Annotation of imach/src/imach.c, revision 1.247
1.247 ! brouard 1: /* $Id: imach.c,v 1.246 2016/09/02 08:49:22 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.247 ! brouard 4: Revision 1.246 2016/09/02 08:49:22 brouard
! 5: *** empty log message ***
! 6:
1.246 brouard 7: Revision 1.245 2016/09/02 07:25:01 brouard
8: *** empty log message ***
9:
1.245 brouard 10: Revision 1.244 2016/09/02 07:17:34 brouard
11: *** empty log message ***
12:
1.244 brouard 13: Revision 1.243 2016/09/02 06:45:35 brouard
14: *** empty log message ***
15:
1.243 brouard 16: Revision 1.242 2016/08/30 15:01:20 brouard
17: Summary: Fixing a lots
18:
1.242 brouard 19: Revision 1.241 2016/08/29 17:17:25 brouard
20: Summary: gnuplot problem in Back projection to fix
21:
1.241 brouard 22: Revision 1.240 2016/08/29 07:53:18 brouard
23: Summary: Better
24:
1.240 brouard 25: Revision 1.239 2016/08/26 15:51:03 brouard
26: Summary: Improvement in Powell output in order to copy and paste
27:
28: Author:
29:
1.239 brouard 30: Revision 1.238 2016/08/26 14:23:35 brouard
31: Summary: Starting tests of 0.99
32:
1.238 brouard 33: Revision 1.237 2016/08/26 09:20:19 brouard
34: Summary: to valgrind
35:
1.237 brouard 36: Revision 1.236 2016/08/25 10:50:18 brouard
37: *** empty log message ***
38:
1.236 brouard 39: Revision 1.235 2016/08/25 06:59:23 brouard
40: *** empty log message ***
41:
1.235 brouard 42: Revision 1.234 2016/08/23 16:51:20 brouard
43: *** empty log message ***
44:
1.234 brouard 45: Revision 1.233 2016/08/23 07:40:50 brouard
46: Summary: not working
47:
1.233 brouard 48: Revision 1.232 2016/08/22 14:20:21 brouard
49: Summary: not working
50:
1.232 brouard 51: Revision 1.231 2016/08/22 07:17:15 brouard
52: Summary: not working
53:
1.231 brouard 54: Revision 1.230 2016/08/22 06:55:53 brouard
55: Summary: Not working
56:
1.230 brouard 57: Revision 1.229 2016/07/23 09:45:53 brouard
58: Summary: Completing for func too
59:
1.229 brouard 60: Revision 1.228 2016/07/22 17:45:30 brouard
61: Summary: Fixing some arrays, still debugging
62:
1.227 brouard 63: Revision 1.226 2016/07/12 18:42:34 brouard
64: Summary: temp
65:
1.226 brouard 66: Revision 1.225 2016/07/12 08:40:03 brouard
67: Summary: saving but not running
68:
1.225 brouard 69: Revision 1.224 2016/07/01 13:16:01 brouard
70: Summary: Fixes
71:
1.224 brouard 72: Revision 1.223 2016/02/19 09:23:35 brouard
73: Summary: temporary
74:
1.223 brouard 75: Revision 1.222 2016/02/17 08:14:50 brouard
76: Summary: Probably last 0.98 stable version 0.98r6
77:
1.222 brouard 78: Revision 1.221 2016/02/15 23:35:36 brouard
79: Summary: minor bug
80:
1.220 brouard 81: Revision 1.219 2016/02/15 00:48:12 brouard
82: *** empty log message ***
83:
1.219 brouard 84: Revision 1.218 2016/02/12 11:29:23 brouard
85: Summary: 0.99 Back projections
86:
1.218 brouard 87: Revision 1.217 2015/12/23 17:18:31 brouard
88: Summary: Experimental backcast
89:
1.217 brouard 90: Revision 1.216 2015/12/18 17:32:11 brouard
91: Summary: 0.98r4 Warning and status=-2
92:
93: Version 0.98r4 is now:
94: - displaying an error when status is -1, date of interview unknown and date of death known;
95: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
96: Older changes concerning s=-2, dating from 2005 have been supersed.
97:
1.216 brouard 98: Revision 1.215 2015/12/16 08:52:24 brouard
99: Summary: 0.98r4 working
100:
1.215 brouard 101: Revision 1.214 2015/12/16 06:57:54 brouard
102: Summary: temporary not working
103:
1.214 brouard 104: Revision 1.213 2015/12/11 18:22:17 brouard
105: Summary: 0.98r4
106:
1.213 brouard 107: Revision 1.212 2015/11/21 12:47:24 brouard
108: Summary: minor typo
109:
1.212 brouard 110: Revision 1.211 2015/11/21 12:41:11 brouard
111: Summary: 0.98r3 with some graph of projected cross-sectional
112:
113: Author: Nicolas Brouard
114:
1.211 brouard 115: Revision 1.210 2015/11/18 17:41:20 brouard
116: Summary: Start working on projected prevalences
117:
1.210 brouard 118: Revision 1.209 2015/11/17 22:12:03 brouard
119: Summary: Adding ftolpl parameter
120: Author: N Brouard
121:
122: We had difficulties to get smoothed confidence intervals. It was due
123: to the period prevalence which wasn't computed accurately. The inner
124: parameter ftolpl is now an outer parameter of the .imach parameter
125: file after estepm. If ftolpl is small 1.e-4 and estepm too,
126: computation are long.
127:
1.209 brouard 128: Revision 1.208 2015/11/17 14:31:57 brouard
129: Summary: temporary
130:
1.208 brouard 131: Revision 1.207 2015/10/27 17:36:57 brouard
132: *** empty log message ***
133:
1.207 brouard 134: Revision 1.206 2015/10/24 07:14:11 brouard
135: *** empty log message ***
136:
1.206 brouard 137: Revision 1.205 2015/10/23 15:50:53 brouard
138: Summary: 0.98r3 some clarification for graphs on likelihood contributions
139:
1.205 brouard 140: Revision 1.204 2015/10/01 16:20:26 brouard
141: Summary: Some new graphs of contribution to likelihood
142:
1.204 brouard 143: Revision 1.203 2015/09/30 17:45:14 brouard
144: Summary: looking at better estimation of the hessian
145:
146: Also a better criteria for convergence to the period prevalence And
147: therefore adding the number of years needed to converge. (The
148: prevalence in any alive state shold sum to one
149:
1.203 brouard 150: Revision 1.202 2015/09/22 19:45:16 brouard
151: Summary: Adding some overall graph on contribution to likelihood. Might change
152:
1.202 brouard 153: Revision 1.201 2015/09/15 17:34:58 brouard
154: Summary: 0.98r0
155:
156: - Some new graphs like suvival functions
157: - Some bugs fixed like model=1+age+V2.
158:
1.201 brouard 159: Revision 1.200 2015/09/09 16:53:55 brouard
160: Summary: Big bug thanks to Flavia
161:
162: Even model=1+age+V2. did not work anymore
163:
1.200 brouard 164: Revision 1.199 2015/09/07 14:09:23 brouard
165: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
166:
1.199 brouard 167: Revision 1.198 2015/09/03 07:14:39 brouard
168: Summary: 0.98q5 Flavia
169:
1.198 brouard 170: Revision 1.197 2015/09/01 18:24:39 brouard
171: *** empty log message ***
172:
1.197 brouard 173: Revision 1.196 2015/08/18 23:17:52 brouard
174: Summary: 0.98q5
175:
1.196 brouard 176: Revision 1.195 2015/08/18 16:28:39 brouard
177: Summary: Adding a hack for testing purpose
178:
179: After reading the title, ftol and model lines, if the comment line has
180: a q, starting with #q, the answer at the end of the run is quit. It
181: permits to run test files in batch with ctest. The former workaround was
182: $ echo q | imach foo.imach
183:
1.195 brouard 184: Revision 1.194 2015/08/18 13:32:00 brouard
185: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
186:
1.194 brouard 187: Revision 1.193 2015/08/04 07:17:42 brouard
188: Summary: 0.98q4
189:
1.193 brouard 190: Revision 1.192 2015/07/16 16:49:02 brouard
191: Summary: Fixing some outputs
192:
1.192 brouard 193: Revision 1.191 2015/07/14 10:00:33 brouard
194: Summary: Some fixes
195:
1.191 brouard 196: Revision 1.190 2015/05/05 08:51:13 brouard
197: Summary: Adding digits in output parameters (7 digits instead of 6)
198:
199: Fix 1+age+.
200:
1.190 brouard 201: Revision 1.189 2015/04/30 14:45:16 brouard
202: Summary: 0.98q2
203:
1.189 brouard 204: Revision 1.188 2015/04/30 08:27:53 brouard
205: *** empty log message ***
206:
1.188 brouard 207: Revision 1.187 2015/04/29 09:11:15 brouard
208: *** empty log message ***
209:
1.187 brouard 210: Revision 1.186 2015/04/23 12:01:52 brouard
211: Summary: V1*age is working now, version 0.98q1
212:
213: Some codes had been disabled in order to simplify and Vn*age was
214: working in the optimization phase, ie, giving correct MLE parameters,
215: but, as usual, outputs were not correct and program core dumped.
216:
1.186 brouard 217: Revision 1.185 2015/03/11 13:26:42 brouard
218: Summary: Inclusion of compile and links command line for Intel Compiler
219:
1.185 brouard 220: Revision 1.184 2015/03/11 11:52:39 brouard
221: Summary: Back from Windows 8. Intel Compiler
222:
1.184 brouard 223: Revision 1.183 2015/03/10 20:34:32 brouard
224: Summary: 0.98q0, trying with directest, mnbrak fixed
225:
226: We use directest instead of original Powell test; probably no
227: incidence on the results, but better justifications;
228: We fixed Numerical Recipes mnbrak routine which was wrong and gave
229: wrong results.
230:
1.183 brouard 231: Revision 1.182 2015/02/12 08:19:57 brouard
232: Summary: Trying to keep directest which seems simpler and more general
233: Author: Nicolas Brouard
234:
1.182 brouard 235: Revision 1.181 2015/02/11 23:22:24 brouard
236: Summary: Comments on Powell added
237:
238: Author:
239:
1.181 brouard 240: Revision 1.180 2015/02/11 17:33:45 brouard
241: Summary: Finishing move from main to function (hpijx and prevalence_limit)
242:
1.180 brouard 243: Revision 1.179 2015/01/04 09:57:06 brouard
244: Summary: back to OS/X
245:
1.179 brouard 246: Revision 1.178 2015/01/04 09:35:48 brouard
247: *** empty log message ***
248:
1.178 brouard 249: Revision 1.177 2015/01/03 18:40:56 brouard
250: Summary: Still testing ilc32 on OSX
251:
1.177 brouard 252: Revision 1.176 2015/01/03 16:45:04 brouard
253: *** empty log message ***
254:
1.176 brouard 255: Revision 1.175 2015/01/03 16:33:42 brouard
256: *** empty log message ***
257:
1.175 brouard 258: Revision 1.174 2015/01/03 16:15:49 brouard
259: Summary: Still in cross-compilation
260:
1.174 brouard 261: Revision 1.173 2015/01/03 12:06:26 brouard
262: Summary: trying to detect cross-compilation
263:
1.173 brouard 264: Revision 1.172 2014/12/27 12:07:47 brouard
265: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
266:
1.172 brouard 267: Revision 1.171 2014/12/23 13:26:59 brouard
268: Summary: Back from Visual C
269:
270: Still problem with utsname.h on Windows
271:
1.171 brouard 272: Revision 1.170 2014/12/23 11:17:12 brouard
273: Summary: Cleaning some \%% back to %%
274:
275: The escape was mandatory for a specific compiler (which one?), but too many warnings.
276:
1.170 brouard 277: Revision 1.169 2014/12/22 23:08:31 brouard
278: Summary: 0.98p
279:
280: Outputs some informations on compiler used, OS etc. Testing on different platforms.
281:
1.169 brouard 282: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 283: Summary: update
1.169 brouard 284:
1.168 brouard 285: Revision 1.167 2014/12/22 13:50:56 brouard
286: Summary: Testing uname and compiler version and if compiled 32 or 64
287:
288: Testing on Linux 64
289:
1.167 brouard 290: Revision 1.166 2014/12/22 11:40:47 brouard
291: *** empty log message ***
292:
1.166 brouard 293: Revision 1.165 2014/12/16 11:20:36 brouard
294: Summary: After compiling on Visual C
295:
296: * imach.c (Module): Merging 1.61 to 1.162
297:
1.165 brouard 298: Revision 1.164 2014/12/16 10:52:11 brouard
299: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
300:
301: * imach.c (Module): Merging 1.61 to 1.162
302:
1.164 brouard 303: Revision 1.163 2014/12/16 10:30:11 brouard
304: * imach.c (Module): Merging 1.61 to 1.162
305:
1.163 brouard 306: Revision 1.162 2014/09/25 11:43:39 brouard
307: Summary: temporary backup 0.99!
308:
1.162 brouard 309: Revision 1.1 2014/09/16 11:06:58 brouard
310: Summary: With some code (wrong) for nlopt
311:
312: Author:
313:
314: Revision 1.161 2014/09/15 20:41:41 brouard
315: Summary: Problem with macro SQR on Intel compiler
316:
1.161 brouard 317: Revision 1.160 2014/09/02 09:24:05 brouard
318: *** empty log message ***
319:
1.160 brouard 320: Revision 1.159 2014/09/01 10:34:10 brouard
321: Summary: WIN32
322: Author: Brouard
323:
1.159 brouard 324: Revision 1.158 2014/08/27 17:11:51 brouard
325: *** empty log message ***
326:
1.158 brouard 327: Revision 1.157 2014/08/27 16:26:55 brouard
328: Summary: Preparing windows Visual studio version
329: Author: Brouard
330:
331: In order to compile on Visual studio, time.h is now correct and time_t
332: and tm struct should be used. difftime should be used but sometimes I
333: just make the differences in raw time format (time(&now).
334: Trying to suppress #ifdef LINUX
335: Add xdg-open for __linux in order to open default browser.
336:
1.157 brouard 337: Revision 1.156 2014/08/25 20:10:10 brouard
338: *** empty log message ***
339:
1.156 brouard 340: Revision 1.155 2014/08/25 18:32:34 brouard
341: Summary: New compile, minor changes
342: Author: Brouard
343:
1.155 brouard 344: Revision 1.154 2014/06/20 17:32:08 brouard
345: Summary: Outputs now all graphs of convergence to period prevalence
346:
1.154 brouard 347: Revision 1.153 2014/06/20 16:45:46 brouard
348: Summary: If 3 live state, convergence to period prevalence on same graph
349: Author: Brouard
350:
1.153 brouard 351: Revision 1.152 2014/06/18 17:54:09 brouard
352: Summary: open browser, use gnuplot on same dir than imach if not found in the path
353:
1.152 brouard 354: Revision 1.151 2014/06/18 16:43:30 brouard
355: *** empty log message ***
356:
1.151 brouard 357: Revision 1.150 2014/06/18 16:42:35 brouard
358: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
359: Author: brouard
360:
1.150 brouard 361: Revision 1.149 2014/06/18 15:51:14 brouard
362: Summary: Some fixes in parameter files errors
363: Author: Nicolas Brouard
364:
1.149 brouard 365: Revision 1.148 2014/06/17 17:38:48 brouard
366: Summary: Nothing new
367: Author: Brouard
368:
369: Just a new packaging for OS/X version 0.98nS
370:
1.148 brouard 371: Revision 1.147 2014/06/16 10:33:11 brouard
372: *** empty log message ***
373:
1.147 brouard 374: Revision 1.146 2014/06/16 10:20:28 brouard
375: Summary: Merge
376: Author: Brouard
377:
378: Merge, before building revised version.
379:
1.146 brouard 380: Revision 1.145 2014/06/10 21:23:15 brouard
381: Summary: Debugging with valgrind
382: Author: Nicolas Brouard
383:
384: Lot of changes in order to output the results with some covariates
385: After the Edimburgh REVES conference 2014, it seems mandatory to
386: improve the code.
387: No more memory valgrind error but a lot has to be done in order to
388: continue the work of splitting the code into subroutines.
389: Also, decodemodel has been improved. Tricode is still not
390: optimal. nbcode should be improved. Documentation has been added in
391: the source code.
392:
1.144 brouard 393: Revision 1.143 2014/01/26 09:45:38 brouard
394: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
395:
396: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
397: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
398:
1.143 brouard 399: Revision 1.142 2014/01/26 03:57:36 brouard
400: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
401:
402: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
403:
1.142 brouard 404: Revision 1.141 2014/01/26 02:42:01 brouard
405: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
406:
1.141 brouard 407: Revision 1.140 2011/09/02 10:37:54 brouard
408: Summary: times.h is ok with mingw32 now.
409:
1.140 brouard 410: Revision 1.139 2010/06/14 07:50:17 brouard
411: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
412: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
413:
1.139 brouard 414: Revision 1.138 2010/04/30 18:19:40 brouard
415: *** empty log message ***
416:
1.138 brouard 417: Revision 1.137 2010/04/29 18:11:38 brouard
418: (Module): Checking covariates for more complex models
419: than V1+V2. A lot of change to be done. Unstable.
420:
1.137 brouard 421: Revision 1.136 2010/04/26 20:30:53 brouard
422: (Module): merging some libgsl code. Fixing computation
423: of likelione (using inter/intrapolation if mle = 0) in order to
424: get same likelihood as if mle=1.
425: Some cleaning of code and comments added.
426:
1.136 brouard 427: Revision 1.135 2009/10/29 15:33:14 brouard
428: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
429:
1.135 brouard 430: Revision 1.134 2009/10/29 13:18:53 brouard
431: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
432:
1.134 brouard 433: Revision 1.133 2009/07/06 10:21:25 brouard
434: just nforces
435:
1.133 brouard 436: Revision 1.132 2009/07/06 08:22:05 brouard
437: Many tings
438:
1.132 brouard 439: Revision 1.131 2009/06/20 16:22:47 brouard
440: Some dimensions resccaled
441:
1.131 brouard 442: Revision 1.130 2009/05/26 06:44:34 brouard
443: (Module): Max Covariate is now set to 20 instead of 8. A
444: lot of cleaning with variables initialized to 0. Trying to make
445: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
446:
1.130 brouard 447: Revision 1.129 2007/08/31 13:49:27 lievre
448: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
449:
1.129 lievre 450: Revision 1.128 2006/06/30 13:02:05 brouard
451: (Module): Clarifications on computing e.j
452:
1.128 brouard 453: Revision 1.127 2006/04/28 18:11:50 brouard
454: (Module): Yes the sum of survivors was wrong since
455: imach-114 because nhstepm was no more computed in the age
456: loop. Now we define nhstepma in the age loop.
457: (Module): In order to speed up (in case of numerous covariates) we
458: compute health expectancies (without variances) in a first step
459: and then all the health expectancies with variances or standard
460: deviation (needs data from the Hessian matrices) which slows the
461: computation.
462: In the future we should be able to stop the program is only health
463: expectancies and graph are needed without standard deviations.
464:
1.127 brouard 465: Revision 1.126 2006/04/28 17:23:28 brouard
466: (Module): Yes the sum of survivors was wrong since
467: imach-114 because nhstepm was no more computed in the age
468: loop. Now we define nhstepma in the age loop.
469: Version 0.98h
470:
1.126 brouard 471: Revision 1.125 2006/04/04 15:20:31 lievre
472: Errors in calculation of health expectancies. Age was not initialized.
473: Forecasting file added.
474:
475: Revision 1.124 2006/03/22 17:13:53 lievre
476: Parameters are printed with %lf instead of %f (more numbers after the comma).
477: The log-likelihood is printed in the log file
478:
479: Revision 1.123 2006/03/20 10:52:43 brouard
480: * imach.c (Module): <title> changed, corresponds to .htm file
481: name. <head> headers where missing.
482:
483: * imach.c (Module): Weights can have a decimal point as for
484: English (a comma might work with a correct LC_NUMERIC environment,
485: otherwise the weight is truncated).
486: Modification of warning when the covariates values are not 0 or
487: 1.
488: Version 0.98g
489:
490: Revision 1.122 2006/03/20 09:45:41 brouard
491: (Module): Weights can have a decimal point as for
492: English (a comma might work with a correct LC_NUMERIC environment,
493: otherwise the weight is truncated).
494: Modification of warning when the covariates values are not 0 or
495: 1.
496: Version 0.98g
497:
498: Revision 1.121 2006/03/16 17:45:01 lievre
499: * imach.c (Module): Comments concerning covariates added
500:
501: * imach.c (Module): refinements in the computation of lli if
502: status=-2 in order to have more reliable computation if stepm is
503: not 1 month. Version 0.98f
504:
505: Revision 1.120 2006/03/16 15:10:38 lievre
506: (Module): refinements in the computation of lli if
507: status=-2 in order to have more reliable computation if stepm is
508: not 1 month. Version 0.98f
509:
510: Revision 1.119 2006/03/15 17:42:26 brouard
511: (Module): Bug if status = -2, the loglikelihood was
512: computed as likelihood omitting the logarithm. Version O.98e
513:
514: Revision 1.118 2006/03/14 18:20:07 brouard
515: (Module): varevsij Comments added explaining the second
516: table of variances if popbased=1 .
517: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
518: (Module): Function pstamp added
519: (Module): Version 0.98d
520:
521: Revision 1.117 2006/03/14 17:16:22 brouard
522: (Module): varevsij Comments added explaining the second
523: table of variances if popbased=1 .
524: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
525: (Module): Function pstamp added
526: (Module): Version 0.98d
527:
528: Revision 1.116 2006/03/06 10:29:27 brouard
529: (Module): Variance-covariance wrong links and
530: varian-covariance of ej. is needed (Saito).
531:
532: Revision 1.115 2006/02/27 12:17:45 brouard
533: (Module): One freematrix added in mlikeli! 0.98c
534:
535: Revision 1.114 2006/02/26 12:57:58 brouard
536: (Module): Some improvements in processing parameter
537: filename with strsep.
538:
539: Revision 1.113 2006/02/24 14:20:24 brouard
540: (Module): Memory leaks checks with valgrind and:
541: datafile was not closed, some imatrix were not freed and on matrix
542: allocation too.
543:
544: Revision 1.112 2006/01/30 09:55:26 brouard
545: (Module): Back to gnuplot.exe instead of wgnuplot.exe
546:
547: Revision 1.111 2006/01/25 20:38:18 brouard
548: (Module): Lots of cleaning and bugs added (Gompertz)
549: (Module): Comments can be added in data file. Missing date values
550: can be a simple dot '.'.
551:
552: Revision 1.110 2006/01/25 00:51:50 brouard
553: (Module): Lots of cleaning and bugs added (Gompertz)
554:
555: Revision 1.109 2006/01/24 19:37:15 brouard
556: (Module): Comments (lines starting with a #) are allowed in data.
557:
558: Revision 1.108 2006/01/19 18:05:42 lievre
559: Gnuplot problem appeared...
560: To be fixed
561:
562: Revision 1.107 2006/01/19 16:20:37 brouard
563: Test existence of gnuplot in imach path
564:
565: Revision 1.106 2006/01/19 13:24:36 brouard
566: Some cleaning and links added in html output
567:
568: Revision 1.105 2006/01/05 20:23:19 lievre
569: *** empty log message ***
570:
571: Revision 1.104 2005/09/30 16:11:43 lievre
572: (Module): sump fixed, loop imx fixed, and simplifications.
573: (Module): If the status is missing at the last wave but we know
574: that the person is alive, then we can code his/her status as -2
575: (instead of missing=-1 in earlier versions) and his/her
576: contributions to the likelihood is 1 - Prob of dying from last
577: health status (= 1-p13= p11+p12 in the easiest case of somebody in
578: the healthy state at last known wave). Version is 0.98
579:
580: Revision 1.103 2005/09/30 15:54:49 lievre
581: (Module): sump fixed, loop imx fixed, and simplifications.
582:
583: Revision 1.102 2004/09/15 17:31:30 brouard
584: Add the possibility to read data file including tab characters.
585:
586: Revision 1.101 2004/09/15 10:38:38 brouard
587: Fix on curr_time
588:
589: Revision 1.100 2004/07/12 18:29:06 brouard
590: Add version for Mac OS X. Just define UNIX in Makefile
591:
592: Revision 1.99 2004/06/05 08:57:40 brouard
593: *** empty log message ***
594:
595: Revision 1.98 2004/05/16 15:05:56 brouard
596: New version 0.97 . First attempt to estimate force of mortality
597: directly from the data i.e. without the need of knowing the health
598: state at each age, but using a Gompertz model: log u =a + b*age .
599: This is the basic analysis of mortality and should be done before any
600: other analysis, in order to test if the mortality estimated from the
601: cross-longitudinal survey is different from the mortality estimated
602: from other sources like vital statistic data.
603:
604: The same imach parameter file can be used but the option for mle should be -3.
605:
1.133 brouard 606: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 607: former routines in order to include the new code within the former code.
608:
609: The output is very simple: only an estimate of the intercept and of
610: the slope with 95% confident intervals.
611:
612: Current limitations:
613: A) Even if you enter covariates, i.e. with the
614: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
615: B) There is no computation of Life Expectancy nor Life Table.
616:
617: Revision 1.97 2004/02/20 13:25:42 lievre
618: Version 0.96d. Population forecasting command line is (temporarily)
619: suppressed.
620:
621: Revision 1.96 2003/07/15 15:38:55 brouard
622: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
623: rewritten within the same printf. Workaround: many printfs.
624:
625: Revision 1.95 2003/07/08 07:54:34 brouard
626: * imach.c (Repository):
627: (Repository): Using imachwizard code to output a more meaningful covariance
628: matrix (cov(a12,c31) instead of numbers.
629:
630: Revision 1.94 2003/06/27 13:00:02 brouard
631: Just cleaning
632:
633: Revision 1.93 2003/06/25 16:33:55 brouard
634: (Module): On windows (cygwin) function asctime_r doesn't
635: exist so I changed back to asctime which exists.
636: (Module): Version 0.96b
637:
638: Revision 1.92 2003/06/25 16:30:45 brouard
639: (Module): On windows (cygwin) function asctime_r doesn't
640: exist so I changed back to asctime which exists.
641:
642: Revision 1.91 2003/06/25 15:30:29 brouard
643: * imach.c (Repository): Duplicated warning errors corrected.
644: (Repository): Elapsed time after each iteration is now output. It
645: helps to forecast when convergence will be reached. Elapsed time
646: is stamped in powell. We created a new html file for the graphs
647: concerning matrix of covariance. It has extension -cov.htm.
648:
649: Revision 1.90 2003/06/24 12:34:15 brouard
650: (Module): Some bugs corrected for windows. Also, when
651: mle=-1 a template is output in file "or"mypar.txt with the design
652: of the covariance matrix to be input.
653:
654: Revision 1.89 2003/06/24 12:30:52 brouard
655: (Module): Some bugs corrected for windows. Also, when
656: mle=-1 a template is output in file "or"mypar.txt with the design
657: of the covariance matrix to be input.
658:
659: Revision 1.88 2003/06/23 17:54:56 brouard
660: * 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.
661:
662: Revision 1.87 2003/06/18 12:26:01 brouard
663: Version 0.96
664:
665: Revision 1.86 2003/06/17 20:04:08 brouard
666: (Module): Change position of html and gnuplot routines and added
667: routine fileappend.
668:
669: Revision 1.85 2003/06/17 13:12:43 brouard
670: * imach.c (Repository): Check when date of death was earlier that
671: current date of interview. It may happen when the death was just
672: prior to the death. In this case, dh was negative and likelihood
673: was wrong (infinity). We still send an "Error" but patch by
674: assuming that the date of death was just one stepm after the
675: interview.
676: (Repository): Because some people have very long ID (first column)
677: we changed int to long in num[] and we added a new lvector for
678: memory allocation. But we also truncated to 8 characters (left
679: truncation)
680: (Repository): No more line truncation errors.
681:
682: Revision 1.84 2003/06/13 21:44:43 brouard
683: * imach.c (Repository): Replace "freqsummary" at a correct
684: place. It differs from routine "prevalence" which may be called
685: many times. Probs is memory consuming and must be used with
686: parcimony.
687: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
688:
689: Revision 1.83 2003/06/10 13:39:11 lievre
690: *** empty log message ***
691:
692: Revision 1.82 2003/06/05 15:57:20 brouard
693: Add log in imach.c and fullversion number is now printed.
694:
695: */
696: /*
697: Interpolated Markov Chain
698:
699: Short summary of the programme:
700:
1.227 brouard 701: This program computes Healthy Life Expectancies or State-specific
702: (if states aren't health statuses) Expectancies from
703: cross-longitudinal data. Cross-longitudinal data consist in:
704:
705: -1- a first survey ("cross") where individuals from different ages
706: are interviewed on their health status or degree of disability (in
707: the case of a health survey which is our main interest)
708:
709: -2- at least a second wave of interviews ("longitudinal") which
710: measure each change (if any) in individual health status. Health
711: expectancies are computed from the time spent in each health state
712: according to a model. More health states you consider, more time is
713: necessary to reach the Maximum Likelihood of the parameters involved
714: in the model. The simplest model is the multinomial logistic model
715: where pij is the probability to be observed in state j at the second
716: wave conditional to be observed in state i at the first
717: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
718: etc , where 'age' is age and 'sex' is a covariate. If you want to
719: have a more complex model than "constant and age", you should modify
720: the program where the markup *Covariates have to be included here
721: again* invites you to do it. More covariates you add, slower the
1.126 brouard 722: convergence.
723:
724: The advantage of this computer programme, compared to a simple
725: multinomial logistic model, is clear when the delay between waves is not
726: identical for each individual. Also, if a individual missed an
727: intermediate interview, the information is lost, but taken into
728: account using an interpolation or extrapolation.
729:
730: hPijx is the probability to be observed in state i at age x+h
731: conditional to the observed state i at age x. The delay 'h' can be
732: split into an exact number (nh*stepm) of unobserved intermediate
733: states. This elementary transition (by month, quarter,
734: semester or year) is modelled as a multinomial logistic. The hPx
735: matrix is simply the matrix product of nh*stepm elementary matrices
736: and the contribution of each individual to the likelihood is simply
737: hPijx.
738:
739: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 740: of the life expectancies. It also computes the period (stable) prevalence.
741:
742: Back prevalence and projections:
1.227 brouard 743:
744: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
745: double agemaxpar, double ftolpl, int *ncvyearp, double
746: dateprev1,double dateprev2, int firstpass, int lastpass, int
747: mobilavproj)
748:
749: Computes the back prevalence limit for any combination of
750: covariate values k at any age between ageminpar and agemaxpar and
751: returns it in **bprlim. In the loops,
752:
753: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
754: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
755:
756: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 757: Computes for any combination of covariates k and any age between bage and fage
758: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
759: oldm=oldms;savm=savms;
1.227 brouard 760:
761: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 762: Computes the transition matrix starting at age 'age' over
763: 'nhstepm*hstepm*stepm' months (i.e. until
764: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 765: nhstepm*hstepm matrices.
766:
767: Returns p3mat[i][j][h] after calling
768: p3mat[i][j][h]=matprod2(newm,
769: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
770: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
771: oldm);
1.226 brouard 772:
773: Important routines
774:
775: - func (or funcone), computes logit (pij) distinguishing
776: o fixed variables (single or product dummies or quantitative);
777: o varying variables by:
778: (1) wave (single, product dummies, quantitative),
779: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
780: % fixed dummy (treated) or quantitative (not done because time-consuming);
781: % varying dummy (not done) or quantitative (not done);
782: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
783: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
784: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
785: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
786: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 787:
1.226 brouard 788:
789:
1.133 brouard 790: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
791: Institut national d'études démographiques, Paris.
1.126 brouard 792: This software have been partly granted by Euro-REVES, a concerted action
793: from the European Union.
794: It is copyrighted identically to a GNU software product, ie programme and
795: software can be distributed freely for non commercial use. Latest version
796: can be accessed at http://euroreves.ined.fr/imach .
797:
798: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
799: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
800:
801: **********************************************************************/
802: /*
803: main
804: read parameterfile
805: read datafile
806: concatwav
807: freqsummary
808: if (mle >= 1)
809: mlikeli
810: print results files
811: if mle==1
812: computes hessian
813: read end of parameter file: agemin, agemax, bage, fage, estepm
814: begin-prev-date,...
815: open gnuplot file
816: open html file
1.145 brouard 817: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
818: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
819: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
820: freexexit2 possible for memory heap.
