: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8030: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8031:
8032: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8033: if (wav[i] > 1 ) { /* ??? */
8034: L=L+A*weight[i];
8035: /* 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]);*/
8036: }
8037: }
8038:
8039: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8040:
8041: return -2*L*num/sump;
8042: }
8043:
1.136 brouard 8044: #ifdef GSL
8045: /******************* Gompertz_f Likelihood ******************************/
8046: double gompertz_f(const gsl_vector *v, void *params)
8047: {
8048: double A,B,LL=0.0,sump=0.,num=0.;
8049: double *x= (double *) v->data;
8050: int i,n=0; /* n is the size of the sample */
8051:
8052: for (i=0;i<=imx-1 ; i++) {
8053: sump=sump+weight[i];
8054: /* sump=sump+1;*/
8055: num=num+1;
8056: }
8057:
8058:
8059: /* for (i=0; i<=imx; i++)
8060: 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]);*/
8061: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8062: for (i=1;i<=imx ; i++)
8063: {
8064: if (cens[i] == 1 && wav[i]>1)
8065: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8066:
8067: if (cens[i] == 0 && wav[i]>1)
8068: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8069: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8070:
8071: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8072: if (wav[i] > 1 ) { /* ??? */
8073: LL=LL+A*weight[i];
8074: /* 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]);*/
8075: }
8076: }
8077:
8078: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8079: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8080:
8081: return -2*LL*num/sump;
8082: }
8083: #endif
8084:
1.126 brouard 8085: /******************* Printing html file ***********/
1.201 brouard 8086: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8087: int lastpass, int stepm, int weightopt, char model[],\
8088: int imx, double p[],double **matcov,double agemortsup){
8089: int i,k;
8090:
8091: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8092: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8093: for (i=1;i<=2;i++)
8094: 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 8095: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8096: fprintf(fichtm,"</ul>");
8097:
8098: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8099:
8100: 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>");
8101:
8102: for (k=agegomp;k<(agemortsup-2);k++)
8103: 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]);
8104:
8105:
8106: fflush(fichtm);
8107: }
8108:
8109: /******************* Gnuplot file **************/
1.201 brouard 8110: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8111:
8112: char dirfileres[132],optfileres[132];
1.164 brouard 8113:
1.126 brouard 8114: int ng;
8115:
8116:
8117: /*#ifdef windows */
8118: fprintf(ficgp,"cd \"%s\" \n",pathc);
8119: /*#endif */
8120:
8121:
8122: strcpy(dirfileres,optionfilefiname);
8123: strcpy(optfileres,"vpl");
1.199 brouard 8124: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8125: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8126: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8127: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8128: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8129:
8130: }
8131:
1.136 brouard 8132: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8133: {
1.126 brouard 8134:
1.136 brouard 8135: /*-------- data file ----------*/
8136: FILE *fic;
8137: char dummy[]=" ";
1.240 brouard 8138: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8139: int lstra;
1.136 brouard 8140: int linei, month, year,iout;
8141: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8142: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8143: char *stratrunc;
1.223 brouard 8144:
1.240 brouard 8145: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8146: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8147:
1.240 brouard 8148: for(v=1; v <=ncovcol;v++){
8149: DummyV[v]=0;
8150: FixedV[v]=0;
8151: }
8152: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8153: DummyV[v]=1;
8154: FixedV[v]=0;
8155: }
8156: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8157: DummyV[v]=0;
8158: FixedV[v]=1;
8159: }
8160: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8161: DummyV[v]=1;
8162: FixedV[v]=1;
8163: }
8164: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8165: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8166: 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]);
8167: }
1.126 brouard 8168:
1.136 brouard 8169: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8170: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8171: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8172: }
1.126 brouard 8173:
1.136 brouard 8174: i=1;
8175: linei=0;
8176: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8177: linei=linei+1;
8178: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8179: if(line[j] == '\t')
8180: line[j] = ' ';
8181: }
8182: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8183: ;
8184: };
8185: line[j+1]=0; /* Trims blanks at end of line */
8186: if(line[0]=='#'){
8187: fprintf(ficlog,"Comment line\n%s\n",line);
8188: printf("Comment line\n%s\n",line);
8189: continue;
8190: }
8191: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8192: strcpy(line, linetmp);
1.223 brouard 8193:
8194: /* Loops on waves */
8195: for (j=maxwav;j>=1;j--){
8196: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8197: cutv(stra, strb, line, ' ');
8198: if(strb[0]=='.') { /* Missing value */
8199: lval=-1;
8200: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8201: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8202: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8203: 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);
8204: 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);
8205: return 1;
8206: }
8207: }else{
8208: errno=0;
8209: /* what_kind_of_number(strb); */
8210: dval=strtod(strb,&endptr);
8211: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8212: /* if(strb != endptr && *endptr == '\0') */
8213: /* dval=dlval; */
8214: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8215: if( strb[0]=='\0' || (*endptr != '\0')){
8216: 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);
8217: 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);
8218: return 1;
8219: }
8220: cotqvar[j][iv][i]=dval;
8221: cotvar[j][ntv+iv][i]=dval;
8222: }
8223: strcpy(line,stra);
1.223 brouard 8224: }/* end loop ntqv */
1.225 brouard 8225:
1.223 brouard 8226: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8227: cutv(stra, strb, line, ' ');
8228: if(strb[0]=='.') { /* Missing value */
8229: lval=-1;
8230: }else{
8231: errno=0;
8232: lval=strtol(strb,&endptr,10);
8233: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8234: if( strb[0]=='\0' || (*endptr != '\0')){
8235: 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);
8236: 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);
8237: return 1;
8238: }
8239: }
8240: if(lval <-1 || lval >1){
8241: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8242: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8243: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8244: For example, for multinomial values like 1, 2 and 3,\n \
8245: build V1=0 V2=0 for the reference value (1),\n \
8246: V1=1 V2=0 for (2) \n \
1.223 brouard 8247: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8248: output of IMaCh is often meaningless.\n \
1.223 brouard 8249: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8250: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8251: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8252: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8253: For example, for multinomial values like 1, 2 and 3,\n \
8254: build V1=0 V2=0 for the reference value (1),\n \
8255: V1=1 V2=0 for (2) \n \
1.223 brouard 8256: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8257: output of IMaCh is often meaningless.\n \
1.223 brouard 8258: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8259: return 1;
8260: }
8261: cotvar[j][iv][i]=(double)(lval);
8262: strcpy(line,stra);
1.223 brouard 8263: }/* end loop ntv */
1.225 brouard 8264:
1.223 brouard 8265: /* Statuses at wave */
1.137 brouard 8266: cutv(stra, strb, line, ' ');
1.223 brouard 8267: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8268: lval=-1;
1.136 brouard 8269: }else{
1.238 brouard 8270: errno=0;
8271: lval=strtol(strb,&endptr,10);
8272: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8273: if( strb[0]=='\0' || (*endptr != '\0')){
8274: 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);
8275: 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);
8276: return 1;
8277: }
1.136 brouard 8278: }
1.225 brouard 8279:
1.136 brouard 8280: s[j][i]=lval;
1.225 brouard 8281:
1.223 brouard 8282: /* Date of Interview */
1.136 brouard 8283: strcpy(line,stra);
8284: cutv(stra, strb,line,' ');
1.169 brouard 8285: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8286: }
1.169 brouard 8287: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8288: month=99;
8289: year=9999;
1.136 brouard 8290: }else{
1.225 brouard 8291: 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);
8292: 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);
8293: return 1;
1.136 brouard 8294: }
8295: anint[j][i]= (double) year;
8296: mint[j][i]= (double)month;
8297: strcpy(line,stra);
1.223 brouard 8298: } /* End loop on waves */
1.225 brouard 8299:
1.223 brouard 8300: /* Date of death */
1.136 brouard 8301: cutv(stra, strb,line,' ');
1.169 brouard 8302: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8303: }
1.169 brouard 8304: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8305: month=99;
8306: year=9999;
8307: }else{
1.141 brouard 8308: 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 8309: 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);
8310: return 1;
1.136 brouard 8311: }
8312: andc[i]=(double) year;
8313: moisdc[i]=(double) month;
8314: strcpy(line,stra);
8315:
1.223 brouard 8316: /* Date of birth */
1.136 brouard 8317: cutv(stra, strb,line,' ');
1.169 brouard 8318: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8319: }
1.169 brouard 8320: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8321: month=99;
8322: year=9999;
8323: }else{
1.141 brouard 8324: 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);
8325: 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 8326: return 1;
1.136 brouard 8327: }
8328: if (year==9999) {
1.141 brouard 8329: 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);
8330: 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 8331: return 1;
8332:
1.136 brouard 8333: }
8334: annais[i]=(double)(year);
8335: moisnais[i]=(double)(month);
8336: strcpy(line,stra);
1.225 brouard 8337:
1.223 brouard 8338: /* Sample weight */
1.136 brouard 8339: cutv(stra, strb,line,' ');
8340: errno=0;
8341: dval=strtod(strb,&endptr);
8342: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8343: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8344: 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 8345: fflush(ficlog);
8346: return 1;
8347: }
8348: weight[i]=dval;
8349: strcpy(line,stra);
1.225 brouard 8350:
1.223 brouard 8351: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8352: cutv(stra, strb, line, ' ');
8353: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8354: lval=-1;
1.223 brouard 8355: }else{
1.225 brouard 8356: errno=0;
8357: /* what_kind_of_number(strb); */
8358: dval=strtod(strb,&endptr);
8359: /* if(strb != endptr && *endptr == '\0') */
8360: /* dval=dlval; */
8361: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8362: if( strb[0]=='\0' || (*endptr != '\0')){
8363: 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);
8364: 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);
8365: return 1;
8366: }
8367: coqvar[iv][i]=dval;
1.226 brouard 8368: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8369: }
8370: strcpy(line,stra);
8371: }/* end loop nqv */
1.136 brouard 8372:
1.223 brouard 8373: /* Covariate values */
1.136 brouard 8374: for (j=ncovcol;j>=1;j--){
8375: cutv(stra, strb,line,' ');
1.223 brouard 8376: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8377: lval=-1;
1.136 brouard 8378: }else{
1.225 brouard 8379: errno=0;
8380: lval=strtol(strb,&endptr,10);
8381: if( strb[0]=='\0' || (*endptr != '\0')){
8382: 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);
8383: 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);
8384: return 1;
8385: }
1.136 brouard 8386: }
8387: if(lval <-1 || lval >1){
1.225 brouard 8388: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8389: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8390: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8391: For example, for multinomial values like 1, 2 and 3,\n \
8392: build V1=0 V2=0 for the reference value (1),\n \
8393: V1=1 V2=0 for (2) \n \
1.136 brouard 8394: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8395: output of IMaCh is often meaningless.\n \
1.136 brouard 8396: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8397: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8398: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8399: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8400: For example, for multinomial values like 1, 2 and 3,\n \
8401: build V1=0 V2=0 for the reference value (1),\n \
8402: V1=1 V2=0 for (2) \n \
1.136 brouard 8403: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8404: output of IMaCh is often meaningless.\n \
1.136 brouard 8405: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8406: return 1;
1.136 brouard 8407: }
8408: covar[j][i]=(double)(lval);
8409: strcpy(line,stra);
8410: }
8411: lstra=strlen(stra);
1.225 brouard 8412:
1.136 brouard 8413: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8414: stratrunc = &(stra[lstra-9]);
8415: num[i]=atol(stratrunc);
8416: }
8417: else
8418: num[i]=atol(stra);
8419: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8420: 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;}*/
8421:
8422: i=i+1;
8423: } /* End loop reading data */
1.225 brouard 8424:
1.136 brouard 8425: *imax=i-1; /* Number of individuals */
8426: fclose(fic);
1.225 brouard 8427:
1.136 brouard 8428: return (0);
1.164 brouard 8429: /* endread: */
1.225 brouard 8430: printf("Exiting readdata: ");
8431: fclose(fic);
8432: return (1);
1.223 brouard 8433: }
1.126 brouard 8434:
1.234 brouard 8435: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8436: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8437: while (*p2 == ' ')
1.234 brouard 8438: p2++;
8439: /* while ((*p1++ = *p2++) !=0) */
8440: /* ; */
8441: /* do */
8442: /* while (*p2 == ' ') */
8443: /* p2++; */
8444: /* while (*p1++ == *p2++); */
8445: *stri=p2;
1.145 brouard 8446: }
8447:
1.235 brouard 8448: int decoderesult ( char resultline[], int nres)
1.230 brouard 8449: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8450: {
1.235 brouard 8451: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8452: char resultsav[MAXLINE];
1.234 brouard 8453: int resultmodel[MAXLINE];
8454: int modelresult[MAXLINE];
1.230 brouard 8455: char stra[80], strb[80], strc[80], strd[80],stre[80];
8456:
1.234 brouard 8457: removefirstspace(&resultline);
1.233 brouard 8458: printf("decoderesult:%s\n",resultline);
1.230 brouard 8459:
8460: if (strstr(resultline,"v") !=0){
8461: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8462: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8463: return 1;
8464: }
8465: trimbb(resultsav, resultline);
8466: if (strlen(resultsav) >1){
8467: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8468: }
1.253 brouard 8469: if(j == 0){ /* Resultline but no = */
8470: TKresult[nres]=0; /* Combination for the nresult and the model */
8471: return (0);
8472: }
8473:
1.234 brouard 8474: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8475: 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);
8476: 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);
8477: }
8478: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8479: if(nbocc(resultsav,'=') >1){
8480: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8481: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8482: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8483: }else
8484: cutl(strc,strd,resultsav,'=');
1.230 brouard 8485: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8486:
1.230 brouard 8487: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8488: Tvarsel[k]=atoi(strc);
8489: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8490: /* cptcovsel++; */
8491: if (nbocc(stra,'=') >0)
8492: strcpy(resultsav,stra); /* and analyzes it */
8493: }
1.235 brouard 8494: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8495: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8496: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8497: match=0;
1.236 brouard 8498: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8499: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8500: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8501: match=1;
8502: break;
8503: }
8504: }
8505: if(match == 0){
8506: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8507: }
8508: }
8509: }
1.235 brouard 8510: /* Checking for missing or useless values in comparison of current model needs */
8511: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8512: match=0;
1.235 brouard 8513: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8514: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8515: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8516: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8517: ++match;
8518: }
8519: }
8520: }
8521: if(match == 0){
8522: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8523: }else if(match > 1){
8524: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8525: }
8526: }
1.235 brouard 8527:
1.234 brouard 8528: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8529: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8530: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8531: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8532: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8533: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8534: /* 1 0 0 0 */
8535: /* 2 1 0 0 */
8536: /* 3 0 1 0 */
8537: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8538: /* 5 0 0 1 */
8539: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8540: /* 7 0 1 1 */
8541: /* 8 1 1 1 */
1.237 brouard 8542: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8543: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8544: /* V5*age V5 known which value for nres? */
8545: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8546: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8547: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8548: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8549: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8550: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8551: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8552: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8553: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8554: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8555: k4++;;
8556: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8557: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8558: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8559: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8560: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8561: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8562: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8563: k4q++;;
8564: }
8565: }
1.234 brouard 8566:
1.235 brouard 8567: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8568: return (0);
8569: }
1.235 brouard 8570:
1.230 brouard 8571: int decodemodel( char model[], int lastobs)
8572: /**< This routine decodes the model and returns:
1.224 brouard 8573: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8574: * - nagesqr = 1 if age*age in the model, otherwise 0.