821:
822: h Pij x | pij_nom ficrestpij
823: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
824: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
825: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
826:
827: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
828: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
829: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
830: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
831: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
832:
1.126 brouard 833: forecasting if prevfcast==1 prevforecast call prevalence()
834: health expectancies
835: Variance-covariance of DFLE
836: prevalence()
837: movingaverage()
838: varevsij()
839: if popbased==1 varevsij(,popbased)
840: total life expectancies
841: Variance of period (stable) prevalence
842: end
843: */
844:
1.187 brouard 845: /* #define DEBUG */
846: /* #define DEBUGBRENT */
1.203 brouard 847: /* #define DEBUGLINMIN */
848: /* #define DEBUGHESS */
849: #define DEBUGHESSIJ
1.224 brouard 850: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 851: #define POWELL /* Instead of NLOPT */
1.224 brouard 852: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 853: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
854: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 855:
856: #include <math.h>
857: #include <stdio.h>
858: #include <stdlib.h>
859: #include <string.h>
1.226 brouard 860: #include <ctype.h>
1.159 brouard 861:
862: #ifdef _WIN32
863: #include <io.h>
1.172 brouard 864: #include <windows.h>
865: #include <tchar.h>
1.159 brouard 866: #else
1.126 brouard 867: #include <unistd.h>
1.159 brouard 868: #endif
1.126 brouard 869:
870: #include <limits.h>
871: #include <sys/types.h>
1.171 brouard 872:
873: #if defined(__GNUC__)
874: #include <sys/utsname.h> /* Doesn't work on Windows */
875: #endif
876:
1.126 brouard 877: #include <sys/stat.h>
878: #include <errno.h>
1.159 brouard 879: /* extern int errno; */
1.126 brouard 880:
1.157 brouard 881: /* #ifdef LINUX */
882: /* #include <time.h> */
883: /* #include "timeval.h" */
884: /* #else */
885: /* #include <sys/time.h> */
886: /* #endif */
887:
1.126 brouard 888: #include <time.h>
889:
1.136 brouard 890: #ifdef GSL
891: #include <gsl/gsl_errno.h>
892: #include <gsl/gsl_multimin.h>
893: #endif
894:
1.167 brouard 895:
1.162 brouard 896: #ifdef NLOPT
897: #include <nlopt.h>
898: typedef struct {
899: double (* function)(double [] );
900: } myfunc_data ;
901: #endif
902:
1.126 brouard 903: /* #include <libintl.h> */
904: /* #define _(String) gettext (String) */
905:
1.141 brouard 906: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 907:
908: #define GNUPLOTPROGRAM "gnuplot"
909: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
910: #define FILENAMELENGTH 132
911:
912: #define GLOCK_ERROR_NOPATH -1 /* empty path */
913: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
914:
1.144 brouard 915: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
916: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 917:
918: #define NINTERVMAX 8
1.144 brouard 919: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
920: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
921: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 922: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 923: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
924: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 925: #define MAXN 20000
1.144 brouard 926: #define YEARM 12. /**< Number of months per year */
1.218 brouard 927: /* #define AGESUP 130 */
928: #define AGESUP 150
929: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 930: #define AGEBASE 40
1.194 brouard 931: #define AGEOVERFLOW 1.e20
1.164 brouard 932: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 933: #ifdef _WIN32
934: #define DIRSEPARATOR '\\'
935: #define CHARSEPARATOR "\\"
936: #define ODIRSEPARATOR '/'
937: #else
1.126 brouard 938: #define DIRSEPARATOR '/'
939: #define CHARSEPARATOR "/"
940: #define ODIRSEPARATOR '\\'
941: #endif
942:
1.247 ! brouard 943: /* $Id: imach.c,v 1.246 2016/09/02 08:49:22 brouard Exp $ */
1.126 brouard 944: /* $State: Exp $ */
1.196 brouard 945: #include "version.h"
946: char version[]=__IMACH_VERSION__;
1.224 brouard 947: 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";
1.247 ! brouard 948: char fullversion[]="$Revision: 1.246 $ $Date: 2016/09/02 08:49:22 $";
1.126 brouard 949: char strstart[80];
950: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 951: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 952: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 953: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
954: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
955: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 956: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
957: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 958: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
959: int cptcovprodnoage=0; /**< Number of covariate products without age */
960: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 961: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
962: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 963: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 964: int nsd=0; /**< Total number of single dummy variables (output) */
965: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 966: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 967: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 968: int ntveff=0; /**< ntveff number of effective time varying variables */
969: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 970: int cptcov=0; /* Working variable */
1.218 brouard 971: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 972: int npar=NPARMAX;
973: int nlstate=2; /* Number of live states */
974: int ndeath=1; /* Number of dead states */
1.130 brouard 975: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 976: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 977: int popbased=0;
978:
979: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 980: int maxwav=0; /* Maxim number of waves */
981: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
982: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
983: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 984: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 985: int mle=1, weightopt=0;
1.126 brouard 986: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
987: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
988: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
989: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 990: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 991: int selected(int kvar); /* Is covariate kvar selected for printing results */
992:
1.130 brouard 993: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 994: double **matprod2(); /* test */
1.126 brouard 995: double **oldm, **newm, **savm; /* Working pointers to matrices */
996: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 997: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
998:
1.136 brouard 999: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1000: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1001: FILE *ficlog, *ficrespow;
1.130 brouard 1002: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1003: double fretone; /* Only one call to likelihood */
1.130 brouard 1004: long ipmx=0; /* Number of contributions */
1.126 brouard 1005: double sw; /* Sum of weights */
1006: char filerespow[FILENAMELENGTH];
1007: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1008: FILE *ficresilk;
1009: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1010: FILE *ficresprobmorprev;
1011: FILE *fichtm, *fichtmcov; /* Html File */
1012: FILE *ficreseij;
1013: char filerese[FILENAMELENGTH];
1014: FILE *ficresstdeij;
1015: char fileresstde[FILENAMELENGTH];
1016: FILE *ficrescveij;
1017: char filerescve[FILENAMELENGTH];
1018: FILE *ficresvij;
1019: char fileresv[FILENAMELENGTH];
1020: FILE *ficresvpl;
1021: char fileresvpl[FILENAMELENGTH];
1022: char title[MAXLINE];
1.234 brouard 1023: char model[MAXLINE]; /**< The model line */
1.217 brouard 1024: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1025: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1026: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1027: char command[FILENAMELENGTH];
1028: int outcmd=0;
1029:
1.217 brouard 1030: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1031: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1032: char filelog[FILENAMELENGTH]; /* Log file */
1033: char filerest[FILENAMELENGTH];
1034: char fileregp[FILENAMELENGTH];
1035: char popfile[FILENAMELENGTH];
1036:
1037: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1038:
1.157 brouard 1039: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1040: /* struct timezone tzp; */
1041: /* extern int gettimeofday(); */
1042: struct tm tml, *gmtime(), *localtime();
1043:
1044: extern time_t time();
1045:
1046: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1047: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1048: struct tm tm;
1049:
1.126 brouard 1050: char strcurr[80], strfor[80];
1051:
1052: char *endptr;
1053: long lval;
1054: double dval;
1055:
1056: #define NR_END 1
1057: #define FREE_ARG char*
1058: #define FTOL 1.0e-10
1059:
1060: #define NRANSI
1.240 brouard 1061: #define ITMAX 200
1062: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1063:
1064: #define TOL 2.0e-4
1065:
1066: #define CGOLD 0.3819660
1067: #define ZEPS 1.0e-10
1068: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1069:
1070: #define GOLD 1.618034
1071: #define GLIMIT 100.0
1072: #define TINY 1.0e-20
1073:
1074: static double maxarg1,maxarg2;
1075: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1076: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1077:
1078: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1079: #define rint(a) floor(a+0.5)
1.166 brouard 1080: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1081: #define mytinydouble 1.0e-16
1.166 brouard 1082: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1083: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1084: /* static double dsqrarg; */
1085: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1086: static double sqrarg;
1087: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1088: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1089: int agegomp= AGEGOMP;
1090:
1091: int imx;
1092: int stepm=1;
1093: /* Stepm, step in month: minimum step interpolation*/
1094:
1095: int estepm;
1096: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1097:
1098: int m,nb;
1099: long *num;
1.197 brouard 1100: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1101: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1102: covariate for which somebody answered excluding
1103: undefined. Usually 2: 0 and 1. */
1104: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1105: covariate for which somebody answered including
1106: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1107: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1108: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1109: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1110: double *ageexmed,*agecens;
1111: double dateintmean=0;
1112:
1113: double *weight;
1114: int **s; /* Status */
1.141 brouard 1115: double *agedc;
1.145 brouard 1116: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1117: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1118: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1119: double **coqvar; /* Fixed quantitative covariate iqv */
1120: double ***cotvar; /* Time varying covariate itv */
1121: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1122: double idx;
1123: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1124: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1125: /*k 1 2 3 4 5 6 7 8 9 */
1126: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1127: /* Tndvar[k] 1 2 3 4 5 */
1128: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1129: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1130: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1131: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1132: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1133: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1134: /* Tprod[i]=k 4 7 */
1135: /* Tage[i]=k 5 8 */
1136: /* */
1137: /* Type */
1138: /* V 1 2 3 4 5 */
1139: /* F F V V V */
1140: /* D Q D D Q */
1141: /* */
1142: int *TvarsD;
1143: int *TvarsDind;
1144: int *TvarsQ;
1145: int *TvarsQind;
1146:
1.235 brouard 1147: #define MAXRESULTLINES 10
1148: int nresult=0;
1149: int TKresult[MAXRESULTLINES];
1.237 brouard 1150: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1151: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1152: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1153: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1154: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1155: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1156:
1.234 brouard 1157: /* 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 1158: 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 */
1159: 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 */
1160: 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 */
1161: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1162: 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 */
1163: 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 1164: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1165: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1166: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1167: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1168: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1169: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1170: 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 */
1171: 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 */
1172:
1.230 brouard 1173: int *Tvarsel; /**< Selected covariates for output */
1174: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1175: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1176: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1177: 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 1178: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1179: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1180: int *Tage;
1.227 brouard 1181: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1182: 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 1183: 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*/
1184: 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 1185: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1186: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1187: int **Tvard;
1188: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1189: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1190: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1191: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1192: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1193: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1194: double *lsurv, *lpop, *tpop;
1195:
1.231 brouard 1196: #define FD 1; /* Fixed dummy covariate */
1197: #define FQ 2; /* Fixed quantitative covariate */
1198: #define FP 3; /* Fixed product covariate */
1199: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1200: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1201: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1202: #define VD 10; /* Varying dummy covariate */
1203: #define VQ 11; /* Varying quantitative covariate */
1204: #define VP 12; /* Varying product covariate */
1205: #define VPDD 13; /* Varying product dummy*dummy covariate */
1206: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1207: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1208: #define APFD 16; /* Age product * fixed dummy covariate */
1209: #define APFQ 17; /* Age product * fixed quantitative covariate */
1210: #define APVD 18; /* Age product * varying dummy covariate */
1211: #define APVQ 19; /* Age product * varying quantitative covariate */
1212:
1213: #define FTYPE 1; /* Fixed covariate */
1214: #define VTYPE 2; /* Varying covariate (loop in wave) */
1215: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1216:
1217: struct kmodel{
1218: int maintype; /* main type */
1219: int subtype; /* subtype */
1220: };
1221: struct kmodel modell[NCOVMAX];
1222:
1.143 brouard 1223: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1224: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1225:
1226: /**************** split *************************/
1227: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1228: {
1229: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1230: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1231: */
1232: char *ss; /* pointer */
1.186 brouard 1233: int l1=0, l2=0; /* length counters */
1.126 brouard 1234:
1235: l1 = strlen(path ); /* length of path */
1236: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1237: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1238: if ( ss == NULL ) { /* no directory, so determine current directory */
1239: strcpy( name, path ); /* we got the fullname name because no directory */
1240: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1241: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1242: /* get current working directory */
1243: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1244: #ifdef WIN32
1245: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1246: #else
1247: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1248: #endif
1.126 brouard 1249: return( GLOCK_ERROR_GETCWD );
1250: }
1251: /* got dirc from getcwd*/
1252: printf(" DIRC = %s \n",dirc);
1.205 brouard 1253: } else { /* strip directory from path */
1.126 brouard 1254: ss++; /* after this, the filename */
1255: l2 = strlen( ss ); /* length of filename */
1256: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1257: strcpy( name, ss ); /* save file name */
1258: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1259: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1260: printf(" DIRC2 = %s \n",dirc);
1261: }
1262: /* We add a separator at the end of dirc if not exists */
1263: l1 = strlen( dirc ); /* length of directory */
1264: if( dirc[l1-1] != DIRSEPARATOR ){
1265: dirc[l1] = DIRSEPARATOR;
1266: dirc[l1+1] = 0;
1267: printf(" DIRC3 = %s \n",dirc);
1268: }
1269: ss = strrchr( name, '.' ); /* find last / */
1270: if (ss >0){
1271: ss++;
1272: strcpy(ext,ss); /* save extension */
1273: l1= strlen( name);
1274: l2= strlen(ss)+1;
1275: strncpy( finame, name, l1-l2);
1276: finame[l1-l2]= 0;
1277: }
1278:
1279: return( 0 ); /* we're done */
1280: }
1281:
1282:
1283: /******************************************/
1284:
1285: void replace_back_to_slash(char *s, char*t)
1286: {
1287: int i;
1288: int lg=0;
1289: i=0;
1290: lg=strlen(t);
1291: for(i=0; i<= lg; i++) {
1292: (s[i] = t[i]);
1293: if (t[i]== '\\') s[i]='/';
1294: }
1295: }
1296:
1.132 brouard 1297: char *trimbb(char *out, char *in)
1.137 brouard 1298: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1299: char *s;
1300: s=out;
1301: while (*in != '\0'){
1.137 brouard 1302: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1303: in++;
1304: }
1305: *out++ = *in++;
1306: }
1307: *out='\0';
1308: return s;
1309: }
1310:
1.187 brouard 1311: /* char *substrchaine(char *out, char *in, char *chain) */
1312: /* { */
1313: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1314: /* char *s, *t; */
1315: /* t=in;s=out; */
1316: /* while ((*in != *chain) && (*in != '\0')){ */
1317: /* *out++ = *in++; */
1318: /* } */
1319:
1320: /* /\* *in matches *chain *\/ */
1321: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1322: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1323: /* } */
1324: /* in--; chain--; */
1325: /* while ( (*in != '\0')){ */
1326: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1327: /* *out++ = *in++; */
1328: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1329: /* } */
1330: /* *out='\0'; */
1331: /* out=s; */
1332: /* return out; */
1333: /* } */
1334: char *substrchaine(char *out, char *in, char *chain)
1335: {
1336: /* Substract chain 'chain' from 'in', return and output 'out' */
1337: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1338:
1339: char *strloc;
1340:
1341: strcpy (out, in);
1342: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1343: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1344: if(strloc != NULL){
1345: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1346: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1347: /* strcpy (strloc, strloc +strlen(chain));*/
1348: }
1349: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1350: return out;
1351: }
1352:
1353:
1.145 brouard 1354: char *cutl(char *blocc, char *alocc, char *in, char occ)
1355: {
1.187 brouard 1356: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1357: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1358: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1359: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1360: */
1.160 brouard 1361: char *s, *t;
1.145 brouard 1362: t=in;s=in;
1363: while ((*in != occ) && (*in != '\0')){
1364: *alocc++ = *in++;
1365: }
1366: if( *in == occ){
1367: *(alocc)='\0';
1368: s=++in;
1369: }
1370:
1371: if (s == t) {/* occ not found */
1372: *(alocc-(in-s))='\0';
1373: in=s;
1374: }
1375: while ( *in != '\0'){
1376: *blocc++ = *in++;
1377: }
1378:
1379: *blocc='\0';
1380: return t;
1381: }
1.137 brouard 1382: char *cutv(char *blocc, char *alocc, char *in, char occ)
1383: {
1.187 brouard 1384: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1385: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1386: gives blocc="abcdef2ghi" and alocc="j".
1387: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1388: */
1389: char *s, *t;
1390: t=in;s=in;
1391: while (*in != '\0'){
1392: while( *in == occ){
1393: *blocc++ = *in++;
1394: s=in;
1395: }
1396: *blocc++ = *in++;
1397: }
1398: if (s == t) /* occ not found */
1399: *(blocc-(in-s))='\0';
1400: else
1401: *(blocc-(in-s)-1)='\0';
1402: in=s;
1403: while ( *in != '\0'){
1404: *alocc++ = *in++;
1405: }
1406:
1407: *alocc='\0';
1408: return s;
1409: }
1410:
1.126 brouard 1411: int nbocc(char *s, char occ)
1412: {
1413: int i,j=0;
1414: int lg=20;
1415: i=0;
1416: lg=strlen(s);
1417: for(i=0; i<= lg; i++) {
1.234 brouard 1418: if (s[i] == occ ) j++;
1.126 brouard 1419: }
1420: return j;
1421: }
1422:
1.137 brouard 1423: /* void cutv(char *u,char *v, char*t, char occ) */
1424: /* { */
1425: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1426: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1427: /* gives u="abcdef2ghi" and v="j" *\/ */
1428: /* int i,lg,j,p=0; */
1429: /* i=0; */
1430: /* lg=strlen(t); */
1431: /* for(j=0; j<=lg-1; j++) { */
1432: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1433: /* } */
1.126 brouard 1434:
1.137 brouard 1435: /* for(j=0; j<p; j++) { */
1436: /* (u[j] = t[j]); */
1437: /* } */
1438: /* u[p]='\0'; */
1.126 brouard 1439:
1.137 brouard 1440: /* for(j=0; j<= lg; j++) { */
1441: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1442: /* } */
1443: /* } */
1.126 brouard 1444:
1.160 brouard 1445: #ifdef _WIN32
1446: char * strsep(char **pp, const char *delim)
1447: {
1448: char *p, *q;
1449:
1450: if ((p = *pp) == NULL)
1451: return 0;
1452: if ((q = strpbrk (p, delim)) != NULL)
1453: {
1454: *pp = q + 1;
1455: *q = '\0';
1456: }
1457: else
1458: *pp = 0;
1459: return p;
1460: }
1461: #endif
1462:
1.126 brouard 1463: /********************** nrerror ********************/
1464:
1465: void nrerror(char error_text[])
1466: {
1467: fprintf(stderr,"ERREUR ...\n");
1468: fprintf(stderr,"%s\n",error_text);
1469: exit(EXIT_FAILURE);
1470: }
1471: /*********************** vector *******************/
1472: double *vector(int nl, int nh)
1473: {
1474: double *v;
1475: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1476: if (!v) nrerror("allocation failure in vector");
1477: return v-nl+NR_END;
1478: }
1479:
1480: /************************ free vector ******************/
1481: void free_vector(double*v, int nl, int nh)
1482: {
1483: free((FREE_ARG)(v+nl-NR_END));
1484: }
1485:
1486: /************************ivector *******************************/
1487: int *ivector(long nl,long nh)
1488: {
1489: int *v;
1490: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1491: if (!v) nrerror("allocation failure in ivector");
1492: return v-nl+NR_END;
1493: }
1494:
1495: /******************free ivector **************************/
1496: void free_ivector(int *v, long nl, long nh)
1497: {
1498: free((FREE_ARG)(v+nl-NR_END));
1499: }
1500:
1501: /************************lvector *******************************/
1502: long *lvector(long nl,long nh)
1503: {
1504: long *v;
1505: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1506: if (!v) nrerror("allocation failure in ivector");
1507: return v-nl+NR_END;
1508: }
1509:
1510: /******************free lvector **************************/
1511: void free_lvector(long *v, long nl, long nh)
1512: {
1513: free((FREE_ARG)(v+nl-NR_END));
1514: }
1515:
1516: /******************* imatrix *******************************/
1517: int **imatrix(long nrl, long nrh, long ncl, long nch)
1518: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1519: {
1520: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1521: int **m;
1522:
1523: /* allocate pointers to rows */
1524: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1525: if (!m) nrerror("allocation failure 1 in matrix()");
1526: m += NR_END;
1527: m -= nrl;
1528:
1529:
1530: /* allocate rows and set pointers to them */
1531: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1532: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1533: m[nrl] += NR_END;
1534: m[nrl] -= ncl;
1535:
1536: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1537:
1538: /* return pointer to array of pointers to rows */
1539: return m;
1540: }
1541:
1542: /****************** free_imatrix *************************/
1543: void free_imatrix(m,nrl,nrh,ncl,nch)
1544: int **m;
1545: long nch,ncl,nrh,nrl;
1546: /* free an int matrix allocated by imatrix() */
1547: {
1548: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1549: free((FREE_ARG) (m+nrl-NR_END));
1550: }
1551:
1552: /******************* matrix *******************************/
1553: double **matrix(long nrl, long nrh, long ncl, long nch)
1554: {
1555: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1556: double **m;
1557:
1558: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1559: if (!m) nrerror("allocation failure 1 in matrix()");
1560: m += NR_END;
1561: m -= nrl;
1562:
1563: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1564: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1565: m[nrl] += NR_END;
1566: m[nrl] -= ncl;
1567:
1568: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1569: return m;
1.145 brouard 1570: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1571: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1572: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1573: */
1574: }
1575:
1576: /*************************free matrix ************************/
1577: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1578: {
1579: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1580: free((FREE_ARG)(m+nrl-NR_END));
1581: }
1582:
1583: /******************* ma3x *******************************/
1584: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1585: {
1586: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1587: double ***m;
1588:
1589: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1590: if (!m) nrerror("allocation failure 1 in matrix()");
1591: m += NR_END;
1592: m -= nrl;
1593:
1594: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1595: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1596: m[nrl] += NR_END;
1597: m[nrl] -= ncl;
1598:
1599: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1600:
1601: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1602: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1603: m[nrl][ncl] += NR_END;
1604: m[nrl][ncl] -= nll;
1605: for (j=ncl+1; j<=nch; j++)
1606: m[nrl][j]=m[nrl][j-1]+nlay;
1607:
1608: for (i=nrl+1; i<=nrh; i++) {
1609: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1610: for (j=ncl+1; j<=nch; j++)
1611: m[i][j]=m[i][j-1]+nlay;
1612: }
1613: return m;
1614: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1615: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1616: */
1617: }
1618:
1619: /*************************free ma3x ************************/
1620: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1621: {
1622: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1623: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1624: free((FREE_ARG)(m+nrl-NR_END));
1625: }
1626:
1627: /*************** function subdirf ***********/
1628: char *subdirf(char fileres[])
1629: {
1630: /* Caution optionfilefiname is hidden */
1631: strcpy(tmpout,optionfilefiname);
1632: strcat(tmpout,"/"); /* Add to the right */
1633: strcat(tmpout,fileres);
1634: return tmpout;
1635: }
1636:
1637: /*************** function subdirf2 ***********/
1638: char *subdirf2(char fileres[], char *preop)
1639: {
1640:
1641: /* Caution optionfilefiname is hidden */
1642: strcpy(tmpout,optionfilefiname);
1643: strcat(tmpout,"/");
1644: strcat(tmpout,preop);
1645: strcat(tmpout,fileres);
1646: return tmpout;
1647: }
1648:
1649: /*************** function subdirf3 ***********/
1650: char *subdirf3(char fileres[], char *preop, char *preop2)
1651: {
1652:
1653: /* Caution optionfilefiname is hidden */
1654: strcpy(tmpout,optionfilefiname);
1655: strcat(tmpout,"/");
1656: strcat(tmpout,preop);
1657: strcat(tmpout,preop2);
1658: strcat(tmpout,fileres);
1659: return tmpout;
1660: }
1.213 brouard 1661:
1662: /*************** function subdirfext ***********/
1663: char *subdirfext(char fileres[], char *preop, char *postop)
1664: {
1665:
1666: strcpy(tmpout,preop);
1667: strcat(tmpout,fileres);
1668: strcat(tmpout,postop);
1669: return tmpout;
1670: }
1.126 brouard 1671:
1.213 brouard 1672: /*************** function subdirfext3 ***********/
1673: char *subdirfext3(char fileres[], char *preop, char *postop)
1674: {
1675:
1676: /* Caution optionfilefiname is hidden */
1677: strcpy(tmpout,optionfilefiname);
1678: strcat(tmpout,"/");
1679: strcat(tmpout,preop);
1680: strcat(tmpout,fileres);
1681: strcat(tmpout,postop);
1682: return tmpout;
1683: }
1684:
1.162 brouard 1685: char *asc_diff_time(long time_sec, char ascdiff[])
1686: {
1687: long sec_left, days, hours, minutes;
1688: days = (time_sec) / (60*60*24);
1689: sec_left = (time_sec) % (60*60*24);
1690: hours = (sec_left) / (60*60) ;
1691: sec_left = (sec_left) %(60*60);
1692: minutes = (sec_left) /60;
1693: sec_left = (sec_left) % (60);
1694: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1695: return ascdiff;
1696: }
1697:
1.126 brouard 1698: /***************** f1dim *************************/
1699: extern int ncom;
1700: extern double *pcom,*xicom;
1701: extern double (*nrfunc)(double []);
1702:
1703: double f1dim(double x)
1704: {
1705: int j;
1706: double f;
1707: double *xt;
1708:
1709: xt=vector(1,ncom);
1710: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1711: f=(*nrfunc)(xt);
1712: free_vector(xt,1,ncom);
1713: return f;
1714: }
1715:
1716: /*****************brent *************************/
1717: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1718: {
1719: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1720: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1721: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1722: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1723: * returned function value.
1724: */
1.126 brouard 1725: int iter;
1726: double a,b,d,etemp;
1.159 brouard 1727: double fu=0,fv,fw,fx;
1.164 brouard 1728: double ftemp=0.;
1.126 brouard 1729: double p,q,r,tol1,tol2,u,v,w,x,xm;
1730: double e=0.0;
1731:
1732: a=(ax < cx ? ax : cx);
1733: b=(ax > cx ? ax : cx);
1734: x=w=v=bx;
1735: fw=fv=fx=(*f)(x);
1736: for (iter=1;iter<=ITMAX;iter++) {
1737: xm=0.5*(a+b);
1738: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1739: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1740: printf(".");fflush(stdout);
1741: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1742: #ifdef DEBUGBRENT
1.126 brouard 1743: 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);
1744: 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);
1745: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1746: #endif
1747: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1748: *xmin=x;
1749: return fx;
1750: }
1751: ftemp=fu;
1752: if (fabs(e) > tol1) {
1753: r=(x-w)*(fx-fv);
1754: q=(x-v)*(fx-fw);
1755: p=(x-v)*q-(x-w)*r;
1756: q=2.0*(q-r);
1757: if (q > 0.0) p = -p;
1758: q=fabs(q);
1759: etemp=e;
1760: e=d;
1761: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1762: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1763: else {
1.224 brouard 1764: d=p/q;
1765: u=x+d;
1766: if (u-a < tol2 || b-u < tol2)
1767: d=SIGN(tol1,xm-x);
1.126 brouard 1768: }
1769: } else {
1770: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1771: }
1772: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1773: fu=(*f)(u);
1774: if (fu <= fx) {
1775: if (u >= x) a=x; else b=x;
1776: SHFT(v,w,x,u)
1.183 brouard 1777: SHFT(fv,fw,fx,fu)
1778: } else {
1779: if (u < x) a=u; else b=u;
1780: if (fu <= fw || w == x) {
1.224 brouard 1781: v=w;
1782: w=u;
1783: fv=fw;
1784: fw=fu;
1.183 brouard 1785: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1786: v=u;
1787: fv=fu;
1.183 brouard 1788: }
1789: }
1.126 brouard 1790: }
1791: nrerror("Too many iterations in brent");
1792: *xmin=x;
1793: return fx;
1794: }
1795:
1796: /****************** mnbrak ***********************/
1797:
1798: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1799: double (*func)(double))
1.183 brouard 1800: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1801: the downhill direction (defined by the function as evaluated at the initial points) and returns
1802: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1803: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1804: */
1.126 brouard 1805: double ulim,u,r,q, dum;
1806: double fu;
1.187 brouard 1807:
1808: double scale=10.;
1809: int iterscale=0;
1810:
1811: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1812: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1813:
1814:
1815: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1816: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1817: /* *bx = *ax - (*ax - *bx)/scale; */
1818: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1819: /* } */
1820:
1.126 brouard 1821: if (*fb > *fa) {
1822: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1823: SHFT(dum,*fb,*fa,dum)
1824: }
1.126 brouard 1825: *cx=(*bx)+GOLD*(*bx-*ax);
1826: *fc=(*func)(*cx);
1.183 brouard 1827: #ifdef DEBUG
1.224 brouard 1828: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1829: 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 1830: #endif
1.224 brouard 1831: 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 1832: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1833: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1834: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1835: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1836: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1837: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1838: fu=(*func)(u);
1.163 brouard 1839: #ifdef DEBUG
1840: /* f(x)=A(x-u)**2+f(u) */
1841: double A, fparabu;
1842: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1843: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1844: 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);
1845: 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 1846: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1847: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1848: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1849: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1850: #endif
1.184 brouard 1851: #ifdef MNBRAKORIGINAL
1.183 brouard 1852: #else
1.191 brouard 1853: /* if (fu > *fc) { */
1854: /* #ifdef DEBUG */
1855: /* printf("mnbrak4 fu > fc \n"); */
1856: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1857: /* #endif */
1858: /* /\* 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 *\\/ *\/ */
1859: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1860: /* dum=u; /\* Shifting c and u *\/ */
1861: /* u = *cx; */
1862: /* *cx = dum; */
1863: /* dum = fu; */
1864: /* fu = *fc; */
1865: /* *fc =dum; */
1866: /* } else { /\* end *\/ */
1867: /* #ifdef DEBUG */
1868: /* printf("mnbrak3 fu < fc \n"); */
1869: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1870: /* #endif */
1871: /* dum=u; /\* Shifting c and u *\/ */
1872: /* u = *cx; */
1873: /* *cx = dum; */
1874: /* dum = fu; */
1875: /* fu = *fc; */
1876: /* *fc =dum; */
1877: /* } */
1.224 brouard 1878: #ifdef DEBUGMNBRAK
1879: double A, fparabu;
1880: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1881: fparabu= *fa - A*(*ax-u)*(*ax-u);
1882: 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);
1883: 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 1884: #endif
1.191 brouard 1885: dum=u; /* Shifting c and u */
1886: u = *cx;
1887: *cx = dum;
1888: dum = fu;
1889: fu = *fc;
1890: *fc =dum;
1.183 brouard 1891: #endif
1.162 brouard 1892: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1893: #ifdef DEBUG
1.224 brouard 1894: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1895: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1896: #endif
1.126 brouard 1897: fu=(*func)(u);
1898: if (fu < *fc) {
1.183 brouard 1899: #ifdef DEBUG
1.224 brouard 1900: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1901: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1902: #endif
1903: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1904: SHFT(*fb,*fc,fu,(*func)(u))
1905: #ifdef DEBUG
1906: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1907: #endif
1908: }
1.162 brouard 1909: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1910: #ifdef DEBUG
1.224 brouard 1911: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1912: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1913: #endif
1.126 brouard 1914: u=ulim;
1915: fu=(*func)(u);
1.183 brouard 1916: } else { /* u could be left to b (if r > q parabola has a maximum) */
1917: #ifdef DEBUG
1.224 brouard 1918: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1919: 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 1920: #endif
1.126 brouard 1921: u=(*cx)+GOLD*(*cx-*bx);
1922: fu=(*func)(u);
1.224 brouard 1923: #ifdef DEBUG
1924: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1925: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1926: #endif
1.183 brouard 1927: } /* end tests */
1.126 brouard 1928: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1929: SHFT(*fa,*fb,*fc,fu)
1930: #ifdef DEBUG
1.224 brouard 1931: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1932: 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 1933: #endif
1934: } /* 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 1935: }
1936:
1937: /*************** linmin ************************/
1.162 brouard 1938: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1939: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1940: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1941: the value of func at the returned location p . This is actually all accomplished by calling the
1942: routines mnbrak and brent .*/
1.126 brouard 1943: int ncom;
1944: double *pcom,*xicom;
1945: double (*nrfunc)(double []);
1946:
1.224 brouard 1947: #ifdef LINMINORIGINAL
1.126 brouard 1948: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1949: #else
1950: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1951: #endif
1.126 brouard 1952: {
1953: double brent(double ax, double bx, double cx,
1954: double (*f)(double), double tol, double *xmin);
1955: double f1dim(double x);
1956: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1957: double *fc, double (*func)(double));
1958: int j;
1959: double xx,xmin,bx,ax;
1960: double fx,fb,fa;
1.187 brouard 1961:
1.203 brouard 1962: #ifdef LINMINORIGINAL
1963: #else
1964: double scale=10., axs, xxs; /* Scale added for infinity */
1965: #endif
1966:
1.126 brouard 1967: ncom=n;
1968: pcom=vector(1,n);
1969: xicom=vector(1,n);
1970: nrfunc=func;
1971: for (j=1;j<=n;j++) {
1972: pcom[j]=p[j];
1.202 brouard 1973: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1974: }
1.187 brouard 1975:
1.203 brouard 1976: #ifdef LINMINORIGINAL
1977: xx=1.;
1978: #else
1979: axs=0.0;
1980: xxs=1.;
1981: do{
1982: xx= xxs;
1983: #endif
1.187 brouard 1984: ax=0.;
1985: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1986: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1987: /* 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)) */
1988: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1989: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1990: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
1991: /* 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 1992: #ifdef LINMINORIGINAL
1993: #else
1994: if (fx != fx){
1.224 brouard 1995: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
1996: printf("|");
1997: fprintf(ficlog,"|");
1.203 brouard 1998: #ifdef DEBUGLINMIN
1.224 brouard 1999: 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 2000: #endif
2001: }
1.224 brouard 2002: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2003: #endif
2004:
1.191 brouard 2005: #ifdef DEBUGLINMIN
2006: 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 2007: 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 2008: #endif
1.224 brouard 2009: #ifdef LINMINORIGINAL
2010: #else
2011: if(fb == fx){ /* Flat function in the direction */
2012: xmin=xx;
2013: *flat=1;
2014: }else{
2015: *flat=0;
2016: #endif
2017: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2018: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2019: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2020: /* fmin = f(p[j] + xmin * xi[j]) */
2021: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2022: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2023: #ifdef DEBUG
1.224 brouard 2024: 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);
2025: 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);
2026: #endif
2027: #ifdef LINMINORIGINAL
2028: #else
2029: }
1.126 brouard 2030: #endif
1.191 brouard 2031: #ifdef DEBUGLINMIN
2032: printf("linmin end ");
1.202 brouard 2033: fprintf(ficlog,"linmin end ");
1.191 brouard 2034: #endif
1.126 brouard 2035: for (j=1;j<=n;j++) {
1.203 brouard 2036: #ifdef LINMINORIGINAL
2037: xi[j] *= xmin;
2038: #else
2039: #ifdef DEBUGLINMIN
2040: if(xxs <1.0)
2041: printf(" before xi[%d]=%12.8f", j,xi[j]);
2042: #endif
2043: 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) */
2044: #ifdef DEBUGLINMIN
2045: if(xxs <1.0)
2046: 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 );
2047: #endif
2048: #endif
1.187 brouard 2049: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2050: }
1.191 brouard 2051: #ifdef DEBUGLINMIN
1.203 brouard 2052: printf("\n");
1.191 brouard 2053: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2054: 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 2055: for (j=1;j<=n;j++) {
1.202 brouard 2056: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2057: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2058: if(j % ncovmodel == 0){
1.191 brouard 2059: printf("\n");
1.202 brouard 2060: fprintf(ficlog,"\n");
2061: }
1.191 brouard 2062: }
1.203 brouard 2063: #else
1.191 brouard 2064: #endif
1.126 brouard 2065: free_vector(xicom,1,n);
2066: free_vector(pcom,1,n);
2067: }
2068:
2069:
2070: /*************** powell ************************/
1.162 brouard 2071: /*
2072: Minimization of a function func of n variables. Input consists of an initial starting point
2073: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2074: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2075: such that failure to decrease by more than this amount on one iteration signals doneness. On
2076: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2077: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2078: */
1.224 brouard 2079: #ifdef LINMINORIGINAL
2080: #else
2081: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2082: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2083: #endif
1.126 brouard 2084: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2085: double (*func)(double []))
2086: {
1.224 brouard 2087: #ifdef LINMINORIGINAL
2088: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2089: double (*func)(double []));
1.224 brouard 2090: #else
1.241 brouard 2091: void linmin(double p[], double xi[], int n, double *fret,
2092: double (*func)(double []),int *flat);
1.224 brouard 2093: #endif
1.239 brouard 2094: int i,ibig,j,jk,k;
1.126 brouard 2095: double del,t,*pt,*ptt,*xit;
1.181 brouard 2096: double directest;
1.126 brouard 2097: double fp,fptt;
2098: double *xits;
2099: int niterf, itmp;
1.224 brouard 2100: #ifdef LINMINORIGINAL
2101: #else
2102:
2103: flatdir=ivector(1,n);
2104: for (j=1;j<=n;j++) flatdir[j]=0;
2105: #endif
1.126 brouard 2106:
2107: pt=vector(1,n);
2108: ptt=vector(1,n);
2109: xit=vector(1,n);
2110: xits=vector(1,n);
2111: *fret=(*func)(p);
2112: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2113: rcurr_time = time(NULL);
1.126 brouard 2114: for (*iter=1;;++(*iter)) {
1.187 brouard 2115: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2116: ibig=0;
2117: del=0.0;
1.157 brouard 2118: rlast_time=rcurr_time;
2119: /* (void) gettimeofday(&curr_time,&tzp); */
2120: rcurr_time = time(NULL);
2121: curr_time = *localtime(&rcurr_time);
2122: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2123: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2124: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2125: for (i=1;i<=n;i++) {
1.126 brouard 2126: fprintf(ficrespow," %.12lf", p[i]);
2127: }
1.239 brouard 2128: fprintf(ficrespow,"\n");fflush(ficrespow);
2129: printf("\n#model= 1 + age ");
2130: fprintf(ficlog,"\n#model= 1 + age ");
2131: if(nagesqr==1){
1.241 brouard 2132: printf(" + age*age ");
2133: fprintf(ficlog," + age*age ");
1.239 brouard 2134: }
2135: for(j=1;j <=ncovmodel-2;j++){
2136: if(Typevar[j]==0) {
2137: printf(" + V%d ",Tvar[j]);
2138: fprintf(ficlog," + V%d ",Tvar[j]);
2139: }else if(Typevar[j]==1) {
2140: printf(" + V%d*age ",Tvar[j]);
2141: fprintf(ficlog," + V%d*age ",Tvar[j]);
2142: }else if(Typevar[j]==2) {
2143: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2144: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2145: }
2146: }
1.126 brouard 2147: printf("\n");
1.239 brouard 2148: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2149: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2150: fprintf(ficlog,"\n");
1.239 brouard 2151: for(i=1,jk=1; i <=nlstate; i++){
2152: for(k=1; k <=(nlstate+ndeath); k++){
2153: if (k != i) {
2154: printf("%d%d ",i,k);
2155: fprintf(ficlog,"%d%d ",i,k);
2156: for(j=1; j <=ncovmodel; j++){
2157: printf("%12.7f ",p[jk]);
2158: fprintf(ficlog,"%12.7f ",p[jk]);
2159: jk++;
2160: }
2161: printf("\n");
2162: fprintf(ficlog,"\n");
2163: }
2164: }
2165: }
1.241 brouard 2166: if(*iter <=3 && *iter >1){
1.157 brouard 2167: tml = *localtime(&rcurr_time);
2168: strcpy(strcurr,asctime(&tml));
2169: rforecast_time=rcurr_time;
1.126 brouard 2170: itmp = strlen(strcurr);
2171: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2172: strcurr[itmp-1]='\0';
1.162 brouard 2173: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2174: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2175: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2176: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2177: forecast_time = *localtime(&rforecast_time);
2178: strcpy(strfor,asctime(&forecast_time));
2179: itmp = strlen(strfor);
2180: if(strfor[itmp-1]=='\n')
2181: strfor[itmp-1]='\0';
2182: 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);
2183: 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 2184: }
2185: }
1.187 brouard 2186: for (i=1;i<=n;i++) { /* For each direction i */
2187: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2188: fptt=(*fret);
2189: #ifdef DEBUG
1.203 brouard 2190: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2191: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2192: #endif
1.203 brouard 2193: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2194: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2195: #ifdef LINMINORIGINAL
1.188 brouard 2196: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2197: #else
2198: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2199: flatdir[i]=flat; /* Function is vanishing in that direction i */
2200: #endif
2201: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2202: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2203: /* because that direction will be replaced unless the gain del is small */
2204: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2205: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2206: /* with the new direction. */
2207: del=fabs(fptt-(*fret));
2208: ibig=i;
1.126 brouard 2209: }
2210: #ifdef DEBUG
2211: printf("%d %.12e",i,(*fret));
2212: fprintf(ficlog,"%d %.12e",i,(*fret));
2213: for (j=1;j<=n;j++) {
1.224 brouard 2214: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2215: printf(" x(%d)=%.12e",j,xit[j]);
2216: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2217: }
2218: for(j=1;j<=n;j++) {
1.225 brouard 2219: printf(" p(%d)=%.12e",j,p[j]);
2220: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2221: }
2222: printf("\n");
2223: fprintf(ficlog,"\n");
2224: #endif
1.187 brouard 2225: } /* end loop on each direction i */
2226: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2227: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2228: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2229: for(j=1;j<=n;j++) {
1.225 brouard 2230: if(flatdir[j] >0){
2231: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2232: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2233: }
2234: /* printf("\n"); */
2235: /* fprintf(ficlog,"\n"); */
2236: }
1.243 brouard 2237: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2238: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2239: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2240: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2241: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2242: /* decreased of more than 3.84 */
2243: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2244: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2245: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2246:
1.188 brouard 2247: /* Starting the program with initial values given by a former maximization will simply change */
2248: /* the scales of the directions and the directions, because the are reset to canonical directions */
2249: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2250: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2251: #ifdef DEBUG
2252: int k[2],l;
2253: k[0]=1;
2254: k[1]=-1;
2255: printf("Max: %.12e",(*func)(p));
2256: fprintf(ficlog,"Max: %.12e",(*func)(p));
2257: for (j=1;j<=n;j++) {
2258: printf(" %.12e",p[j]);
2259: fprintf(ficlog," %.12e",p[j]);
2260: }
2261: printf("\n");
2262: fprintf(ficlog,"\n");
2263: for(l=0;l<=1;l++) {
2264: for (j=1;j<=n;j++) {
2265: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2266: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2267: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2268: }
2269: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2270: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2271: }
2272: #endif
2273:
1.224 brouard 2274: #ifdef LINMINORIGINAL
2275: #else
2276: free_ivector(flatdir,1,n);
2277: #endif
1.126 brouard 2278: free_vector(xit,1,n);
2279: free_vector(xits,1,n);
2280: free_vector(ptt,1,n);
2281: free_vector(pt,1,n);
2282: return;
1.192 brouard 2283: } /* enough precision */
1.240 brouard 2284: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2285: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2286: ptt[j]=2.0*p[j]-pt[j];
2287: xit[j]=p[j]-pt[j];
2288: pt[j]=p[j];
2289: }
1.181 brouard 2290: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2291: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2292: if (*iter <=4) {
1.225 brouard 2293: #else
2294: #endif
1.224 brouard 2295: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2296: #else
1.161 brouard 2297: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2298: #endif
1.162 brouard 2299: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2300: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2301: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2302: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2303: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2304: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2305: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2306: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2307: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2308: /* Even if f3 <f1, directest can be negative and t >0 */
2309: /* mu² and del² are equal when f3=f1 */
2310: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2311: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2312: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2313: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2314: #ifdef NRCORIGINAL
2315: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2316: #else
2317: 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 2318: t= t- del*SQR(fp-fptt);
1.183 brouard 2319: #endif
1.202 brouard 2320: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2321: #ifdef DEBUG
1.181 brouard 2322: 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);
2323: 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 2324: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2325: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2326: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2327: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2328: 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);
2329: 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);
2330: #endif
1.183 brouard 2331: #ifdef POWELLORIGINAL
2332: if (t < 0.0) { /* Then we use it for new direction */
2333: #else
1.182 brouard 2334: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2335: 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 2336: 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 2337: 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 2338: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2339: }
1.181 brouard 2340: if (directest < 0.0) { /* Then we use it for new direction */
2341: #endif
1.191 brouard 2342: #ifdef DEBUGLINMIN
1.234 brouard 2343: printf("Before linmin in direction P%d-P0\n",n);
2344: for (j=1;j<=n;j++) {
2345: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2346: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2347: if(j % ncovmodel == 0){
2348: printf("\n");
2349: fprintf(ficlog,"\n");
2350: }
2351: }
1.224 brouard 2352: #endif
2353: #ifdef LINMINORIGINAL
1.234 brouard 2354: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2355: #else
1.234 brouard 2356: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2357: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2358: #endif
1.234 brouard 2359:
1.191 brouard 2360: #ifdef DEBUGLINMIN
1.234 brouard 2361: for (j=1;j<=n;j++) {
2362: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2363: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2364: if(j % ncovmodel == 0){
2365: printf("\n");
2366: fprintf(ficlog,"\n");
2367: }
2368: }
1.224 brouard 2369: #endif
1.234 brouard 2370: for (j=1;j<=n;j++) {
2371: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2372: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2373: }
1.224 brouard 2374: #ifdef LINMINORIGINAL
2375: #else
1.234 brouard 2376: for (j=1, flatd=0;j<=n;j++) {
2377: if(flatdir[j]>0)
2378: flatd++;
2379: }
2380: if(flatd >0){
2381: printf("%d flat directions\n",flatd);
2382: fprintf(ficlog,"%d flat directions\n",flatd);
2383: for (j=1;j<=n;j++) {
2384: if(flatdir[j]>0){
2385: printf("%d ",j);
2386: fprintf(ficlog,"%d ",j);
2387: }
2388: }
2389: printf("\n");
2390: fprintf(ficlog,"\n");
2391: }
1.191 brouard 2392: #endif
1.234 brouard 2393: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2394: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2395:
1.126 brouard 2396: #ifdef DEBUG
1.234 brouard 2397: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2398: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2399: for(j=1;j<=n;j++){
2400: printf(" %lf",xit[j]);
2401: fprintf(ficlog," %lf",xit[j]);
2402: }
2403: printf("\n");
2404: fprintf(ficlog,"\n");
1.126 brouard 2405: #endif
1.192 brouard 2406: } /* end of t or directest negative */
1.224 brouard 2407: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2408: #else
1.234 brouard 2409: } /* end if (fptt < fp) */
1.192 brouard 2410: #endif
1.225 brouard 2411: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2412: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2413: #else
1.224 brouard 2414: #endif
1.234 brouard 2415: } /* loop iteration */
1.126 brouard 2416: }
1.234 brouard 2417:
1.126 brouard 2418: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2419:
1.235 brouard 2420: 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 2421: {
1.235 brouard 2422: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2423: (and selected quantitative values in nres)
2424: by left multiplying the unit
1.234 brouard 2425: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2426: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2427: /* Wx is row vector: population in state 1, population in state 2, population dead */
2428: /* or prevalence in state 1, prevalence in state 2, 0 */
2429: /* newm is the matrix after multiplications, its rows are identical at a factor */
2430: /* Initial matrix pimij */
2431: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2432: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2433: /* 0, 0 , 1} */
2434: /*
2435: * and after some iteration: */
2436: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2437: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2438: /* 0, 0 , 1} */
2439: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2440: /* {0.51571254859325999, 0.4842874514067399, */
2441: /* 0.51326036147820708, 0.48673963852179264} */
2442: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2443:
1.126 brouard 2444: int i, ii,j,k;
1.209 brouard 2445: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2446: /* double **matprod2(); */ /* test */
1.218 brouard 2447: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2448: double **newm;
1.209 brouard 2449: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2450: int ncvloop=0;
1.169 brouard 2451:
1.209 brouard 2452: min=vector(1,nlstate);
2453: max=vector(1,nlstate);
2454: meandiff=vector(1,nlstate);
2455:
1.218 brouard 2456: /* Starting with matrix unity */
1.126 brouard 2457: for (ii=1;ii<=nlstate+ndeath;ii++)
2458: for (j=1;j<=nlstate+ndeath;j++){
2459: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2460: }
1.169 brouard 2461:
2462: cov[1]=1.;
2463:
2464: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2465: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2466: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2467: ncvloop++;
1.126 brouard 2468: newm=savm;
2469: /* Covariates have to be included here again */
1.138 brouard 2470: cov[2]=agefin;
1.187 brouard 2471: if(nagesqr==1)
2472: cov[3]= agefin*agefin;;
1.234 brouard 2473: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2474: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2475: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2476: /* 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 2477: }
2478: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2479: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2480: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2481: /* 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 2482: }
1.237 brouard 2483: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2484: if(Dummy[Tvar[Tage[k]]]){
2485: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2486: } else{
1.235 brouard 2487: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2488: }
1.235 brouard 2489: /* 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 2490: }
1.237 brouard 2491: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2492: /* 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 2493: if(Dummy[Tvard[k][1]==0]){
2494: if(Dummy[Tvard[k][2]==0]){
2495: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2496: }else{
2497: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2498: }
2499: }else{
2500: if(Dummy[Tvard[k][2]==0]){
2501: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2502: }else{
2503: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2504: }
2505: }
1.234 brouard 2506: }
1.138 brouard 2507: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2508: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2509: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2510: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2511: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2512: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2513: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2514:
1.126 brouard 2515: savm=oldm;
2516: oldm=newm;
1.209 brouard 2517:
2518: for(j=1; j<=nlstate; j++){
2519: max[j]=0.;
2520: min[j]=1.;
2521: }
2522: for(i=1;i<=nlstate;i++){
2523: sumnew=0;
2524: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2525: for(j=1; j<=nlstate; j++){
2526: prlim[i][j]= newm[i][j]/(1-sumnew);
2527: max[j]=FMAX(max[j],prlim[i][j]);
2528: min[j]=FMIN(min[j],prlim[i][j]);
2529: }
2530: }
2531:
1.126 brouard 2532: maxmax=0.;
1.209 brouard 2533: for(j=1; j<=nlstate; j++){
2534: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2535: maxmax=FMAX(maxmax,meandiff[j]);
2536: /* 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 2537: } /* j loop */
1.203 brouard 2538: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2539: /* 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 2540: if(maxmax < ftolpl){
1.209 brouard 2541: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2542: free_vector(min,1,nlstate);
2543: free_vector(max,1,nlstate);
2544: free_vector(meandiff,1,nlstate);
1.126 brouard 2545: return prlim;
2546: }
1.169 brouard 2547: } /* age loop */
1.208 brouard 2548: /* After some age loop it doesn't converge */
1.209 brouard 2549: 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 2550: 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 2551: /* 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); */
2552: free_vector(min,1,nlstate);
2553: free_vector(max,1,nlstate);
2554: free_vector(meandiff,1,nlstate);
1.208 brouard 2555:
1.169 brouard 2556: return prlim; /* should not reach here */
1.126 brouard 2557: }
2558:
1.217 brouard 2559:
2560: /**** Back Prevalence limit (stable or period prevalence) ****************/
2561:
1.218 brouard 2562: /* 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) */
2563: /* 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 2564: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2565: {
1.218 brouard 2566: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2567: matrix by transitions matrix until convergence is reached with precision ftolpl */
2568: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2569: /* Wx is row vector: population in state 1, population in state 2, population dead */
2570: /* or prevalence in state 1, prevalence in state 2, 0 */
2571: /* newm is the matrix after multiplications, its rows are identical at a factor */
2572: /* Initial matrix pimij */
2573: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2574: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2575: /* 0, 0 , 1} */
2576: /*
2577: * and after some iteration: */
2578: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2579: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2580: /* 0, 0 , 1} */
2581: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2582: /* {0.51571254859325999, 0.4842874514067399, */
2583: /* 0.51326036147820708, 0.48673963852179264} */
2584: /* If we start from prlim again, prlim tends to a constant matrix */
2585:
2586: int i, ii,j,k;
1.247 ! brouard 2587: int first=0;
1.217 brouard 2588: double *min, *max, *meandiff, maxmax,sumnew=0.;
2589: /* double **matprod2(); */ /* test */
2590: double **out, cov[NCOVMAX+1], **bmij();
2591: double **newm;
1.218 brouard 2592: double **dnewm, **doldm, **dsavm; /* for use */
2593: double **oldm, **savm; /* for use */
2594:
1.217 brouard 2595: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2596: int ncvloop=0;
2597:
2598: min=vector(1,nlstate);
2599: max=vector(1,nlstate);
2600: meandiff=vector(1,nlstate);
2601:
1.218 brouard 2602: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2603: oldm=oldms; savm=savms;
2604:
2605: /* Starting with matrix unity */
2606: for (ii=1;ii<=nlstate+ndeath;ii++)
2607: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2608: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2609: }
2610:
2611: cov[1]=1.;
2612:
2613: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2614: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2615: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2616: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2617: ncvloop++;
1.218 brouard 2618: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2619: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2620: /* Covariates have to be included here again */
2621: cov[2]=agefin;
2622: if(nagesqr==1)
2623: cov[3]= agefin*agefin;;
1.242 brouard 2624: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2625: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2626: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2627: /* 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)); */
2628: }
2629: /* for (k=1; k<=cptcovn;k++) { */
2630: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2631: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2632: /* /\* 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])]); *\/ */
2633: /* } */
2634: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2635: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2636: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2637: /* 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]); */
2638: }
2639: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2640: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2641: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2642: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2643: for (k=1; k<=cptcovage;k++){ /* For product with age */
2644: if(Dummy[Tvar[Tage[k]]]){
2645: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2646: } else{
2647: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2648: }
2649: /* 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]); */
2650: }
2651: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2652: /* 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]); */
2653: if(Dummy[Tvard[k][1]==0]){
2654: if(Dummy[Tvard[k][2]==0]){
2655: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2656: }else{
2657: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2658: }
2659: }else{
2660: if(Dummy[Tvard[k][2]==0]){
2661: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2662: }else{
2663: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2664: }
2665: }
1.217 brouard 2666: }
2667:
2668: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2669: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2670: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2671: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2672: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2673: /* ij should be linked to the correct index of cov */
2674: /* age and covariate values ij are in 'cov', but we need to pass
2675: * ij for the observed prevalence at age and status and covariate
2676: * number: prevacurrent[(int)agefin][ii][ij]
2677: */
2678: /* 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 *\/ */
2679: /* 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 *\/ */
2680: 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 2681: savm=oldm;
2682: oldm=newm;
2683: for(j=1; j<=nlstate; j++){
2684: max[j]=0.;
2685: min[j]=1.;
2686: }
2687: for(j=1; j<=nlstate; j++){
2688: for(i=1;i<=nlstate;i++){
1.234 brouard 2689: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2690: bprlim[i][j]= newm[i][j];
2691: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2692: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2693: }
2694: }
1.218 brouard 2695:
1.217 brouard 2696: maxmax=0.;
2697: for(i=1; i<=nlstate; i++){
2698: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2699: maxmax=FMAX(maxmax,meandiff[i]);
2700: /* 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); */
2701: } /* j loop */
2702: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2703: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2704: if(maxmax < ftolpl){
1.220 brouard 2705: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2706: free_vector(min,1,nlstate);
2707: free_vector(max,1,nlstate);
2708: free_vector(meandiff,1,nlstate);
2709: return bprlim;
2710: }
2711: } /* age loop */
2712: /* After some age loop it doesn't converge */
1.247 ! brouard 2713: if(first){
! 2714: first=1;
! 2715: 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\
! 2716: 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);
! 2717: }
! 2718: 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 2719: 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);
2720: /* 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); */
2721: free_vector(min,1,nlstate);
2722: free_vector(max,1,nlstate);
2723: free_vector(meandiff,1,nlstate);
2724:
2725: return bprlim; /* should not reach here */
2726: }
2727:
1.126 brouard 2728: /*************** transition probabilities ***************/
2729:
2730: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2731: {
1.138 brouard 2732: /* According to parameters values stored in x and the covariate's values stored in cov,
2733: computes the probability to be observed in state j being in state i by appying the
2734: model to the ncovmodel covariates (including constant and age).
2735: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2736: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2737: ncth covariate in the global vector x is given by the formula:
2738: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2739: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2740: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2741: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2742: Outputs ps[i][j] the probability to be observed in j being in j according to
2743: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2744: */
2745: double s1, lnpijopii;
1.126 brouard 2746: /*double t34;*/
1.164 brouard 2747: int i,j, nc, ii, jj;
1.126 brouard 2748:
1.223 brouard 2749: for(i=1; i<= nlstate; i++){
2750: for(j=1; j<i;j++){
2751: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2752: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2753: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2754: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2755: }
2756: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2757: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2758: }
2759: for(j=i+1; j<=nlstate+ndeath;j++){
2760: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2761: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2762: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2763: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2764: }
2765: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2766: }
2767: }
1.218 brouard 2768:
1.223 brouard 2769: for(i=1; i<= nlstate; i++){
2770: s1=0;
2771: for(j=1; j<i; j++){
2772: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2773: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2774: }
2775: for(j=i+1; j<=nlstate+ndeath; j++){
2776: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2777: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2778: }
2779: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2780: ps[i][i]=1./(s1+1.);
2781: /* Computing other pijs */
2782: for(j=1; j<i; j++)
2783: ps[i][j]= exp(ps[i][j])*ps[i][i];
2784: for(j=i+1; j<=nlstate+ndeath; j++)
2785: ps[i][j]= exp(ps[i][j])*ps[i][i];
2786: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2787: } /* end i */
1.218 brouard 2788:
1.223 brouard 2789: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2790: for(jj=1; jj<= nlstate+ndeath; jj++){
2791: ps[ii][jj]=0;
2792: ps[ii][ii]=1;
2793: }
2794: }
1.218 brouard 2795:
2796:
1.223 brouard 2797: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2798: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2799: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2800: /* } */
2801: /* printf("\n "); */
2802: /* } */
2803: /* printf("\n ");printf("%lf ",cov[2]);*/
2804: /*
2805: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2806: goto end;*/
1.223 brouard 2807: return ps;
1.126 brouard 2808: }
2809:
1.218 brouard 2810: /*************** backward transition probabilities ***************/
2811:
2812: /* 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 ) */
2813: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2814: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2815: {
1.222 brouard 2816: /* Computes the backward probability at age agefin and covariate ij
2817: * and returns in **ps as well as **bmij.
2818: */
1.218 brouard 2819: int i, ii, j,k;
1.222 brouard 2820:
2821: double **out, **pmij();
2822: double sumnew=0.;
1.218 brouard 2823: double agefin;
1.222 brouard 2824:
2825: double **dnewm, **dsavm, **doldm;
2826: double **bbmij;
2827:
1.218 brouard 2828: doldm=ddoldms; /* global pointers */
1.222 brouard 2829: dnewm=ddnewms;
2830: dsavm=ddsavms;
2831:
2832: agefin=cov[2];
2833: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2834: the observed prevalence (with this covariate ij) */
2835: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2836: /* We do have the matrix Px in savm and we need pij */
2837: for (j=1;j<=nlstate+ndeath;j++){
2838: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2839: for (ii=1;ii<=nlstate;ii++){
2840: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2841: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2842: for (ii=1;ii<=nlstate+ndeath;ii++){
2843: if(sumnew >= 1.e-10){
2844: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2845: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2846: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2847: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2848: /* }else */
2849: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2850: }else{
1.242 brouard 2851: ;
2852: /* 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 2853: }
2854: } /*End ii */
2855: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2856: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2857: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2858: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2859: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2860: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2861: /* left Product of this matrix by diag matrix of prevalences (savm) */
2862: for (j=1;j<=nlstate+ndeath;j++){
2863: for (ii=1;ii<=nlstate+ndeath;ii++){
2864: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2865: }
2866: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2867: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2868: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2869: /* end bmij */
2870: return ps;
1.218 brouard 2871: }
1.217 brouard 2872: /*************** transition probabilities ***************/
2873:
1.218 brouard 2874: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2875: {
2876: /* According to parameters values stored in x and the covariate's values stored in cov,
2877: computes the probability to be observed in state j being in state i by appying the
2878: model to the ncovmodel covariates (including constant and age).
2879: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2880: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2881: ncth covariate in the global vector x is given by the formula:
2882: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2883: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2884: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2885: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2886: Outputs ps[i][j] the probability to be observed in j being in j according to
2887: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2888: */
2889: double s1, lnpijopii;
2890: /*double t34;*/
2891: int i,j, nc, ii, jj;
2892:
1.234 brouard 2893: for(i=1; i<= nlstate; i++){
2894: for(j=1; j<i;j++){
2895: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2896: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2897: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2898: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2899: }
2900: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2901: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2902: }
2903: for(j=i+1; j<=nlstate+ndeath;j++){
2904: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2905: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2906: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2907: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2908: }
2909: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2910: }
2911: }
2912:
2913: for(i=1; i<= nlstate; i++){
2914: s1=0;
2915: for(j=1; j<i; j++){
2916: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2917: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2918: }
2919: for(j=i+1; j<=nlstate+ndeath; j++){
2920: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2921: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2922: }
2923: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2924: ps[i][i]=1./(s1+1.);
2925: /* Computing other pijs */
2926: for(j=1; j<i; j++)
2927: ps[i][j]= exp(ps[i][j])*ps[i][i];
2928: for(j=i+1; j<=nlstate+ndeath; j++)
2929: ps[i][j]= exp(ps[i][j])*ps[i][i];
2930: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2931: } /* end i */
2932:
2933: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2934: for(jj=1; jj<= nlstate+ndeath; jj++){
2935: ps[ii][jj]=0;
2936: ps[ii][ii]=1;
2937: }
2938: }
2939: /* Added for backcast */ /* Transposed matrix too */
2940: for(jj=1; jj<= nlstate+ndeath; jj++){
2941: s1=0.;
2942: for(ii=1; ii<= nlstate+ndeath; ii++){
2943: s1+=ps[ii][jj];
2944: }
2945: for(ii=1; ii<= nlstate; ii++){
2946: ps[ii][jj]=ps[ii][jj]/s1;
2947: }
2948: }
2949: /* Transposition */
2950: for(jj=1; jj<= nlstate+ndeath; jj++){
2951: for(ii=jj; ii<= nlstate+ndeath; ii++){
2952: s1=ps[ii][jj];
2953: ps[ii][jj]=ps[jj][ii];
2954: ps[jj][ii]=s1;
2955: }
2956: }
2957: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2958: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2959: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2960: /* } */
2961: /* printf("\n "); */
2962: /* } */
2963: /* printf("\n ");printf("%lf ",cov[2]);*/
2964: /*
2965: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2966: goto end;*/
2967: return ps;
1.217 brouard 2968: }
2969:
2970:
1.126 brouard 2971: /**************** Product of 2 matrices ******************/
2972:
1.145 brouard 2973: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2974: {
2975: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2976: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2977: /* in, b, out are matrice of pointers which should have been initialized
2978: before: only the contents of out is modified. The function returns
2979: a pointer to pointers identical to out */
1.145 brouard 2980: int i, j, k;
1.126 brouard 2981: for(i=nrl; i<= nrh; i++)
1.145 brouard 2982: for(k=ncolol; k<=ncoloh; k++){
2983: out[i][k]=0.;
2984: for(j=ncl; j<=nch; j++)
2985: out[i][k] +=in[i][j]*b[j][k];
2986: }
1.126 brouard 2987: return out;
2988: }
2989:
2990:
2991: /************* Higher Matrix Product ***************/
2992:
1.235 brouard 2993: 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 2994: {
1.218 brouard 2995: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 2996: 'nhstepm*hstepm*stepm' months (i.e. until
2997: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
2998: nhstepm*hstepm matrices.
2999: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3000: (typically every 2 years instead of every month which is too big
3001: for the memory).
3002: Model is determined by parameters x and covariates have to be
3003: included manually here.
3004:
3005: */
3006:
3007: int i, j, d, h, k;
1.131 brouard 3008: double **out, cov[NCOVMAX+1];
1.126 brouard 3009: double **newm;
1.187 brouard 3010: double agexact;
1.214 brouard 3011: double agebegin, ageend;
1.126 brouard 3012:
3013: /* Hstepm could be zero and should return the unit matrix */
3014: for (i=1;i<=nlstate+ndeath;i++)
3015: for (j=1;j<=nlstate+ndeath;j++){
3016: oldm[i][j]=(i==j ? 1.0 : 0.0);
3017: po[i][j][0]=(i==j ? 1.0 : 0.0);
3018: }
3019: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3020: for(h=1; h <=nhstepm; h++){
3021: for(d=1; d <=hstepm; d++){
3022: newm=savm;
3023: /* Covariates have to be included here again */
3024: cov[1]=1.;
1.214 brouard 3025: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3026: cov[2]=agexact;
3027: if(nagesqr==1)
1.227 brouard 3028: cov[3]= agexact*agexact;
1.235 brouard 3029: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3030: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3031: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3032: /* 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)); */
3033: }
3034: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3035: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3036: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3037: /* 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]); */
3038: }
3039: for (k=1; k<=cptcovage;k++){
3040: if(Dummy[Tvar[Tage[k]]]){
3041: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3042: } else{
3043: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3044: }
3045: /* 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]); */
3046: }
3047: for (k=1; k<=cptcovprod;k++){ /* */
3048: /* 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]); */
3049: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3050: }
3051: /* for (k=1; k<=cptcovn;k++) */
3052: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3053: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3054: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3055: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3056: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3057:
3058:
1.126 brouard 3059: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3060: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3061: /* right multiplication of oldm by the current matrix */
1.126 brouard 3062: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3063: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3064: /* if((int)age == 70){ */
3065: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3066: /* for(i=1; i<=nlstate+ndeath; i++) { */
3067: /* printf("%d pmmij ",i); */
3068: /* for(j=1;j<=nlstate+ndeath;j++) { */
3069: /* printf("%f ",pmmij[i][j]); */
3070: /* } */
3071: /* printf(" oldm "); */
3072: /* for(j=1;j<=nlstate+ndeath;j++) { */
3073: /* printf("%f ",oldm[i][j]); */
3074: /* } */
3075: /* printf("\n"); */
3076: /* } */
3077: /* } */
1.126 brouard 3078: savm=oldm;
3079: oldm=newm;
3080: }
3081: for(i=1; i<=nlstate+ndeath; i++)
3082: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3083: po[i][j][h]=newm[i][j];
3084: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3085: }
1.128 brouard 3086: /*printf("h=%d ",h);*/
1.126 brouard 3087: } /* end h */
1.218 brouard 3088: /* printf("\n H=%d \n",h); */
1.126 brouard 3089: return po;
3090: }
3091:
1.217 brouard 3092: /************* Higher Back Matrix Product ***************/
1.218 brouard 3093: /* 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 3094: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3095: {
1.218 brouard 3096: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3097: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3098: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3099: nhstepm*hstepm matrices.