8575: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8576: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8577: * - cptcovage number of covariates with age*products =2
8578: * - cptcovs number of simple covariates
8579: * - 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
8580: * which is a new column after the 9 (ncovcol) variables.
8581: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8582: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8583: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8584: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8585: */
1.136 brouard 8586: {
1.238 brouard 8587: int i, j, k, ks, v;
1.227 brouard 8588: int j1, k1, k2, k3, k4;
1.136 brouard 8589: char modelsav[80];
1.145 brouard 8590: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8591: char *strpt;
1.136 brouard 8592:
1.145 brouard 8593: /*removespace(model);*/
1.136 brouard 8594: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8595: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8596: if (strstr(model,"AGE") !=0){
1.192 brouard 8597: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8598: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8599: return 1;
8600: }
1.141 brouard 8601: if (strstr(model,"v") !=0){
8602: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8603: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8604: return 1;
8605: }
1.187 brouard 8606: strcpy(modelsav,model);
8607: if ((strpt=strstr(model,"age*age")) !=0){
8608: printf(" strpt=%s, model=%s\n",strpt, model);
8609: if(strpt != model){
1.234 brouard 8610: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8611: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8612: corresponding column of parameters.\n",model);
1.234 brouard 8613: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8614: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8615: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8616: return 1;
1.225 brouard 8617: }
1.187 brouard 8618: nagesqr=1;
8619: if (strstr(model,"+age*age") !=0)
1.234 brouard 8620: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8621: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8622: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8623: else
1.234 brouard 8624: substrchaine(modelsav, model, "age*age");
1.187 brouard 8625: }else
8626: nagesqr=0;
8627: if (strlen(modelsav) >1){
8628: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8629: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8630: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8631: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8632: * cst, age and age*age
8633: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8634: /* including age products which are counted in cptcovage.
8635: * but the covariates which are products must be treated
8636: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8637: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8638: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8639:
8640:
1.187 brouard 8641: /* Design
8642: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8643: * < ncovcol=8 >
8644: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8645: * k= 1 2 3 4 5 6 7 8
8646: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8647: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8648: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8649: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8650: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8651: * Tage[++cptcovage]=k
8652: * if products, new covar are created after ncovcol with k1
8653: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8654: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8655: * 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
8656: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8657: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8658: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8659: * < ncovcol=8 >
8660: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8661: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8662: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8663: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8664: * p Tprod[1]@2={ 6, 5}
8665: *p Tvard[1][1]@4= {7, 8, 5, 6}
8666: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8667: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8668: *How to reorganize?
8669: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8670: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8671: * {2, 1, 4, 8, 5, 6, 3, 7}
8672: * Struct []
8673: */
1.225 brouard 8674:
1.187 brouard 8675: /* This loop fills the array Tvar from the string 'model'.*/
8676: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8677: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8678: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8679: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8680: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8681: /* k=1 Tvar[1]=2 (from V2) */
8682: /* k=5 Tvar[5] */
8683: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8684: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8685: /* } */
1.198 brouard 8686: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8687: /*
8688: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8689: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8690: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8691: }
1.187 brouard 8692: cptcovage=0;
8693: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8694: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8695: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8696: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8697: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8698: /*scanf("%d",i);*/
8699: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8700: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8701: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8702: /* covar is not filled and then is empty */
8703: cptcovprod--;
8704: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8705: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8706: Typevar[k]=1; /* 1 for age product */
8707: cptcovage++; /* Sums the number of covariates which include age as a product */
8708: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8709: /*printf("stre=%s ", stre);*/
8710: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8711: cptcovprod--;
8712: cutl(stre,strb,strc,'V');
8713: Tvar[k]=atoi(stre);
8714: Typevar[k]=1; /* 1 for age product */
8715: cptcovage++;
8716: Tage[cptcovage]=k;
8717: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8718: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8719: cptcovn++;
8720: cptcovprodnoage++;k1++;
8721: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8722: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8723: because this model-covariate is a construction we invent a new column
8724: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8725: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8726: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8727: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8728: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8729: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8730: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8731: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8732: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8733: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8734: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8735: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8736: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8737: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8738: for (i=1; i<=lastobs;i++){
8739: /* Computes the new covariate which is a product of
8740: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8741: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8742: }
8743: } /* End age is not in the model */
8744: } /* End if model includes a product */
8745: else { /* no more sum */
8746: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8747: /* scanf("%d",i);*/
8748: cutl(strd,strc,strb,'V');
8749: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8750: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8751: Tvar[k]=atoi(strd);
8752: Typevar[k]=0; /* 0 for simple covariates */
8753: }
8754: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8755: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8756: scanf("%d",i);*/
1.187 brouard 8757: } /* end of loop + on total covariates */
8758: } /* end if strlen(modelsave == 0) age*age might exist */
8759: } /* end if strlen(model == 0) */
1.136 brouard 8760:
8761: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8762: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8763:
1.136 brouard 8764: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8765: printf("cptcovprod=%d ", cptcovprod);
8766: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8767: scanf("%d ",i);*/
8768:
8769:
1.230 brouard 8770: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8771: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8772: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8773: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8774: k = 1 2 3 4 5 6 7 8 9
8775: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8776: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8777: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8778: Dummy[k] 1 0 0 0 3 1 1 2 3
8779: Tmodelind[combination of covar]=k;
1.225 brouard 8780: */
8781: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8782: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8783: /* 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 8784: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8785: printf("Model=%s\n\
8786: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8787: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8788: 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);
8789: fprintf(ficlog,"Model=%s\n\
8790: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8791: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8792: 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 8793: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8794: 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 */
8795: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8796: Fixed[k]= 0;
8797: Dummy[k]= 0;
1.225 brouard 8798: ncoveff++;
1.232 brouard 8799: ncovf++;
1.234 brouard 8800: nsd++;
8801: modell[k].maintype= FTYPE;
8802: TvarsD[nsd]=Tvar[k];
8803: TvarsDind[nsd]=k;
8804: TvarF[ncovf]=Tvar[k];
8805: TvarFind[ncovf]=k;
8806: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8807: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8808: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8809: Fixed[k]= 0;
8810: Dummy[k]= 0;
8811: ncoveff++;
8812: ncovf++;
8813: modell[k].maintype= FTYPE;
8814: TvarF[ncovf]=Tvar[k];
8815: TvarFind[ncovf]=k;
1.230 brouard 8816: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8817: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8818: }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 8819: Fixed[k]= 0;
8820: Dummy[k]= 1;
1.230 brouard 8821: nqfveff++;
1.234 brouard 8822: modell[k].maintype= FTYPE;
8823: modell[k].subtype= FQ;
8824: nsq++;
8825: TvarsQ[nsq]=Tvar[k];
8826: TvarsQind[nsq]=k;
1.232 brouard 8827: ncovf++;
1.234 brouard 8828: TvarF[ncovf]=Tvar[k];
8829: TvarFind[ncovf]=k;
1.231 brouard 8830: 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 8831: 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 8832: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8833: Fixed[k]= 1;
8834: Dummy[k]= 0;
1.225 brouard 8835: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8836: modell[k].maintype= VTYPE;
8837: modell[k].subtype= VD;
8838: nsd++;
8839: TvarsD[nsd]=Tvar[k];
8840: TvarsDind[nsd]=k;
8841: ncovv++; /* Only simple time varying variables */
8842: TvarV[ncovv]=Tvar[k];
1.242 brouard 8843: 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 8844: 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 */
8845: 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 8846: 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);
8847: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8848: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8849: Fixed[k]= 1;
8850: Dummy[k]= 1;
8851: nqtveff++;
8852: modell[k].maintype= VTYPE;
8853: modell[k].subtype= VQ;
8854: ncovv++; /* Only simple time varying variables */
8855: nsq++;
8856: TvarsQ[nsq]=Tvar[k];
8857: TvarsQind[nsq]=k;
8858: TvarV[ncovv]=Tvar[k];
1.242 brouard 8859: 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 8860: 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 */
8861: 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 8862: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8863: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8864: 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 8865: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8866: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8867: ncova++;
8868: TvarA[ncova]=Tvar[k];
8869: TvarAind[ncova]=k;
1.231 brouard 8870: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8871: Fixed[k]= 2;
8872: Dummy[k]= 2;
8873: modell[k].maintype= ATYPE;
8874: modell[k].subtype= APFD;
8875: /* ncoveff++; */
1.227 brouard 8876: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8877: Fixed[k]= 2;
8878: Dummy[k]= 3;
8879: modell[k].maintype= ATYPE;
8880: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8881: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8882: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8883: Fixed[k]= 3;
8884: Dummy[k]= 2;
8885: modell[k].maintype= ATYPE;
8886: modell[k].subtype= APVD; /* Product age * varying dummy */
8887: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8888: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8889: Fixed[k]= 3;
8890: Dummy[k]= 3;
8891: modell[k].maintype= ATYPE;
8892: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8893: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8894: }
8895: }else if (Typevar[k] == 2) { /* product without age */
8896: k1=Tposprod[k];
8897: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8898: if(Tvard[k1][2] <=ncovcol){
8899: Fixed[k]= 1;
8900: Dummy[k]= 0;
8901: modell[k].maintype= FTYPE;
8902: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8903: ncovf++; /* Fixed variables without age */
8904: TvarF[ncovf]=Tvar[k];
8905: TvarFind[ncovf]=k;
8906: }else if(Tvard[k1][2] <=ncovcol+nqv){
8907: Fixed[k]= 0; /* or 2 ?*/
8908: Dummy[k]= 1;
8909: modell[k].maintype= FTYPE;
8910: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8911: ncovf++; /* Varying variables without age */
8912: TvarF[ncovf]=Tvar[k];
8913: TvarFind[ncovf]=k;
8914: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8915: Fixed[k]= 1;
8916: Dummy[k]= 0;
8917: modell[k].maintype= VTYPE;
8918: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8919: ncovv++; /* Varying variables without age */
8920: TvarV[ncovv]=Tvar[k];
8921: TvarVind[ncovv]=k;
8922: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8923: Fixed[k]= 1;
8924: Dummy[k]= 1;
8925: modell[k].maintype= VTYPE;
8926: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8927: ncovv++; /* Varying variables without age */
8928: TvarV[ncovv]=Tvar[k];
8929: TvarVind[ncovv]=k;
8930: }
1.227 brouard 8931: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8932: if(Tvard[k1][2] <=ncovcol){
8933: Fixed[k]= 0; /* or 2 ?*/
8934: Dummy[k]= 1;
8935: modell[k].maintype= FTYPE;
8936: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8937: ncovf++; /* Fixed variables without age */
8938: TvarF[ncovf]=Tvar[k];
8939: TvarFind[ncovf]=k;
8940: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8941: Fixed[k]= 1;
8942: Dummy[k]= 1;
8943: modell[k].maintype= VTYPE;
8944: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8945: ncovv++; /* Varying variables without age */
8946: TvarV[ncovv]=Tvar[k];
8947: TvarVind[ncovv]=k;
8948: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8949: Fixed[k]= 1;
8950: Dummy[k]= 1;
8951: modell[k].maintype= VTYPE;
8952: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8953: ncovv++; /* Varying variables without age */
8954: TvarV[ncovv]=Tvar[k];
8955: TvarVind[ncovv]=k;
8956: ncovv++; /* Varying variables without age */
8957: TvarV[ncovv]=Tvar[k];
8958: TvarVind[ncovv]=k;
8959: }
1.227 brouard 8960: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8961: if(Tvard[k1][2] <=ncovcol){
8962: Fixed[k]= 1;
8963: Dummy[k]= 1;
8964: modell[k].maintype= VTYPE;
8965: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8966: ncovv++; /* Varying variables without age */
8967: TvarV[ncovv]=Tvar[k];
8968: TvarVind[ncovv]=k;
8969: }else if(Tvard[k1][2] <=ncovcol+nqv){
8970: Fixed[k]= 1;
8971: Dummy[k]= 1;
8972: modell[k].maintype= VTYPE;
8973: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8974: ncovv++; /* Varying variables without age */
8975: TvarV[ncovv]=Tvar[k];
8976: TvarVind[ncovv]=k;
8977: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8978: Fixed[k]= 1;
8979: Dummy[k]= 0;
8980: modell[k].maintype= VTYPE;
8981: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8982: ncovv++; /* Varying variables without age */
8983: TvarV[ncovv]=Tvar[k];
8984: TvarVind[ncovv]=k;
8985: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8986: Fixed[k]= 1;
8987: Dummy[k]= 1;
8988: modell[k].maintype= VTYPE;
8989: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8990: ncovv++; /* Varying variables without age */
8991: TvarV[ncovv]=Tvar[k];
8992: TvarVind[ncovv]=k;
8993: }
1.227 brouard 8994: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8995: if(Tvard[k1][2] <=ncovcol){
8996: Fixed[k]= 1;
8997: Dummy[k]= 1;
8998: modell[k].maintype= VTYPE;
8999: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9000: ncovv++; /* Varying variables without age */
9001: TvarV[ncovv]=Tvar[k];
9002: TvarVind[ncovv]=k;
9003: }else if(Tvard[k1][2] <=ncovcol+nqv){
9004: Fixed[k]= 1;
9005: Dummy[k]= 1;
9006: modell[k].maintype= VTYPE;
9007: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9008: ncovv++; /* Varying variables without age */
9009: TvarV[ncovv]=Tvar[k];
9010: TvarVind[ncovv]=k;