3100: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3101: (typically every 2 years instead of every month which is too big
1.217 brouard 3102: for the memory).
1.218 brouard 3103: Model is determined by parameters x and covariates have to be
3104: included manually here.
1.217 brouard 3105:
1.222 brouard 3106: */
1.217 brouard 3107:
3108: int i, j, d, h, k;
3109: double **out, cov[NCOVMAX+1];
3110: double **newm;
3111: double agexact;
3112: double agebegin, ageend;
1.222 brouard 3113: double **oldm, **savm;
1.217 brouard 3114:
1.222 brouard 3115: oldm=oldms;savm=savms;
1.217 brouard 3116: /* Hstepm could be zero and should return the unit matrix */
3117: for (i=1;i<=nlstate+ndeath;i++)
3118: for (j=1;j<=nlstate+ndeath;j++){
3119: oldm[i][j]=(i==j ? 1.0 : 0.0);
3120: po[i][j][0]=(i==j ? 1.0 : 0.0);
3121: }
3122: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3123: for(h=1; h <=nhstepm; h++){
3124: for(d=1; d <=hstepm; d++){
3125: newm=savm;
3126: /* Covariates have to be included here again */
3127: cov[1]=1.;
3128: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3129: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3130: cov[2]=agexact;
3131: if(nagesqr==1)
1.222 brouard 3132: cov[3]= agexact*agexact;
1.218 brouard 3133: for (k=1; k<=cptcovn;k++)
1.222 brouard 3134: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3135: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3136: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3137: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3138: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3139: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3140: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3141: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3142: /* 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 3143:
3144:
1.217 brouard 3145: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3146: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3147: /* Careful transposed matrix */
1.222 brouard 3148: /* age is in cov[2] */
1.218 brouard 3149: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3150: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3151: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3152: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3153: /* if((int)age == 70){ */
3154: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3155: /* for(i=1; i<=nlstate+ndeath; i++) { */
3156: /* printf("%d pmmij ",i); */
3157: /* for(j=1;j<=nlstate+ndeath;j++) { */
3158: /* printf("%f ",pmmij[i][j]); */
3159: /* } */
3160: /* printf(" oldm "); */
3161: /* for(j=1;j<=nlstate+ndeath;j++) { */
3162: /* printf("%f ",oldm[i][j]); */
3163: /* } */
3164: /* printf("\n"); */
3165: /* } */
3166: /* } */
3167: savm=oldm;
3168: oldm=newm;
3169: }
3170: for(i=1; i<=nlstate+ndeath; i++)
3171: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3172: po[i][j][h]=newm[i][j];
3173: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3174: }
3175: /*printf("h=%d ",h);*/
3176: } /* end h */
1.222 brouard 3177: /* printf("\n H=%d \n",h); */
1.217 brouard 3178: return po;
3179: }
3180:
3181:
1.162 brouard 3182: #ifdef NLOPT
3183: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3184: double fret;
3185: double *xt;
3186: int j;
3187: myfunc_data *d2 = (myfunc_data *) pd;
3188: /* xt = (p1-1); */
3189: xt=vector(1,n);
3190: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3191:
3192: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3193: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3194: printf("Function = %.12lf ",fret);
3195: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3196: printf("\n");
3197: free_vector(xt,1,n);
3198: return fret;
3199: }
3200: #endif
1.126 brouard 3201:
3202: /*************** log-likelihood *************/
3203: double func( double *x)
3204: {
1.226 brouard 3205: int i, ii, j, k, mi, d, kk;
3206: int ioffset=0;
3207: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3208: double **out;
3209: double lli; /* Individual log likelihood */
3210: int s1, s2;
1.228 brouard 3211: 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 3212: double bbh, survp;
3213: long ipmx;
3214: double agexact;
3215: /*extern weight */
3216: /* We are differentiating ll according to initial status */
3217: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3218: /*for(i=1;i<imx;i++)
3219: printf(" %d\n",s[4][i]);
3220: */
1.162 brouard 3221:
1.226 brouard 3222: ++countcallfunc;
1.162 brouard 3223:
1.226 brouard 3224: cov[1]=1.;
1.126 brouard 3225:
1.226 brouard 3226: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3227: ioffset=0;
1.226 brouard 3228: if(mle==1){
3229: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3230: /* Computes the values of the ncovmodel covariates of the model
3231: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3232: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3233: to be observed in j being in i according to the model.
3234: */
1.243 brouard 3235: ioffset=2+nagesqr ;
1.233 brouard 3236: /* Fixed */
1.234 brouard 3237: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3238: 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)*/
3239: }
1.226 brouard 3240: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3241: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3242: has been calculated etc */
3243: /* For an individual i, wav[i] gives the number of effective waves */
3244: /* We compute the contribution to Likelihood of each effective transition
3245: mw[mi][i] is real wave of the mi th effectve wave */
3246: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3247: s2=s[mw[mi+1][i]][i];
3248: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3249: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3250: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3251: */
3252: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3253: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3254: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3255: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3256: }
3257: for (ii=1;ii<=nlstate+ndeath;ii++)
3258: for (j=1;j<=nlstate+ndeath;j++){
3259: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3260: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3261: }
3262: for(d=0; d<dh[mi][i]; d++){
3263: newm=savm;
3264: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3265: cov[2]=agexact;
3266: if(nagesqr==1)
3267: cov[3]= agexact*agexact; /* Should be changed here */
3268: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3269: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3270: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3271: else
3272: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3273: }
3274: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3275: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3276: savm=oldm;
3277: oldm=newm;
3278: } /* end mult */
3279:
3280: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3281: /* But now since version 0.9 we anticipate for bias at large stepm.
3282: * If stepm is larger than one month (smallest stepm) and if the exact delay
3283: * (in months) between two waves is not a multiple of stepm, we rounded to
3284: * the nearest (and in case of equal distance, to the lowest) interval but now
3285: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3286: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3287: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3288: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3289: * -stepm/2 to stepm/2 .
3290: * For stepm=1 the results are the same as for previous versions of Imach.
3291: * For stepm > 1 the results are less biased than in previous versions.
3292: */
1.234 brouard 3293: s1=s[mw[mi][i]][i];
3294: s2=s[mw[mi+1][i]][i];
3295: bbh=(double)bh[mi][i]/(double)stepm;
3296: /* bias bh is positive if real duration
3297: * is higher than the multiple of stepm and negative otherwise.
3298: */
3299: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3300: if( s2 > nlstate){
3301: /* i.e. if s2 is a death state and if the date of death is known
3302: then the contribution to the likelihood is the probability to
3303: die between last step unit time and current step unit time,
3304: which is also equal to probability to die before dh
3305: minus probability to die before dh-stepm .
3306: In version up to 0.92 likelihood was computed
3307: as if date of death was unknown. Death was treated as any other
3308: health state: the date of the interview describes the actual state
3309: and not the date of a change in health state. The former idea was
3310: to consider that at each interview the state was recorded
3311: (healthy, disable or death) and IMaCh was corrected; but when we
3312: introduced the exact date of death then we should have modified
3313: the contribution of an exact death to the likelihood. This new
3314: contribution is smaller and very dependent of the step unit
3315: stepm. It is no more the probability to die between last interview
3316: and month of death but the probability to survive from last
3317: interview up to one month before death multiplied by the
3318: probability to die within a month. Thanks to Chris
3319: Jackson for correcting this bug. Former versions increased
3320: mortality artificially. The bad side is that we add another loop
3321: which slows down the processing. The difference can be up to 10%
3322: lower mortality.
3323: */
3324: /* If, at the beginning of the maximization mostly, the
3325: cumulative probability or probability to be dead is
3326: constant (ie = 1) over time d, the difference is equal to
3327: 0. out[s1][3] = savm[s1][3]: probability, being at state
3328: s1 at precedent wave, to be dead a month before current
3329: wave is equal to probability, being at state s1 at
3330: precedent wave, to be dead at mont of the current
3331: wave. Then the observed probability (that this person died)
3332: is null according to current estimated parameter. In fact,
3333: it should be very low but not zero otherwise the log go to
3334: infinity.
3335: */
1.183 brouard 3336: /* #ifdef INFINITYORIGINAL */
3337: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3338: /* #else */
3339: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3340: /* lli=log(mytinydouble); */
3341: /* else */
3342: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3343: /* #endif */
1.226 brouard 3344: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3345:
1.226 brouard 3346: } else if ( s2==-1 ) { /* alive */
3347: for (j=1,survp=0. ; j<=nlstate; j++)
3348: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3349: /*survp += out[s1][j]; */
3350: lli= log(survp);
3351: }
3352: else if (s2==-4) {
3353: for (j=3,survp=0. ; j<=nlstate; j++)
3354: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3355: lli= log(survp);
3356: }
3357: else if (s2==-5) {
3358: for (j=1,survp=0. ; j<=2; j++)
3359: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3360: lli= log(survp);
3361: }
3362: else{
3363: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3364: /* 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 */
3365: }
3366: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3367: /*if(lli ==000.0)*/
3368: /*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); */
3369: ipmx +=1;
3370: sw += weight[i];
3371: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3372: /* if (lli < log(mytinydouble)){ */
3373: /* 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); */
3374: /* 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]); */
3375: /* } */
3376: } /* end of wave */
3377: } /* end of individual */
3378: } else if(mle==2){
3379: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3380: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3381: for(mi=1; mi<= wav[i]-1; mi++){
3382: for (ii=1;ii<=nlstate+ndeath;ii++)
3383: for (j=1;j<=nlstate+ndeath;j++){
3384: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3385: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3386: }
3387: for(d=0; d<=dh[mi][i]; d++){
3388: newm=savm;
3389: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3390: cov[2]=agexact;
3391: if(nagesqr==1)
3392: cov[3]= agexact*agexact;
3393: for (kk=1; kk<=cptcovage;kk++) {
3394: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3395: }
3396: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3397: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3398: savm=oldm;
3399: oldm=newm;
3400: } /* end mult */
3401:
3402: s1=s[mw[mi][i]][i];
3403: s2=s[mw[mi+1][i]][i];
3404: bbh=(double)bh[mi][i]/(double)stepm;
3405: 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 */
3406: ipmx +=1;
3407: sw += weight[i];
3408: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3409: } /* end of wave */
3410: } /* end of individual */
3411: } else if(mle==3){ /* exponential inter-extrapolation */
3412: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3413: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3414: for(mi=1; mi<= wav[i]-1; mi++){
3415: for (ii=1;ii<=nlstate+ndeath;ii++)
3416: for (j=1;j<=nlstate+ndeath;j++){
3417: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3418: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3419: }
3420: for(d=0; d<dh[mi][i]; d++){
3421: newm=savm;
3422: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3423: cov[2]=agexact;
3424: if(nagesqr==1)
3425: cov[3]= agexact*agexact;
3426: for (kk=1; kk<=cptcovage;kk++) {
3427: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3428: }
3429: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3430: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3431: savm=oldm;
3432: oldm=newm;
3433: } /* end mult */
3434:
3435: s1=s[mw[mi][i]][i];
3436: s2=s[mw[mi+1][i]][i];
3437: bbh=(double)bh[mi][i]/(double)stepm;
3438: 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 */
3439: ipmx +=1;
3440: sw += weight[i];
3441: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3442: } /* end of wave */
3443: } /* end of individual */
3444: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3445: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3446: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3447: for(mi=1; mi<= wav[i]-1; mi++){
3448: for (ii=1;ii<=nlstate+ndeath;ii++)
3449: for (j=1;j<=nlstate+ndeath;j++){
3450: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3451: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3452: }
3453: for(d=0; d<dh[mi][i]; d++){
3454: newm=savm;
3455: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3456: cov[2]=agexact;
3457: if(nagesqr==1)
3458: cov[3]= agexact*agexact;
3459: for (kk=1; kk<=cptcovage;kk++) {
3460: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3461: }
1.126 brouard 3462:
1.226 brouard 3463: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3464: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3465: savm=oldm;
3466: oldm=newm;
3467: } /* end mult */
3468:
3469: s1=s[mw[mi][i]][i];
3470: s2=s[mw[mi+1][i]][i];
3471: if( s2 > nlstate){
3472: lli=log(out[s1][s2] - savm[s1][s2]);
3473: } else if ( s2==-1 ) { /* alive */
3474: for (j=1,survp=0. ; j<=nlstate; j++)
3475: survp += out[s1][j];
3476: lli= log(survp);
3477: }else{
3478: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3479: }
3480: ipmx +=1;
3481: sw += weight[i];
3482: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3483: /* 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 3484: } /* end of wave */
3485: } /* end of individual */
3486: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3487: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3488: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3489: for(mi=1; mi<= wav[i]-1; mi++){
3490: for (ii=1;ii<=nlstate+ndeath;ii++)
3491: for (j=1;j<=nlstate+ndeath;j++){
3492: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3493: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3494: }
3495: for(d=0; d<dh[mi][i]; d++){
3496: newm=savm;
3497: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3498: cov[2]=agexact;
3499: if(nagesqr==1)
3500: cov[3]= agexact*agexact;
3501: for (kk=1; kk<=cptcovage;kk++) {
3502: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3503: }
1.126 brouard 3504:
1.226 brouard 3505: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3506: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3507: savm=oldm;
3508: oldm=newm;
3509: } /* end mult */
3510:
3511: s1=s[mw[mi][i]][i];
3512: s2=s[mw[mi+1][i]][i];
3513: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3514: ipmx +=1;
3515: sw += weight[i];
3516: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3517: /*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]);*/
3518: } /* end of wave */
3519: } /* end of individual */
3520: } /* End of if */
3521: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3522: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3523: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3524: return -l;
1.126 brouard 3525: }
3526:
3527: /*************** log-likelihood *************/
3528: double funcone( double *x)
3529: {
1.228 brouard 3530: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3531: int i, ii, j, k, mi, d, kk;
1.228 brouard 3532: int ioffset=0;
1.131 brouard 3533: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3534: double **out;
3535: double lli; /* Individual log likelihood */
3536: double llt;
3537: int s1, s2;
1.228 brouard 3538: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3539:
1.126 brouard 3540: double bbh, survp;
1.187 brouard 3541: double agexact;
1.214 brouard 3542: double agebegin, ageend;
1.126 brouard 3543: /*extern weight */
3544: /* We are differentiating ll according to initial status */
3545: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3546: /*for(i=1;i<imx;i++)
3547: printf(" %d\n",s[4][i]);
3548: */
3549: cov[1]=1.;
3550:
3551: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3552: ioffset=0;
3553: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3554: /* ioffset=2+nagesqr+cptcovage; */
3555: ioffset=2+nagesqr;
1.232 brouard 3556: /* Fixed */
1.224 brouard 3557: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3558: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3559: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3560: 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)*/
3561: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3562: /* cov[2+6]=covar[Tvar[6]][i]; */
3563: /* cov[2+6]=covar[2][i]; V2 */
3564: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3565: /* cov[2+7]=covar[Tvar[7]][i]; */
3566: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3567: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3568: /* cov[2+9]=covar[Tvar[9]][i]; */
3569: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3570: }
1.232 brouard 3571: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3572: /* 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?)*\/ */
3573: /* } */
1.231 brouard 3574: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3575: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3576: /* } */
1.225 brouard 3577:
1.233 brouard 3578:
3579: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3580: /* Wave varying (but not age varying) */
3581: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3582: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3583: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3584: }
1.232 brouard 3585: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3586: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3587: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3588: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3589: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3590: /* 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 3591: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3592: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3593: /* /\* 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]); *\/ */
3594: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3595: /* } */
1.126 brouard 3596: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3597: for (j=1;j<=nlstate+ndeath;j++){
3598: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3599: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3600: }
1.214 brouard 3601:
3602: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3603: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3604: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 ! brouard 3605: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3606: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3607: and mw[mi+1][i]. dh depends on stepm.*/
3608: newm=savm;
1.247 ! brouard 3609: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3610: cov[2]=agexact;
3611: if(nagesqr==1)
3612: cov[3]= agexact*agexact;
3613: for (kk=1; kk<=cptcovage;kk++) {
3614: if(!FixedV[Tvar[Tage[kk]]])
3615: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3616: else
3617: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3618: }
3619: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3620: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3621: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3622: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3623: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3624: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3625: savm=oldm;
3626: oldm=newm;
1.126 brouard 3627: } /* end mult */
3628:
3629: s1=s[mw[mi][i]][i];
3630: s2=s[mw[mi+1][i]][i];
1.217 brouard 3631: /* if(s2==-1){ */
3632: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3633: /* /\* exit(1); *\/ */
3634: /* } */
1.126 brouard 3635: bbh=(double)bh[mi][i]/(double)stepm;
3636: /* bias is positive if real duration
3637: * is higher than the multiple of stepm and negative otherwise.
3638: */
3639: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3640: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3641: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3642: for (j=1,survp=0. ; j<=nlstate; j++)
3643: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3644: lli= log(survp);
1.126 brouard 3645: }else if (mle==1){
1.242 brouard 3646: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3647: } else if(mle==2){
1.242 brouard 3648: 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 3649: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3650: 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 3651: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3652: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3653: } else{ /* mle=0 back to 1 */
1.242 brouard 3654: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3655: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3656: } /* End of if */
3657: ipmx +=1;
3658: sw += weight[i];
3659: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3660: /*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 3661: if(globpr){
1.246 brouard 3662: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3663: %11.6f %11.6f %11.6f ", \
1.242 brouard 3664: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3665: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3666: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3667: llt +=ll[k]*gipmx/gsw;
3668: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3669: }
3670: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3671: }
1.232 brouard 3672: } /* end of wave */
3673: } /* end of individual */
3674: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3675: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3676: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3677: if(globpr==0){ /* First time we count the contributions and weights */
3678: gipmx=ipmx;
3679: gsw=sw;
3680: }
3681: return -l;
1.126 brouard 3682: }
3683:
3684:
3685: /*************** function likelione ***********/
3686: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3687: {
3688: /* This routine should help understanding what is done with
3689: the selection of individuals/waves and
3690: to check the exact contribution to the likelihood.
3691: Plotting could be done.
3692: */
3693: int k;
3694:
3695: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3696: strcpy(fileresilk,"ILK_");
1.202 brouard 3697: strcat(fileresilk,fileresu);
1.126 brouard 3698: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3699: printf("Problem with resultfile: %s\n", fileresilk);
3700: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3701: }
1.214 brouard 3702: 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");
3703: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3704: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3705: for(k=1; k<=nlstate; k++)
3706: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3707: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3708: }
3709:
3710: *fretone=(*funcone)(p);
3711: if(*globpri !=0){
3712: fclose(ficresilk);
1.205 brouard 3713: if (mle ==0)
3714: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3715: else if(mle >=1)
3716: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3717: 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 3718:
1.208 brouard 3719:
3720: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3721: 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 3722: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3723: }
1.207 brouard 3724: 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 3725: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3726: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3727: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3728: fflush(fichtm);
1.205 brouard 3729: }
1.126 brouard 3730: return;
3731: }
3732:
3733:
3734: /*********** Maximum Likelihood Estimation ***************/
3735:
3736: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3737: {
1.165 brouard 3738: int i,j, iter=0;
1.126 brouard 3739: double **xi;
3740: double fret;
3741: double fretone; /* Only one call to likelihood */
3742: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3743:
3744: #ifdef NLOPT
3745: int creturn;
3746: nlopt_opt opt;
3747: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3748: double *lb;
3749: double minf; /* the minimum objective value, upon return */
3750: double * p1; /* Shifted parameters from 0 instead of 1 */
3751: myfunc_data dinst, *d = &dinst;
3752: #endif
3753:
3754:
1.126 brouard 3755: xi=matrix(1,npar,1,npar);
3756: for (i=1;i<=npar;i++)
3757: for (j=1;j<=npar;j++)
3758: xi[i][j]=(i==j ? 1.0 : 0.0);
3759: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3760: strcpy(filerespow,"POW_");
1.126 brouard 3761: strcat(filerespow,fileres);
3762: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3763: printf("Problem with resultfile: %s\n", filerespow);
3764: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3765: }
3766: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3767: for (i=1;i<=nlstate;i++)
3768: for(j=1;j<=nlstate+ndeath;j++)
3769: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3770: fprintf(ficrespow,"\n");
1.162 brouard 3771: #ifdef POWELL
1.126 brouard 3772: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3773: #endif
1.126 brouard 3774:
1.162 brouard 3775: #ifdef NLOPT
3776: #ifdef NEWUOA
3777: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3778: #else
3779: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3780: #endif
3781: lb=vector(0,npar-1);
3782: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3783: nlopt_set_lower_bounds(opt, lb);
3784: nlopt_set_initial_step1(opt, 0.1);
3785:
3786: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3787: d->function = func;
3788: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3789: nlopt_set_min_objective(opt, myfunc, d);
3790: nlopt_set_xtol_rel(opt, ftol);
3791: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3792: printf("nlopt failed! %d\n",creturn);
3793: }
3794: else {
3795: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3796: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3797: iter=1; /* not equal */
3798: }
3799: nlopt_destroy(opt);
3800: #endif
1.126 brouard 3801: free_matrix(xi,1,npar,1,npar);
3802: fclose(ficrespow);
1.203 brouard 3803: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3804: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3805: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3806:
3807: }
3808:
3809: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3810: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3811: {
3812: double **a,**y,*x,pd;
1.203 brouard 3813: /* double **hess; */
1.164 brouard 3814: int i, j;
1.126 brouard 3815: int *indx;
3816:
3817: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3818: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3819: void lubksb(double **a, int npar, int *indx, double b[]) ;
3820: void ludcmp(double **a, int npar, int *indx, double *d) ;
3821: double gompertz(double p[]);
1.203 brouard 3822: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3823:
3824: printf("\nCalculation of the hessian matrix. Wait...\n");
3825: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3826: for (i=1;i<=npar;i++){
1.203 brouard 3827: printf("%d-",i);fflush(stdout);
3828: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3829:
3830: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3831:
3832: /* printf(" %f ",p[i]);
3833: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3834: }
3835:
3836: for (i=1;i<=npar;i++) {
3837: for (j=1;j<=npar;j++) {
3838: if (j>i) {
1.203 brouard 3839: printf(".%d-%d",i,j);fflush(stdout);
3840: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3841: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3842:
3843: hess[j][i]=hess[i][j];
3844: /*printf(" %lf ",hess[i][j]);*/
3845: }
3846: }
3847: }
3848: printf("\n");
3849: fprintf(ficlog,"\n");
3850:
3851: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3852: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3853:
3854: a=matrix(1,npar,1,npar);
3855: y=matrix(1,npar,1,npar);
3856: x=vector(1,npar);
3857: indx=ivector(1,npar);
3858: for (i=1;i<=npar;i++)
3859: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3860: ludcmp(a,npar,indx,&pd);
3861:
3862: for (j=1;j<=npar;j++) {
3863: for (i=1;i<=npar;i++) x[i]=0;
3864: x[j]=1;
3865: lubksb(a,npar,indx,x);
3866: for (i=1;i<=npar;i++){
3867: matcov[i][j]=x[i];
3868: }
3869: }
3870:
3871: printf("\n#Hessian matrix#\n");
3872: fprintf(ficlog,"\n#Hessian matrix#\n");
3873: for (i=1;i<=npar;i++) {
3874: for (j=1;j<=npar;j++) {
1.203 brouard 3875: printf("%.6e ",hess[i][j]);
3876: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3877: }
3878: printf("\n");
3879: fprintf(ficlog,"\n");
3880: }
3881:
1.203 brouard 3882: /* printf("\n#Covariance matrix#\n"); */
3883: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3884: /* for (i=1;i<=npar;i++) { */
3885: /* for (j=1;j<=npar;j++) { */
3886: /* printf("%.6e ",matcov[i][j]); */
3887: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3888: /* } */
3889: /* printf("\n"); */
3890: /* fprintf(ficlog,"\n"); */
3891: /* } */
3892:
1.126 brouard 3893: /* Recompute Inverse */
1.203 brouard 3894: /* for (i=1;i<=npar;i++) */
3895: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3896: /* ludcmp(a,npar,indx,&pd); */
3897:
3898: /* printf("\n#Hessian matrix recomputed#\n"); */
3899:
3900: /* for (j=1;j<=npar;j++) { */
3901: /* for (i=1;i<=npar;i++) x[i]=0; */
3902: /* x[j]=1; */
3903: /* lubksb(a,npar,indx,x); */
3904: /* for (i=1;i<=npar;i++){ */
3905: /* y[i][j]=x[i]; */
3906: /* printf("%.3e ",y[i][j]); */
3907: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3908: /* } */
3909: /* printf("\n"); */
3910: /* fprintf(ficlog,"\n"); */
3911: /* } */
3912:
3913: /* Verifying the inverse matrix */
3914: #ifdef DEBUGHESS
3915: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3916:
1.203 brouard 3917: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3918: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3919:
3920: for (j=1;j<=npar;j++) {
3921: for (i=1;i<=npar;i++){
1.203 brouard 3922: printf("%.2f ",y[i][j]);
3923: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3924: }
3925: printf("\n");
3926: fprintf(ficlog,"\n");
3927: }
1.203 brouard 3928: #endif
1.126 brouard 3929:
3930: free_matrix(a,1,npar,1,npar);
3931: free_matrix(y,1,npar,1,npar);
3932: free_vector(x,1,npar);
3933: free_ivector(indx,1,npar);
1.203 brouard 3934: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3935:
3936:
3937: }
3938:
3939: /*************** hessian matrix ****************/
3940: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3941: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3942: int i;
3943: int l=1, lmax=20;
1.203 brouard 3944: double k1,k2, res, fx;
1.132 brouard 3945: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3946: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3947: int k=0,kmax=10;
3948: double l1;
3949:
3950: fx=func(x);
3951: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3952: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3953: l1=pow(10,l);
3954: delts=delt;
3955: for(k=1 ; k <kmax; k=k+1){
3956: delt = delta*(l1*k);
3957: p2[theta]=x[theta] +delt;
1.145 brouard 3958: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3959: p2[theta]=x[theta]-delt;
3960: k2=func(p2)-fx;
3961: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3962: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3963:
1.203 brouard 3964: #ifdef DEBUGHESSII
1.126 brouard 3965: 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);
3966: 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);
3967: #endif
3968: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3969: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3970: k=kmax;
3971: }
3972: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3973: k=kmax; l=lmax*10;
1.126 brouard 3974: }
3975: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3976: delts=delt;
3977: }
1.203 brouard 3978: } /* End loop k */
1.126 brouard 3979: }
3980: delti[theta]=delts;
3981: return res;
3982:
3983: }
3984:
1.203 brouard 3985: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3986: {
3987: int i;
1.164 brouard 3988: int l=1, lmax=20;
1.126 brouard 3989: double k1,k2,k3,k4,res,fx;
1.132 brouard 3990: double p2[MAXPARM+1];
1.203 brouard 3991: int k, kmax=1;
3992: double v1, v2, cv12, lc1, lc2;
1.208 brouard 3993:
3994: int firstime=0;
1.203 brouard 3995:
1.126 brouard 3996: fx=func(x);
1.203 brouard 3997: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 3998: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 3999: p2[thetai]=x[thetai]+delti[thetai]*k;
4000: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4001: k1=func(p2)-fx;
4002:
1.203 brouard 4003: p2[thetai]=x[thetai]+delti[thetai]*k;
4004: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4005: k2=func(p2)-fx;
4006:
1.203 brouard 4007: p2[thetai]=x[thetai]-delti[thetai]*k;
4008: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4009: k3=func(p2)-fx;
4010:
1.203 brouard 4011: p2[thetai]=x[thetai]-delti[thetai]*k;
4012: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4013: k4=func(p2)-fx;
1.203 brouard 4014: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4015: if(k1*k2*k3*k4 <0.){
1.208 brouard 4016: firstime=1;
1.203 brouard 4017: kmax=kmax+10;
1.208 brouard 4018: }
4019: if(kmax >=10 || firstime ==1){
1.246 brouard 4020: 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);
4021: 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 4022: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4023: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4024: }
4025: #ifdef DEBUGHESSIJ
4026: v1=hess[thetai][thetai];
4027: v2=hess[thetaj][thetaj];
4028: cv12=res;
4029: /* Computing eigen value of Hessian matrix */
4030: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4031: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4032: if ((lc2 <0) || (lc1 <0) ){
4033: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4034: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4035: 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);
4036: 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);
4037: }
1.126 brouard 4038: #endif
4039: }
4040: return res;
4041: }
4042:
1.203 brouard 4043: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4044: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4045: /* { */
4046: /* int i; */
4047: /* int l=1, lmax=20; */
4048: /* double k1,k2,k3,k4,res,fx; */
4049: /* double p2[MAXPARM+1]; */
4050: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4051: /* int k=0,kmax=10; */
4052: /* double l1; */
4053:
4054: /* fx=func(x); */
4055: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4056: /* l1=pow(10,l); */
4057: /* delts=delt; */
4058: /* for(k=1 ; k <kmax; k=k+1){ */
4059: /* delt = delti*(l1*k); */
4060: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4061: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4062: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4063: /* k1=func(p2)-fx; */
4064:
4065: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4066: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4067: /* k2=func(p2)-fx; */
4068:
4069: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4070: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4071: /* k3=func(p2)-fx; */
4072:
4073: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4074: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4075: /* k4=func(p2)-fx; */
4076: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4077: /* #ifdef DEBUGHESSIJ */
4078: /* 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); */
4079: /* 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); */
4080: /* #endif */
4081: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4082: /* k=kmax; */
4083: /* } */
4084: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4085: /* k=kmax; l=lmax*10; */
4086: /* } */
4087: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4088: /* delts=delt; */
4089: /* } */
4090: /* } /\* End loop k *\/ */
4091: /* } */
4092: /* delti[theta]=delts; */
4093: /* return res; */
4094: /* } */
4095:
4096:
1.126 brouard 4097: /************** Inverse of matrix **************/
4098: void ludcmp(double **a, int n, int *indx, double *d)
4099: {
4100: int i,imax,j,k;
4101: double big,dum,sum,temp;
4102: double *vv;
4103:
4104: vv=vector(1,n);
4105: *d=1.0;
4106: for (i=1;i<=n;i++) {
4107: big=0.0;
4108: for (j=1;j<=n;j++)
4109: if ((temp=fabs(a[i][j])) > big) big=temp;
4110: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4111: vv[i]=1.0/big;
4112: }
4113: for (j=1;j<=n;j++) {
4114: for (i=1;i<j;i++) {
4115: sum=a[i][j];
4116: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4117: a[i][j]=sum;
4118: }
4119: big=0.0;
4120: for (i=j;i<=n;i++) {
4121: sum=a[i][j];
4122: for (k=1;k<j;k++)
4123: sum -= a[i][k]*a[k][j];
4124: a[i][j]=sum;
4125: if ( (dum=vv[i]*fabs(sum)) >= big) {
4126: big=dum;
4127: imax=i;
4128: }
4129: }
4130: if (j != imax) {
4131: for (k=1;k<=n;k++) {
4132: dum=a[imax][k];
4133: a[imax][k]=a[j][k];
4134: a[j][k]=dum;
4135: }
4136: *d = -(*d);
4137: vv[imax]=vv[j];
4138: }
4139: indx[j]=imax;
4140: if (a[j][j] == 0.0) a[j][j]=TINY;
4141: if (j != n) {
4142: dum=1.0/(a[j][j]);
4143: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4144: }
4145: }
4146: free_vector(vv,1,n); /* Doesn't work */
4147: ;
4148: }
4149:
4150: void lubksb(double **a, int n, int *indx, double b[])
4151: {
4152: int i,ii=0,ip,j;
4153: double sum;
4154:
4155: for (i=1;i<=n;i++) {
4156: ip=indx[i];
4157: sum=b[ip];
4158: b[ip]=b[i];
4159: if (ii)
4160: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4161: else if (sum) ii=i;
4162: b[i]=sum;
4163: }
4164: for (i=n;i>=1;i--) {
4165: sum=b[i];
4166: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4167: b[i]=sum/a[i][i];
4168: }
4169: }
4170:
4171: void pstamp(FILE *fichier)
4172: {
1.196 brouard 4173: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4174: }
4175:
4176: /************ Frequencies ********************/
1.226 brouard 4177: void freqsummary(char fileres[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
4178: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4179: int firstpass, int lastpass, int stepm, int weightopt, char model[])
4180: { /* Some frequencies */
4181:
1.227 brouard 4182: int i, m, jk, j1, bool, z1,j, k, iv;
1.226 brouard 4183: int iind=0, iage=0;
4184: int mi; /* Effective wave */
4185: int first;
4186: double ***freq; /* Frequencies */
4187: double *meanq;
4188: double **meanqt;
4189: double *pp, **prop, *posprop, *pospropt;
4190: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4191: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4192: double agebegin, ageend;
4193:
4194: pp=vector(1,nlstate);
4195: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4196: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4197: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4198: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4199: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4200: meanqt=matrix(1,lastpass,1,nqtveff);
4201: strcpy(fileresp,"P_");
4202: strcat(fileresp,fileresu);
4203: /*strcat(fileresphtm,fileresu);*/
4204: if((ficresp=fopen(fileresp,"w"))==NULL) {
4205: printf("Problem with prevalence resultfile: %s\n", fileresp);
4206: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4207: exit(0);
4208: }
1.240 brouard 4209:
1.226 brouard 4210: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4211: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4212: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4213: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4214: fflush(ficlog);
4215: exit(70);
4216: }
4217: else{
4218: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4219: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4220: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4221: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4222: }
1.237 brouard 4223: 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 4224:
1.226 brouard 4225: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4226: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4227: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4228: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4229: fflush(ficlog);
4230: exit(70);
1.240 brouard 4231: } else{
1.226 brouard 4232: 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 4233: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4234: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4235: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4236: }
1.240 brouard 4237: 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);
4238:
1.226 brouard 4239: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4240: j1=0;
1.126 brouard 4241:
1.227 brouard 4242: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4243: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4244: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4245:
1.226 brouard 4246: first=1;
1.240 brouard 4247:
1.226 brouard 4248: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4249: reference=low_education V1=0,V2=0
4250: med_educ V1=1 V2=0,
4251: high_educ V1=0 V2=1
4252: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4253: */
1.240 brouard 4254:
1.227 brouard 4255: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on covariates combination in order of model, excluding quantitatives V4=0, V3=0 for example, fixed or varying covariates */
1.226 brouard 4256: posproptt=0.;
4257: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4258: scanf("%d", i);*/
4259: for (i=-5; i<=nlstate+ndeath; i++)
4260: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4261: for(m=iagemin; m <= iagemax+3; m++)
4262: freq[i][jk][m]=0;
4263:
1.226 brouard 4264: for (i=1; i<=nlstate; i++) {
4265: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4266: prop[i][m]=0;
1.226 brouard 4267: posprop[i]=0;
4268: pospropt[i]=0;
4269: }
1.227 brouard 4270: /* for (z1=1; z1<= nqfveff; z1++) { */
4271: /* meanq[z1]+=0.; */
4272: /* for(m=1;m<=lastpass;m++){ */
4273: /* meanqt[m][z1]=0.; */
4274: /* } */
4275: /* } */
1.240 brouard 4276:
1.226 brouard 4277: dateintsum=0;
4278: k2cpt=0;
1.227 brouard 4279: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4280: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4281: bool=1;
1.227 brouard 4282: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4283: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4284: /* for (z1=1; z1<= nqfveff; z1++) { */
4285: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4286: /* } */
1.234 brouard 4287: for (z1=1; z1<=cptcoveff; z1++) {
4288: /* if(Tvaraff[z1] ==-20){ */
4289: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4290: /* }else if(Tvaraff[z1] ==-10){ */
4291: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4292: /* }else */
4293: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){
4294: /* Tests if this individual iind responded to j1 (V4=1 V3=0) */
4295: bool=0;
4296: /* 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",
4297: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4298: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4299: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4300: } /* Onlyf fixed */
4301: } /* end z1 */
4302: } /* cptcovn > 0 */
1.227 brouard 4303: } /* end any */
4304: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4305: /* for(m=firstpass; m<=lastpass; m++){ */
4306: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4307: m=mw[mi][iind];
4308: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4309: for (z1=1; z1<=cptcoveff; z1++) {
4310: if( Fixed[Tmodelind[z1]]==1){
4311: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4312: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4313: bool=0;
4314: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4315: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4316: bool=0;
4317: }
4318: }
4319: }
4320: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4321: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4322: if(bool==1){
4323: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4324: and mw[mi+1][iind]. dh depends on stepm. */
4325: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4326: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4327: if(m >=firstpass && m <=lastpass){
4328: k2=anint[m][iind]+(mint[m][iind]/12.);
4329: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4330: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4331: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4332: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4333: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4334: if (m<lastpass) {
4335: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4336: /* 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]); */
4337: if(s[m][iind]==-1)
4338: 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.));
4339: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4340: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4341: 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 */
4342: }
4343: } /* end if between passes */
4344: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99)) {
4345: dateintsum=dateintsum+k2;
4346: k2cpt++;
4347: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4348: }
4349: } /* end bool 2 */
4350: } /* end m */
1.226 brouard 4351: } /* end bool */
4352: } /* end iind = 1 to imx */
4353: /* prop[s][age] is feeded for any initial and valid live state as well as
4354: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4355:
4356:
1.226 brouard 4357: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4358: pstamp(ficresp);
1.240 brouard 4359: if (cptcoveff>0){
1.226 brouard 4360: fprintf(ficresp, "\n#********** Variable ");
4361: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4362: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4363: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4364: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4365: if(DummyV[z1]){
4366: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4367: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4368: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4369: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4370: }else{
4371: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4372: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4373: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4374: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4375: }
1.226 brouard 4376: }
4377: fprintf(ficresp, "**********\n#");
4378: fprintf(ficresphtm, "**********</h3>\n");
4379: fprintf(ficresphtmfr, "**********</h3>\n");
4380: fprintf(ficlog, "**********\n");
4381: }
4382: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4383: for(i=1; i<=nlstate;i++) {
1.240 brouard 4384: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4385: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4386: }
4387: fprintf(ficresp, "\n");
4388: fprintf(ficresphtm, "\n");
1.240 brouard 4389:
1.226 brouard 4390: /* Header of frequency table by age */
4391: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4392: fprintf(ficresphtmfr,"<th>Age</th> ");
4393: for(jk=-1; jk <=nlstate+ndeath; jk++){
4394: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4395: if(jk!=0 && m!=0)
4396: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4397: }
4398: }
4399: fprintf(ficresphtmfr, "\n");
1.240 brouard 4400:
1.226 brouard 4401: /* For each age */
4402: for(iage=iagemin; iage <= iagemax+3; iage++){
4403: fprintf(ficresphtm,"<tr>");
4404: if(iage==iagemax+1){
1.240 brouard 4405: fprintf(ficlog,"1");
4406: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4407: }else if(iage==iagemax+2){
1.240 brouard 4408: fprintf(ficlog,"0");
4409: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4410: }else if(iage==iagemax+3){
1.240 brouard 4411: fprintf(ficlog,"Total");
4412: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4413: }else{
1.240 brouard 4414: if(first==1){
4415: first=0;
4416: printf("See log file for details...\n");
4417: }
4418: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4419: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4420: }
4421: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4422: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4423: pp[jk] += freq[jk][m][iage];
1.226 brouard 4424: }
4425: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4426: for(m=-1, pos=0; m <=0 ; m++)
4427: pos += freq[jk][m][iage];
4428: if(pp[jk]>=1.e-10){
4429: if(first==1){
4430: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4431: }
4432: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4433: }else{
4434: if(first==1)
4435: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4436: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4437: }
1.226 brouard 4438: }
1.240 brouard 4439:
1.226 brouard 4440: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4441: /* posprop[jk]=0; */
4442: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4443: pp[jk] += freq[jk][m][iage];
1.226 brouard 4444: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4445:
1.226 brouard 4446: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4447: pos += pp[jk]; /* pos is the total number of transitions until this age */
4448: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4449: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4450: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4451: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4452: }
4453: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4454: if(pos>=1.e-5){
4455: if(first==1)
4456: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4457: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4458: }else{
4459: if(first==1)
4460: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4461: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4462: }
4463: if( iage <= iagemax){
4464: if(pos>=1.e-5){
4465: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4466: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4467: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4468: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4469: }
4470: else{
4471: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4472: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4473: }
4474: }
4475: pospropt[jk] +=posprop[jk];
1.226 brouard 4476: } /* end loop jk */
4477: /* pospropt=0.; */
4478: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4479: for(m=-1; m <=nlstate+ndeath; m++){
4480: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4481: if(first==1){
4482: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4483: }
4484: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4485: }
4486: if(jk!=0 && m!=0)
4487: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4488: }
1.226 brouard 4489: } /* end loop jk */
4490: posproptt=0.;
4491: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4492: posproptt += pospropt[jk];
1.226 brouard 4493: }
4494: fprintf(ficresphtmfr,"</tr>\n ");
4495: if(iage <= iagemax){
1.240 brouard 4496: fprintf(ficresp,"\n");
4497: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4498: }
4499: if(first==1)
1.240 brouard 4500: printf("Others in log...\n");
1.226 brouard 4501: fprintf(ficlog,"\n");
4502: } /* end loop age iage */
4503: fprintf(ficresphtm,"<tr><th>Tot</th>");
4504: for(jk=1; jk <=nlstate ; jk++){
4505: if(posproptt < 1.e-5){
1.240 brouard 4506: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4507: }else{
1.240 brouard 4508: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4509: }
4510: }
4511: fprintf(ficresphtm,"</tr>\n");
4512: fprintf(ficresphtm,"</table>\n");
4513: fprintf(ficresphtmfr,"</table>\n");
4514: if(posproptt < 1.e-5){
4515: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4516: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4517: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4518: invalidvarcomb[j1]=1;
4519: }else{
4520: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4521: invalidvarcomb[j1]=0;
4522: }
4523: fprintf(ficresphtmfr,"</table>\n");
4524: } /* end selected combination of covariate j1 */
4525: dateintmean=dateintsum/k2cpt;
1.240 brouard 4526:
1.226 brouard 4527: fclose(ficresp);
4528: fclose(ficresphtm);
4529: fclose(ficresphtmfr);
4530: free_vector(meanq,1,nqfveff);
4531: free_matrix(meanqt,1,lastpass,1,nqtveff);
4532: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4533: free_vector(pospropt,1,nlstate);
4534: free_vector(posprop,1,nlstate);
4535: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4536: free_vector(pp,1,nlstate);
4537: /* End of freqsummary */
4538: }
1.126 brouard 4539:
4540: /************ Prevalence ********************/
1.227 brouard 4541: 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)
4542: {
4543: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4544: in each health status at the date of interview (if between dateprev1 and dateprev2).
4545: We still use firstpass and lastpass as another selection.
4546: */
1.126 brouard 4547:
1.227 brouard 4548: int i, m, jk, j1, bool, z1,j, iv;
4549: int mi; /* Effective wave */
4550: int iage;
4551: double agebegin, ageend;
4552:
4553: double **prop;
4554: double posprop;
4555: double y2; /* in fractional years */
4556: int iagemin, iagemax;
4557: int first; /** to stop verbosity which is redirected to log file */
4558:
4559: iagemin= (int) agemin;
4560: iagemax= (int) agemax;
4561: /*pp=vector(1,nlstate);*/
4562: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4563: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4564: j1=0;
1.222 brouard 4565:
1.227 brouard 4566: /*j=cptcoveff;*/
4567: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4568:
1.227 brouard 4569: first=1;
4570: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4571: for (i=1; i<=nlstate; i++)
4572: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4573: prop[i][iage]=0.0;
4574: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4575: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4576: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4577:
4578: for (i=1; i<=imx; i++) { /* Each individual */
4579: bool=1;
4580: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4581: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4582: m=mw[mi][i];
4583: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4584: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4585: for (z1=1; z1<=cptcoveff; z1++){
4586: if( Fixed[Tmodelind[z1]]==1){
4587: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4588: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4589: bool=0;
4590: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4591: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4592: bool=0;
4593: }
4594: }
4595: if(bool==1){ /* Otherwise we skip that wave/person */
4596: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4597: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4598: if(m >=firstpass && m <=lastpass){
4599: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4600: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4601: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4602: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4603: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4604: 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);
4605: exit(1);
4606: }
4607: if (s[m][i]>0 && s[m][i]<=nlstate) {
4608: /*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]]);*/
4609: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4610: prop[s[m][i]][iagemax+3] += weight[i];
4611: } /* end valid statuses */
4612: } /* end selection of dates */
4613: } /* end selection of waves */
4614: } /* end bool */
4615: } /* end wave */
4616: } /* end individual */
4617: for(i=iagemin; i <= iagemax+3; i++){
4618: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4619: posprop += prop[jk][i];
4620: }
4621:
4622: for(jk=1; jk <=nlstate ; jk++){
4623: if( i <= iagemax){
4624: if(posprop>=1.e-5){
4625: probs[i][jk][j1]= prop[jk][i]/posprop;
4626: } else{
4627: if(first==1){
4628: first=0;
4629: 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]);
4630: }
4631: }
4632: }
4633: }/* end jk */
4634: }/* end i */
1.222 brouard 4635: /*} *//* end i1 */
1.227 brouard 4636: } /* end j1 */
1.222 brouard 4637:
1.227 brouard 4638: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4639: /*free_vector(pp,1,nlstate);*/
4640: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4641: } /* End of prevalence */
1.126 brouard 4642:
4643: /************* Waves Concatenation ***************/
4644:
4645: 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)
4646: {
4647: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4648: Death is a valid wave (if date is known).
4649: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4650: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4651: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4652: */
1.126 brouard 4653:
1.224 brouard 4654: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4655: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4656: double sum=0., jmean=0.;*/
1.224 brouard 4657: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4658: int j, k=0,jk, ju, jl;
4659: double sum=0.;
4660: first=0;
1.214 brouard 4661: firstwo=0;
1.217 brouard 4662: firsthree=0;
1.218 brouard 4663: firstfour=0;
1.164 brouard 4664: jmin=100000;
1.126 brouard 4665: jmax=-1;
4666: jmean=0.;
1.224 brouard 4667:
4668: /* Treating live states */
1.214 brouard 4669: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4670: mi=0; /* First valid wave */
1.227 brouard 4671: mli=0; /* Last valid wave */
1.126 brouard 4672: m=firstpass;
1.214 brouard 4673: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4674: 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 */
4675: mli=m-1;/* mw[++mi][i]=m-1; */
4676: }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 */
4677: mw[++mi][i]=m;
4678: mli=m;
1.224 brouard 4679: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4680: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4681: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4682: }
1.227 brouard 4683: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4684: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4685: break;
1.224 brouard 4686: #else
1.227 brouard 4687: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4688: if(firsthree == 0){
4689: 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);
4690: firsthree=1;
4691: }
4692: 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);
4693: mw[++mi][i]=m;
4694: mli=m;
4695: }
4696: if(s[m][i]==-2){ /* Vital status is really unknown */
4697: nbwarn++;
4698: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4699: 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);
4700: 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);
4701: }
4702: break;
4703: }
4704: break;
1.224 brouard 4705: #endif
1.227 brouard 4706: }/* End m >= lastpass */
1.126 brouard 4707: }/* end while */
1.224 brouard 4708:
1.227 brouard 4709: /* 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 4710: /* After last pass */
1.224 brouard 4711: /* Treating death states */
1.214 brouard 4712: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4713: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4714: /* } */
1.126 brouard 4715: mi++; /* Death is another wave */
4716: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4717: /* Only death is a correct wave */
1.126 brouard 4718: mw[mi][i]=m;
1.224 brouard 4719: }
4720: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4721: 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 4722: /* m++; */
4723: /* mi++; */
4724: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4725: /* mw[mi][i]=m; */
1.218 brouard 4726: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4727: 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 */
4728: nbwarn++;
4729: if(firstfiv==0){
4730: 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 );
4731: firstfiv=1;
4732: }else{
4733: 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 );
4734: }
4735: }else{ /* Death occured afer last wave potential bias */
4736: nberr++;
4737: if(firstwo==0){
4738: 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 );
4739: firstwo=1;
4740: }
4741: 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 );
4742: }
1.218 brouard 4743: }else{ /* end date of interview is known */
1.227 brouard 4744: /* death is known but not confirmed by death status at any wave */
4745: if(firstfour==0){
4746: 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 );
4747: firstfour=1;
4748: }
4749: 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 4750: }
1.224 brouard 4751: } /* end if date of death is known */
4752: #endif
4753: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4754: /* wav[i]=mw[mi][i]; */
1.126 brouard 4755: if(mi==0){
4756: nbwarn++;
4757: if(first==0){
1.227 brouard 4758: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4759: first=1;
1.126 brouard 4760: }
4761: if(first==1){
1.227 brouard 4762: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4763: }
4764: } /* end mi==0 */
4765: } /* End individuals */
1.214 brouard 4766: /* wav and mw are no more changed */
1.223 brouard 4767:
1.214 brouard 4768:
1.126 brouard 4769: for(i=1; i<=imx; i++){
4770: for(mi=1; mi<wav[i];mi++){
4771: if (stepm <=0)
1.227 brouard 4772: dh[mi][i]=1;
1.126 brouard 4773: else{
1.227 brouard 4774: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4775: if (agedc[i] < 2*AGESUP) {
4776: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4777: if(j==0) j=1; /* Survives at least one month after exam */
4778: else if(j<0){
4779: nberr++;
4780: 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]);
4781: j=1; /* Temporary Dangerous patch */
4782: 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);
4783: 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]);
4784: 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);
4785: }
4786: k=k+1;
4787: if (j >= jmax){
4788: jmax=j;
4789: ijmax=i;
4790: }
4791: if (j <= jmin){
4792: jmin=j;
4793: ijmin=i;
4794: }
4795: sum=sum+j;
4796: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4797: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4798: }
4799: }
4800: else{
4801: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4802: /* 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 4803:
1.227 brouard 4804: k=k+1;
4805: if (j >= jmax) {
4806: jmax=j;
4807: ijmax=i;
4808: }
4809: else if (j <= jmin){
4810: jmin=j;
4811: ijmin=i;
4812: }
4813: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4814: /*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]);*/
4815: if(j<0){
4816: nberr++;
4817: 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]);
4818: 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]);
4819: }
4820: sum=sum+j;
4821: }
4822: jk= j/stepm;
4823: jl= j -jk*stepm;
4824: ju= j -(jk+1)*stepm;
4825: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4826: if(jl==0){
4827: dh[mi][i]=jk;
4828: bh[mi][i]=0;
4829: }else{ /* We want a negative bias in order to only have interpolation ie
4830: * to avoid the price of an extra matrix product in likelihood */
4831: dh[mi][i]=jk+1;
4832: bh[mi][i]=ju;
4833: }
4834: }else{
4835: if(jl <= -ju){
4836: dh[mi][i]=jk;
4837: bh[mi][i]=jl; /* bias is positive if real duration
4838: * is higher than the multiple of stepm and negative otherwise.
4839: */
4840: }
4841: else{
4842: dh[mi][i]=jk+1;
4843: bh[mi][i]=ju;
4844: }
4845: if(dh[mi][i]==0){
4846: dh[mi][i]=1; /* At least one step */
4847: bh[mi][i]=ju; /* At least one step */
4848: /* 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);*/
4849: }
4850: } /* end if mle */
1.126 brouard 4851: }
4852: } /* end wave */
4853: }
4854: jmean=sum/k;
4855: 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 4856: 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 4857: }
1.126 brouard 4858:
4859: /*********** Tricode ****************************/
1.220 brouard 4860: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4861: {
4862: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4863: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4864: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4865: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4866: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4867: */
1.130 brouard 4868:
1.242 brouard 4869: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
4870: int modmaxcovj=0; /* Modality max of covariates j */
4871: int cptcode=0; /* Modality max of covariates j */
4872: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4873:
4874:
1.242 brouard 4875: /* cptcoveff=0; */
4876: /* *cptcov=0; */
1.126 brouard 4877:
1.242 brouard 4878: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4879:
1.242 brouard 4880: /* Loop on covariates without age and products and no quantitative variable */
4881: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
4882: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4883: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4884: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4885: switch(Fixed[k]) {
4886: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
4887: 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*/
4888: ij=(int)(covar[Tvar[k]][i]);
4889: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4890: * If product of Vn*Vm, still boolean *:
4891: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4892: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4893: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4894: modality of the nth covariate of individual i. */
4895: if (ij > modmaxcovj)
4896: modmaxcovj=ij;
4897: else if (ij < modmincovj)
4898: modmincovj=ij;
4899: if ((ij < -1) && (ij > NCOVMAX)){
4900: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
4901: exit(1);
4902: }else
4903: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
4904: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
4905: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
4906: /* getting the maximum value of the modality of the covariate
4907: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
4908: female ies 1, then modmaxcovj=1.
4909: */
4910: } /* end for loop on individuals i */
4911: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4912: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4913: cptcode=modmaxcovj;
4914: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
4915: /*for (i=0; i<=cptcode; i++) {*/
4916: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
4917: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4918: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4919: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
4920: if( j != -1){
4921: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
4922: covariate for which somebody answered excluding
4923: undefined. Usually 2: 0 and 1. */
4924: }
4925: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
4926: covariate for which somebody answered including
4927: undefined. Usually 3: -1, 0 and 1. */
4928: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
4929: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
4930: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 4931:
1.242 brouard 4932: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
4933: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
4934: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
4935: /* modmincovj=3; modmaxcovj = 7; */
4936: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
4937: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
4938: /* defining two dummy variables: variables V1_1 and V1_2.*/
4939: /* nbcode[Tvar[j]][ij]=k; */
4940: /* nbcode[Tvar[j]][1]=0; */
4941: /* nbcode[Tvar[j]][2]=1; */
4942: /* nbcode[Tvar[j]][3]=2; */
4943: /* To be continued (not working yet). */
4944: ij=0; /* ij is similar to i but can jump over null modalities */
4945: 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*/
4946: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
4947: break;
4948: }
4949: ij++;
4950: 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*/
4951: cptcode = ij; /* New max modality for covar j */
4952: } /* end of loop on modality i=-1 to 1 or more */
4953: break;
4954: case 1: /* Testing on varying covariate, could be simple and
4955: * should look at waves or product of fixed *
4956: * varying. No time to test -1, assuming 0 and 1 only */
4957: ij=0;
4958: for(i=0; i<=1;i++){
4959: nbcode[Tvar[k]][++ij]=i;
4960: }
4961: break;
4962: default:
4963: break;
4964: } /* end switch */
4965: } /* end dummy test */
4966:
4967: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
4968: /* /\*recode from 0 *\/ */
4969: /* k is a modality. If we have model=V1+V1*sex */
4970: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
4971: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
4972: /* } */
4973: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
4974: /* if (ij > ncodemax[j]) { */
4975: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4976: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
4977: /* break; */
4978: /* } */
4979: /* } /\* end of loop on modality k *\/ */
4980: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
4981:
4982: for (k=-1; k< maxncov; k++) Ndum[k]=0;
4983: /* Look at fixed dummy (single or product) covariates to check empty modalities */
4984: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
4985: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
4986: 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 */
4987: 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 */
4988: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
4989: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
4990:
4991: ij=0;
4992: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
4993: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
4994: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
4995: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
4996: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
4997: /* If product not in single variable we don't print results */
4998: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
4999: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5000: 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*/
5001: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5002: 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 */
5003: if(Fixed[k]!=0)
5004: anyvaryingduminmodel=1;
5005: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5006: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5007: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5008: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5009: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5010: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5011: }
5012: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5013: /* ij--; */
5014: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5015: *cptcov=ij; /*Number of total real effective covariates: effective
5016: * because they can be excluded from the model and real
5017: * if in the model but excluded because missing values, but how to get k from ij?*/
5018: for(j=ij+1; j<= cptcovt; j++){
5019: Tvaraff[j]=0;
5020: Tmodelind[j]=0;
5021: }
5022: for(j=ntveff+1; j<= cptcovt; j++){
5023: TmodelInvind[j]=0;
5024: }
5025: /* To be sorted */
5026: ;
5027: }
1.126 brouard 5028:
1.145 brouard 5029:
1.126 brouard 5030: /*********** Health Expectancies ****************/
5031:
1.235 brouard 5032: 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 5033:
5034: {
5035: /* Health expectancies, no variances */
1.164 brouard 5036: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5037: int nhstepma, nstepma; /* Decreasing with age */
5038: double age, agelim, hf;
5039: double ***p3mat;
5040: double eip;
5041:
1.238 brouard 5042: /* pstamp(ficreseij); */
1.126 brouard 5043: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5044: fprintf(ficreseij,"# Age");
5045: for(i=1; i<=nlstate;i++){
5046: for(j=1; j<=nlstate;j++){
5047: fprintf(ficreseij," e%1d%1d ",i,j);
5048: }
5049: fprintf(ficreseij," e%1d. ",i);
5050: }
5051: fprintf(ficreseij,"\n");
5052:
5053:
5054: if(estepm < stepm){
5055: printf ("Problem %d lower than %d\n",estepm, stepm);
5056: }
5057: else hstepm=estepm;
5058: /* We compute the life expectancy from trapezoids spaced every estepm months
5059: * This is mainly to measure the difference between two models: for example
5060: * if stepm=24 months pijx are given only every 2 years and by summing them
5061: * we are calculating an estimate of the Life Expectancy assuming a linear
5062: * progression in between and thus overestimating or underestimating according
5063: * to the curvature of the survival function. If, for the same date, we
5064: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5065: * to compare the new estimate of Life expectancy with the same linear
5066: * hypothesis. A more precise result, taking into account a more precise
5067: * curvature will be obtained if estepm is as small as stepm. */
5068:
5069: /* For example we decided to compute the life expectancy with the smallest unit */
5070: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5071: nhstepm is the number of hstepm from age to agelim
5072: nstepm is the number of stepm from age to agelin.
5073: Look at hpijx to understand the reason of that which relies in memory size
5074: and note for a fixed period like estepm months */
5075: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5076: survival function given by stepm (the optimization length). Unfortunately it
5077: means that if the survival funtion is printed only each two years of age and if
5078: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5079: results. So we changed our mind and took the option of the best precision.
5080: */
5081: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5082:
5083: agelim=AGESUP;
5084: /* If stepm=6 months */
5085: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5086: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5087:
5088: /* nhstepm age range expressed in number of stepm */
5089: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5090: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5091: /* if (stepm >= YEARM) hstepm=1;*/
5092: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5093: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5094:
5095: for (age=bage; age<=fage; age ++){
5096: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5097: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5098: /* if (stepm >= YEARM) hstepm=1;*/
5099: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5100:
5101: /* If stepm=6 months */
5102: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5103: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5104:
1.235 brouard 5105: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5106:
5107: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5108:
5109: printf("%d|",(int)age);fflush(stdout);
5110: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5111:
5112: /* Computing expectancies */
5113: for(i=1; i<=nlstate;i++)
5114: for(j=1; j<=nlstate;j++)
5115: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5116: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5117:
5118: /* 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]);*/
5119:
5120: }
5121:
5122: fprintf(ficreseij,"%3.0f",age );
5123: for(i=1; i<=nlstate;i++){
5124: eip=0;
5125: for(j=1; j<=nlstate;j++){
5126: eip +=eij[i][j][(int)age];
5127: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5128: }
5129: fprintf(ficreseij,"%9.4f", eip );
5130: }
5131: fprintf(ficreseij,"\n");
5132:
5133: }
5134: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5135: printf("\n");
5136: fprintf(ficlog,"\n");
5137:
5138: }
5139:
1.235 brouard 5140: 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 5141:
5142: {
5143: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5144: to initial status i, ei. .
1.126 brouard 5145: */
5146: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5147: int nhstepma, nstepma; /* Decreasing with age */
5148: double age, agelim, hf;
5149: double ***p3matp, ***p3matm, ***varhe;
5150: double **dnewm,**doldm;
5151: double *xp, *xm;
5152: double **gp, **gm;
5153: double ***gradg, ***trgradg;
5154: int theta;
5155:
5156: double eip, vip;
5157:
5158: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5159: xp=vector(1,npar);
5160: xm=vector(1,npar);
5161: dnewm=matrix(1,nlstate*nlstate,1,npar);
5162: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5163:
5164: pstamp(ficresstdeij);
5165: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5166: fprintf(ficresstdeij,"# Age");
5167: for(i=1; i<=nlstate;i++){
5168: for(j=1; j<=nlstate;j++)
5169: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5170: fprintf(ficresstdeij," e%1d. ",i);
5171: }
5172: fprintf(ficresstdeij,"\n");
5173:
5174: pstamp(ficrescveij);
5175: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5176: fprintf(ficrescveij,"# Age");
5177: for(i=1; i<=nlstate;i++)
5178: for(j=1; j<=nlstate;j++){
5179: cptj= (j-1)*nlstate+i;
5180: for(i2=1; i2<=nlstate;i2++)
5181: for(j2=1; j2<=nlstate;j2++){
5182: cptj2= (j2-1)*nlstate+i2;
5183: if(cptj2 <= cptj)
5184: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5185: }
5186: }
5187: fprintf(ficrescveij,"\n");
5188:
5189: if(estepm < stepm){
5190: printf ("Problem %d lower than %d\n",estepm, stepm);
5191: }
5192: else hstepm=estepm;
5193: /* We compute the life expectancy from trapezoids spaced every estepm months
5194: * This is mainly to measure the difference between two models: for example
5195: * if stepm=24 months pijx are given only every 2 years and by summing them
5196: * we are calculating an estimate of the Life Expectancy assuming a linear
5197: * progression in between and thus overestimating or underestimating according
5198: * to the curvature of the survival function. If, for the same date, we
5199: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5200: * to compare the new estimate of Life expectancy with the same linear
5201: * hypothesis. A more precise result, taking into account a more precise
5202: * curvature will be obtained if estepm is as small as stepm. */
5203:
5204: /* For example we decided to compute the life expectancy with the smallest unit */
5205: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5206: nhstepm is the number of hstepm from age to agelim
5207: nstepm is the number of stepm from age to agelin.
5208: Look at hpijx to understand the reason of that which relies in memory size
5209: and note for a fixed period like estepm months */
5210: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5211: survival function given by stepm (the optimization length). Unfortunately it
5212: means that if the survival funtion is printed only each two years of age and if
5213: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5214: results. So we changed our mind and took the option of the best precision.