9011: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9012: Fixed[k]= 1;
9013: Dummy[k]= 1;
9014: modell[k].maintype= VTYPE;
9015: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9016: ncovv++; /* Varying variables without age */
9017: TvarV[ncovv]=Tvar[k];
9018: TvarVind[ncovv]=k;
9019: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9020: Fixed[k]= 1;
9021: Dummy[k]= 1;
9022: modell[k].maintype= VTYPE;
9023: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9024: ncovv++; /* Varying variables without age */
9025: TvarV[ncovv]=Tvar[k];
9026: TvarVind[ncovv]=k;
9027: }
1.227 brouard 9028: }else{
1.240 brouard 9029: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9030: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9031: } /*end k1*/
1.225 brouard 9032: }else{
1.226 brouard 9033: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9034: 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 9035: }
1.227 brouard 9036: 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 9037: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9038: 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]);
9039: }
9040: /* Searching for doublons in the model */
9041: for(k1=1; k1<= cptcovt;k1++){
9042: for(k2=1; k2 <k1;k2++){
9043: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9044: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9045: if(Tvar[k1]==Tvar[k2]){
9046: 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]]);
9047: 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);
9048: return(1);
9049: }
9050: }else if (Typevar[k1] ==2){
9051: k3=Tposprod[k1];
9052: k4=Tposprod[k2];
9053: 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])) ){
9054: 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]]);
9055: 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);
9056: return(1);
9057: }
9058: }
1.227 brouard 9059: }
9060: }
1.225 brouard 9061: }
9062: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9063: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9064: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9065: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9066: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9067: /*endread:*/
1.225 brouard 9068: printf("Exiting decodemodel: ");
9069: return (1);
1.136 brouard 9070: }
9071:
1.169 brouard 9072: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9073: {/* Check ages at death */
1.136 brouard 9074: int i, m;
1.218 brouard 9075: int firstone=0;
9076:
1.136 brouard 9077: for (i=1; i<=imx; i++) {
9078: for(m=2; (m<= maxwav); m++) {
9079: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9080: anint[m][i]=9999;
1.216 brouard 9081: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9082: s[m][i]=-1;
1.136 brouard 9083: }
9084: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 9085: *nberr = *nberr + 1;
1.218 brouard 9086: if(firstone == 0){
9087: firstone=1;
9088: 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);
9089: }
9090: 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 9091: s[m][i]=-1;
9092: }
9093: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9094: (*nberr)++;
1.136 brouard 9095: 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]);
9096: 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]);
9097: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
9098: }
9099: }
9100: }
9101:
9102: for (i=1; i<=imx; i++) {
9103: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9104: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9105: 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 9106: if (s[m][i] >= nlstate+1) {
1.169 brouard 9107: if(agedc[i]>0){
9108: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9109: agev[m][i]=agedc[i];
1.214 brouard 9110: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9111: }else {
1.136 brouard 9112: if ((int)andc[i]!=9999){
9113: nbwarn++;
9114: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9115: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9116: agev[m][i]=-1;
9117: }
9118: }
1.169 brouard 9119: } /* agedc > 0 */
1.214 brouard 9120: } /* end if */
1.136 brouard 9121: else if(s[m][i] !=9){ /* Standard case, age in fractional
9122: years but with the precision of a month */
9123: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9124: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9125: agev[m][i]=1;
9126: else if(agev[m][i] < *agemin){
9127: *agemin=agev[m][i];
9128: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9129: }
9130: else if(agev[m][i] >*agemax){
9131: *agemax=agev[m][i];
1.156 brouard 9132: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9133: }
9134: /*agev[m][i]=anint[m][i]-annais[i];*/
9135: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9136: } /* en if 9*/
1.136 brouard 9137: else { /* =9 */
1.214 brouard 9138: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9139: agev[m][i]=1;
9140: s[m][i]=-1;
9141: }
9142: }
1.214 brouard 9143: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9144: agev[m][i]=1;
1.214 brouard 9145: else{
9146: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9147: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9148: agev[m][i]=0;
9149: }
9150: } /* End for lastpass */
9151: }
1.136 brouard 9152:
9153: for (i=1; i<=imx; i++) {
9154: for(m=firstpass; (m<=lastpass); m++){
9155: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9156: (*nberr)++;
1.136 brouard 9157: 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);
9158: 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);
9159: return 1;
9160: }
9161: }
9162: }
9163:
9164: /*for (i=1; i<=imx; i++){
9165: for (m=firstpass; (m<lastpass); m++){
9166: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9167: }
9168:
9169: }*/
9170:
9171:
1.139 brouard 9172: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9173: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9174:
9175: return (0);
1.164 brouard 9176: /* endread:*/
1.136 brouard 9177: printf("Exiting calandcheckages: ");
9178: return (1);
9179: }
9180:
1.172 brouard 9181: #if defined(_MSC_VER)
9182: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9183: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9184: //#include "stdafx.h"
9185: //#include <stdio.h>
9186: //#include <tchar.h>
9187: //#include <windows.h>
9188: //#include <iostream>
9189: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9190:
9191: LPFN_ISWOW64PROCESS fnIsWow64Process;
9192:
9193: BOOL IsWow64()
9194: {
9195: BOOL bIsWow64 = FALSE;
9196:
9197: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9198: // (HANDLE, PBOOL);
9199:
9200: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9201:
9202: HMODULE module = GetModuleHandle(_T("kernel32"));
9203: const char funcName[] = "IsWow64Process";
9204: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9205: GetProcAddress(module, funcName);
9206:
9207: if (NULL != fnIsWow64Process)
9208: {
9209: if (!fnIsWow64Process(GetCurrentProcess(),
9210: &bIsWow64))
9211: //throw std::exception("Unknown error");
9212: printf("Unknown error\n");
9213: }
9214: return bIsWow64 != FALSE;
9215: }
9216: #endif
1.177 brouard 9217:
1.191 brouard 9218: void syscompilerinfo(int logged)
1.167 brouard 9219: {
9220: /* #include "syscompilerinfo.h"*/
1.185 brouard 9221: /* command line Intel compiler 32bit windows, XP compatible:*/
9222: /* /GS /W3 /Gy
9223: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9224: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9225: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9226: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9227: */
9228: /* 64 bits */
1.185 brouard 9229: /*
9230: /GS /W3 /Gy
9231: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9232: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9233: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9234: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9235: /* Optimization are useless and O3 is slower than O2 */
9236: /*
9237: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9238: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9239: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9240: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9241: */
1.186 brouard 9242: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9243: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9244: /PDB:"visual studio
9245: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9246: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9247: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9248: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9249: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9250: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9251: uiAccess='false'"
9252: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9253: /NOLOGO /TLBID:1
9254: */
1.177 brouard 9255: #if defined __INTEL_COMPILER
1.178 brouard 9256: #if defined(__GNUC__)
9257: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9258: #endif
1.177 brouard 9259: #elif defined(__GNUC__)
1.179 brouard 9260: #ifndef __APPLE__
1.174 brouard 9261: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9262: #endif
1.177 brouard 9263: struct utsname sysInfo;
1.178 brouard 9264: int cross = CROSS;
9265: if (cross){
9266: printf("Cross-");
1.191 brouard 9267: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9268: }
1.174 brouard 9269: #endif
9270:
1.171 brouard 9271: #include <stdint.h>
1.178 brouard 9272:
1.191 brouard 9273: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9274: #if defined(__clang__)
1.191 brouard 9275: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9276: #endif
9277: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9278: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9279: #endif
9280: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9281: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9282: #endif
9283: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9284: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9285: #endif
9286: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9287: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9288: #endif
9289: #if defined(_MSC_VER)
1.191 brouard 9290: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9291: #endif
9292: #if defined(__PGI)
1.191 brouard 9293: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9294: #endif
9295: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9296: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9297: #endif
1.191 brouard 9298: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9299:
1.167 brouard 9300: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9301: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9302: // Windows (x64 and x86)
1.191 brouard 9303: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9304: #elif __unix__ // all unices, not all compilers
9305: // Unix
1.191 brouard 9306: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9307: #elif __linux__
9308: // linux
1.191 brouard 9309: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9310: #elif __APPLE__
1.174 brouard 9311: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9312: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9313: #endif
9314:
9315: /* __MINGW32__ */
9316: /* __CYGWIN__ */
9317: /* __MINGW64__ */
9318: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9319: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9320: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9321: /* _WIN64 // Defined for applications for Win64. */
9322: /* _M_X64 // Defined for compilations that target x64 processors. */
9323: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9324:
1.167 brouard 9325: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9326: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9327: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9328: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9329: #else
1.191 brouard 9330: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9331: #endif
9332:
1.169 brouard 9333: #if defined(__GNUC__)
9334: # if defined(__GNUC_PATCHLEVEL__)
9335: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9336: + __GNUC_MINOR__ * 100 \
9337: + __GNUC_PATCHLEVEL__)
9338: # else
9339: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9340: + __GNUC_MINOR__ * 100)
9341: # endif
1.174 brouard 9342: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9343: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9344:
9345: if (uname(&sysInfo) != -1) {
9346: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9347: 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 9348: }
9349: else
9350: perror("uname() error");
1.179 brouard 9351: //#ifndef __INTEL_COMPILER
9352: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9353: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9354: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9355: #endif
1.169 brouard 9356: #endif
1.172 brouard 9357:
9358: // void main()
9359: // {
1.169 brouard 9360: #if defined(_MSC_VER)
1.174 brouard 9361: if (IsWow64()){
1.191 brouard 9362: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9363: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9364: }
9365: else{
1.191 brouard 9366: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9367: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9368: }
1.172 brouard 9369: // printf("\nPress Enter to continue...");
9370: // getchar();
9371: // }
9372:
1.169 brouard 9373: #endif
9374:
1.167 brouard 9375:
1.219 brouard 9376: }
1.136 brouard 9377:
1.219 brouard 9378: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9379: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9380: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9381: /* double ftolpl = 1.e-10; */
1.180 brouard 9382: double age, agebase, agelim;
1.203 brouard 9383: double tot;
1.180 brouard 9384:
1.202 brouard 9385: strcpy(filerespl,"PL_");
9386: strcat(filerespl,fileresu);
9387: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9388: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9389: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9390: }
1.227 brouard 9391: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9392: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9393: pstamp(ficrespl);
1.203 brouard 9394: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9395: fprintf(ficrespl,"#Age ");
9396: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9397: fprintf(ficrespl,"\n");
1.180 brouard 9398:
1.219 brouard 9399: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9400:
1.219 brouard 9401: agebase=ageminpar;
9402: agelim=agemaxpar;
1.180 brouard 9403:
1.227 brouard 9404: /* i1=pow(2,ncoveff); */
1.234 brouard 9405: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9406: if (cptcovn < 1){i1=1;}
1.180 brouard 9407:
1.238 brouard 9408: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9409: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9410: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9411: continue;
1.235 brouard 9412:
1.238 brouard 9413: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9414: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9415: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9416: /* k=k+1; */
9417: /* to clean */
9418: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9419: fprintf(ficrespl,"#******");
9420: printf("#******");
9421: fprintf(ficlog,"#******");
9422: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9423: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9424: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9425: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9426: }
9427: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9428: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9429: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9430: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9431: }
9432: fprintf(ficrespl,"******\n");
9433: printf("******\n");
9434: fprintf(ficlog,"******\n");
9435: if(invalidvarcomb[k]){
9436: printf("\nCombination (%d) ignored because no case \n",k);
9437: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9438: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9439: continue;
9440: }
1.219 brouard 9441:
1.238 brouard 9442: fprintf(ficrespl,"#Age ");
9443: for(j=1;j<=cptcoveff;j++) {
9444: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9445: }
9446: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9447: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9448:
1.238 brouard 9449: for (age=agebase; age<=agelim; age++){
9450: /* for (age=agebase; age<=agebase; age++){ */
9451: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9452: fprintf(ficrespl,"%.0f ",age );
9453: for(j=1;j<=cptcoveff;j++)
9454: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9455: tot=0.;
9456: for(i=1; i<=nlstate;i++){
9457: tot += prlim[i][i];
9458: fprintf(ficrespl," %.5f", prlim[i][i]);
9459: }
9460: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9461: } /* Age */
9462: /* was end of cptcod */
9463: } /* cptcov */
9464: } /* nres */
1.219 brouard 9465: return 0;
1.180 brouard 9466: }
9467:
1.218 brouard 9468: 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){
9469: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9470:
9471: /* Computes the back prevalence limit for any combination of covariate values
9472: * at any age between ageminpar and agemaxpar
9473: */
1.235 brouard 9474: int i, j, k, i1, nres=0 ;
1.217 brouard 9475: /* double ftolpl = 1.e-10; */
9476: double age, agebase, agelim;
9477: double tot;
1.218 brouard 9478: /* double ***mobaverage; */
9479: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9480:
9481: strcpy(fileresplb,"PLB_");
9482: strcat(fileresplb,fileresu);
9483: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9484: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9485: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9486: }
9487: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9488: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9489: pstamp(ficresplb);
9490: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9491: fprintf(ficresplb,"#Age ");
9492: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9493: fprintf(ficresplb,"\n");
9494:
1.218 brouard 9495:
9496: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9497:
9498: agebase=ageminpar;
9499: agelim=agemaxpar;
9500:
9501:
1.227 brouard 9502: i1=pow(2,cptcoveff);
1.218 brouard 9503: if (cptcovn < 1){i1=1;}
1.227 brouard 9504:
1.238 brouard 9505: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9506: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9507: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9508: continue;
9509: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9510: fprintf(ficresplb,"#******");
9511: printf("#******");
9512: fprintf(ficlog,"#******");
9513: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9514: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9515: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9516: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9517: }
9518: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9519: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9520: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9521: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9522: }
9523: fprintf(ficresplb,"******\n");
9524: printf("******\n");
9525: fprintf(ficlog,"******\n");
9526: if(invalidvarcomb[k]){
9527: printf("\nCombination (%d) ignored because no cases \n",k);
9528: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9529: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9530: continue;
9531: }
1.218 brouard 9532:
1.238 brouard 9533: fprintf(ficresplb,"#Age ");
9534: for(j=1;j<=cptcoveff;j++) {
9535: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9536: }
9537: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9538: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9539:
9540:
1.238 brouard 9541: for (age=agebase; age<=agelim; age++){
9542: /* for (age=agebase; age<=agebase; age++){ */
9543: if(mobilavproj > 0){
9544: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9545: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9546: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9547: }else if (mobilavproj == 0){
9548: 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);
9549: 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);
9550: exit(1);
9551: }else{
9552: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9553: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9554: }
9555: fprintf(ficresplb,"%.0f ",age );
9556: for(j=1;j<=cptcoveff;j++)
9557: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9558: tot=0.;
9559: for(i=1; i<=nlstate;i++){
9560: tot += bprlim[i][i];
9561: fprintf(ficresplb," %.5f", bprlim[i][i]);
9562: }
9563: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9564: } /* Age */
9565: /* was end of cptcod */
9566: } /* end of any combination */
9567: } /* end of nres */
1.218 brouard 9568: /* hBijx(p, bage, fage); */
9569: /* fclose(ficrespijb); */
9570:
9571: return 0;
1.217 brouard 9572: }
1.218 brouard 9573:
1.180 brouard 9574: int hPijx(double *p, int bage, int fage){
9575: /*------------- h Pij x at various ages ------------*/
9576:
9577: int stepsize;
9578: int agelim;
9579: int hstepm;
9580: int nhstepm;
1.235 brouard 9581: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9582:
9583: double agedeb;
9584: double ***p3mat;
9585:
1.201 brouard 9586: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9587: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9588: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9589: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9590: }
9591: printf("Computing pij: result on file '%s' \n", filerespij);
9592: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9593:
9594: stepsize=(int) (stepm+YEARM-1)/YEARM;
9595: /*if (stepm<=24) stepsize=2;*/
9596:
9597: agelim=AGESUP;
9598: hstepm=stepsize*YEARM; /* Every year of age */
9599: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9600:
1.180 brouard 9601: /* hstepm=1; aff par mois*/
9602: pstamp(ficrespij);
9603: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9604: i1= pow(2,cptcoveff);
1.218 brouard 9605: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9606: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9607: /* k=k+1; */
1.235 brouard 9608: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9609: for(k=1; k<=i1;k++){
1.253 brouard 9610: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9611: continue;
1.183 brouard 9612: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9613: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9614: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9615: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9616: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9617: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9618: }
1.183 brouard 9619: fprintf(ficrespij,"******\n");
9620:
9621: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9622: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9623: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9624:
9625: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9626:
1.183 brouard 9627: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9628: oldm=oldms;savm=savms;
1.235 brouard 9629: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9630: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9631: for(i=1; i<=nlstate;i++)
9632: for(j=1; j<=nlstate+ndeath;j++)
9633: fprintf(ficrespij," %1d-%1d",i,j);
9634: fprintf(ficrespij,"\n");
9635: for (h=0; h<=nhstepm; h++){
9636: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9637: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9638: for(i=1; i<=nlstate;i++)
9639: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9640: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9641: fprintf(ficrespij,"\n");
9642: }
1.183 brouard 9643: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9644: fprintf(ficrespij,"\n");
9645: }
1.180 brouard 9646: /*}*/
9647: }
1.218 brouard 9648: return 0;
1.180 brouard 9649: }
1.218 brouard 9650:
9651: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9652: /*------------- h Bij x at various ages ------------*/
9653:
9654: int stepsize;
1.218 brouard 9655: /* int agelim; */
9656: int ageminl;
1.217 brouard 9657: int hstepm;
9658: int nhstepm;
1.238 brouard 9659: int h, i, i1, j, k, nres;
1.218 brouard 9660:
1.217 brouard 9661: double agedeb;
9662: double ***p3mat;
1.218 brouard 9663:
9664: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9665: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9666: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9667: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9668: }
9669: printf("Computing pij back: result on file '%s' \n", filerespijb);
9670: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9671:
9672: stepsize=(int) (stepm+YEARM-1)/YEARM;
9673: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9674:
1.218 brouard 9675: /* agelim=AGESUP; */
9676: ageminl=30;
9677: hstepm=stepsize*YEARM; /* Every year of age */
9678: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9679:
9680: /* hstepm=1; aff par mois*/
9681: pstamp(ficrespijb);
9682: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9683: i1= pow(2,cptcoveff);
1.218 brouard 9684: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9685: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9686: /* k=k+1; */
1.238 brouard 9687: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9688: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9689: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9690: continue;
9691: fprintf(ficrespijb,"\n#****** ");
9692: for(j=1;j<=cptcoveff;j++)
9693: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9694: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9695: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9696: }
9697: fprintf(ficrespijb,"******\n");
9698: if(invalidvarcomb[k]){
9699: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9700: continue;
9701: }
9702:
9703: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9704: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9705: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9706: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9707: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9708:
9709: /* nhstepm=nhstepm*YEARM; aff par mois*/
9710:
9711: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9712: /* oldm=oldms;savm=savms; */
9713: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9714: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9715: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9716: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9717: for(i=1; i<=nlstate;i++)
9718: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9719: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9720: fprintf(ficrespijb,"\n");
1.238 brouard 9721: for (h=0; h<=nhstepm; h++){
9722: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9723: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9724: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9725: for(i=1; i<=nlstate;i++)
9726: for(j=1; j<=nlstate+ndeath;j++)
9727: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9728: fprintf(ficrespijb,"\n");
9729: }
9730: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9731: fprintf(ficrespijb,"\n");
9732: } /* end age deb */
9733: } /* end combination */
9734: } /* end nres */
1.218 brouard 9735: return 0;
9736: } /* hBijx */
1.217 brouard 9737:
1.180 brouard 9738:
1.136 brouard 9739: /***********************************************/
9740: /**************** Main Program *****************/
9741: /***********************************************/
9742:
9743: int main(int argc, char *argv[])
9744: {
9745: #ifdef GSL
9746: const gsl_multimin_fminimizer_type *T;
9747: size_t iteri = 0, it;
9748: int rval = GSL_CONTINUE;
9749: int status = GSL_SUCCESS;
9750: double ssval;
9751: #endif
9752: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9753: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9754: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9755: int jj, ll, li, lj, lk;
1.136 brouard 9756: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9757: int num_filled;
1.136 brouard 9758: int itimes;
9759: int NDIM=2;
9760: int vpopbased=0;
1.235 brouard 9761: int nres=0;
1.136 brouard 9762:
1.164 brouard 9763: char ca[32], cb[32];
1.136 brouard 9764: /* FILE *fichtm; *//* Html File */
9765: /* FILE *ficgp;*/ /*Gnuplot File */
9766: struct stat info;
1.191 brouard 9767: double agedeb=0.;
1.194 brouard 9768:
9769: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9770: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9771:
1.165 brouard 9772: double fret;
1.191 brouard 9773: double dum=0.; /* Dummy variable */
1.136 brouard 9774: double ***p3mat;
1.218 brouard 9775: /* double ***mobaverage; */
1.164 brouard 9776:
9777: char line[MAXLINE];
1.197 brouard 9778: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9779:
1.234 brouard 9780: char modeltemp[MAXLINE];
1.230 brouard 9781: char resultline[MAXLINE];
9782:
1.136 brouard 9783: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9784: char *tok, *val; /* pathtot */
1.136 brouard 9785: int firstobs=1, lastobs=10;
1.195 brouard 9786: int c, h , cpt, c2;
1.191 brouard 9787: int jl=0;
9788: int i1, j1, jk, stepsize=0;
1.194 brouard 9789: int count=0;
9790:
1.164 brouard 9791: int *tab;
1.136 brouard 9792: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9793: int backcast=0;
1.136 brouard 9794: int mobilav=0,popforecast=0;
1.191 brouard 9795: int hstepm=0, nhstepm=0;
1.136 brouard 9796: int agemortsup;
9797: float sumlpop=0.;
9798: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9799: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9800:
1.191 brouard 9801: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9802: double ftolpl=FTOL;
9803: double **prlim;
1.217 brouard 9804: double **bprlim;
1.136 brouard 9805: double ***param; /* Matrix of parameters */
1.251 brouard 9806: double ***paramstart; /* Matrix of starting parameter values */
9807: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9808: double **matcov; /* Matrix of covariance */
1.203 brouard 9809: double **hess; /* Hessian matrix */
1.136 brouard 9810: double ***delti3; /* Scale */
9811: double *delti; /* Scale */
9812: double ***eij, ***vareij;
9813: double **varpl; /* Variances of prevalence limits by age */
9814: double *epj, vepp;
1.164 brouard 9815:
1.136 brouard 9816: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9817: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9818:
1.136 brouard 9819: double **ximort;
1.145 brouard 9820: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9821: int *dcwave;
9822:
1.164 brouard 9823: char z[1]="c";
1.136 brouard 9824:
9825: /*char *strt;*/
9826: char strtend[80];
1.126 brouard 9827:
1.164 brouard 9828:
1.126 brouard 9829: /* setlocale (LC_ALL, ""); */
9830: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9831: /* textdomain (PACKAGE); */
9832: /* setlocale (LC_CTYPE, ""); */
9833: /* setlocale (LC_MESSAGES, ""); */
9834:
9835: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9836: rstart_time = time(NULL);
9837: /* (void) gettimeofday(&start_time,&tzp);*/
9838: start_time = *localtime(&rstart_time);
1.126 brouard 9839: curr_time=start_time;
1.157 brouard 9840: /*tml = *localtime(&start_time.tm_sec);*/
9841: /* strcpy(strstart,asctime(&tml)); */
9842: strcpy(strstart,asctime(&start_time));
1.126 brouard 9843:
9844: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9845: /* tp.tm_sec = tp.tm_sec +86400; */
9846: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9847: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9848: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9849: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9850: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9851: /* strt=asctime(&tmg); */
9852: /* printf("Time(after) =%s",strstart); */
9853: /* (void) time (&time_value);
9854: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9855: * tm = *localtime(&time_value);
9856: * strstart=asctime(&tm);
9857: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9858: */
9859:
9860: nberr=0; /* Number of errors and warnings */
9861: nbwarn=0;
1.184 brouard 9862: #ifdef WIN32
9863: _getcwd(pathcd, size);
9864: #else
1.126 brouard 9865: getcwd(pathcd, size);
1.184 brouard 9866: #endif
1.191 brouard 9867: syscompilerinfo(0);
1.196 brouard 9868: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9869: if(argc <=1){
9870: printf("\nEnter the parameter file name: ");
1.205 brouard 9871: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9872: printf("ERROR Empty parameter file name\n");
9873: goto end;
9874: }
1.126 brouard 9875: i=strlen(pathr);
9876: if(pathr[i-1]=='\n')
9877: pathr[i-1]='\0';
1.156 brouard 9878: i=strlen(pathr);
1.205 brouard 9879: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9880: pathr[i-1]='\0';
1.205 brouard 9881: }
9882: i=strlen(pathr);
9883: if( i==0 ){
9884: printf("ERROR Empty parameter file name\n");
9885: goto end;
9886: }
9887: for (tok = pathr; tok != NULL; ){
1.126 brouard 9888: printf("Pathr |%s|\n",pathr);
9889: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9890: printf("val= |%s| pathr=%s\n",val,pathr);
9891: strcpy (pathtot, val);
9892: if(pathr[0] == '\0') break; /* Dirty */
9893: }
9894: }
9895: else{
9896: strcpy(pathtot,argv[1]);
9897: }
9898: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9899: /*cygwin_split_path(pathtot,path,optionfile);
9900: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9901: /* cutv(path,optionfile,pathtot,'\\');*/
9902:
9903: /* Split argv[0], imach program to get pathimach */
9904: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9905: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9906: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9907: /* strcpy(pathimach,argv[0]); */
9908: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9909: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9910: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9911: #ifdef WIN32
9912: _chdir(path); /* Can be a relative path */
9913: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9914: #else
1.126 brouard 9915: chdir(path); /* Can be a relative path */
1.184 brouard 9916: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9917: #endif
9918: printf("Current directory %s!\n",pathcd);
1.126 brouard 9919: strcpy(command,"mkdir ");
9920: strcat(command,optionfilefiname);
9921: if((outcmd=system(command)) != 0){
1.169 brouard 9922: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9923: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9924: /* fclose(ficlog); */
9925: /* exit(1); */
9926: }
9927: /* if((imk=mkdir(optionfilefiname))<0){ */
9928: /* perror("mkdir"); */
9929: /* } */
9930:
9931: /*-------- arguments in the command line --------*/
9932:
1.186 brouard 9933: /* Main Log file */
1.126 brouard 9934: strcat(filelog, optionfilefiname);
9935: strcat(filelog,".log"); /* */
9936: if((ficlog=fopen(filelog,"w"))==NULL) {
9937: printf("Problem with logfile %s\n",filelog);
9938: goto end;
9939: }
9940: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9941: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9942: fprintf(ficlog,"\nEnter the parameter file name: \n");
9943: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9944: path=%s \n\
9945: optionfile=%s\n\
9946: optionfilext=%s\n\
1.156 brouard 9947: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9948:
1.197 brouard 9949: syscompilerinfo(1);
1.167 brouard 9950:
1.126 brouard 9951: printf("Local time (at start):%s",strstart);
9952: fprintf(ficlog,"Local time (at start): %s",strstart);
9953: fflush(ficlog);
9954: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9955: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9956:
9957: /* */
9958: strcpy(fileres,"r");
9959: strcat(fileres, optionfilefiname);
1.201 brouard 9960: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9961: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9962: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9963:
1.186 brouard 9964: /* Main ---------arguments file --------*/
1.126 brouard 9965:
9966: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9967: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9968: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9969: fflush(ficlog);
1.149 brouard 9970: /* goto end; */
9971: exit(70);
1.126 brouard 9972: }
9973:
9974:
9975:
9976: strcpy(filereso,"o");
1.201 brouard 9977: strcat(filereso,fileresu);
1.126 brouard 9978: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9979: printf("Problem with Output resultfile: %s\n", filereso);
9980: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9981: fflush(ficlog);
9982: goto end;
9983: }
9984:
9985: /* Reads comments: lines beginning with '#' */
9986: numlinepar=0;
1.197 brouard 9987:
9988: /* First parameter line */
9989: while(fgets(line, MAXLINE, ficpar)) {
9990: /* If line starts with a # it is a comment */
9991: if (line[0] == '#') {
9992: numlinepar++;
9993: fputs(line,stdout);
9994: fputs(line,ficparo);
9995: fputs(line,ficlog);
9996: continue;
9997: }else
9998: break;
9999: }
10000: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10001: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10002: if (num_filled != 5) {
10003: printf("Should be 5 parameters\n");
10004: }
1.126 brouard 10005: numlinepar++;
1.197 brouard 10006: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10007: }
10008: /* Second parameter line */
10009: while(fgets(line, MAXLINE, ficpar)) {
10010: /* If line starts with a # it is a comment */
10011: if (line[0] == '#') {
10012: numlinepar++;
10013: fputs(line,stdout);
10014: fputs(line,ficparo);
10015: fputs(line,ficlog);
10016: continue;
10017: }else
10018: break;
10019: }
1.223 brouard 10020: 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", \
10021: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10022: if (num_filled != 11) {
10023: 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 10024: printf("but line=%s\n",line);
1.197 brouard 10025: }
1.223 brouard 10026: 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 10027: }
1.203 brouard 10028: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10029: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10030: /* Third parameter line */
10031: while(fgets(line, MAXLINE, ficpar)) {
10032: /* If line starts with a # it is a comment */
10033: if (line[0] == '#') {
10034: numlinepar++;
10035: fputs(line,stdout);
10036: fputs(line,ficparo);
10037: fputs(line,ficlog);
10038: continue;
10039: }else
10040: break;
10041: }
1.201 brouard 10042: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10043: if (num_filled == 0)
10044: model[0]='\0';
10045: else if (num_filled != 1){
1.197 brouard 10046: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10047: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10048: model[0]='\0';
10049: goto end;
10050: }
10051: else{
10052: if (model[0]=='+'){
10053: for(i=1; i<=strlen(model);i++)
10054: modeltemp[i-1]=model[i];
1.201 brouard 10055: strcpy(model,modeltemp);
1.197 brouard 10056: }
10057: }
1.199 brouard 10058: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10059: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10060: }
10061: /* 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); */
10062: /* numlinepar=numlinepar+3; /\* In general *\/ */
10063: /* 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 10064: 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);
10065: 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 10066: fflush(ficlog);
1.190 brouard 10067: /* if(model[0]=='#'|| model[0]== '\0'){ */
10068: if(model[0]=='#'){
1.187 brouard 10069: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10070: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10071: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10072: if(mle != -1){
10073: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10074: exit(1);
10075: }
10076: }
1.126 brouard 10077: while((c=getc(ficpar))=='#' && c!= EOF){
10078: ungetc(c,ficpar);
10079: fgets(line, MAXLINE, ficpar);
10080: numlinepar++;
1.195 brouard 10081: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10082: z[0]=line[1];
10083: }
10084: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10085: fputs(line, stdout);
10086: //puts(line);
1.126 brouard 10087: fputs(line,ficparo);
10088: fputs(line,ficlog);
10089: }
10090: ungetc(c,ficpar);
10091:
10092:
1.145 brouard 10093: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10094: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10095: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10096: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10097: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10098: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10099: v1+v2*age+v2*v3 makes cptcovn = 3
10100: */
10101: if (strlen(model)>1)
1.187 brouard 10102: 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 10103: else
1.187 brouard 10104: ncovmodel=2; /* Constant and age */
1.133 brouard 10105: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10106: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10107: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10108: 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);
10109: 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);
10110: fflush(stdout);
10111: fclose (ficlog);
10112: goto end;
10113: }
1.126 brouard 10114: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10115: delti=delti3[1][1];
10116: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10117: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10118: /* We could also provide initial parameters values giving by simple logistic regression
10119: * only one way, that is without matrix product. We will have nlstate maximizations */
10120: /* for(i=1;i<nlstate;i++){ */
10121: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10122: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10123: /* } */
1.126 brouard 10124: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10125: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10126: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10127: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10128: fclose (ficparo);
10129: fclose (ficlog);
10130: goto end;
10131: exit(0);
1.220 brouard 10132: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10133: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10134: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10135: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10136: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10137: matcov=matrix(1,npar,1,npar);
1.203 brouard 10138: hess=matrix(1,npar,1,npar);
1.220 brouard 10139: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10140: /* Read guessed parameters */
1.126 brouard 10141: /* Reads comments: lines beginning with '#' */
10142: while((c=getc(ficpar))=='#' && c!= EOF){
10143: ungetc(c,ficpar);
10144: fgets(line, MAXLINE, ficpar);
10145: numlinepar++;
1.141 brouard 10146: fputs(line,stdout);
1.126 brouard 10147: fputs(line,ficparo);
10148: fputs(line,ficlog);
10149: }
10150: ungetc(c,ficpar);
10151:
10152: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10153: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10154: for(i=1; i <=nlstate; i++){
1.234 brouard 10155: j=0;
1.126 brouard 10156: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10157: if(jj==i) continue;
10158: j++;
10159: fscanf(ficpar,"%1d%1d",&i1,&j1);
10160: if ((i1 != i) || (j1 != jj)){
10161: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10162: It might be a problem of design; if ncovcol and the model are correct\n \
10163: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10164: exit(1);
10165: }
10166: fprintf(ficparo,"%1d%1d",i1,j1);
10167: if(mle==1)
10168: printf("%1d%1d",i,jj);
10169: fprintf(ficlog,"%1d%1d",i,jj);
10170: for(k=1; k<=ncovmodel;k++){
10171: fscanf(ficpar," %lf",¶m[i][j][k]);
10172: if(mle==1){
10173: printf(" %lf",param[i][j][k]);
10174: fprintf(ficlog," %lf",param[i][j][k]);
10175: }
10176: else
10177: fprintf(ficlog," %lf",param[i][j][k]);
10178: fprintf(ficparo," %lf",param[i][j][k]);
10179: }
10180: fscanf(ficpar,"\n");
10181: numlinepar++;
10182: if(mle==1)
10183: printf("\n");
10184: fprintf(ficlog,"\n");
10185: fprintf(ficparo,"\n");
1.126 brouard 10186: }
10187: }
10188: fflush(ficlog);
1.234 brouard 10189:
1.251 brouard 10190: /* Reads parameters values */
1.126 brouard 10191: p=param[1][1];
1.251 brouard 10192: pstart=paramstart[1][1];
1.126 brouard 10193:
10194: /* Reads comments: lines beginning with '#' */
10195: while((c=getc(ficpar))=='#' && c!= EOF){
10196: ungetc(c,ficpar);
10197: fgets(line, MAXLINE, ficpar);
10198: numlinepar++;
1.141 brouard 10199: fputs(line,stdout);
1.126 brouard 10200: fputs(line,ficparo);
10201: fputs(line,ficlog);
10202: }
10203: ungetc(c,ficpar);
10204:
10205: for(i=1; i <=nlstate; i++){
10206: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10207: fscanf(ficpar,"%1d%1d",&i1,&j1);
10208: if ( (i1-i) * (j1-j) != 0){
10209: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10210: exit(1);
10211: }
10212: printf("%1d%1d",i,j);
10213: fprintf(ficparo,"%1d%1d",i1,j1);
10214: fprintf(ficlog,"%1d%1d",i1,j1);
10215: for(k=1; k<=ncovmodel;k++){
10216: fscanf(ficpar,"%le",&delti3[i][j][k]);
10217: printf(" %le",delti3[i][j][k]);
10218: fprintf(ficparo," %le",delti3[i][j][k]);
10219: fprintf(ficlog," %le",delti3[i][j][k]);
10220: }
10221: fscanf(ficpar,"\n");
10222: numlinepar++;
10223: printf("\n");
10224: fprintf(ficparo,"\n");
10225: fprintf(ficlog,"\n");
1.126 brouard 10226: }
10227: }
10228: fflush(ficlog);
1.234 brouard 10229:
1.145 brouard 10230: /* Reads covariance matrix */
1.126 brouard 10231: delti=delti3[1][1];
1.220 brouard 10232:
10233:
1.126 brouard 10234: /* 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 10235:
1.126 brouard 10236: /* Reads comments: lines beginning with '#' */
10237: while((c=getc(ficpar))=='#' && c!= EOF){
10238: ungetc(c,ficpar);
10239: fgets(line, MAXLINE, ficpar);
10240: numlinepar++;
1.141 brouard 10241: fputs(line,stdout);
1.126 brouard 10242: fputs(line,ficparo);
10243: fputs(line,ficlog);
10244: }
10245: ungetc(c,ficpar);
1.220 brouard 10246:
1.126 brouard 10247: matcov=matrix(1,npar,1,npar);
1.203 brouard 10248: hess=matrix(1,npar,1,npar);
1.131 brouard 10249: for(i=1; i <=npar; i++)
10250: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10251:
1.194 brouard 10252: /* Scans npar lines */
1.126 brouard 10253: for(i=1; i <=npar; i++){
1.226 brouard 10254: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10255: if(count != 3){
1.226 brouard 10256: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10257: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10258: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10259: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10260: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10261: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10262: exit(1);
1.220 brouard 10263: }else{
1.226 brouard 10264: if(mle==1)
10265: printf("%1d%1d%d",i1,j1,jk);
10266: }
10267: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10268: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10269: for(j=1; j <=i; j++){
1.226 brouard 10270: fscanf(ficpar," %le",&matcov[i][j]);
10271: if(mle==1){
10272: printf(" %.5le",matcov[i][j]);
10273: }
10274: fprintf(ficlog," %.5le",matcov[i][j]);
10275: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10276: }
10277: fscanf(ficpar,"\n");
10278: numlinepar++;
10279: if(mle==1)
1.220 brouard 10280: printf("\n");
1.126 brouard 10281: fprintf(ficlog,"\n");
10282: fprintf(ficparo,"\n");
10283: }
1.194 brouard 10284: /* End of read covariance matrix npar lines */
1.126 brouard 10285: for(i=1; i <=npar; i++)
10286: for(j=i+1;j<=npar;j++)
1.226 brouard 10287: matcov[i][j]=matcov[j][i];
1.126 brouard 10288:
10289: if(mle==1)
10290: printf("\n");
10291: fprintf(ficlog,"\n");
10292:
10293: fflush(ficlog);
10294:
10295: /*-------- Rewriting parameter file ----------*/
10296: strcpy(rfileres,"r"); /* "Rparameterfile */
10297: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10298: strcat(rfileres,"."); /* */
10299: strcat(rfileres,optionfilext); /* Other files have txt extension */
10300: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10301: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10302: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10303: }
10304: fprintf(ficres,"#%s\n",version);
10305: } /* End of mle != -3 */
1.218 brouard 10306:
1.186 brouard 10307: /* Main data
10308: */
1.126 brouard 10309: n= lastobs;
10310: num=lvector(1,n);
10311: moisnais=vector(1,n);
10312: annais=vector(1,n);
10313: moisdc=vector(1,n);
10314: andc=vector(1,n);
1.220 brouard 10315: weight=vector(1,n);
1.126 brouard 10316: agedc=vector(1,n);
10317: cod=ivector(1,n);
1.220 brouard 10318: for(i=1;i<=n;i++){
1.234 brouard 10319: num[i]=0;
10320: moisnais[i]=0;
10321: annais[i]=0;
10322: moisdc[i]=0;
10323: andc[i]=0;
10324: agedc[i]=0;
10325: cod[i]=0;
10326: weight[i]=1.0; /* Equal weights, 1 by default */
10327: }
1.126 brouard 10328: mint=matrix(1,maxwav,1,n);
10329: anint=matrix(1,maxwav,1,n);
1.131 brouard 10330: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10331: tab=ivector(1,NCOVMAX);
1.144 brouard 10332: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10333: 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 10334:
1.136 brouard 10335: /* Reads data from file datafile */
10336: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10337: goto end;
10338:
10339: /* Calculation of the number of parameters from char model */
1.234 brouard 10340: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10341: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10342: k=3 V4 Tvar[k=3]= 4 (from V4)
10343: k=2 V1 Tvar[k=2]= 1 (from V1)
10344: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10345: */
10346:
10347: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10348: TvarsDind=ivector(1,NCOVMAX); /* */
10349: TvarsD=ivector(1,NCOVMAX); /* */
10350: TvarsQind=ivector(1,NCOVMAX); /* */
10351: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10352: TvarF=ivector(1,NCOVMAX); /* */
10353: TvarFind=ivector(1,NCOVMAX); /* */
10354: TvarV=ivector(1,NCOVMAX); /* */
10355: TvarVind=ivector(1,NCOVMAX); /* */
10356: TvarA=ivector(1,NCOVMAX); /* */
10357: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10358: TvarFD=ivector(1,NCOVMAX); /* */
10359: TvarFDind=ivector(1,NCOVMAX); /* */
10360: TvarFQ=ivector(1,NCOVMAX); /* */
10361: TvarFQind=ivector(1,NCOVMAX); /* */
10362: TvarVD=ivector(1,NCOVMAX); /* */
10363: TvarVDind=ivector(1,NCOVMAX); /* */
10364: TvarVQ=ivector(1,NCOVMAX); /* */
10365: TvarVQind=ivector(1,NCOVMAX); /* */
10366:
1.230 brouard 10367: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10368: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10369: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10370: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10371: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10372: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10373: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10374: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10375: */
10376: /* For model-covariate k tells which data-covariate to use but
10377: because this model-covariate is a construction we invent a new column
10378: ncovcol + k1
10379: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10380: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10381: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10382: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10383: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10384: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10385: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10386: */
1.145 brouard 10387: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10388: 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 10389: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10390: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10391: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10392: 4 covariates (3 plus signs)
10393: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10394: */
1.230 brouard 10395: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10396: * individual dummy, fixed or varying:
10397: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10398: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10399: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10400: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10401: * Tmodelind[1]@9={9,0,3,2,}*/
10402: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10403: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10404: * individual quantitative, fixed or varying:
10405: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10406: * 3, 1, 0, 0, 0, 0, 0, 0},
10407: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10408: /* Main decodemodel */
10409:
1.187 brouard 10410:
1.223 brouard 10411: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10412: goto end;
10413:
1.137 brouard 10414: if((double)(lastobs-imx)/(double)imx > 1.10){
10415: nbwarn++;
10416: 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);
10417: 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);
10418: }
1.136 brouard 10419: /* if(mle==1){*/
1.137 brouard 10420: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10421: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10422: }
10423:
10424: /*-calculation of age at interview from date of interview and age at death -*/
10425: agev=matrix(1,maxwav,1,imx);
10426:
10427: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10428: goto end;
10429:
1.126 brouard 10430:
1.136 brouard 10431: agegomp=(int)agemin;
10432: free_vector(moisnais,1,n);
10433: free_vector(annais,1,n);
1.126 brouard 10434: /* free_matrix(mint,1,maxwav,1,n);
10435: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10436: /* free_vector(moisdc,1,n); */
10437: /* free_vector(andc,1,n); */
1.145 brouard 10438: /* */
10439:
1.126 brouard 10440: wav=ivector(1,imx);
1.214 brouard 10441: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10442: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10443: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10444: 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.*/
10445: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10446: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10447:
10448: /* Concatenates waves */
1.214 brouard 10449: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10450: Death is a valid wave (if date is known).