5215: */
5216: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5217:
5218: /* If stepm=6 months */
5219: /* nhstepm age range expressed in number of stepm */
5220: agelim=AGESUP;
5221: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5222: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5223: /* if (stepm >= YEARM) hstepm=1;*/
5224: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5225:
5226: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5227: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5228: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5229: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5230: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5231: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5232:
5233: for (age=bage; age<=fage; age ++){
5234: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5235: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5236: /* if (stepm >= YEARM) hstepm=1;*/
5237: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5238:
1.126 brouard 5239: /* If stepm=6 months */
5240: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5241: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5242:
5243: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5244:
1.126 brouard 5245: /* Computing Variances of health expectancies */
5246: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5247: decrease memory allocation */
5248: for(theta=1; theta <=npar; theta++){
5249: for(i=1; i<=npar; i++){
1.222 brouard 5250: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5251: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5252: }
1.235 brouard 5253: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5254: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5255:
1.126 brouard 5256: for(j=1; j<= nlstate; j++){
1.222 brouard 5257: for(i=1; i<=nlstate; i++){
5258: for(h=0; h<=nhstepm-1; h++){
5259: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5260: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5261: }
5262: }
1.126 brouard 5263: }
1.218 brouard 5264:
1.126 brouard 5265: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5266: for(h=0; h<=nhstepm-1; h++){
5267: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5268: }
1.126 brouard 5269: }/* End theta */
5270:
5271:
5272: for(h=0; h<=nhstepm-1; h++)
5273: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5274: for(theta=1; theta <=npar; theta++)
5275: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5276:
1.218 brouard 5277:
1.222 brouard 5278: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5279: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5280: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5281:
1.222 brouard 5282: printf("%d|",(int)age);fflush(stdout);
5283: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5284: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5285: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5286: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5287: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5288: for(ij=1;ij<=nlstate*nlstate;ij++)
5289: for(ji=1;ji<=nlstate*nlstate;ji++)
5290: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5291: }
5292: }
1.218 brouard 5293:
1.126 brouard 5294: /* Computing expectancies */
1.235 brouard 5295: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5296: for(i=1; i<=nlstate;i++)
5297: for(j=1; j<=nlstate;j++)
1.222 brouard 5298: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5299: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5300:
1.222 brouard 5301: /* 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 5302:
1.222 brouard 5303: }
1.218 brouard 5304:
1.126 brouard 5305: fprintf(ficresstdeij,"%3.0f",age );
5306: for(i=1; i<=nlstate;i++){
5307: eip=0.;
5308: vip=0.;
5309: for(j=1; j<=nlstate;j++){
1.222 brouard 5310: eip += eij[i][j][(int)age];
5311: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5312: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5313: 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 5314: }
5315: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5316: }
5317: fprintf(ficresstdeij,"\n");
1.218 brouard 5318:
1.126 brouard 5319: fprintf(ficrescveij,"%3.0f",age );
5320: for(i=1; i<=nlstate;i++)
5321: for(j=1; j<=nlstate;j++){
1.222 brouard 5322: cptj= (j-1)*nlstate+i;
5323: for(i2=1; i2<=nlstate;i2++)
5324: for(j2=1; j2<=nlstate;j2++){
5325: cptj2= (j2-1)*nlstate+i2;
5326: if(cptj2 <= cptj)
5327: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5328: }
1.126 brouard 5329: }
5330: fprintf(ficrescveij,"\n");
1.218 brouard 5331:
1.126 brouard 5332: }
5333: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5334: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5335: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5336: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5337: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5338: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5339: printf("\n");
5340: fprintf(ficlog,"\n");
1.218 brouard 5341:
1.126 brouard 5342: free_vector(xm,1,npar);
5343: free_vector(xp,1,npar);
5344: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5345: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5346: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5347: }
1.218 brouard 5348:
1.126 brouard 5349: /************ Variance ******************/
1.235 brouard 5350: 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 5351: {
5352: /* Variance of health expectancies */
5353: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5354: /* double **newm;*/
5355: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5356:
5357: /* int movingaverage(); */
5358: double **dnewm,**doldm;
5359: double **dnewmp,**doldmp;
5360: int i, j, nhstepm, hstepm, h, nstepm ;
5361: int k;
5362: double *xp;
5363: double **gp, **gm; /* for var eij */
5364: double ***gradg, ***trgradg; /*for var eij */
5365: double **gradgp, **trgradgp; /* for var p point j */
5366: double *gpp, *gmp; /* for var p point j */
5367: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5368: double ***p3mat;
5369: double age,agelim, hf;
5370: /* double ***mobaverage; */
5371: int theta;
5372: char digit[4];
5373: char digitp[25];
5374:
5375: char fileresprobmorprev[FILENAMELENGTH];
5376:
5377: if(popbased==1){
5378: if(mobilav!=0)
5379: strcpy(digitp,"-POPULBASED-MOBILAV_");
5380: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5381: }
5382: else
5383: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5384:
1.218 brouard 5385: /* if (mobilav!=0) { */
5386: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5387: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5388: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5389: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5390: /* } */
5391: /* } */
5392:
5393: strcpy(fileresprobmorprev,"PRMORPREV-");
5394: sprintf(digit,"%-d",ij);
5395: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5396: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5397: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5398: strcat(fileresprobmorprev,fileresu);
5399: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5400: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5401: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5402: }
5403: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5404: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5405: pstamp(ficresprobmorprev);
5406: 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 5407: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5408: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5409: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5410: }
5411: for(j=1;j<=cptcoveff;j++)
5412: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5413: fprintf(ficresprobmorprev,"\n");
5414:
1.218 brouard 5415: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5416: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5417: fprintf(ficresprobmorprev," p.%-d SE",j);
5418: for(i=1; i<=nlstate;i++)
5419: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5420: }
5421: fprintf(ficresprobmorprev,"\n");
5422:
5423: fprintf(ficgp,"\n# Routine varevsij");
5424: fprintf(ficgp,"\nunset title \n");
5425: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5426: 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");
5427: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5428: /* } */
5429: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5430: pstamp(ficresvij);
5431: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5432: if(popbased==1)
5433: 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);
5434: else
5435: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5436: fprintf(ficresvij,"# Age");
5437: for(i=1; i<=nlstate;i++)
5438: for(j=1; j<=nlstate;j++)
5439: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5440: fprintf(ficresvij,"\n");
5441:
5442: xp=vector(1,npar);
5443: dnewm=matrix(1,nlstate,1,npar);
5444: doldm=matrix(1,nlstate,1,nlstate);
5445: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5446: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5447:
5448: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5449: gpp=vector(nlstate+1,nlstate+ndeath);
5450: gmp=vector(nlstate+1,nlstate+ndeath);
5451: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5452:
1.218 brouard 5453: if(estepm < stepm){
5454: printf ("Problem %d lower than %d\n",estepm, stepm);
5455: }
5456: else hstepm=estepm;
5457: /* For example we decided to compute the life expectancy with the smallest unit */
5458: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5459: nhstepm is the number of hstepm from age to agelim
5460: nstepm is the number of stepm from age to agelim.
5461: Look at function hpijx to understand why because of memory size limitations,
5462: we decided (b) to get a life expectancy respecting the most precise curvature of the
5463: survival function given by stepm (the optimization length). Unfortunately it
5464: means that if the survival funtion is printed every two years of age and if
5465: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5466: results. So we changed our mind and took the option of the best precision.
5467: */
5468: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5469: agelim = AGESUP;
5470: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5471: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5472: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5473: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5474: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5475: gp=matrix(0,nhstepm,1,nlstate);
5476: gm=matrix(0,nhstepm,1,nlstate);
5477:
5478:
5479: for(theta=1; theta <=npar; theta++){
5480: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5481: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5482: }
5483:
1.242 brouard 5484: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5485:
5486: if (popbased==1) {
5487: if(mobilav ==0){
5488: for(i=1; i<=nlstate;i++)
5489: prlim[i][i]=probs[(int)age][i][ij];
5490: }else{ /* mobilav */
5491: for(i=1; i<=nlstate;i++)
5492: prlim[i][i]=mobaverage[(int)age][i][ij];
5493: }
5494: }
5495:
1.235 brouard 5496: 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 5497: for(j=1; j<= nlstate; j++){
5498: for(h=0; h<=nhstepm; h++){
5499: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5500: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5501: }
5502: }
5503: /* Next for computing probability of death (h=1 means
5504: computed over hstepm matrices product = hstepm*stepm months)
5505: as a weighted average of prlim.
5506: */
5507: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5508: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5509: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5510: }
5511: /* end probability of death */
5512:
5513: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5514: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5515:
1.242 brouard 5516: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5517:
5518: if (popbased==1) {
5519: if(mobilav ==0){
5520: for(i=1; i<=nlstate;i++)
5521: prlim[i][i]=probs[(int)age][i][ij];
5522: }else{ /* mobilav */
5523: for(i=1; i<=nlstate;i++)
5524: prlim[i][i]=mobaverage[(int)age][i][ij];
5525: }
5526: }
5527:
1.235 brouard 5528: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5529:
5530: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5531: for(h=0; h<=nhstepm; h++){
5532: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5533: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5534: }
5535: }
5536: /* This for computing probability of death (h=1 means
5537: computed over hstepm matrices product = hstepm*stepm months)
5538: as a weighted average of prlim.
5539: */
5540: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5541: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5542: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5543: }
5544: /* end probability of death */
5545:
5546: for(j=1; j<= nlstate; j++) /* vareij */
5547: for(h=0; h<=nhstepm; h++){
5548: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5549: }
5550:
5551: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5552: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5553: }
5554:
5555: } /* End theta */
5556:
5557: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5558:
5559: for(h=0; h<=nhstepm; h++) /* veij */
5560: for(j=1; j<=nlstate;j++)
5561: for(theta=1; theta <=npar; theta++)
5562: trgradg[h][j][theta]=gradg[h][theta][j];
5563:
5564: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5565: for(theta=1; theta <=npar; theta++)
5566: trgradgp[j][theta]=gradgp[theta][j];
5567:
5568:
5569: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5570: for(i=1;i<=nlstate;i++)
5571: for(j=1;j<=nlstate;j++)
5572: vareij[i][j][(int)age] =0.;
5573:
5574: for(h=0;h<=nhstepm;h++){
5575: for(k=0;k<=nhstepm;k++){
5576: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5577: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5578: for(i=1;i<=nlstate;i++)
5579: for(j=1;j<=nlstate;j++)
5580: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5581: }
5582: }
5583:
5584: /* pptj */
5585: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5586: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5587: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5588: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5589: varppt[j][i]=doldmp[j][i];
5590: /* end ppptj */
5591: /* x centered again */
5592:
1.242 brouard 5593: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5594:
5595: if (popbased==1) {
5596: if(mobilav ==0){
5597: for(i=1; i<=nlstate;i++)
5598: prlim[i][i]=probs[(int)age][i][ij];
5599: }else{ /* mobilav */
5600: for(i=1; i<=nlstate;i++)
5601: prlim[i][i]=mobaverage[(int)age][i][ij];
5602: }
5603: }
5604:
5605: /* This for computing probability of death (h=1 means
5606: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5607: as a weighted average of prlim.
5608: */
1.235 brouard 5609: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5610: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5611: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5612: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5613: }
5614: /* end probability of death */
5615:
5616: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5617: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5618: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5619: for(i=1; i<=nlstate;i++){
5620: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5621: }
5622: }
5623: fprintf(ficresprobmorprev,"\n");
5624:
5625: fprintf(ficresvij,"%.0f ",age );
5626: for(i=1; i<=nlstate;i++)
5627: for(j=1; j<=nlstate;j++){
5628: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5629: }
5630: fprintf(ficresvij,"\n");
5631: free_matrix(gp,0,nhstepm,1,nlstate);
5632: free_matrix(gm,0,nhstepm,1,nlstate);
5633: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5634: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5635: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5636: } /* End age */
5637: free_vector(gpp,nlstate+1,nlstate+ndeath);
5638: free_vector(gmp,nlstate+1,nlstate+ndeath);
5639: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5640: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5641: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5642: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5643: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5644: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5645: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5646: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5647: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5648: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5649: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5650: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5651: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5652: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5653: 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);
5654: /* 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 5655: */
1.218 brouard 5656: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5657: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5658:
1.218 brouard 5659: free_vector(xp,1,npar);
5660: free_matrix(doldm,1,nlstate,1,nlstate);
5661: free_matrix(dnewm,1,nlstate,1,npar);
5662: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5663: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5664: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5665: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5666: fclose(ficresprobmorprev);
5667: fflush(ficgp);
5668: fflush(fichtm);
5669: } /* end varevsij */
1.126 brouard 5670:
5671: /************ Variance of prevlim ******************/
1.235 brouard 5672: 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 5673: {
1.205 brouard 5674: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5675: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5676:
1.126 brouard 5677: double **dnewm,**doldm;
5678: int i, j, nhstepm, hstepm;
5679: double *xp;
5680: double *gp, *gm;
5681: double **gradg, **trgradg;
1.208 brouard 5682: double **mgm, **mgp;
1.126 brouard 5683: double age,agelim;
5684: int theta;
5685:
5686: pstamp(ficresvpl);
5687: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5688: fprintf(ficresvpl,"# Age ");
5689: if(nresult >=1)
5690: fprintf(ficresvpl," Result# ");
1.126 brouard 5691: for(i=1; i<=nlstate;i++)
5692: fprintf(ficresvpl," %1d-%1d",i,i);
5693: fprintf(ficresvpl,"\n");
5694:
5695: xp=vector(1,npar);
5696: dnewm=matrix(1,nlstate,1,npar);
5697: doldm=matrix(1,nlstate,1,nlstate);
5698:
5699: hstepm=1*YEARM; /* Every year of age */
5700: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5701: agelim = AGESUP;
5702: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5703: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5704: if (stepm >= YEARM) hstepm=1;
5705: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5706: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5707: mgp=matrix(1,npar,1,nlstate);
5708: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5709: gp=vector(1,nlstate);
5710: gm=vector(1,nlstate);
5711:
5712: for(theta=1; theta <=npar; theta++){
5713: for(i=1; i<=npar; i++){ /* Computes gradient */
5714: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5715: }
1.209 brouard 5716: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5717: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5718: else
1.235 brouard 5719: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5720: for(i=1;i<=nlstate;i++){
1.126 brouard 5721: gp[i] = prlim[i][i];
1.208 brouard 5722: mgp[theta][i] = prlim[i][i];
5723: }
1.126 brouard 5724: for(i=1; i<=npar; i++) /* Computes gradient */
5725: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5726: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5727: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5728: else
1.235 brouard 5729: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5730: for(i=1;i<=nlstate;i++){
1.126 brouard 5731: gm[i] = prlim[i][i];
1.208 brouard 5732: mgm[theta][i] = prlim[i][i];
5733: }
1.126 brouard 5734: for(i=1;i<=nlstate;i++)
5735: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5736: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5737: } /* End theta */
5738:
5739: trgradg =matrix(1,nlstate,1,npar);
5740:
5741: for(j=1; j<=nlstate;j++)
5742: for(theta=1; theta <=npar; theta++)
5743: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5744: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5745: /* printf("\nmgm mgp %d ",(int)age); */
5746: /* for(j=1; j<=nlstate;j++){ */
5747: /* printf(" %d ",j); */
5748: /* for(theta=1; theta <=npar; theta++) */
5749: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5750: /* printf("\n "); */
5751: /* } */
5752: /* } */
5753: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5754: /* printf("\n gradg %d ",(int)age); */
5755: /* for(j=1; j<=nlstate;j++){ */
5756: /* printf("%d ",j); */
5757: /* for(theta=1; theta <=npar; theta++) */
5758: /* printf("%d %lf ",theta,gradg[theta][j]); */
5759: /* printf("\n "); */
5760: /* } */
5761: /* } */
1.126 brouard 5762:
5763: for(i=1;i<=nlstate;i++)
5764: varpl[i][(int)age] =0.;
1.209 brouard 5765: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5766: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5767: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5768: }else{
1.126 brouard 5769: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5770: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5771: }
1.126 brouard 5772: for(i=1;i<=nlstate;i++)
5773: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5774:
5775: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5776: if(nresult >=1)
5777: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5778: for(i=1; i<=nlstate;i++)
5779: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5780: fprintf(ficresvpl,"\n");
5781: free_vector(gp,1,nlstate);
5782: free_vector(gm,1,nlstate);
1.208 brouard 5783: free_matrix(mgm,1,npar,1,nlstate);
5784: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5785: free_matrix(gradg,1,npar,1,nlstate);
5786: free_matrix(trgradg,1,nlstate,1,npar);
5787: } /* End age */
5788:
5789: free_vector(xp,1,npar);
5790: free_matrix(doldm,1,nlstate,1,npar);
5791: free_matrix(dnewm,1,nlstate,1,nlstate);
5792:
5793: }
5794:
5795: /************ Variance of one-step probabilities ******************/
5796: 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 5797: {
5798: int i, j=0, k1, l1, tj;
5799: int k2, l2, j1, z1;
5800: int k=0, l;
5801: int first=1, first1, first2;
5802: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5803: double **dnewm,**doldm;
5804: double *xp;
5805: double *gp, *gm;
5806: double **gradg, **trgradg;
5807: double **mu;
5808: double age, cov[NCOVMAX+1];
5809: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5810: int theta;
5811: char fileresprob[FILENAMELENGTH];
5812: char fileresprobcov[FILENAMELENGTH];
5813: char fileresprobcor[FILENAMELENGTH];
5814: double ***varpij;
5815:
5816: strcpy(fileresprob,"PROB_");
5817: strcat(fileresprob,fileres);
5818: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5819: printf("Problem with resultfile: %s\n", fileresprob);
5820: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5821: }
5822: strcpy(fileresprobcov,"PROBCOV_");
5823: strcat(fileresprobcov,fileresu);
5824: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5825: printf("Problem with resultfile: %s\n", fileresprobcov);
5826: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5827: }
5828: strcpy(fileresprobcor,"PROBCOR_");
5829: strcat(fileresprobcor,fileresu);
5830: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5831: printf("Problem with resultfile: %s\n", fileresprobcor);
5832: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5833: }
5834: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5835: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5836: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5837: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5838: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5839: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5840: pstamp(ficresprob);
5841: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5842: fprintf(ficresprob,"# Age");
5843: pstamp(ficresprobcov);
5844: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5845: fprintf(ficresprobcov,"# Age");
5846: pstamp(ficresprobcor);
5847: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5848: fprintf(ficresprobcor,"# Age");
1.126 brouard 5849:
5850:
1.222 brouard 5851: for(i=1; i<=nlstate;i++)
5852: for(j=1; j<=(nlstate+ndeath);j++){
5853: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5854: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5855: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5856: }
5857: /* fprintf(ficresprob,"\n");
5858: fprintf(ficresprobcov,"\n");
5859: fprintf(ficresprobcor,"\n");
5860: */
5861: xp=vector(1,npar);
5862: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5863: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5864: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5865: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5866: first=1;
5867: fprintf(ficgp,"\n# Routine varprob");
5868: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5869: fprintf(fichtm,"\n");
5870:
5871: 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);
5872: 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);
5873: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5874: and drawn. It helps understanding how is the covariance between two incidences.\
5875: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5876: 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 5877: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5878: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5879: standard deviations wide on each axis. <br>\
5880: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5881: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5882: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5883:
1.222 brouard 5884: cov[1]=1;
5885: /* tj=cptcoveff; */
1.225 brouard 5886: tj = (int) pow(2,cptcoveff);
1.222 brouard 5887: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5888: j1=0;
1.224 brouard 5889: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5890: if (cptcovn>0) {
5891: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5892: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5893: fprintf(ficresprob, "**********\n#\n");
5894: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5895: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5896: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5897:
1.222 brouard 5898: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5899: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5900: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5901:
5902:
1.222 brouard 5903: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5904: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5905: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5906:
1.222 brouard 5907: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5908: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5909: fprintf(ficresprobcor, "**********\n#");
5910: if(invalidvarcomb[j1]){
5911: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5912: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5913: continue;
5914: }
5915: }
5916: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5917: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5918: gp=vector(1,(nlstate)*(nlstate+ndeath));
5919: gm=vector(1,(nlstate)*(nlstate+ndeath));
5920: for (age=bage; age<=fage; age ++){
5921: cov[2]=age;
5922: if(nagesqr==1)
5923: cov[3]= age*age;
5924: for (k=1; k<=cptcovn;k++) {
5925: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5926: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5927: * 1 1 1 1 1
5928: * 2 2 1 1 1
5929: * 3 1 2 1 1
5930: */
5931: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5932: }
5933: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5934: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5935: for (k=1; k<=cptcovprod;k++)
5936: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5937:
5938:
1.222 brouard 5939: for(theta=1; theta <=npar; theta++){
5940: for(i=1; i<=npar; i++)
5941: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 5942:
1.222 brouard 5943: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 5944:
1.222 brouard 5945: k=0;
5946: for(i=1; i<= (nlstate); i++){
5947: for(j=1; j<=(nlstate+ndeath);j++){
5948: k=k+1;
5949: gp[k]=pmmij[i][j];
5950: }
5951: }
1.220 brouard 5952:
1.222 brouard 5953: for(i=1; i<=npar; i++)
5954: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 5955:
1.222 brouard 5956: pmij(pmmij,cov,ncovmodel,xp,nlstate);
5957: k=0;
5958: for(i=1; i<=(nlstate); i++){
5959: for(j=1; j<=(nlstate+ndeath);j++){
5960: k=k+1;
5961: gm[k]=pmmij[i][j];
5962: }
5963: }
1.220 brouard 5964:
1.222 brouard 5965: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
5966: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
5967: }
1.126 brouard 5968:
1.222 brouard 5969: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
5970: for(theta=1; theta <=npar; theta++)
5971: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 5972:
1.222 brouard 5973: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
5974: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 5975:
1.222 brouard 5976: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 5977:
1.222 brouard 5978: k=0;
5979: for(i=1; i<=(nlstate); i++){
5980: for(j=1; j<=(nlstate+ndeath);j++){
5981: k=k+1;
5982: mu[k][(int) age]=pmmij[i][j];
5983: }
5984: }
5985: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
5986: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
5987: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 5988:
1.222 brouard 5989: /*printf("\n%d ",(int)age);
5990: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
5991: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5992: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
5993: }*/
1.220 brouard 5994:
1.222 brouard 5995: fprintf(ficresprob,"\n%d ",(int)age);
5996: fprintf(ficresprobcov,"\n%d ",(int)age);
5997: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 5998:
1.222 brouard 5999: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6000: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6001: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6002: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6003: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6004: }
6005: i=0;
6006: for (k=1; k<=(nlstate);k++){
6007: for (l=1; l<=(nlstate+ndeath);l++){
6008: i++;
6009: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6010: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6011: for (j=1; j<=i;j++){
6012: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6013: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6014: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6015: }
6016: }
6017: }/* end of loop for state */
6018: } /* end of loop for age */
6019: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6020: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6021: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6022: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6023:
6024: /* Confidence intervalle of pij */
6025: /*
6026: fprintf(ficgp,"\nunset parametric;unset label");
6027: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6028: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6029: 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);
6030: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6031: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6032: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6033: */
6034:
6035: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6036: first1=1;first2=2;
6037: for (k2=1; k2<=(nlstate);k2++){
6038: for (l2=1; l2<=(nlstate+ndeath);l2++){
6039: if(l2==k2) continue;
6040: j=(k2-1)*(nlstate+ndeath)+l2;
6041: for (k1=1; k1<=(nlstate);k1++){
6042: for (l1=1; l1<=(nlstate+ndeath);l1++){
6043: if(l1==k1) continue;
6044: i=(k1-1)*(nlstate+ndeath)+l1;
6045: if(i<=j) continue;
6046: for (age=bage; age<=fage; age ++){
6047: if ((int)age %5==0){
6048: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6049: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6050: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6051: mu1=mu[i][(int) age]/stepm*YEARM ;
6052: mu2=mu[j][(int) age]/stepm*YEARM;
6053: c12=cv12/sqrt(v1*v2);
6054: /* Computing eigen value of matrix of covariance */
6055: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6056: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6057: if ((lc2 <0) || (lc1 <0) ){
6058: if(first2==1){
6059: first1=0;
6060: 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);
6061: }
6062: 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);
6063: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6064: /* lc2=fabs(lc2); */
6065: }
1.220 brouard 6066:
1.222 brouard 6067: /* Eigen vectors */
6068: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6069: /*v21=sqrt(1.-v11*v11); *//* error */
6070: v21=(lc1-v1)/cv12*v11;
6071: v12=-v21;
6072: v22=v11;
6073: tnalp=v21/v11;
6074: if(first1==1){
6075: first1=0;
6076: 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);
6077: }
6078: 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);
6079: /*printf(fignu*/
6080: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6081: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6082: if(first==1){
6083: first=0;
6084: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6085: fprintf(ficgp,"\nset parametric;unset label");
6086: 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);
6087: fprintf(ficgp,"\nset ter svg size 640, 480");
6088: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6089: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6090: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6091: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6092: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6093: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6094: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6095: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6096: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6097: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6098: 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", \
6099: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6100: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6101: }else{
6102: first=0;
6103: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6104: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6105: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6106: 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", \
6107: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6108: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6109: }/* if first */
6110: } /* age mod 5 */
6111: } /* end loop age */
6112: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6113: first=1;
6114: } /*l12 */
6115: } /* k12 */
6116: } /*l1 */
6117: }/* k1 */
6118: } /* loop on combination of covariates j1 */
6119: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6120: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6121: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6122: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6123: free_vector(xp,1,npar);
6124: fclose(ficresprob);
6125: fclose(ficresprobcov);
6126: fclose(ficresprobcor);
6127: fflush(ficgp);
6128: fflush(fichtmcov);
6129: }
1.126 brouard 6130:
6131:
6132: /******************* Printing html file ***********/
1.201 brouard 6133: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6134: int lastpass, int stepm, int weightopt, char model[],\
6135: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6136: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6137: double jprev1, double mprev1,double anprev1, double dateprev1, \
6138: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6139: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6140:
6141: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6142: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6143: </ul>");
1.237 brouard 6144: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6145: </ul>", model);
1.214 brouard 6146: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6147: 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",
6148: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6149: 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 6150: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6151: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6152: fprintf(fichtm,"\
6153: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6154: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6155: fprintf(fichtm,"\
1.217 brouard 6156: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6157: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6158: fprintf(fichtm,"\
1.126 brouard 6159: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6160: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6161: fprintf(fichtm,"\
1.217 brouard 6162: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6163: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6164: fprintf(fichtm,"\
1.211 brouard 6165: - (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 6166: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6167: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6168: if(prevfcast==1){
6169: fprintf(fichtm,"\
6170: - Prevalence projections by age and states: \
1.201 brouard 6171: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6172: }
1.126 brouard 6173:
1.222 brouard 6174: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6175:
1.225 brouard 6176: m=pow(2,cptcoveff);
1.222 brouard 6177: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6178:
1.222 brouard 6179: jj1=0;
1.237 brouard 6180:
6181: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6182: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6183: if(TKresult[nres]!= k1)
6184: continue;
1.220 brouard 6185:
1.222 brouard 6186: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6187: jj1++;
6188: if (cptcovn > 0) {
6189: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6190: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6191: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6192: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6193: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6194: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6195: }
1.237 brouard 6196: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6197: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6198: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6199: }
6200:
1.230 brouard 6201: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6202: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6203: if(invalidvarcomb[k1]){
6204: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6205: printf("\nCombination (%d) ignored because no cases \n",k1);
6206: continue;
6207: }
6208: }
6209: /* aij, bij */
1.241 brouard 6210: 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> \
6211: <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 6212: /* Pij */
1.241 brouard 6213: 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> \
6214: <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 6215: /* Quasi-incidences */
6216: 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 6217: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6218: 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 6219: 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> \
6220: <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 6221: /* Survival functions (period) in state j */
6222: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6223: 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> \
6224: <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 6225: }
6226: /* State specific survival functions (period) */
6227: for(cpt=1; cpt<=nlstate;cpt++){
6228: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6229: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6230: <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 6231: }
6232: /* Period (stable) prevalence in each health state */
6233: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6234: 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> \
6235: <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 6236: }
6237: if(backcast==1){
6238: /* Period (stable) back prevalence in each health state */
6239: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6240: 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> \
6241: <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 6242: }
1.217 brouard 6243: }
1.222 brouard 6244: if(prevfcast==1){
6245: /* Projection of prevalence up to period (stable) prevalence in each health state */
6246: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6247: 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> \
6248: <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 6249: }
6250: }
1.220 brouard 6251:
1.222 brouard 6252: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6253: 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> \
6254: <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 6255: }
6256: /* } /\* end i1 *\/ */
6257: }/* End k1 */
6258: fprintf(fichtm,"</ul>");
1.126 brouard 6259:
1.222 brouard 6260: fprintf(fichtm,"\
1.126 brouard 6261: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6262: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6263: - 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 6264: But because parameters are usually highly correlated (a higher incidence of disability \
6265: and a higher incidence of recovery can give very close observed transition) it might \
6266: be very useful to look not only at linear confidence intervals estimated from the \
6267: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6268: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6269: covariance matrix of the one-step probabilities. \
6270: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6271:
1.222 brouard 6272: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6273: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6274: fprintf(fichtm,"\
1.126 brouard 6275: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6276: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6277:
1.222 brouard 6278: fprintf(fichtm,"\
1.126 brouard 6279: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6280: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6281: fprintf(fichtm,"\
1.126 brouard 6282: - 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): \
6283: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6284: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6285: fprintf(fichtm,"\
1.126 brouard 6286: - (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): \
6287: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6288: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6289: fprintf(fichtm,"\
1.128 brouard 6290: - 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 6291: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6292: fprintf(fichtm,"\
1.128 brouard 6293: - 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 6294: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6295: fprintf(fichtm,"\
1.126 brouard 6296: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6297: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6298:
6299: /* if(popforecast==1) fprintf(fichtm,"\n */
6300: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6301: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6302: /* <br>",fileres,fileres,fileres,fileres); */
6303: /* else */
6304: /* 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 6305: fflush(fichtm);
6306: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6307:
1.225 brouard 6308: m=pow(2,cptcoveff);
1.222 brouard 6309: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6310:
1.222 brouard 6311: jj1=0;
1.237 brouard 6312:
1.241 brouard 6313: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6314: for(k1=1; k1<=m;k1++){
1.237 brouard 6315: if(TKresult[nres]!= k1)
6316: continue;
1.222 brouard 6317: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6318: jj1++;
1.126 brouard 6319: if (cptcovn > 0) {
6320: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6321: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6322: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6323: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6324: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6325: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6326: }
6327:
1.126 brouard 6328: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6329:
1.222 brouard 6330: if(invalidvarcomb[k1]){
6331: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6332: continue;
6333: }
1.126 brouard 6334: }
6335: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6336: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6337: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6338: <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 6339: }
6340: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6341: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6342: true period expectancies (those weighted with period prevalences are also\
6343: drawn in addition to the population based expectancies computed using\
1.241 brouard 6344: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6345: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6346: /* } /\* end i1 *\/ */
6347: }/* End k1 */
1.241 brouard 6348: }/* End nres */
1.222 brouard 6349: fprintf(fichtm,"</ul>");
6350: fflush(fichtm);
1.126 brouard 6351: }
6352:
6353: /******************* Gnuplot file **************/
1.223 brouard 6354: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6355:
6356: char dirfileres[132],optfileres[132];
1.223 brouard 6357: char gplotcondition[132];
1.237 brouard 6358: 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 6359: int lv=0, vlv=0, kl=0;
1.130 brouard 6360: int ng=0;
1.201 brouard 6361: int vpopbased;
1.223 brouard 6362: int ioffset; /* variable offset for columns */
1.235 brouard 6363: int nres=0; /* Index of resultline */
1.219 brouard 6364:
1.126 brouard 6365: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6366: /* printf("Problem with file %s",optionfilegnuplot); */
6367: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6368: /* } */
6369:
6370: /*#ifdef windows */
6371: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6372: /*#endif */
1.225 brouard 6373: m=pow(2,cptcoveff);
1.126 brouard 6374:
1.202 brouard 6375: /* Contribution to likelihood */
6376: /* Plot the probability implied in the likelihood */
1.223 brouard 6377: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6378: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6379: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6380: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6381: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6382: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6383: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6384: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6385: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6386: 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));
6387: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6388: 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));
6389: for (i=1; i<= nlstate ; i ++) {