10451: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10452: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10453: and mw[mi+1][i]. dh depends on stepm.
10454: */
10455:
1.126 brouard 10456: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10457: /* Concatenates waves */
1.145 brouard 10458:
1.215 brouard 10459: free_vector(moisdc,1,n);
10460: free_vector(andc,1,n);
10461:
1.126 brouard 10462: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10463: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10464: ncodemax[1]=1;
1.145 brouard 10465: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10466: cptcoveff=0;
1.220 brouard 10467: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10468: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10469: }
10470:
10471: ncovcombmax=pow(2,cptcoveff);
10472: invalidvarcomb=ivector(1, ncovcombmax);
10473: for(i=1;i<ncovcombmax;i++)
10474: invalidvarcomb[i]=0;
10475:
1.211 brouard 10476: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10477: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10478: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10479:
1.200 brouard 10480: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10481: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10482: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10483: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10484: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10485: * (currently 0 or 1) in the data.
10486: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10487: * corresponding modality (h,j).
10488: */
10489:
1.145 brouard 10490: h=0;
10491: /*if (cptcovn > 0) */
1.126 brouard 10492: m=pow(2,cptcoveff);
10493:
1.144 brouard 10494: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10495: * For k=4 covariates, h goes from 1 to m=2**k
10496: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10497: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10498: * h\k 1 2 3 4
1.143 brouard 10499: *______________________________
10500: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10501: * 2 2 1 1 1
10502: * 3 i=2 1 2 1 1
10503: * 4 2 2 1 1
10504: * 5 i=3 1 i=2 1 2 1
10505: * 6 2 1 2 1
10506: * 7 i=4 1 2 2 1
10507: * 8 2 2 2 1
1.197 brouard 10508: * 9 i=5 1 i=3 1 i=2 1 2
10509: * 10 2 1 1 2
10510: * 11 i=6 1 2 1 2
10511: * 12 2 2 1 2
10512: * 13 i=7 1 i=4 1 2 2
10513: * 14 2 1 2 2
10514: * 15 i=8 1 2 2 2
10515: * 16 2 2 2 2
1.143 brouard 10516: */
1.212 brouard 10517: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10518: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10519: * and the value of each covariate?
10520: * V1=1, V2=1, V3=2, V4=1 ?
10521: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10522: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10523: * In order to get the real value in the data, we use nbcode
10524: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10525: * We are keeping this crazy system in order to be able (in the future?)
10526: * to have more than 2 values (0 or 1) for a covariate.
10527: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10528: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10529: * bbbbbbbb
10530: * 76543210
10531: * h-1 00000101 (6-1=5)
1.219 brouard 10532: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10533: * &
10534: * 1 00000001 (1)
1.219 brouard 10535: * 00000000 = 1 & ((h-1) >> (k-1))
10536: * +1= 00000001 =1
1.211 brouard 10537: *
10538: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10539: * h' 1101 =2^3+2^2+0x2^1+2^0
10540: * >>k' 11
10541: * & 00000001
10542: * = 00000001
10543: * +1 = 00000010=2 = codtabm(14,3)
10544: * Reverse h=6 and m=16?
10545: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10546: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10547: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10548: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10549: * V3=decodtabm(14,3,2**4)=2
10550: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10551: *(h-1) >> (j-1) 0011 =13 >> 2
10552: * &1 000000001
10553: * = 000000001
10554: * +1= 000000010 =2
10555: * 2211
10556: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10557: * V3=2
1.220 brouard 10558: * codtabm and decodtabm are identical
1.211 brouard 10559: */
10560:
1.145 brouard 10561:
10562: free_ivector(Ndum,-1,NCOVMAX);
10563:
10564:
1.126 brouard 10565:
1.186 brouard 10566: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10567: strcpy(optionfilegnuplot,optionfilefiname);
10568: if(mle==-3)
1.201 brouard 10569: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10570: strcat(optionfilegnuplot,".gp");
10571:
10572: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10573: printf("Problem with file %s",optionfilegnuplot);
10574: }
10575: else{
1.204 brouard 10576: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10577: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10578: //fprintf(ficgp,"set missing 'NaNq'\n");
10579: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10580: }
10581: /* fclose(ficgp);*/
1.186 brouard 10582:
10583:
10584: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10585:
10586: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10587: if(mle==-3)
1.201 brouard 10588: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10589: strcat(optionfilehtm,".htm");
10590: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10591: printf("Problem with %s \n",optionfilehtm);
10592: exit(0);
1.126 brouard 10593: }
10594:
10595: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10596: strcat(optionfilehtmcov,"-cov.htm");
10597: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10598: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10599: }
10600: else{
10601: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10602: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10603: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10604: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10605: }
10606:
1.213 brouard 10607: 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 10608: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10609: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10610: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10611: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10612: \n\
10613: <hr size=\"2\" color=\"#EC5E5E\">\
10614: <ul><li><h4>Parameter files</h4>\n\
10615: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10616: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10617: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10618: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10619: - Date and time at start: %s</ul>\n",\
10620: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10621: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10622: fileres,fileres,\
10623: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10624: fflush(fichtm);
10625:
10626: strcpy(pathr,path);
10627: strcat(pathr,optionfilefiname);
1.184 brouard 10628: #ifdef WIN32
10629: _chdir(optionfilefiname); /* Move to directory named optionfile */
10630: #else
1.126 brouard 10631: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10632: #endif
10633:
1.126 brouard 10634:
1.220 brouard 10635: /* Calculates basic frequencies. Computes observed prevalence at single age
10636: and for any valid combination of covariates
1.126 brouard 10637: and prints on file fileres'p'. */
1.251 brouard 10638: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10639: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10640:
10641: fprintf(fichtm,"\n");
10642: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10643: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10644: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10645: imx,agemin,agemax,jmin,jmax,jmean);
10646: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10647: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10648: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10649: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10650: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10651:
1.126 brouard 10652: /* For Powell, parameters are in a vector p[] starting at p[1]
10653: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10654: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10655:
10656: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10657: /* For mortality only */
1.126 brouard 10658: if (mle==-3){
1.136 brouard 10659: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10660: for(i=1;i<=NDIM;i++)
10661: for(j=1;j<=NDIM;j++)
10662: ximort[i][j]=0.;
1.186 brouard 10663: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10664: cens=ivector(1,n);
10665: ageexmed=vector(1,n);
10666: agecens=vector(1,n);
10667: dcwave=ivector(1,n);
1.223 brouard 10668:
1.126 brouard 10669: for (i=1; i<=imx; i++){
10670: dcwave[i]=-1;
10671: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10672: if (s[m][i]>nlstate) {
10673: dcwave[i]=m;
10674: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10675: break;
10676: }
1.126 brouard 10677: }
1.226 brouard 10678:
1.126 brouard 10679: for (i=1; i<=imx; i++) {
10680: if (wav[i]>0){
1.226 brouard 10681: ageexmed[i]=agev[mw[1][i]][i];
10682: j=wav[i];
10683: agecens[i]=1.;
10684:
10685: if (ageexmed[i]> 1 && wav[i] > 0){
10686: agecens[i]=agev[mw[j][i]][i];
10687: cens[i]= 1;
10688: }else if (ageexmed[i]< 1)
10689: cens[i]= -1;
10690: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10691: cens[i]=0 ;
1.126 brouard 10692: }
10693: else cens[i]=-1;
10694: }
10695:
10696: for (i=1;i<=NDIM;i++) {
10697: for (j=1;j<=NDIM;j++)
1.226 brouard 10698: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10699: }
10700:
1.145 brouard 10701: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10702: /*printf("%lf %lf", p[1], p[2]);*/
10703:
10704:
1.136 brouard 10705: #ifdef GSL
10706: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10707: #else
1.126 brouard 10708: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10709: #endif
1.201 brouard 10710: strcpy(filerespow,"POW-MORT_");
10711: strcat(filerespow,fileresu);
1.126 brouard 10712: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10713: printf("Problem with resultfile: %s\n", filerespow);
10714: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10715: }
1.136 brouard 10716: #ifdef GSL
10717: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10718: #else
1.126 brouard 10719: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10720: #endif
1.126 brouard 10721: /* for (i=1;i<=nlstate;i++)
10722: for(j=1;j<=nlstate+ndeath;j++)
10723: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10724: */
10725: fprintf(ficrespow,"\n");
1.136 brouard 10726: #ifdef GSL
10727: /* gsl starts here */
10728: T = gsl_multimin_fminimizer_nmsimplex;
10729: gsl_multimin_fminimizer *sfm = NULL;
10730: gsl_vector *ss, *x;
10731: gsl_multimin_function minex_func;
10732:
10733: /* Initial vertex size vector */
10734: ss = gsl_vector_alloc (NDIM);
10735:
10736: if (ss == NULL){
10737: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10738: }
10739: /* Set all step sizes to 1 */
10740: gsl_vector_set_all (ss, 0.001);
10741:
10742: /* Starting point */
1.126 brouard 10743:
1.136 brouard 10744: x = gsl_vector_alloc (NDIM);
10745:
10746: if (x == NULL){
10747: gsl_vector_free(ss);
10748: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10749: }
10750:
10751: /* Initialize method and iterate */
10752: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10753: /* gsl_vector_set(x, 0, 0.0268); */
10754: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10755: gsl_vector_set(x, 0, p[1]);
10756: gsl_vector_set(x, 1, p[2]);
10757:
10758: minex_func.f = &gompertz_f;
10759: minex_func.n = NDIM;
10760: minex_func.params = (void *)&p; /* ??? */
10761:
10762: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10763: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10764:
10765: printf("Iterations beginning .....\n\n");
10766: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10767:
10768: iteri=0;
10769: while (rval == GSL_CONTINUE){
10770: iteri++;
10771: status = gsl_multimin_fminimizer_iterate(sfm);
10772:
10773: if (status) printf("error: %s\n", gsl_strerror (status));
10774: fflush(0);
10775:
10776: if (status)
10777: break;
10778:
10779: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10780: ssval = gsl_multimin_fminimizer_size (sfm);
10781:
10782: if (rval == GSL_SUCCESS)
10783: printf ("converged to a local maximum at\n");
10784:
10785: printf("%5d ", iteri);
10786: for (it = 0; it < NDIM; it++){
10787: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10788: }
10789: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10790: }
10791:
10792: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10793:
10794: gsl_vector_free(x); /* initial values */
10795: gsl_vector_free(ss); /* inital step size */
10796: for (it=0; it<NDIM; it++){
10797: p[it+1]=gsl_vector_get(sfm->x,it);
10798: fprintf(ficrespow," %.12lf", p[it]);
10799: }
10800: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10801: #endif
10802: #ifdef POWELL
10803: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10804: #endif
1.126 brouard 10805: fclose(ficrespow);
10806:
1.203 brouard 10807: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10808:
10809: for(i=1; i <=NDIM; i++)
10810: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10811: matcov[i][j]=matcov[j][i];
1.126 brouard 10812:
10813: printf("\nCovariance matrix\n ");
1.203 brouard 10814: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10815: for(i=1; i <=NDIM; i++) {
10816: for(j=1;j<=NDIM;j++){
1.220 brouard 10817: printf("%f ",matcov[i][j]);
10818: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10819: }
1.203 brouard 10820: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10821: }
10822:
10823: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10824: for (i=1;i<=NDIM;i++) {
1.126 brouard 10825: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10826: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10827: }
1.126 brouard 10828: lsurv=vector(1,AGESUP);
10829: lpop=vector(1,AGESUP);
10830: tpop=vector(1,AGESUP);
10831: lsurv[agegomp]=100000;
10832:
10833: for (k=agegomp;k<=AGESUP;k++) {
10834: agemortsup=k;
10835: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10836: }
10837:
10838: for (k=agegomp;k<agemortsup;k++)
10839: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10840:
10841: for (k=agegomp;k<agemortsup;k++){
10842: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10843: sumlpop=sumlpop+lpop[k];
10844: }
10845:
10846: tpop[agegomp]=sumlpop;
10847: for (k=agegomp;k<(agemortsup-3);k++){
10848: /* tpop[k+1]=2;*/
10849: tpop[k+1]=tpop[k]-lpop[k];
10850: }
10851:
10852:
10853: printf("\nAge lx qx dx Lx Tx e(x)\n");
10854: for (k=agegomp;k<(agemortsup-2);k++)
10855: 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]);
10856:
10857:
10858: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10859: ageminpar=50;
10860: agemaxpar=100;
1.194 brouard 10861: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10862: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10863: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10864: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10865: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10866: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10867: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10868: }else{
10869: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10870: 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 10871: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10872: }
1.201 brouard 10873: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10874: stepm, weightopt,\
10875: model,imx,p,matcov,agemortsup);
10876:
10877: free_vector(lsurv,1,AGESUP);
10878: free_vector(lpop,1,AGESUP);
10879: free_vector(tpop,1,AGESUP);
1.220 brouard 10880: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10881: free_ivector(cens,1,n);
10882: free_vector(agecens,1,n);
10883: free_ivector(dcwave,1,n);
1.220 brouard 10884: #ifdef GSL
1.136 brouard 10885: #endif
1.186 brouard 10886: } /* Endof if mle==-3 mortality only */
1.205 brouard 10887: /* Standard */
10888: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10889: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10890: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10891: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10892: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10893: for (k=1; k<=npar;k++)
10894: printf(" %d %8.5f",k,p[k]);
10895: printf("\n");
1.205 brouard 10896: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10897: /* mlikeli uses func not funcone */
1.247 brouard 10898: /* for(i=1;i<nlstate;i++){ */
10899: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10900: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10901: /* } */
1.205 brouard 10902: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10903: }
10904: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10905: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10906: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10907: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10908: }
10909: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10910: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10911: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10912: for (k=1; k<=npar;k++)
10913: printf(" %d %8.5f",k,p[k]);
10914: printf("\n");
10915:
10916: /*--------- results files --------------*/
1.224 brouard 10917: 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 10918:
10919:
10920: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10921: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10922: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10923: for(i=1,jk=1; i <=nlstate; i++){
10924: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10925: if (k != i) {
10926: printf("%d%d ",i,k);
10927: fprintf(ficlog,"%d%d ",i,k);
10928: fprintf(ficres,"%1d%1d ",i,k);
10929: for(j=1; j <=ncovmodel; j++){
10930: printf("%12.7f ",p[jk]);
10931: fprintf(ficlog,"%12.7f ",p[jk]);
10932: fprintf(ficres,"%12.7f ",p[jk]);
10933: jk++;
10934: }
10935: printf("\n");
10936: fprintf(ficlog,"\n");
10937: fprintf(ficres,"\n");
10938: }
1.126 brouard 10939: }
10940: }