6390: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6391: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6392: 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);
6393: for (j=2; j<= nlstate+ndeath ; j ++) {
6394: 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);
6395: }
6396: fprintf(ficgp,";\nset out; unset ylabel;\n");
6397: }
6398: /* 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 */
6399: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6400: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6401: fprintf(ficgp,"\nset out;unset log\n");
6402: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6403:
1.126 brouard 6404: strcpy(dirfileres,optionfilefiname);
6405: strcpy(optfileres,"vpl");
1.223 brouard 6406: /* 1eme*/
1.238 brouard 6407: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6408: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6409: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6410: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6411: if(TKresult[nres]!= k1)
6412: continue;
6413: /* We are interested in selected combination by the resultline */
1.246 brouard 6414: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6415: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6416: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6417: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6418: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6419: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6420: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6421: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6422: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6423: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6424: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6425: }
6426: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6427: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6428: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6429: }
1.246 brouard 6430: /* printf("\n#\n"); */
1.238 brouard 6431: fprintf(ficgp,"\n#\n");
6432: if(invalidvarcomb[k1]){
6433: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6434: continue;
6435: }
1.235 brouard 6436:
1.241 brouard 6437: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6438: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6439: 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 6440:
1.238 brouard 6441: for (i=1; i<= nlstate ; i ++) {
6442: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6443: else fprintf(ficgp," %%*lf (%%*lf)");
6444: }
1.242 brouard 6445: 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 6446: for (i=1; i<= nlstate ; i ++) {
6447: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6448: else fprintf(ficgp," %%*lf (%%*lf)");
6449: }
1.242 brouard 6450: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6451: for (i=1; i<= nlstate ; i ++) {
6452: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6453: else fprintf(ficgp," %%*lf (%%*lf)");
6454: }
6455: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6456: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6457: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6458: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6459: if(cptcoveff ==0){
1.245 brouard 6460: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6461: }else{
6462: kl=0;
6463: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6464: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6465: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6466: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6467: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6468: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6469: kl++;
1.238 brouard 6470: /* 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 *\/ */
6471: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6472: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6473: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6474: if(k==cptcoveff){
1.245 brouard 6475: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6476: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6477: }else{
6478: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6479: kl++;
6480: }
6481: } /* end covariate */
6482: } /* end if no covariate */
6483: } /* end if backcast */
6484: fprintf(ficgp,"\nset out \n");
6485: } /* nres */
1.201 brouard 6486: } /* k1 */
6487: } /* cpt */
1.235 brouard 6488:
6489:
1.126 brouard 6490: /*2 eme*/
1.238 brouard 6491: for (k1=1; k1<= m ; k1 ++){
6492: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6493: if(TKresult[nres]!= k1)
6494: continue;
6495: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6496: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6497: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6498: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6499: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6500: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6501: vlv= nbcode[Tvaraff[k]][lv];
6502: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6503: }
1.237 brouard 6504: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6505: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6506: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6507: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6508: }
1.211 brouard 6509: fprintf(ficgp,"\n#\n");
1.223 brouard 6510: if(invalidvarcomb[k1]){
6511: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6512: continue;
6513: }
1.219 brouard 6514:
1.241 brouard 6515: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6516: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6517: if(vpopbased==0)
6518: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6519: else
6520: fprintf(ficgp,"\nreplot ");
6521: for (i=1; i<= nlstate+1 ; i ++) {
6522: k=2*i;
6523: 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);
6524: for (j=1; j<= nlstate+1 ; j ++) {
6525: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6526: else fprintf(ficgp," %%*lf (%%*lf)");
6527: }
6528: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6529: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6530: 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);
6531: for (j=1; j<= nlstate+1 ; j ++) {
6532: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6533: else fprintf(ficgp," %%*lf (%%*lf)");
6534: }
6535: fprintf(ficgp,"\" t\"\" w l lt 0,");
6536: 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);
6537: for (j=1; j<= nlstate+1 ; j ++) {
6538: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6539: else fprintf(ficgp," %%*lf (%%*lf)");
6540: }
6541: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6542: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6543: } /* state */
6544: } /* vpopbased */
1.244 brouard 6545: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6546: } /* end nres */
6547: } /* k1 end 2 eme*/
6548:
6549:
6550: /*3eme*/
6551: for (k1=1; k1<= m ; k1 ++){
6552: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6553: if(TKresult[nres]!= k1)
1.238 brouard 6554: continue;
6555:
6556: for (cpt=1; cpt<= nlstate ; cpt ++) {
6557: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6558: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6559: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6560: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6561: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6562: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6563: vlv= nbcode[Tvaraff[k]][lv];
6564: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6565: }
6566: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6567: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6568: }
6569: fprintf(ficgp,"\n#\n");
6570: if(invalidvarcomb[k1]){
6571: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6572: continue;
6573: }
6574:
6575: /* k=2+nlstate*(2*cpt-2); */
6576: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6577: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6578: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6579: 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 6580: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6581: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6582: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6583: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6584: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6585: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6586:
1.238 brouard 6587: */
6588: for (i=1; i< nlstate ; i ++) {
6589: 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);
6590: /* 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 6591:
1.238 brouard 6592: }
6593: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6594: }
6595: } /* end nres */
6596: } /* end kl 3eme */
1.126 brouard 6597:
1.223 brouard 6598: /* 4eme */
1.201 brouard 6599: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6600: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6601: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6602: if(TKresult[nres]!= k1)
1.223 brouard 6603: continue;
1.238 brouard 6604: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6605: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6606: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6607: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6608: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6609: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6610: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6611: vlv= nbcode[Tvaraff[k]][lv];
6612: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6613: }
6614: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6615: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6616: }
6617: fprintf(ficgp,"\n#\n");
6618: if(invalidvarcomb[k1]){
6619: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6620: continue;
1.223 brouard 6621: }
1.238 brouard 6622:
1.241 brouard 6623: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6624: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6625: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6626: k=3;
6627: for (i=1; i<= nlstate ; i ++){
6628: if(i==1){
6629: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6630: }else{
6631: fprintf(ficgp,", '' ");
6632: }
6633: l=(nlstate+ndeath)*(i-1)+1;
6634: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6635: for (j=2; j<= nlstate+ndeath ; j ++)
6636: fprintf(ficgp,"+$%d",k+l+j-1);
6637: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6638: } /* nlstate */
6639: fprintf(ficgp,"\nset out\n");
6640: } /* end cpt state*/
6641: } /* end nres */
6642: } /* end covariate k1 */
6643:
1.220 brouard 6644: /* 5eme */
1.201 brouard 6645: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6646: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6647: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6648: if(TKresult[nres]!= k1)
1.227 brouard 6649: continue;
1.238 brouard 6650: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6651: 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);
6652: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6653: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6654: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6655: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6656: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6657: vlv= nbcode[Tvaraff[k]][lv];
6658: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6659: }
6660: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6661: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6662: }
6663: fprintf(ficgp,"\n#\n");
6664: if(invalidvarcomb[k1]){
6665: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6666: continue;
6667: }
1.227 brouard 6668:
1.241 brouard 6669: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6670: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6671: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6672: k=3;
6673: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6674: if(j==1)
6675: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6676: else
6677: fprintf(ficgp,", '' ");
6678: l=(nlstate+ndeath)*(cpt-1) +j;
6679: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6680: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6681: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6682: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6683: } /* nlstate */
6684: fprintf(ficgp,", '' ");
6685: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6686: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6687: l=(nlstate+ndeath)*(cpt-1) +j;
6688: if(j < nlstate)
6689: fprintf(ficgp,"$%d +",k+l);
6690: else
6691: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6692: }
6693: fprintf(ficgp,"\nset out\n");
6694: } /* end cpt state*/
6695: } /* end covariate */
6696: } /* end nres */
1.227 brouard 6697:
1.220 brouard 6698: /* 6eme */
1.202 brouard 6699: /* CV preval stable (period) for each covariate */
1.237 brouard 6700: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6701: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6702: if(TKresult[nres]!= k1)
6703: continue;
1.153 brouard 6704: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6705:
1.211 brouard 6706: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6707: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6708: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6709: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6710: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6711: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6712: vlv= nbcode[Tvaraff[k]][lv];
6713: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6714: }
1.237 brouard 6715: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6716: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6717: }
1.211 brouard 6718: fprintf(ficgp,"\n#\n");
1.223 brouard 6719: if(invalidvarcomb[k1]){
1.227 brouard 6720: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6721: continue;
1.223 brouard 6722: }
1.227 brouard 6723:
1.241 brouard 6724: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6725: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6726: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6727: k=3; /* Offset */
1.153 brouard 6728: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6729: if(i==1)
6730: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6731: else
6732: fprintf(ficgp,", '' ");
6733: l=(nlstate+ndeath)*(i-1)+1;
6734: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6735: for (j=2; j<= nlstate ; j ++)
6736: fprintf(ficgp,"+$%d",k+l+j-1);
6737: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6738: } /* nlstate */
1.201 brouard 6739: fprintf(ficgp,"\nset out\n");
1.153 brouard 6740: } /* end cpt state*/
6741: } /* end covariate */
1.227 brouard 6742:
6743:
1.220 brouard 6744: /* 7eme */
1.218 brouard 6745: if(backcast == 1){
1.217 brouard 6746: /* CV back preval stable (period) for each covariate */
1.237 brouard 6747: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6748: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6749: if(TKresult[nres]!= k1)
6750: continue;
1.218 brouard 6751: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6752: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6753: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6754: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6755: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6756: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6757: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6758: vlv= nbcode[Tvaraff[k]][lv];
6759: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6760: }
1.237 brouard 6761: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6762: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6763: }
1.227 brouard 6764: fprintf(ficgp,"\n#\n");
6765: if(invalidvarcomb[k1]){
6766: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6767: continue;
6768: }
6769:
1.241 brouard 6770: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6771: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6772: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6773: k=3; /* Offset */
6774: for (i=1; i<= nlstate ; i ++){
6775: if(i==1)
6776: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6777: else
6778: fprintf(ficgp,", '' ");
6779: /* l=(nlstate+ndeath)*(i-1)+1; */
6780: l=(nlstate+ndeath)*(cpt-1)+1;
6781: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6782: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6783: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6784: /* for (j=2; j<= nlstate ; j ++) */
6785: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6786: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6787: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6788: } /* nlstate */
6789: fprintf(ficgp,"\nset out\n");
1.218 brouard 6790: } /* end cpt state*/
6791: } /* end covariate */
6792: } /* End if backcast */
6793:
1.223 brouard 6794: /* 8eme */
1.218 brouard 6795: if(prevfcast==1){
6796: /* Projection from cross-sectional to stable (period) for each covariate */
6797:
1.237 brouard 6798: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6799: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6800: if(TKresult[nres]!= k1)
6801: continue;
1.211 brouard 6802: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6803: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6804: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6805: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6806: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6807: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6808: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6809: vlv= nbcode[Tvaraff[k]][lv];
6810: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6811: }
1.237 brouard 6812: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6813: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6814: }
1.227 brouard 6815: fprintf(ficgp,"\n#\n");
6816: if(invalidvarcomb[k1]){
6817: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6818: continue;
6819: }
6820:
6821: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6822: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6823: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6824: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6825: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6826: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6827: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6828: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6829: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6830: if(i==1){
6831: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6832: }else{
6833: fprintf(ficgp,",\\\n '' ");
6834: }
6835: if(cptcoveff ==0){ /* No covariate */
6836: ioffset=2; /* Age is in 2 */
6837: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6838: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6839: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6840: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6841: fprintf(ficgp," u %d:(", ioffset);
6842: if(i==nlstate+1)
6843: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6844: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6845: else
6846: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6847: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6848: }else{ /* more than 2 covariates */
6849: if(cptcoveff ==1){
6850: ioffset=4; /* Age is in 4 */
6851: }else{
6852: ioffset=6; /* Age is in 6 */
6853: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6854: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6855: }
6856: fprintf(ficgp," u %d:(",ioffset);
6857: kl=0;
6858: strcpy(gplotcondition,"(");
6859: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6860: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6861: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6862: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6863: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6864: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6865: kl++;
6866: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6867: kl++;
6868: if(k <cptcoveff && cptcoveff>1)
6869: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6870: }
6871: strcpy(gplotcondition+strlen(gplotcondition),")");
6872: /* 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 *\/ */
6873: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6874: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6875: /* '' 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*/
6876: if(i==nlstate+1){
6877: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6878: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6879: }else{
6880: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6881: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6882: }
6883: } /* end if covariate */
6884: } /* nlstate */
6885: fprintf(ficgp,"\nset out\n");
1.223 brouard 6886: } /* end cpt state*/
6887: } /* end covariate */
6888: } /* End if prevfcast */
1.227 brouard 6889:
6890:
1.238 brouard 6891: /* 9eme writing MLE parameters */
6892: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6893: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6894: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6895: for(k=1; k <=(nlstate+ndeath); k++){
6896: if (k != i) {
1.227 brouard 6897: fprintf(ficgp,"# current state %d\n",k);
6898: for(j=1; j <=ncovmodel; j++){
6899: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6900: jk++;
6901: }
6902: fprintf(ficgp,"\n");
1.126 brouard 6903: }
6904: }
1.223 brouard 6905: }
1.187 brouard 6906: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6907:
1.145 brouard 6908: /*goto avoid;*/
1.238 brouard 6909: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6910: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6911: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6912: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6913: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6914: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6915: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6916: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6917: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6918: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6919: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6920: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6921: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6922: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6923: fprintf(ficgp,"#\n");
1.223 brouard 6924: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6925: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6926: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6927: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6928: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6929: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6930: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6931: if(TKresult[nres]!= jk)
6932: continue;
6933: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6934: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6935: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6936: }
6937: fprintf(ficgp,"\n#\n");
1.241 brouard 6938: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 6939: fprintf(ficgp,"\nset ter svg size 640, 480 ");
6940: if (ng==1){
6941: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
6942: fprintf(ficgp,"\nunset log y");
6943: }else if (ng==2){
6944: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
6945: fprintf(ficgp,"\nset log y");
6946: }else if (ng==3){
6947: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
6948: fprintf(ficgp,"\nset log y");
6949: }else
6950: fprintf(ficgp,"\nunset title ");
6951: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
6952: i=1;
6953: for(k2=1; k2<=nlstate; k2++) {
6954: k3=i;
6955: for(k=1; k<=(nlstate+ndeath); k++) {
6956: if (k != k2){
6957: switch( ng) {
6958: case 1:
6959: if(nagesqr==0)
6960: fprintf(ficgp," p%d+p%d*x",i,i+1);
6961: else /* nagesqr =1 */
6962: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6963: break;
6964: case 2: /* ng=2 */
6965: if(nagesqr==0)
6966: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
6967: else /* nagesqr =1 */
6968: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
6969: break;
6970: case 3:
6971: if(nagesqr==0)
6972: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
6973: else /* nagesqr =1 */
6974: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
6975: break;
6976: }
6977: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 6978: ijp=1; /* product no age */
6979: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
6980: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 6981: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 6982: if(j==Tage[ij]) { /* Product by age */
6983: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 6984: if(DummyV[j]==0){
1.237 brouard 6985: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
6986: }else{ /* quantitative */
6987: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
6988: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
6989: }
6990: ij++;
6991: }
6992: }else if(j==Tprod[ijp]) { /* */
6993: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
6994: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 6995: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
6996: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 6997: /* 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)]); */
6998: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
6999: }else{ /* Vn is dummy and Vm is quanti */
7000: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7001: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7002: }
7003: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7004: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7005: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7006: }else{ /* Both quanti */
7007: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7008: }
7009: }
1.238 brouard 7010: ijp++;
1.237 brouard 7011: }
7012: } else{ /* simple covariate */
7013: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7014: if(Dummy[j]==0){
7015: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7016: }else{ /* quantitative */
7017: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7018: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7019: }
1.237 brouard 7020: } /* end simple */
7021: } /* end j */
1.223 brouard 7022: }else{
7023: i=i-ncovmodel;
7024: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7025: fprintf(ficgp," (1.");
7026: }
1.227 brouard 7027:
1.223 brouard 7028: if(ng != 1){
7029: fprintf(ficgp,")/(1");
1.227 brouard 7030:
1.223 brouard 7031: for(k1=1; k1 <=nlstate; k1++){
7032: if(nagesqr==0)
7033: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7034: else /* nagesqr =1 */
7035: 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 7036:
1.223 brouard 7037: ij=1;
7038: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7039: if((j-2)==Tage[ij]) { /* Bug valgrind */
7040: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7041: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7042: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7043: ij++;
7044: }
7045: }
7046: else
1.225 brouard 7047: 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 7048: }
7049: fprintf(ficgp,")");
7050: }
7051: fprintf(ficgp,")");
7052: if(ng ==2)
7053: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7054: else /* ng= 3 */
7055: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7056: }else{ /* end ng <> 1 */
7057: if( k !=k2) /* logit p11 is hard to draw */
7058: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7059: }
7060: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7061: fprintf(ficgp,",");
7062: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7063: fprintf(ficgp,",");
7064: i=i+ncovmodel;
7065: } /* end k */
7066: } /* end k2 */
7067: fprintf(ficgp,"\n set out\n");
7068: } /* end jk */
7069: } /* end ng */
7070: /* avoid: */
7071: fflush(ficgp);
1.126 brouard 7072: } /* end gnuplot */
7073:
7074:
7075: /*************** Moving average **************/
1.219 brouard 7076: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7077: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7078:
1.222 brouard 7079: int i, cpt, cptcod;
7080: int modcovmax =1;
7081: int mobilavrange, mob;
7082: int iage=0;
7083:
7084: double sum=0.;
7085: double age;
7086: double *sumnewp, *sumnewm;
7087: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7088:
7089:
1.225 brouard 7090: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7091: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7092:
7093: sumnewp = vector(1,ncovcombmax);
7094: sumnewm = vector(1,ncovcombmax);
7095: agemingood = vector(1,ncovcombmax);
7096: agemaxgood = vector(1,ncovcombmax);
7097:
7098: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7099: sumnewm[cptcod]=0.;
7100: sumnewp[cptcod]=0.;
7101: agemingood[cptcod]=0;
7102: agemaxgood[cptcod]=0;
7103: }
7104: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7105:
7106: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7107: if(mobilav==1) mobilavrange=5; /* default */
7108: else mobilavrange=mobilav;
7109: for (age=bage; age<=fage; age++)
7110: for (i=1; i<=nlstate;i++)
7111: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7112: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7113: /* We keep the original values on the extreme ages bage, fage and for
7114: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7115: we use a 5 terms etc. until the borders are no more concerned.
7116: */
7117: for (mob=3;mob <=mobilavrange;mob=mob+2){
7118: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7119: for (i=1; i<=nlstate;i++){
7120: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7121: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7122: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7123: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7124: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7125: }
7126: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7127: }
7128: }
7129: }/* end age */
7130: }/* end mob */
7131: }else
7132: return -1;
7133: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7134: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7135: if(invalidvarcomb[cptcod]){
7136: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7137: continue;
7138: }
1.219 brouard 7139:
1.222 brouard 7140: agemingood[cptcod]=fage-(mob-1)/2;
7141: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7142: sumnewm[cptcod]=0.;
7143: for (i=1; i<=nlstate;i++){
7144: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7145: }
7146: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7147: agemingood[cptcod]=age;
7148: }else{ /* bad */
7149: for (i=1; i<=nlstate;i++){
7150: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7151: } /* i */
7152: } /* end bad */
7153: }/* age */
7154: sum=0.;
7155: for (i=1; i<=nlstate;i++){
7156: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7157: }
7158: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7159: 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);
7160: /* for (i=1; i<=nlstate;i++){ */
7161: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7162: /* } /\* i *\/ */
7163: } /* end bad */
7164: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7165: /* From youngest, finding the oldest wrong */
7166: agemaxgood[cptcod]=bage+(mob-1)/2;
7167: for (age=bage+(mob-1)/2; age<=fage; age++){
7168: sumnewm[cptcod]=0.;
7169: for (i=1; i<=nlstate;i++){
7170: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7171: }
7172: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7173: agemaxgood[cptcod]=age;
7174: }else{ /* bad */
7175: for (i=1; i<=nlstate;i++){
7176: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7177: } /* i */
7178: } /* end bad */
7179: }/* age */
7180: sum=0.;
7181: for (i=1; i<=nlstate;i++){
7182: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7183: }
7184: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7185: 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);
7186: /* for (i=1; i<=nlstate;i++){ */
7187: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7188: /* } /\* i *\/ */
7189: } /* end bad */
7190:
7191: for (age=bage; age<=fage; age++){
1.235 brouard 7192: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7193: sumnewp[cptcod]=0.;
7194: sumnewm[cptcod]=0.;
7195: for (i=1; i<=nlstate;i++){
7196: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7197: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7198: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7199: }
7200: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7201: }
7202: /* printf("\n"); */
7203: /* } */
7204: /* brutal averaging */
7205: for (i=1; i<=nlstate;i++){
7206: for (age=1; age<=bage; age++){
7207: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7208: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7209: }
7210: for (age=fage; age<=AGESUP; age++){
7211: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7212: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7213: }
7214: } /* end i status */
7215: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7216: for (age=1; age<=AGESUP; age++){
7217: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7218: mobaverage[(int)age][i][cptcod]=0.;
7219: }
7220: }
7221: }/* end cptcod */
7222: free_vector(sumnewm,1, ncovcombmax);
7223: free_vector(sumnewp,1, ncovcombmax);
7224: free_vector(agemaxgood,1, ncovcombmax);
7225: free_vector(agemingood,1, ncovcombmax);
7226: return 0;
7227: }/* End movingaverage */
1.218 brouard 7228:
1.126 brouard 7229:
7230: /************** Forecasting ******************/
1.235 brouard 7231: 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 7232: /* proj1, year, month, day of starting projection
7233: agemin, agemax range of age
7234: dateprev1 dateprev2 range of dates during which prevalence is computed
7235: anproj2 year of en of projection (same day and month as proj1).
7236: */
1.235 brouard 7237: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7238: double agec; /* generic age */
7239: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7240: double *popeffectif,*popcount;
7241: double ***p3mat;
1.218 brouard 7242: /* double ***mobaverage; */
1.126 brouard 7243: char fileresf[FILENAMELENGTH];
7244:
7245: agelim=AGESUP;
1.211 brouard 7246: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7247: in each health status at the date of interview (if between dateprev1 and dateprev2).
7248: We still use firstpass and lastpass as another selection.
7249: */
1.214 brouard 7250: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7251: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7252:
1.201 brouard 7253: strcpy(fileresf,"F_");
7254: strcat(fileresf,fileresu);
1.126 brouard 7255: if((ficresf=fopen(fileresf,"w"))==NULL) {
7256: printf("Problem with forecast resultfile: %s\n", fileresf);
7257: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7258: }
1.235 brouard 7259: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7260: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7261:
1.225 brouard 7262: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7263:
7264:
7265: stepsize=(int) (stepm+YEARM-1)/YEARM;
7266: if (stepm<=12) stepsize=1;
7267: if(estepm < stepm){
7268: printf ("Problem %d lower than %d\n",estepm, stepm);
7269: }
7270: else hstepm=estepm;
7271:
7272: hstepm=hstepm/stepm;
7273: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7274: fractional in yp1 */
7275: anprojmean=yp;
7276: yp2=modf((yp1*12),&yp);
7277: mprojmean=yp;
7278: yp1=modf((yp2*30.5),&yp);
7279: jprojmean=yp;
7280: if(jprojmean==0) jprojmean=1;
7281: if(mprojmean==0) jprojmean=1;
7282:
1.227 brouard 7283: i1=pow(2,cptcoveff);
1.126 brouard 7284: if (cptcovn < 1){i1=1;}
7285:
7286: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7287:
7288: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7289:
1.126 brouard 7290: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7291: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7292: for(k=1; k<=i1;k++){
7293: if(TKresult[nres]!= k)
7294: continue;
1.227 brouard 7295: if(invalidvarcomb[k]){
7296: printf("\nCombination (%d) projection ignored because no cases \n",k);
7297: continue;
7298: }
7299: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7300: for(j=1;j<=cptcoveff;j++) {
7301: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7302: }
1.235 brouard 7303: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7304: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7305: }
1.227 brouard 7306: fprintf(ficresf," yearproj age");
7307: for(j=1; j<=nlstate+ndeath;j++){
7308: for(i=1; i<=nlstate;i++)
7309: fprintf(ficresf," p%d%d",i,j);
7310: fprintf(ficresf," wp.%d",j);
7311: }
7312: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7313: fprintf(ficresf,"\n");
7314: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7315: for (agec=fage; agec>=(ageminpar-1); agec--){
7316: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7317: nhstepm = nhstepm/hstepm;
7318: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7319: oldm=oldms;savm=savms;
1.235 brouard 7320: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7321:
7322: for (h=0; h<=nhstepm; h++){
7323: if (h*hstepm/YEARM*stepm ==yearp) {
7324: fprintf(ficresf,"\n");
7325: for(j=1;j<=cptcoveff;j++)
7326: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7327: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7328: }
7329: for(j=1; j<=nlstate+ndeath;j++) {
7330: ppij=0.;
7331: for(i=1; i<=nlstate;i++) {
7332: if (mobilav==1)
7333: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7334: else {
7335: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7336: }
7337: if (h*hstepm/YEARM*stepm== yearp) {
7338: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7339: }
7340: } /* end i */
7341: if (h*hstepm/YEARM*stepm==yearp) {
7342: fprintf(ficresf," %.3f", ppij);
7343: }
7344: }/* end j */
7345: } /* end h */
7346: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7347: } /* end agec */
7348: } /* end yearp */
7349: } /* end k */
1.219 brouard 7350:
1.126 brouard 7351: fclose(ficresf);
1.215 brouard 7352: printf("End of Computing forecasting \n");
7353: fprintf(ficlog,"End of Computing forecasting\n");
7354:
1.126 brouard 7355: }
7356:
1.218 brouard 7357: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7358: /* 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 7359: /* /\* back1, year, month, day of starting backection */
7360: /* agemin, agemax range of age */
7361: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7362: /* anback2 year of en of backection (same day and month as back1). */
7363: /* *\/ */
7364: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7365: /* double agec; /\* generic age *\/ */
7366: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7367: /* double *popeffectif,*popcount; */
7368: /* double ***p3mat; */
7369: /* /\* double ***mobaverage; *\/ */
7370: /* char fileresfb[FILENAMELENGTH]; */
7371:
7372: /* agelim=AGESUP; */
7373: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7374: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7375: /* We still use firstpass and lastpass as another selection. */
7376: /* *\/ */
7377: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7378: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7379: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7380:
7381: /* strcpy(fileresfb,"FB_"); */
7382: /* strcat(fileresfb,fileresu); */
7383: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7384: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7385: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7386: /* } */
7387: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7388: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7389:
1.225 brouard 7390: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7391:
7392: /* /\* if (mobilav!=0) { *\/ */
7393: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7394: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7395: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7396: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7397: /* /\* } *\/ */
7398: /* /\* } *\/ */
7399:
7400: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7401: /* if (stepm<=12) stepsize=1; */
7402: /* if(estepm < stepm){ */
7403: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7404: /* } */
7405: /* else hstepm=estepm; */
7406:
7407: /* hstepm=hstepm/stepm; */
7408: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7409: /* fractional in yp1 *\/ */
7410: /* anprojmean=yp; */
7411: /* yp2=modf((yp1*12),&yp); */
7412: /* mprojmean=yp; */
7413: /* yp1=modf((yp2*30.5),&yp); */
7414: /* jprojmean=yp; */
7415: /* if(jprojmean==0) jprojmean=1; */
7416: /* if(mprojmean==0) jprojmean=1; */
7417:
1.225 brouard 7418: /* i1=cptcoveff; */
1.218 brouard 7419: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7420:
1.218 brouard 7421: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7422:
1.218 brouard 7423: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7424:
7425: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7426: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7427: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7428: /* k=k+1; */
7429: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7430: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7431: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7432: /* } */
7433: /* fprintf(ficresfb," yearbproj age"); */
7434: /* for(j=1; j<=nlstate+ndeath;j++){ */
7435: /* for(i=1; i<=nlstate;i++) */
7436: /* fprintf(ficresfb," p%d%d",i,j); */
7437: /* fprintf(ficresfb," p.%d",j); */
7438: /* } */
7439: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7440: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7441: /* fprintf(ficresfb,"\n"); */
7442: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7443: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7444: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7445: /* nhstepm = nhstepm/hstepm; */
7446: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7447: /* oldm=oldms;savm=savms; */
7448: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7449: /* for (h=0; h<=nhstepm; h++){ */
7450: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7451: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7452: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7453: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7454: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7455: /* } */
7456: /* for(j=1; j<=nlstate+ndeath;j++) { */
7457: /* ppij=0.; */
7458: /* for(i=1; i<=nlstate;i++) { */
7459: /* if (mobilav==1) */
7460: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7461: /* else { */
7462: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7463: /* } */
7464: /* if (h*hstepm/YEARM*stepm== yearp) { */
7465: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7466: /* } */
7467: /* } /\* end i *\/ */
7468: /* if (h*hstepm/YEARM*stepm==yearp) { */
7469: /* fprintf(ficresfb," %.3f", ppij); */
7470: /* } */
7471: /* }/\* end j *\/ */
7472: /* } /\* end h *\/ */
7473: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7474: /* } /\* end agec *\/ */
7475: /* } /\* end yearp *\/ */
7476: /* } /\* end cptcod *\/ */
7477: /* } /\* end cptcov *\/ */
7478:
7479: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7480:
7481: /* fclose(ficresfb); */
7482: /* printf("End of Computing Back forecasting \n"); */
7483: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7484:
1.218 brouard 7485: /* } */
1.217 brouard 7486:
1.126 brouard 7487: /************** Forecasting *****not tested NB*************/
1.227 brouard 7488: /* 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 7489:
1.227 brouard 7490: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7491: /* int *popage; */
7492: /* double calagedatem, agelim, kk1, kk2; */
7493: /* double *popeffectif,*popcount; */
7494: /* double ***p3mat,***tabpop,***tabpopprev; */
7495: /* /\* double ***mobaverage; *\/ */
7496: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7497:
1.227 brouard 7498: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7499: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7500: /* agelim=AGESUP; */
7501: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7502:
1.227 brouard 7503: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7504:
7505:
1.227 brouard 7506: /* strcpy(filerespop,"POP_"); */
7507: /* strcat(filerespop,fileresu); */
7508: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7509: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7510: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7511: /* } */
7512: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7513: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7514:
1.227 brouard 7515: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7516:
1.227 brouard 7517: /* /\* if (mobilav!=0) { *\/ */
7518: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7519: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7520: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7521: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7522: /* /\* } *\/ */
7523: /* /\* } *\/ */
1.126 brouard 7524:
1.227 brouard 7525: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7526: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7527:
1.227 brouard 7528: /* agelim=AGESUP; */
1.126 brouard 7529:
1.227 brouard 7530: /* hstepm=1; */
7531: /* hstepm=hstepm/stepm; */
1.218 brouard 7532:
1.227 brouard 7533: /* if (popforecast==1) { */
7534: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7535: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7536: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7537: /* } */
7538: /* popage=ivector(0,AGESUP); */
7539: /* popeffectif=vector(0,AGESUP); */
7540: /* popcount=vector(0,AGESUP); */
1.126 brouard 7541:
1.227 brouard 7542: /* i=1; */
7543: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7544:
1.227 brouard 7545: /* imx=i; */
7546: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7547: /* } */
1.218 brouard 7548:
1.227 brouard 7549: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7550: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7551: /* k=k+1; */
7552: /* fprintf(ficrespop,"\n#******"); */
7553: /* for(j=1;j<=cptcoveff;j++) { */
7554: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7555: /* } */
7556: /* fprintf(ficrespop,"******\n"); */
7557: /* fprintf(ficrespop,"# Age"); */
7558: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7559: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7560:
1.227 brouard 7561: /* for (cpt=0; cpt<=0;cpt++) { */
7562: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7563:
1.227 brouard 7564: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7565: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7566: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7567:
1.227 brouard 7568: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7569: /* oldm=oldms;savm=savms; */
7570: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7571:
1.227 brouard 7572: /* for (h=0; h<=nhstepm; h++){ */
7573: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7574: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7575: /* } */
7576: /* for(j=1; j<=nlstate+ndeath;j++) { */
7577: /* kk1=0.;kk2=0; */
7578: /* for(i=1; i<=nlstate;i++) { */
7579: /* if (mobilav==1) */
7580: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7581: /* else { */
7582: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7583: /* } */
7584: /* } */
7585: /* if (h==(int)(calagedatem+12*cpt)){ */
7586: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7587: /* /\*fprintf(ficrespop," %.3f", kk1); */