1.203 brouard 10941: if(mle != 0){
10942: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10943: ftolhess=ftol; /* Usually correct */
1.203 brouard 10944: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10945: 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");
10946: 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");
10947: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10948: for(k=1; k <=(nlstate+ndeath); k++){
10949: if (k != i) {
10950: printf("%d%d ",i,k);
10951: fprintf(ficlog,"%d%d ",i,k);
10952: for(j=1; j <=ncovmodel; j++){
10953: 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]));
10954: 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]));
10955: jk++;
10956: }
10957: printf("\n");
10958: fprintf(ficlog,"\n");
10959: }
10960: }
1.193 brouard 10961: }
1.203 brouard 10962: } /* end of hesscov and Wald tests */
1.225 brouard 10963:
1.203 brouard 10964: /* */
1.126 brouard 10965: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10966: printf("# Scales (for hessian or gradient estimation)\n");
10967: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10968: for(i=1,jk=1; i <=nlstate; i++){
10969: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10970: if (j!=i) {
10971: fprintf(ficres,"%1d%1d",i,j);
10972: printf("%1d%1d",i,j);
10973: fprintf(ficlog,"%1d%1d",i,j);
10974: for(k=1; k<=ncovmodel;k++){
10975: printf(" %.5e",delti[jk]);
10976: fprintf(ficlog," %.5e",delti[jk]);
10977: fprintf(ficres," %.5e",delti[jk]);
10978: jk++;
10979: }
10980: printf("\n");
10981: fprintf(ficlog,"\n");
10982: fprintf(ficres,"\n");
10983: }
1.126 brouard 10984: }
10985: }
10986:
10987: 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 10988: if(mle >= 1) /* To big for the screen */
1.126 brouard 10989: 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");
10990: 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");
10991: /* # 121 Var(a12)\n\ */
10992: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10993: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10994: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10995: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10996: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10997: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10998: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10999:
11000:
11001: /* Just to have a covariance matrix which will be more understandable
11002: even is we still don't want to manage dictionary of variables
11003: */
11004: for(itimes=1;itimes<=2;itimes++){
11005: jj=0;
11006: for(i=1; i <=nlstate; i++){
1.225 brouard 11007: for(j=1; j <=nlstate+ndeath; j++){
11008: if(j==i) continue;
11009: for(k=1; k<=ncovmodel;k++){
11010: jj++;
11011: ca[0]= k+'a'-1;ca[1]='\0';
11012: if(itimes==1){
11013: if(mle>=1)
11014: printf("#%1d%1d%d",i,j,k);
11015: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11016: fprintf(ficres,"#%1d%1d%d",i,j,k);
11017: }else{
11018: if(mle>=1)
11019: printf("%1d%1d%d",i,j,k);
11020: fprintf(ficlog,"%1d%1d%d",i,j,k);
11021: fprintf(ficres,"%1d%1d%d",i,j,k);
11022: }
11023: ll=0;
11024: for(li=1;li <=nlstate; li++){
11025: for(lj=1;lj <=nlstate+ndeath; lj++){
11026: if(lj==li) continue;
11027: for(lk=1;lk<=ncovmodel;lk++){
11028: ll++;
11029: if(ll<=jj){
11030: cb[0]= lk +'a'-1;cb[1]='\0';
11031: if(ll<jj){
11032: if(itimes==1){
11033: if(mle>=1)
11034: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11035: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11036: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11037: }else{
11038: if(mle>=1)
11039: printf(" %.5e",matcov[jj][ll]);
11040: fprintf(ficlog," %.5e",matcov[jj][ll]);
11041: fprintf(ficres," %.5e",matcov[jj][ll]);
11042: }
11043: }else{
11044: if(itimes==1){
11045: if(mle>=1)
11046: printf(" Var(%s%1d%1d)",ca,i,j);
11047: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11048: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11049: }else{
11050: if(mle>=1)
11051: printf(" %.7e",matcov[jj][ll]);
11052: fprintf(ficlog," %.7e",matcov[jj][ll]);
11053: fprintf(ficres," %.7e",matcov[jj][ll]);
11054: }
11055: }
11056: }
11057: } /* end lk */
11058: } /* end lj */
11059: } /* end li */
11060: if(mle>=1)
11061: printf("\n");
11062: fprintf(ficlog,"\n");
11063: fprintf(ficres,"\n");
11064: numlinepar++;
11065: } /* end k*/
11066: } /*end j */
1.126 brouard 11067: } /* end i */
11068: } /* end itimes */
11069:
11070: fflush(ficlog);
11071: fflush(ficres);
1.225 brouard 11072: while(fgets(line, MAXLINE, ficpar)) {
11073: /* If line starts with a # it is a comment */
11074: if (line[0] == '#') {
11075: numlinepar++;
11076: fputs(line,stdout);
11077: fputs(line,ficparo);
11078: fputs(line,ficlog);
11079: continue;
11080: }else
11081: break;
11082: }
11083:
1.209 brouard 11084: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11085: /* ungetc(c,ficpar); */
11086: /* fgets(line, MAXLINE, ficpar); */
11087: /* fputs(line,stdout); */
11088: /* fputs(line,ficparo); */
11089: /* } */
11090: /* ungetc(c,ficpar); */
1.126 brouard 11091:
11092: estepm=0;
1.209 brouard 11093: 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 11094:
11095: if (num_filled != 6) {
11096: 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);
11097: 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);
11098: goto end;
11099: }
11100: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11101: }
11102: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11103: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11104:
1.209 brouard 11105: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11106: if (estepm==0 || estepm < stepm) estepm=stepm;
11107: if (fage <= 2) {
11108: bage = ageminpar;
11109: fage = agemaxpar;
11110: }
11111:
11112: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11113: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11114: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11115:
1.186 brouard 11116: /* Other stuffs, more or less useful */
1.254 ! brouard 11117: while(fgets(line, MAXLINE, ficpar)) {
! 11118: /* If line starts with a # it is a comment */
! 11119: if (line[0] == '#') {
! 11120: numlinepar++;
! 11121: fputs(line,stdout);
! 11122: fputs(line,ficparo);
! 11123: fputs(line,ficlog);
! 11124: continue;
! 11125: }else
! 11126: break;
! 11127: }
! 11128:
! 11129: if((num_filled=sscanf(line,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav)) !=EOF){
! 11130:
! 11131: if (num_filled != 7) {
! 11132: printf("Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11133: fprintf(ficlog,"Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11134: goto end;
! 11135: }
! 11136: /* 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); */
! 11137: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
! 11138: 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);
! 11139: 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);
! 11140: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
! 11141: fprintf(ficlog,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
1.126 brouard 11142: }
1.254 ! brouard 11143:
! 11144: while(fgets(line, MAXLINE, ficpar)) {
! 11145: /* If line starts with a # it is a comment */
! 11146: if (line[0] == '#') {
! 11147: numlinepar++;
! 11148: fputs(line,stdout);
! 11149: fputs(line,ficparo);
! 11150: fputs(line,ficlog);
! 11151: continue;
! 11152: }else
! 11153: break;
1.126 brouard 11154: }
11155:
11156:
11157: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11158: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11159:
1.254 ! brouard 11160: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
! 11161: if (num_filled != 1) {
! 11162: printf("Error: Not 1 (data)parameters in line but %d, for example:pop_based=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11163: fprintf(ficlog,"Error: Not 1 (data)parameters in line but %d, for example: pop_based=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11164: goto end;
! 11165: }
! 11166: printf("pop_based=%d\n",popbased);
! 11167: fprintf(ficlog,"pop_based=%d\n",popbased);
! 11168: fprintf(ficparo,"pop_based=%d\n",popbased);
! 11169: fprintf(ficres,"pop_based=%d\n",popbased);
! 11170: }
! 11171:
! 11172: while(fgets(line, MAXLINE, ficpar)) {
! 11173: /* If line starts with a # it is a comment */
! 11174: if (line[0] == '#') {
! 11175: numlinepar++;
! 11176: fputs(line,stdout);
! 11177: fputs(line,ficparo);
! 11178: fputs(line,ficlog);
! 11179: continue;
! 11180: }else
! 11181: break;
1.126 brouard 11182: }
1.254 ! brouard 11183: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
! 11184: /* ungetc(c,ficpar); */
! 11185: /* fgets(line, MAXLINE, ficpar); */
! 11186: /* fputs(line,stdout); */
! 11187: /* fputs(line,ficres); */
! 11188: /* fputs(line,ficparo); */
! 11189: /* } */
! 11190: /* ungetc(c,ficpar); */
! 11191:
! 11192: /* 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); */
! 11193: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF){
! 11194: if (num_filled != 8) {
! 11195: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11196: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11197: goto end;
! 11198: }
! 11199: 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);
! 11200: 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);
! 11201: 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);
! 11202: 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);
1.126 brouard 11203: /* day and month of proj2 are not used but only year anproj2.*/
1.217 brouard 11204: }
1.254 ! brouard 11205: while(fgets(line, MAXLINE, ficpar)) {
! 11206: /* If line starts with a # it is a comment */
! 11207: if (line[0] == '#') {
! 11208: numlinepar++;
! 11209: fputs(line,stdout);
! 11210: fputs(line,ficparo);
! 11211: fputs(line,ficlog);
! 11212: continue;
! 11213: }else
! 11214: break;
! 11215: }
! 11216: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
! 11217: /* ungetc(c,ficpar); */
! 11218: /* fgets(line, MAXLINE, ficpar); */
! 11219: /* fputs(line,stdout); */
! 11220: /* fputs(line,ficparo); */
! 11221: /* fputs(line,ficres); */
! 11222: /* } */
! 11223: /* ungetc(c,ficpar); */
1.217 brouard 11224:
11225: 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);
1.254 ! brouard 11226: if((num_filled=sscanf(line,"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)) !=EOF){
! 11227: if (num_filled != 8) {
! 11228: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11229: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
! 11230: goto end;
! 11231: }
! 11232: printf("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);
! 11233: 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);
! 11234: 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);
! 11235: 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 11236: /* day and month of proj2 are not used but only year anproj2.*/
1.254 ! brouard 11237: }
1.230 brouard 11238: /* Results */
1.235 brouard 11239: nresult=0;
1.230 brouard 11240: while(fgets(line, MAXLINE, ficpar)) {
11241: /* If line starts with a # it is a comment */
11242: if (line[0] == '#') {
11243: numlinepar++;
11244: fputs(line,stdout);
11245: fputs(line,ficparo);
11246: fputs(line,ficlog);
1.238 brouard 11247: fputs(line,ficres);
1.230 brouard 11248: continue;
11249: }else
11250: break;
11251: }
1.240 brouard 11252: if (!feof(ficpar))
1.230 brouard 11253: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11254: if (num_filled == 0){
1.230 brouard 11255: resultline[0]='\0';
1.253 brouard 11256: printf("Warning %d: no result line should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
1.240 brouard 11257: break;
11258: } else if (num_filled != 1){
1.253 brouard 11259: printf("ERROR %d: result line should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
1.230 brouard 11260: }
1.235 brouard 11261: nresult++; /* Sum of resultlines */
11262: printf("Result %d: result=%s\n",nresult, resultline);
11263: if(nresult > MAXRESULTLINES){
11264: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11265: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11266: goto end;
11267: }
11268: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11269: fprintf(ficparo,"result: %s\n",resultline);
11270: fprintf(ficres,"result: %s\n",resultline);
11271: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11272: while(fgets(line, MAXLINE, ficpar)) {
11273: /* If line starts with a # it is a comment */
11274: if (line[0] == '#') {
11275: numlinepar++;
11276: fputs(line,stdout);
11277: fputs(line,ficparo);
1.238 brouard 11278: fputs(line,ficres);
1.230 brouard 11279: fputs(line,ficlog);
11280: continue;
11281: }else
11282: break;
11283: }
11284: if (feof(ficpar))
11285: break;
11286: else{ /* Processess output results for this combination of covariate values */
11287: }
1.240 brouard 11288: } /* end while */
1.230 brouard 11289:
11290:
1.126 brouard 11291:
1.230 brouard 11292: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11293: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11294:
11295: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11296: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11297: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11298: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11299: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11300: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11301: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11302: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11303: }else{
1.218 brouard 11304: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11305: }
11306: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11307: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11308: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11309:
1.225 brouard 11310: /*------------ free_vector -------------*/
11311: /* chdir(path); */
1.220 brouard 11312:
1.215 brouard 11313: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11314: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11315: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11316: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11317: free_lvector(num,1,n);
11318: free_vector(agedc,1,n);
11319: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11320: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11321: fclose(ficparo);
11322: fclose(ficres);
1.220 brouard 11323:
11324:
1.186 brouard 11325: /* Other results (useful)*/
1.220 brouard 11326:
11327:
1.126 brouard 11328: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11329: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11330: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11331: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11332: fclose(ficrespl);
11333:
11334: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11335: /*#include "hpijx.h"*/
11336: hPijx(p, bage, fage);
1.145 brouard 11337: fclose(ficrespij);
1.227 brouard 11338:
1.220 brouard 11339: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11340: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11341: k=1;
1.126 brouard 11342: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11343:
1.219 brouard 11344: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11345: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11346: for(i=1;i<=AGESUP;i++)
1.219 brouard 11347: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11348: for(k=1;k<=ncovcombmax;k++)
11349: probs[i][j][k]=0.;
1.219 brouard 11350: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11351: if (mobilav!=0 ||mobilavproj !=0 ) {
11352: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11353: for(i=1;i<=AGESUP;i++)
11354: for(j=1;j<=nlstate;j++)
11355: for(k=1;k<=ncovcombmax;k++)
11356: mobaverages[i][j][k]=0.;
1.219 brouard 11357: mobaverage=mobaverages;
11358: if (mobilav!=0) {
1.235 brouard 11359: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11360: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11361: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11362: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11363: }
1.219 brouard 11364: }
11365: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11366: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11367: else if (mobilavproj !=0) {
1.235 brouard 11368: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11369: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11370: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11371: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11372: }
1.219 brouard 11373: }
11374: }/* end if moving average */
1.227 brouard 11375:
1.126 brouard 11376: /*---------- Forecasting ------------------*/
11377: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11378: if(prevfcast==1){
11379: /* if(stepm ==1){*/
1.225 brouard 11380: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11381: }
1.217 brouard 11382: if(backcast==1){
1.219 brouard 11383: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11384: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11385: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11386:
11387: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11388:
11389: bprlim=matrix(1,nlstate,1,nlstate);
11390: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11391: fclose(ficresplb);
11392:
1.222 brouard 11393: hBijx(p, bage, fage, mobaverage);
11394: fclose(ficrespijb);
1.219 brouard 11395: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11396:
11397: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11398: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11399: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11400: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11401: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11402: }
1.217 brouard 11403:
1.186 brouard 11404:
11405: /* ------ Other prevalence ratios------------ */
1.126 brouard 11406:
1.215 brouard 11407: free_ivector(wav,1,imx);
11408: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11409: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11410: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11411:
11412:
1.127 brouard 11413: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11414:
1.201 brouard 11415: strcpy(filerese,"E_");
11416: strcat(filerese,fileresu);
1.126 brouard 11417: if((ficreseij=fopen(filerese,"w"))==NULL) {
11418: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11419: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11420: }
1.208 brouard 11421: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11422: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11423:
11424: pstamp(ficreseij);
1.219 brouard 11425:
1.235 brouard 11426: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11427: if (cptcovn < 1){i1=1;}
11428:
11429: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11430: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11431: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11432: continue;
1.219 brouard 11433: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11434: printf("\n#****** ");
1.225 brouard 11435: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11436: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11437: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11438: }
11439: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11440: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11441: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11442: }
11443: fprintf(ficreseij,"******\n");
1.235 brouard 11444: printf("******\n");
1.219 brouard 11445:
11446: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11447: oldm=oldms;savm=savms;
1.235 brouard 11448: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11449:
1.219 brouard 11450: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11451: }
11452: fclose(ficreseij);
1.208 brouard 11453: printf("done evsij\n");fflush(stdout);
11454: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11455:
1.227 brouard 11456: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11457:
11458:
1.201 brouard 11459: strcpy(filerest,"T_");
11460: strcat(filerest,fileresu);
1.127 brouard 11461: if((ficrest=fopen(filerest,"w"))==NULL) {
11462: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11463: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11464: }
1.208 brouard 11465: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11466: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11467:
1.126 brouard 11468:
1.201 brouard 11469: strcpy(fileresstde,"STDE_");
11470: strcat(fileresstde,fileresu);
1.126 brouard 11471: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11472: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11473: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11474: }
1.227 brouard 11475: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11476: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11477:
1.201 brouard 11478: strcpy(filerescve,"CVE_");
11479: strcat(filerescve,fileresu);
1.126 brouard 11480: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11481: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11482: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11483: }
1.227 brouard 11484: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11485: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11486:
1.201 brouard 11487: strcpy(fileresv,"V_");
11488: strcat(fileresv,fileresu);
1.126 brouard 11489: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11490: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11491: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11492: }
1.227 brouard 11493: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11494: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11495:
1.145 brouard 11496: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11497: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11498:
1.235 brouard 11499: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11500: if (cptcovn < 1){i1=1;}
11501:
11502: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11503: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11504: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11505: continue;
1.242 brouard 11506: printf("\n#****** Result for:");
11507: fprintf(ficrest,"\n#****** Result for:");
11508: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11509: for(j=1;j<=cptcoveff;j++){
11510: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11511: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11512: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11513: }
1.235 brouard 11514: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11515: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11516: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11517: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11518: }
1.208 brouard 11519: fprintf(ficrest,"******\n");
1.227 brouard 11520: fprintf(ficlog,"******\n");
11521: printf("******\n");
1.208 brouard 11522:
11523: fprintf(ficresstdeij,"\n#****** ");
11524: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11525: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11526: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11527: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11528: }
1.235 brouard 11529: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11530: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11531: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11532: }
1.208 brouard 11533: fprintf(ficresstdeij,"******\n");
11534: fprintf(ficrescveij,"******\n");
11535:
11536: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11537: /* pstamp(ficresvij); */
1.225 brouard 11538: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11539: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11540: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11541: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11542: }
1.208 brouard 11543: fprintf(ficresvij,"******\n");
11544:
11545: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11546: oldm=oldms;savm=savms;
1.235 brouard 11547: printf(" cvevsij ");
11548: fprintf(ficlog, " cvevsij ");
11549: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11550: printf(" end cvevsij \n ");
11551: fprintf(ficlog, " end cvevsij \n ");
11552:
11553: /*
11554: */
11555: /* goto endfree; */
11556:
11557: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11558: pstamp(ficrest);
11559:
11560:
11561: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11562: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11563: cptcod= 0; /* To be deleted */
11564: printf("varevsij vpopbased=%d \n",vpopbased);
11565: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11566: 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 11567: 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 ");
11568: if(vpopbased==1)
11569: 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);
11570: else
11571: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11572: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11573: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11574: fprintf(ficrest,"\n");
11575: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11576: epj=vector(1,nlstate+1);
11577: printf("Computing age specific period (stable) prevalences in each health state \n");
11578: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11579: for(age=bage; age <=fage ;age++){
1.235 brouard 11580: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11581: if (vpopbased==1) {
11582: if(mobilav ==0){
11583: for(i=1; i<=nlstate;i++)
11584: prlim[i][i]=probs[(int)age][i][k];
11585: }else{ /* mobilav */
11586: for(i=1; i<=nlstate;i++)
11587: prlim[i][i]=mobaverage[(int)age][i][k];
11588: }
11589: }
1.219 brouard 11590:
1.227 brouard 11591: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11592: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11593: /* printf(" age %4.0f ",age); */
11594: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11595: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11596: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11597: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11598: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11599: }
11600: epj[nlstate+1] +=epj[j];
11601: }
11602: /* printf(" age %4.0f \n",age); */
1.219 brouard 11603:
1.227 brouard 11604: for(i=1, vepp=0.;i <=nlstate;i++)
11605: for(j=1;j <=nlstate;j++)
11606: vepp += vareij[i][j][(int)age];
11607: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11608: for(j=1;j <=nlstate;j++){
11609: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11610: }
11611: fprintf(ficrest,"\n");
11612: }
1.208 brouard 11613: } /* End vpopbased */
11614: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11615: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11616: free_vector(epj,1,nlstate+1);
1.235 brouard 11617: printf("done selection\n");fflush(stdout);
11618: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11619:
1.145 brouard 11620: /*}*/
1.235 brouard 11621: } /* End k selection */
1.227 brouard 11622:
11623: printf("done State-specific expectancies\n");fflush(stdout);
11624: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11625:
1.126 brouard 11626: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11627:
1.201 brouard 11628: strcpy(fileresvpl,"VPL_");
11629: strcat(fileresvpl,fileresu);
1.126 brouard 11630: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11631: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11632: exit(0);
11633: }
1.208 brouard 11634: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11635: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11636:
1.145 brouard 11637: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11638: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11639:
1.235 brouard 11640: i1=pow(2,cptcoveff);
11641: if (cptcovn < 1){i1=1;}
11642:
11643: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11644: for(k=1; k<=i1;k++){
1.253 brouard 11645: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11646: continue;
1.227 brouard 11647: fprintf(ficresvpl,"\n#****** ");
11648: printf("\n#****** ");
11649: fprintf(ficlog,"\n#****** ");
11650: for(j=1;j<=cptcoveff;j++) {
11651: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11652: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11653: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11654: }
1.235 brouard 11655: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11656: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11657: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11658: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11659: }
1.227 brouard 11660: fprintf(ficresvpl,"******\n");
11661: printf("******\n");
11662: fprintf(ficlog,"******\n");
11663:
11664: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11665: oldm=oldms;savm=savms;
1.235 brouard 11666: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11667: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11668: /*}*/
1.126 brouard 11669: }
1.227 brouard 11670:
1.126 brouard 11671: fclose(ficresvpl);
1.208 brouard 11672: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11673: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11674:
11675: free_vector(weight,1,n);
11676: free_imatrix(Tvard,1,NCOVMAX,1,2);
11677: free_imatrix(s,1,maxwav+1,1,n);
11678: free_matrix(anint,1,maxwav,1,n);
11679: free_matrix(mint,1,maxwav,1,n);
11680: free_ivector(cod,1,n);
11681: free_ivector(tab,1,NCOVMAX);
11682: fclose(ficresstdeij);
11683: fclose(ficrescveij);
11684: fclose(ficresvij);
11685: fclose(ficrest);
11686: fclose(ficpar);
11687:
11688:
1.126 brouard 11689: /*---------- End : free ----------------*/
1.219 brouard 11690: if (mobilav!=0 ||mobilavproj !=0)
11691: 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 11692: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11693: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11694: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11695: } /* mle==-3 arrives here for freeing */
1.227 brouard 11696: /* endfree:*/
11697: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11698: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11699: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11700: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11701: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11702: free_matrix(coqvar,1,maxwav,1,n);
11703: free_matrix(covar,0,NCOVMAX,1,n);
11704: free_matrix(matcov,1,npar,1,npar);
11705: free_matrix(hess,1,npar,1,npar);
11706: /*free_vector(delti,1,npar);*/
11707: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11708: free_matrix(agev,1,maxwav,1,imx);
11709: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11710:
11711: free_ivector(ncodemax,1,NCOVMAX);
11712: free_ivector(ncodemaxwundef,1,NCOVMAX);
11713: free_ivector(Dummy,-1,NCOVMAX);
11714: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11715: free_ivector(DummyV,1,NCOVMAX);
11716: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11717: free_ivector(Typevar,-1,NCOVMAX);
11718: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11719: free_ivector(TvarsQ,1,NCOVMAX);
11720: free_ivector(TvarsQind,1,NCOVMAX);
11721: free_ivector(TvarsD,1,NCOVMAX);
11722: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11723: free_ivector(TvarFD,1,NCOVMAX);
11724: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11725: free_ivector(TvarF,1,NCOVMAX);
11726: free_ivector(TvarFind,1,NCOVMAX);
11727: free_ivector(TvarV,1,NCOVMAX);
11728: free_ivector(TvarVind,1,NCOVMAX);
11729: free_ivector(TvarA,1,NCOVMAX);
11730: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11731: free_ivector(TvarFQ,1,NCOVMAX);
11732: free_ivector(TvarFQind,1,NCOVMAX);
11733: free_ivector(TvarVD,1,NCOVMAX);
11734: free_ivector(TvarVDind,1,NCOVMAX);
11735: free_ivector(TvarVQ,1,NCOVMAX);
11736: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11737: free_ivector(Tvarsel,1,NCOVMAX);
11738: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11739: free_ivector(Tposprod,1,NCOVMAX);
11740: free_ivector(Tprod,1,NCOVMAX);
11741: free_ivector(Tvaraff,1,NCOVMAX);
11742: free_ivector(invalidvarcomb,1,ncovcombmax);
11743: free_ivector(Tage,1,NCOVMAX);
11744: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11745: free_ivector(TmodelInvind,1,NCOVMAX);
11746: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11747:
11748: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11749: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11750: fflush(fichtm);
11751: fflush(ficgp);
11752:
1.227 brouard 11753:
1.126 brouard 11754: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11755: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11756: 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 11757: }else{
11758: printf("End of Imach\n");
11759: fprintf(ficlog,"End of Imach\n");
11760: }
11761: printf("See log file on %s\n",filelog);
11762: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11763: /*(void) gettimeofday(&end_time,&tzp);*/
11764: rend_time = time(NULL);
11765: end_time = *localtime(&rend_time);
11766: /* tml = *localtime(&end_time.tm_sec); */
11767: strcpy(strtend,asctime(&end_time));
1.126 brouard 11768: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11769: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11770: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11771:
1.157 brouard 11772: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11773: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11774: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11775: /* printf("Total time was %d uSec.\n", total_usecs);*/
11776: /* if(fileappend(fichtm,optionfilehtm)){ */
11777: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11778: fclose(fichtm);
11779: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11780: fclose(fichtmcov);
11781: fclose(ficgp);
11782: fclose(ficlog);
11783: /*------ End -----------*/
1.227 brouard 11784:
11785:
11786: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11787: #ifdef WIN32
1.227 brouard 11788: if (_chdir(pathcd) != 0)
11789: printf("Can't move to directory %s!\n",path);
11790: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11791: #else
1.227 brouard 11792: if(chdir(pathcd) != 0)
11793: printf("Can't move to directory %s!\n", path);
11794: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11795: #endif
1.126 brouard 11796: printf("Current directory %s!\n",pathcd);
11797: /*strcat(plotcmd,CHARSEPARATOR);*/
11798: sprintf(plotcmd,"gnuplot");
1.157 brouard 11799: #ifdef _WIN32
1.126 brouard 11800: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11801: #endif
11802: if(!stat(plotcmd,&info)){
1.158 brouard 11803: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11804: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11805: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11806: }else
11807: strcpy(pplotcmd,plotcmd);
1.157 brouard 11808: #ifdef __unix
1.126 brouard 11809: strcpy(plotcmd,GNUPLOTPROGRAM);
11810: if(!stat(plotcmd,&info)){
1.158 brouard 11811: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11812: }else
11813: strcpy(pplotcmd,plotcmd);
11814: #endif
11815: }else
11816: strcpy(pplotcmd,plotcmd);
11817:
11818: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11819: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11820:
1.126 brouard 11821: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11822: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11823: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11824: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11825: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11826: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11827: }
1.158 brouard 11828: printf(" Successful, please wait...");
1.126 brouard 11829: while (z[0] != 'q') {
11830: /* chdir(path); */
1.154 brouard 11831: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11832: scanf("%s",z);
11833: /* if (z[0] == 'c') system("./imach"); */
11834: if (z[0] == 'e') {
1.158 brouard 11835: #ifdef __APPLE__
1.152 brouard 11836: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11837: #elif __linux
11838: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11839: #else
1.152 brouard 11840: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11841: #endif
11842: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11843: system(pplotcmd);
1.126 brouard 11844: }
11845: else if (z[0] == 'g') system(plotcmd);
11846: else if (z[0] == 'q') exit(0);
11847: }
1.227 brouard 11848: end:
1.126 brouard 11849: while (z[0] != 'q') {
1.195 brouard 11850: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11851: scanf("%s",z);
11852: }
11853: }
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