7588: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7589: /* } */
7590: /* } */
7591: /* for(i=1; i<=nlstate;i++){ */
7592: /* kk1=0.; */
7593: /* for(j=1; j<=nlstate;j++){ */
7594: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7595: /* } */
7596: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7597: /* } */
1.218 brouard 7598:
1.227 brouard 7599: /* if (h==(int)(calagedatem+12*cpt)) */
7600: /* for(j=1; j<=nlstate;j++) */
7601: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7602: /* } */
7603: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7604: /* } */
7605: /* } */
1.218 brouard 7606:
1.227 brouard 7607: /* /\******\/ */
1.218 brouard 7608:
1.227 brouard 7609: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7610: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7611: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7612: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7613: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7614:
1.227 brouard 7615: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7616: /* oldm=oldms;savm=savms; */
7617: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7618: /* for (h=0; h<=nhstepm; h++){ */
7619: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7620: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7621: /* } */
7622: /* for(j=1; j<=nlstate+ndeath;j++) { */
7623: /* kk1=0.;kk2=0; */
7624: /* for(i=1; i<=nlstate;i++) { */
7625: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7626: /* } */
7627: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7628: /* } */
7629: /* } */
7630: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7631: /* } */
7632: /* } */
7633: /* } */
7634: /* } */
1.218 brouard 7635:
1.227 brouard 7636: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7637:
1.227 brouard 7638: /* if (popforecast==1) { */
7639: /* free_ivector(popage,0,AGESUP); */
7640: /* free_vector(popeffectif,0,AGESUP); */
7641: /* free_vector(popcount,0,AGESUP); */
7642: /* } */
7643: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7644: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7645: /* fclose(ficrespop); */
7646: /* } /\* End of popforecast *\/ */
1.218 brouard 7647:
1.126 brouard 7648: int fileappend(FILE *fichier, char *optionfich)
7649: {
7650: if((fichier=fopen(optionfich,"a"))==NULL) {
7651: printf("Problem with file: %s\n", optionfich);
7652: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7653: return (0);
7654: }
7655: fflush(fichier);
7656: return (1);
7657: }
7658:
7659:
7660: /**************** function prwizard **********************/
7661: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7662: {
7663:
7664: /* Wizard to print covariance matrix template */
7665:
1.164 brouard 7666: char ca[32], cb[32];
7667: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7668: int numlinepar;
7669:
7670: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7671: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7672: for(i=1; i <=nlstate; i++){
7673: jj=0;
7674: for(j=1; j <=nlstate+ndeath; j++){
7675: if(j==i) continue;
7676: jj++;
7677: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7678: printf("%1d%1d",i,j);
7679: fprintf(ficparo,"%1d%1d",i,j);
7680: for(k=1; k<=ncovmodel;k++){
7681: /* printf(" %lf",param[i][j][k]); */
7682: /* fprintf(ficparo," %lf",param[i][j][k]); */
7683: printf(" 0.");
7684: fprintf(ficparo," 0.");
7685: }
7686: printf("\n");
7687: fprintf(ficparo,"\n");
7688: }
7689: }
7690: printf("# Scales (for hessian or gradient estimation)\n");
7691: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7692: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7693: for(i=1; i <=nlstate; i++){
7694: jj=0;
7695: for(j=1; j <=nlstate+ndeath; j++){
7696: if(j==i) continue;
7697: jj++;
7698: fprintf(ficparo,"%1d%1d",i,j);
7699: printf("%1d%1d",i,j);
7700: fflush(stdout);
7701: for(k=1; k<=ncovmodel;k++){
7702: /* printf(" %le",delti3[i][j][k]); */
7703: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7704: printf(" 0.");
7705: fprintf(ficparo," 0.");
7706: }
7707: numlinepar++;
7708: printf("\n");
7709: fprintf(ficparo,"\n");
7710: }
7711: }
7712: printf("# Covariance matrix\n");
7713: /* # 121 Var(a12)\n\ */
7714: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7715: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7716: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7717: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7718: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7719: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7720: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7721: fflush(stdout);
7722: fprintf(ficparo,"# Covariance matrix\n");
7723: /* # 121 Var(a12)\n\ */
7724: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7725: /* # ...\n\ */
7726: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7727:
7728: for(itimes=1;itimes<=2;itimes++){
7729: jj=0;
7730: for(i=1; i <=nlstate; i++){
7731: for(j=1; j <=nlstate+ndeath; j++){
7732: if(j==i) continue;
7733: for(k=1; k<=ncovmodel;k++){
7734: jj++;
7735: ca[0]= k+'a'-1;ca[1]='\0';
7736: if(itimes==1){
7737: printf("#%1d%1d%d",i,j,k);
7738: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7739: }else{
7740: printf("%1d%1d%d",i,j,k);
7741: fprintf(ficparo,"%1d%1d%d",i,j,k);
7742: /* printf(" %.5le",matcov[i][j]); */
7743: }
7744: ll=0;
7745: for(li=1;li <=nlstate; li++){
7746: for(lj=1;lj <=nlstate+ndeath; lj++){
7747: if(lj==li) continue;
7748: for(lk=1;lk<=ncovmodel;lk++){
7749: ll++;
7750: if(ll<=jj){
7751: cb[0]= lk +'a'-1;cb[1]='\0';
7752: if(ll<jj){
7753: if(itimes==1){
7754: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7755: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7756: }else{
7757: printf(" 0.");
7758: fprintf(ficparo," 0.");
7759: }
7760: }else{
7761: if(itimes==1){
7762: printf(" Var(%s%1d%1d)",ca,i,j);
7763: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7764: }else{
7765: printf(" 0.");
7766: fprintf(ficparo," 0.");
7767: }
7768: }
7769: }
7770: } /* end lk */
7771: } /* end lj */
7772: } /* end li */
7773: printf("\n");
7774: fprintf(ficparo,"\n");
7775: numlinepar++;
7776: } /* end k*/
7777: } /*end j */
7778: } /* end i */
7779: } /* end itimes */
7780:
7781: } /* end of prwizard */
7782: /******************* Gompertz Likelihood ******************************/
7783: double gompertz(double x[])
7784: {
7785: double A,B,L=0.0,sump=0.,num=0.;
7786: int i,n=0; /* n is the size of the sample */
7787:
1.220 brouard 7788: for (i=1;i<=imx ; i++) {
1.126 brouard 7789: sump=sump+weight[i];
7790: /* sump=sump+1;*/
7791: num=num+1;
7792: }
7793:
7794:
7795: /* for (i=0; i<=imx; i++)
7796: 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]);*/
7797:
7798: for (i=1;i<=imx ; i++)
7799: {
7800: if (cens[i] == 1 && wav[i]>1)
7801: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7802:
7803: if (cens[i] == 0 && wav[i]>1)
7804: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7805: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7806:
7807: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7808: if (wav[i] > 1 ) { /* ??? */
7809: L=L+A*weight[i];
7810: /* 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]);*/
7811: }
7812: }
7813:
7814: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7815:
7816: return -2*L*num/sump;
7817: }
7818:
1.136 brouard 7819: #ifdef GSL
7820: /******************* Gompertz_f Likelihood ******************************/
7821: double gompertz_f(const gsl_vector *v, void *params)
7822: {
7823: double A,B,LL=0.0,sump=0.,num=0.;
7824: double *x= (double *) v->data;
7825: int i,n=0; /* n is the size of the sample */
7826:
7827: for (i=0;i<=imx-1 ; i++) {
7828: sump=sump+weight[i];
7829: /* sump=sump+1;*/
7830: num=num+1;
7831: }
7832:
7833:
7834: /* for (i=0; i<=imx; i++)
7835: 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]);*/
7836: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7837: for (i=1;i<=imx ; i++)
7838: {
7839: if (cens[i] == 1 && wav[i]>1)
7840: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7841:
7842: if (cens[i] == 0 && wav[i]>1)
7843: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7844: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7845:
7846: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7847: if (wav[i] > 1 ) { /* ??? */
7848: LL=LL+A*weight[i];
7849: /* 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]);*/
7850: }
7851: }
7852:
7853: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7854: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7855:
7856: return -2*LL*num/sump;
7857: }
7858: #endif
7859:
1.126 brouard 7860: /******************* Printing html file ***********/
1.201 brouard 7861: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7862: int lastpass, int stepm, int weightopt, char model[],\
7863: int imx, double p[],double **matcov,double agemortsup){
7864: int i,k;
7865:
7866: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7867: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7868: for (i=1;i<=2;i++)
7869: 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 7870: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7871: fprintf(fichtm,"</ul>");
7872:
7873: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7874:
7875: 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>");
7876:
7877: for (k=agegomp;k<(agemortsup-2);k++)
7878: 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]);
7879:
7880:
7881: fflush(fichtm);
7882: }
7883:
7884: /******************* Gnuplot file **************/
1.201 brouard 7885: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7886:
7887: char dirfileres[132],optfileres[132];
1.164 brouard 7888:
1.126 brouard 7889: int ng;
7890:
7891:
7892: /*#ifdef windows */
7893: fprintf(ficgp,"cd \"%s\" \n",pathc);
7894: /*#endif */
7895:
7896:
7897: strcpy(dirfileres,optionfilefiname);
7898: strcpy(optfileres,"vpl");
1.199 brouard 7899: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7900: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7901: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7902: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7903: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7904:
7905: }
7906:
1.136 brouard 7907: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7908: {
1.126 brouard 7909:
1.136 brouard 7910: /*-------- data file ----------*/
7911: FILE *fic;
7912: char dummy[]=" ";
1.240 brouard 7913: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7914: int lstra;
1.136 brouard 7915: int linei, month, year,iout;
7916: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7917: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7918: char *stratrunc;
1.223 brouard 7919:
1.240 brouard 7920: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7921: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7922:
1.240 brouard 7923: for(v=1; v <=ncovcol;v++){
7924: DummyV[v]=0;
7925: FixedV[v]=0;
7926: }
7927: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7928: DummyV[v]=1;
7929: FixedV[v]=0;
7930: }
7931: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7932: DummyV[v]=0;
7933: FixedV[v]=1;
7934: }
7935: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7936: DummyV[v]=1;
7937: FixedV[v]=1;
7938: }
7939: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
7940: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
7941: 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]);
7942: }
1.126 brouard 7943:
1.136 brouard 7944: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 7945: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
7946: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 7947: }
1.126 brouard 7948:
1.136 brouard 7949: i=1;
7950: linei=0;
7951: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
7952: linei=linei+1;
7953: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
7954: if(line[j] == '\t')
7955: line[j] = ' ';
7956: }
7957: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
7958: ;
7959: };
7960: line[j+1]=0; /* Trims blanks at end of line */
7961: if(line[0]=='#'){
7962: fprintf(ficlog,"Comment line\n%s\n",line);
7963: printf("Comment line\n%s\n",line);
7964: continue;
7965: }
7966: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 7967: strcpy(line, linetmp);
1.223 brouard 7968:
7969: /* Loops on waves */
7970: for (j=maxwav;j>=1;j--){
7971: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 7972: cutv(stra, strb, line, ' ');
7973: if(strb[0]=='.') { /* Missing value */
7974: lval=-1;
7975: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
7976: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
7977: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
7978: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);
7979: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);fflush(ficlog);
7980: return 1;
7981: }
7982: }else{
7983: errno=0;
7984: /* what_kind_of_number(strb); */
7985: dval=strtod(strb,&endptr);
7986: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
7987: /* if(strb != endptr && *endptr == '\0') */
7988: /* dval=dlval; */
7989: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
7990: if( strb[0]=='\0' || (*endptr != '\0')){
7991: 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);
7992: 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);
7993: return 1;
7994: }
7995: cotqvar[j][iv][i]=dval;
7996: cotvar[j][ntv+iv][i]=dval;
7997: }
7998: strcpy(line,stra);
1.223 brouard 7999: }/* end loop ntqv */
1.225 brouard 8000:
1.223 brouard 8001: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8002: cutv(stra, strb, line, ' ');
8003: if(strb[0]=='.') { /* Missing value */
8004: lval=-1;
8005: }else{
8006: errno=0;
8007: lval=strtol(strb,&endptr,10);
8008: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8009: if( strb[0]=='\0' || (*endptr != '\0')){
8010: 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);
8011: 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);
8012: return 1;
8013: }
8014: }
8015: if(lval <-1 || lval >1){
8016: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8017: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8018: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8019: For example, for multinomial values like 1, 2 and 3,\n \
8020: build V1=0 V2=0 for the reference value (1),\n \
8021: V1=1 V2=0 for (2) \n \
1.223 brouard 8022: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8023: output of IMaCh is often meaningless.\n \
1.223 brouard 8024: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8025: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8026: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8027: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8028: For example, for multinomial values like 1, 2 and 3,\n \
8029: build V1=0 V2=0 for the reference value (1),\n \
8030: V1=1 V2=0 for (2) \n \
1.223 brouard 8031: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8032: output of IMaCh is often meaningless.\n \
1.223 brouard 8033: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8034: return 1;
8035: }
8036: cotvar[j][iv][i]=(double)(lval);
8037: strcpy(line,stra);
1.223 brouard 8038: }/* end loop ntv */
1.225 brouard 8039:
1.223 brouard 8040: /* Statuses at wave */
1.137 brouard 8041: cutv(stra, strb, line, ' ');
1.223 brouard 8042: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8043: lval=-1;
1.136 brouard 8044: }else{
1.238 brouard 8045: errno=0;
8046: lval=strtol(strb,&endptr,10);
8047: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8048: if( strb[0]=='\0' || (*endptr != '\0')){
8049: 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);
8050: 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);
8051: return 1;
8052: }
1.136 brouard 8053: }
1.225 brouard 8054:
1.136 brouard 8055: s[j][i]=lval;
1.225 brouard 8056:
1.223 brouard 8057: /* Date of Interview */
1.136 brouard 8058: strcpy(line,stra);
8059: cutv(stra, strb,line,' ');
1.169 brouard 8060: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8061: }
1.169 brouard 8062: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8063: month=99;
8064: year=9999;
1.136 brouard 8065: }else{
1.225 brouard 8066: 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);
8067: 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);
8068: return 1;
1.136 brouard 8069: }
8070: anint[j][i]= (double) year;
8071: mint[j][i]= (double)month;
8072: strcpy(line,stra);
1.223 brouard 8073: } /* End loop on waves */
1.225 brouard 8074:
1.223 brouard 8075: /* Date of death */
1.136 brouard 8076: cutv(stra, strb,line,' ');
1.169 brouard 8077: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8078: }
1.169 brouard 8079: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8080: month=99;
8081: year=9999;
8082: }else{
1.141 brouard 8083: 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 8084: 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);
8085: return 1;
1.136 brouard 8086: }
8087: andc[i]=(double) year;
8088: moisdc[i]=(double) month;
8089: strcpy(line,stra);
8090:
1.223 brouard 8091: /* Date of birth */
1.136 brouard 8092: cutv(stra, strb,line,' ');
1.169 brouard 8093: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8094: }
1.169 brouard 8095: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8096: month=99;
8097: year=9999;
8098: }else{
1.141 brouard 8099: 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);
8100: 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 8101: return 1;
1.136 brouard 8102: }
8103: if (year==9999) {
1.141 brouard 8104: 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);
8105: 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 8106: return 1;
8107:
1.136 brouard 8108: }
8109: annais[i]=(double)(year);
8110: moisnais[i]=(double)(month);
8111: strcpy(line,stra);
1.225 brouard 8112:
1.223 brouard 8113: /* Sample weight */
1.136 brouard 8114: cutv(stra, strb,line,' ');
8115: errno=0;
8116: dval=strtod(strb,&endptr);
8117: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8118: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8119: 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 8120: fflush(ficlog);
8121: return 1;
8122: }
8123: weight[i]=dval;
8124: strcpy(line,stra);
1.225 brouard 8125:
1.223 brouard 8126: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8127: cutv(stra, strb, line, ' ');
8128: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8129: lval=-1;
1.223 brouard 8130: }else{
1.225 brouard 8131: errno=0;
8132: /* what_kind_of_number(strb); */
8133: dval=strtod(strb,&endptr);
8134: /* if(strb != endptr && *endptr == '\0') */
8135: /* dval=dlval; */
8136: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8137: if( strb[0]=='\0' || (*endptr != '\0')){
8138: 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);
8139: 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);
8140: return 1;
8141: }
8142: coqvar[iv][i]=dval;
1.226 brouard 8143: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8144: }
8145: strcpy(line,stra);
8146: }/* end loop nqv */
1.136 brouard 8147:
1.223 brouard 8148: /* Covariate values */
1.136 brouard 8149: for (j=ncovcol;j>=1;j--){
8150: cutv(stra, strb,line,' ');
1.223 brouard 8151: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8152: lval=-1;
1.136 brouard 8153: }else{
1.225 brouard 8154: errno=0;
8155: lval=strtol(strb,&endptr,10);
8156: if( strb[0]=='\0' || (*endptr != '\0')){
8157: 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);
8158: 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);
8159: return 1;
8160: }
1.136 brouard 8161: }
8162: if(lval <-1 || lval >1){
1.225 brouard 8163: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8164: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8165: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8166: For example, for multinomial values like 1, 2 and 3,\n \
8167: build V1=0 V2=0 for the reference value (1),\n \
8168: V1=1 V2=0 for (2) \n \
1.136 brouard 8169: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8170: output of IMaCh is often meaningless.\n \
1.136 brouard 8171: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8172: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8173: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8174: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8175: For example, for multinomial values like 1, 2 and 3,\n \
8176: build V1=0 V2=0 for the reference value (1),\n \
8177: V1=1 V2=0 for (2) \n \
1.136 brouard 8178: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8179: output of IMaCh is often meaningless.\n \
1.136 brouard 8180: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8181: return 1;
1.136 brouard 8182: }
8183: covar[j][i]=(double)(lval);
8184: strcpy(line,stra);
8185: }
8186: lstra=strlen(stra);
1.225 brouard 8187:
1.136 brouard 8188: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8189: stratrunc = &(stra[lstra-9]);
8190: num[i]=atol(stratrunc);
8191: }
8192: else
8193: num[i]=atol(stra);
8194: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8195: 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;}*/
8196:
8197: i=i+1;
8198: } /* End loop reading data */
1.225 brouard 8199:
1.136 brouard 8200: *imax=i-1; /* Number of individuals */
8201: fclose(fic);
1.225 brouard 8202:
1.136 brouard 8203: return (0);
1.164 brouard 8204: /* endread: */
1.225 brouard 8205: printf("Exiting readdata: ");
8206: fclose(fic);
8207: return (1);
1.223 brouard 8208: }
1.126 brouard 8209:
1.234 brouard 8210: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8211: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8212: while (*p2 == ' ')
1.234 brouard 8213: p2++;
8214: /* while ((*p1++ = *p2++) !=0) */
8215: /* ; */
8216: /* do */
8217: /* while (*p2 == ' ') */
8218: /* p2++; */
8219: /* while (*p1++ == *p2++); */
8220: *stri=p2;
1.145 brouard 8221: }
8222:
1.235 brouard 8223: int decoderesult ( char resultline[], int nres)
1.230 brouard 8224: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8225: {
1.235 brouard 8226: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8227: char resultsav[MAXLINE];
1.234 brouard 8228: int resultmodel[MAXLINE];
8229: int modelresult[MAXLINE];
1.230 brouard 8230: char stra[80], strb[80], strc[80], strd[80],stre[80];
8231:
1.234 brouard 8232: removefirstspace(&resultline);
1.233 brouard 8233: printf("decoderesult:%s\n",resultline);
1.230 brouard 8234:
8235: if (strstr(resultline,"v") !=0){
8236: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8237: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8238: return 1;
8239: }
8240: trimbb(resultsav, resultline);
8241: if (strlen(resultsav) >1){
8242: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8243: }
1.234 brouard 8244: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8245: 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);
8246: 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);
8247: }
8248: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8249: if(nbocc(resultsav,'=') >1){
8250: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8251: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8252: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8253: }else
8254: cutl(strc,strd,resultsav,'=');
1.230 brouard 8255: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8256:
1.230 brouard 8257: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8258: Tvarsel[k]=atoi(strc);
8259: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8260: /* cptcovsel++; */
8261: if (nbocc(stra,'=') >0)
8262: strcpy(resultsav,stra); /* and analyzes it */
8263: }
1.235 brouard 8264: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8265: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8266: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8267: match=0;
1.236 brouard 8268: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8269: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8270: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8271: match=1;
8272: break;
8273: }
8274: }
8275: if(match == 0){
8276: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8277: }
8278: }
8279: }
1.235 brouard 8280: /* Checking for missing or useless values in comparison of current model needs */
8281: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8282: match=0;
1.235 brouard 8283: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8284: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8285: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8286: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8287: ++match;
8288: }
8289: }
8290: }
8291: if(match == 0){
8292: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8293: }else if(match > 1){
8294: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8295: }
8296: }
1.235 brouard 8297:
1.234 brouard 8298: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8299: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8300: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8301: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8302: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8303: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8304: /* 1 0 0 0 */
8305: /* 2 1 0 0 */
8306: /* 3 0 1 0 */
8307: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8308: /* 5 0 0 1 */
8309: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8310: /* 7 0 1 1 */
8311: /* 8 1 1 1 */
1.237 brouard 8312: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8313: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8314: /* V5*age V5 known which value for nres? */
8315: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8316: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8317: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8318: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8319: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8320: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8321: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8322: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8323: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8324: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8325: k4++;;
8326: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8327: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8328: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8329: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8330: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8331: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8332: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8333: k4q++;;
8334: }
8335: }
1.234 brouard 8336:
1.235 brouard 8337: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8338: return (0);
8339: }
1.235 brouard 8340:
1.230 brouard 8341: int decodemodel( char model[], int lastobs)
8342: /**< This routine decodes the model and returns:
1.224 brouard 8343: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8344: * - nagesqr = 1 if age*age in the model, otherwise 0.
8345: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8346: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8347: * - cptcovage number of covariates with age*products =2
8348: * - cptcovs number of simple covariates
8349: * - 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
8350: * which is a new column after the 9 (ncovcol) variables.
8351: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8352: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8353: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8354: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8355: */
1.136 brouard 8356: {
1.238 brouard 8357: int i, j, k, ks, v;
1.227 brouard 8358: int j1, k1, k2, k3, k4;
1.136 brouard 8359: char modelsav[80];
1.145 brouard 8360: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8361: char *strpt;
1.136 brouard 8362:
1.145 brouard 8363: /*removespace(model);*/
1.136 brouard 8364: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8365: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8366: if (strstr(model,"AGE") !=0){
1.192 brouard 8367: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8368: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8369: return 1;
8370: }
1.141 brouard 8371: if (strstr(model,"v") !=0){
8372: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8373: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8374: return 1;
8375: }
1.187 brouard 8376: strcpy(modelsav,model);
8377: if ((strpt=strstr(model,"age*age")) !=0){
8378: printf(" strpt=%s, model=%s\n",strpt, model);
8379: if(strpt != model){
1.234 brouard 8380: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8381: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8382: corresponding column of parameters.\n",model);
1.234 brouard 8383: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8384: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8385: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8386: return 1;
1.225 brouard 8387: }
1.187 brouard 8388: nagesqr=1;
8389: if (strstr(model,"+age*age") !=0)
1.234 brouard 8390: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8391: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8392: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8393: else
1.234 brouard 8394: substrchaine(modelsav, model, "age*age");
1.187 brouard 8395: }else
8396: nagesqr=0;
8397: if (strlen(modelsav) >1){
8398: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8399: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8400: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8401: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8402: * cst, age and age*age
8403: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8404: /* including age products which are counted in cptcovage.
8405: * but the covariates which are products must be treated
8406: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8407: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8408: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8409:
8410:
1.187 brouard 8411: /* Design
8412: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8413: * < ncovcol=8 >
8414: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8415: * k= 1 2 3 4 5 6 7 8
8416: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8417: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8418: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8419: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8420: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8421: * Tage[++cptcovage]=k
8422: * if products, new covar are created after ncovcol with k1
8423: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8424: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8425: * 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
8426: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8427: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8428: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8429: * < ncovcol=8 >
8430: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8431: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8432: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8433: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8434: * p Tprod[1]@2={ 6, 5}
8435: *p Tvard[1][1]@4= {7, 8, 5, 6}
8436: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8437: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8438: *How to reorganize?
8439: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8440: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8441: * {2, 1, 4, 8, 5, 6, 3, 7}
8442: * Struct []
8443: */
1.225 brouard 8444:
1.187 brouard 8445: /* This loop fills the array Tvar from the string 'model'.*/
8446: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8447: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8448: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8449: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8450: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8451: /* k=1 Tvar[1]=2 (from V2) */
8452: /* k=5 Tvar[5] */
8453: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8454: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8455: /* } */
1.198 brouard 8456: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8457: /*
8458: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8459: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8460: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8461: }
1.187 brouard 8462: cptcovage=0;
8463: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8464: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8465: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8466: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8467: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8468: /*scanf("%d",i);*/
8469: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8470: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8471: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8472: /* covar is not filled and then is empty */
8473: cptcovprod--;
8474: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8475: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8476: Typevar[k]=1; /* 1 for age product */
8477: cptcovage++; /* Sums the number of covariates which include age as a product */
8478: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8479: /*printf("stre=%s ", stre);*/
8480: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8481: cptcovprod--;
8482: cutl(stre,strb,strc,'V');
8483: Tvar[k]=atoi(stre);
8484: Typevar[k]=1; /* 1 for age product */
8485: cptcovage++;
8486: Tage[cptcovage]=k;
8487: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8488: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8489: cptcovn++;
8490: cptcovprodnoage++;k1++;
8491: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8492: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8493: because this model-covariate is a construction we invent a new column
8494: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8495: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8496: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8497: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8498: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8499: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8500: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8501: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8502: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8503: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8504: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8505: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8506: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8507: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8508: for (i=1; i<=lastobs;i++){
8509: /* Computes the new covariate which is a product of
8510: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8511: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8512: }
8513: } /* End age is not in the model */
8514: } /* End if model includes a product */
8515: else { /* no more sum */
8516: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8517: /* scanf("%d",i);*/
8518: cutl(strd,strc,strb,'V');
8519: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8520: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8521: Tvar[k]=atoi(strd);
8522: Typevar[k]=0; /* 0 for simple covariates */
8523: }
8524: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8525: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8526: scanf("%d",i);*/
1.187 brouard 8527: } /* end of loop + on total covariates */
8528: } /* end if strlen(modelsave == 0) age*age might exist */
8529: } /* end if strlen(model == 0) */
1.136 brouard 8530:
8531: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8532: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8533:
1.136 brouard 8534: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8535: printf("cptcovprod=%d ", cptcovprod);
8536: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8537: scanf("%d ",i);*/
8538:
8539:
1.230 brouard 8540: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8541: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8542: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8543: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8544: k = 1 2 3 4 5 6 7 8 9
8545: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8546: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8547: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8548: Dummy[k] 1 0 0 0 3 1 1 2 3
8549: Tmodelind[combination of covar]=k;
1.225 brouard 8550: */
8551: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8552: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8553: /* 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 8554: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8555: printf("Model=%s\n\
8556: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8557: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8558: 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);
8559: fprintf(ficlog,"Model=%s\n\
8560: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8561: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8562: 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 8563: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8564: 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 */
8565: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8566: Fixed[k]= 0;
8567: Dummy[k]= 0;
1.225 brouard 8568: ncoveff++;
1.232 brouard 8569: ncovf++;
1.234 brouard 8570: nsd++;
8571: modell[k].maintype= FTYPE;
8572: TvarsD[nsd]=Tvar[k];
8573: TvarsDind[nsd]=k;
8574: TvarF[ncovf]=Tvar[k];
8575: TvarFind[ncovf]=k;
8576: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8577: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8578: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8579: Fixed[k]= 0;
8580: Dummy[k]= 0;
8581: ncoveff++;
8582: ncovf++;
8583: modell[k].maintype= FTYPE;
8584: TvarF[ncovf]=Tvar[k];
8585: TvarFind[ncovf]=k;
1.230 brouard 8586: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8587: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8588: }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 8589: Fixed[k]= 0;
8590: Dummy[k]= 1;
1.230 brouard 8591: nqfveff++;
1.234 brouard 8592: modell[k].maintype= FTYPE;
8593: modell[k].subtype= FQ;
8594: nsq++;
8595: TvarsQ[nsq]=Tvar[k];
8596: TvarsQind[nsq]=k;
1.232 brouard 8597: ncovf++;
1.234 brouard 8598: TvarF[ncovf]=Tvar[k];
8599: TvarFind[ncovf]=k;
1.231 brouard 8600: 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 8601: 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 8602: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8603: Fixed[k]= 1;
8604: Dummy[k]= 0;
1.225 brouard 8605: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8606: modell[k].maintype= VTYPE;
8607: modell[k].subtype= VD;
8608: nsd++;
8609: TvarsD[nsd]=Tvar[k];
8610: TvarsDind[nsd]=k;
8611: ncovv++; /* Only simple time varying variables */
8612: TvarV[ncovv]=Tvar[k];
1.242 brouard 8613: 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 8614: 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 */
8615: 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 8616: 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);
8617: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8618: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8619: Fixed[k]= 1;
8620: Dummy[k]= 1;
8621: nqtveff++;
8622: modell[k].maintype= VTYPE;
8623: modell[k].subtype= VQ;
8624: ncovv++; /* Only simple time varying variables */
8625: nsq++;
8626: TvarsQ[nsq]=Tvar[k];
8627: TvarsQind[nsq]=k;
8628: TvarV[ncovv]=Tvar[k];
1.242 brouard 8629: 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 8630: 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 */
8631: 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 8632: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8633: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8634: 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 8635: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8636: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8637: ncova++;
8638: TvarA[ncova]=Tvar[k];
8639: TvarAind[ncova]=k;
1.231 brouard 8640: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8641: Fixed[k]= 2;
8642: Dummy[k]= 2;
8643: modell[k].maintype= ATYPE;
8644: modell[k].subtype= APFD;
8645: /* ncoveff++; */
1.227 brouard 8646: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8647: Fixed[k]= 2;
8648: Dummy[k]= 3;
8649: modell[k].maintype= ATYPE;
8650: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8651: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8652: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8653: Fixed[k]= 3;
8654: Dummy[k]= 2;
8655: modell[k].maintype= ATYPE;
8656: modell[k].subtype= APVD; /* Product age * varying dummy */
8657: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8658: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8659: Fixed[k]= 3;
8660: Dummy[k]= 3;
8661: modell[k].maintype= ATYPE;
8662: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8663: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8664: }
8665: }else if (Typevar[k] == 2) { /* product without age */
8666: k1=Tposprod[k];
8667: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8668: if(Tvard[k1][2] <=ncovcol){
8669: Fixed[k]= 1;
8670: Dummy[k]= 0;
8671: modell[k].maintype= FTYPE;
8672: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8673: ncovf++; /* Fixed variables without age */
8674: TvarF[ncovf]=Tvar[k];
8675: TvarFind[ncovf]=k;
8676: }else if(Tvard[k1][2] <=ncovcol+nqv){
8677: Fixed[k]= 0; /* or 2 ?*/
8678: Dummy[k]= 1;
8679: modell[k].maintype= FTYPE;
8680: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8681: ncovf++; /* Varying variables without age */
8682: TvarF[ncovf]=Tvar[k];
8683: TvarFind[ncovf]=k;
8684: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8685: Fixed[k]= 1;
8686: Dummy[k]= 0;
8687: modell[k].maintype= VTYPE;
8688: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8689: ncovv++; /* Varying variables without age */
8690: TvarV[ncovv]=Tvar[k];
8691: TvarVind[ncovv]=k;
8692: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8693: Fixed[k]= 1;
8694: Dummy[k]= 1;
8695: modell[k].maintype= VTYPE;
8696: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8697: ncovv++; /* Varying variables without age */
8698: TvarV[ncovv]=Tvar[k];
8699: TvarVind[ncovv]=k;
8700: }
1.227 brouard 8701: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8702: if(Tvard[k1][2] <=ncovcol){
8703: Fixed[k]= 0; /* or 2 ?*/
8704: Dummy[k]= 1;
8705: modell[k].maintype= FTYPE;
8706: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8707: ncovf++; /* Fixed variables without age */
8708: TvarF[ncovf]=Tvar[k];
8709: TvarFind[ncovf]=k;
8710: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8711: Fixed[k]= 1;
8712: Dummy[k]= 1;
8713: modell[k].maintype= VTYPE;
8714: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8715: ncovv++; /* Varying variables without age */
8716: TvarV[ncovv]=Tvar[k];
8717: TvarVind[ncovv]=k;
8718: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8719: Fixed[k]= 1;
8720: Dummy[k]= 1;
8721: modell[k].maintype= VTYPE;
8722: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8723: ncovv++; /* Varying variables without age */
8724: TvarV[ncovv]=Tvar[k];
8725: TvarVind[ncovv]=k;
8726: ncovv++; /* Varying variables without age */
8727: TvarV[ncovv]=Tvar[k];
8728: TvarVind[ncovv]=k;
8729: }
1.227 brouard 8730: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8731: if(Tvard[k1][2] <=ncovcol){
8732: Fixed[k]= 1;
8733: Dummy[k]= 1;
8734: modell[k].maintype= VTYPE;
8735: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8736: ncovv++; /* Varying variables without age */
8737: TvarV[ncovv]=Tvar[k];
8738: TvarVind[ncovv]=k;
8739: }else if(Tvard[k1][2] <=ncovcol+nqv){
8740: Fixed[k]= 1;
8741: Dummy[k]= 1;
8742: modell[k].maintype= VTYPE;
8743: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8744: ncovv++; /* Varying variables without age */
8745: TvarV[ncovv]=Tvar[k];
8746: TvarVind[ncovv]=k;
8747: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8748: Fixed[k]= 1;
8749: Dummy[k]= 0;
8750: modell[k].maintype= VTYPE;
8751: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8752: ncovv++; /* Varying variables without age */
8753: TvarV[ncovv]=Tvar[k];
8754: TvarVind[ncovv]=k;
8755: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8756: Fixed[k]= 1;
8757: Dummy[k]= 1;
8758: modell[k].maintype= VTYPE;
8759: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8760: ncovv++; /* Varying variables without age */
8761: TvarV[ncovv]=Tvar[k];
8762: TvarVind[ncovv]=k;
8763: }
1.227 brouard 8764: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8765: if(Tvard[k1][2] <=ncovcol){
8766: Fixed[k]= 1;
8767: Dummy[k]= 1;
8768: modell[k].maintype= VTYPE;
8769: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8770: ncovv++; /* Varying variables without age */
8771: TvarV[ncovv]=Tvar[k];
8772: TvarVind[ncovv]=k;
8773: }else if(Tvard[k1][2] <=ncovcol+nqv){
8774: Fixed[k]= 1;
8775: Dummy[k]= 1;
8776: modell[k].maintype= VTYPE;
8777: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8778: ncovv++; /* Varying variables without age */
8779: TvarV[ncovv]=Tvar[k];
8780: TvarVind[ncovv]=k;
8781: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8782: Fixed[k]= 1;
8783: Dummy[k]= 1;
8784: modell[k].maintype= VTYPE;
8785: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8786: ncovv++; /* Varying variables without age */
8787: TvarV[ncovv]=Tvar[k];
8788: TvarVind[ncovv]=k;
8789: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8790: Fixed[k]= 1;
8791: Dummy[k]= 1;
8792: modell[k].maintype= VTYPE;
8793: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8794: ncovv++; /* Varying variables without age */
8795: TvarV[ncovv]=Tvar[k];
8796: TvarVind[ncovv]=k;
8797: }
1.227 brouard 8798: }else{
1.240 brouard 8799: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8800: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8801: } /*end k1*/
1.225 brouard 8802: }else{
1.226 brouard 8803: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8804: 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 8805: }
1.227 brouard 8806: 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 8807: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8808: 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]);
8809: }
8810: /* Searching for doublons in the model */
8811: for(k1=1; k1<= cptcovt;k1++){
8812: for(k2=1; k2 <k1;k2++){
8813: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8814: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8815: if(Tvar[k1]==Tvar[k2]){
8816: 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]]);
8817: 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);
8818: return(1);
8819: }
8820: }else if (Typevar[k1] ==2){
8821: k3=Tposprod[k1];
8822: k4=Tposprod[k2];
8823: 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])) ){
8824: 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]]);
8825: 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);
8826: return(1);
8827: }
8828: }
1.227 brouard 8829: }
8830: }
1.225 brouard 8831: }
8832: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8833: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8834: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8835: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8836: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8837: /*endread:*/
1.225 brouard 8838: printf("Exiting decodemodel: ");
8839: return (1);
1.136 brouard 8840: }
8841:
1.169 brouard 8842: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.136 brouard 8843: {
8844: int i, m;
1.218 brouard 8845: int firstone=0;
8846:
1.136 brouard 8847: for (i=1; i<=imx; i++) {
8848: for(m=2; (m<= maxwav); m++) {
8849: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8850: anint[m][i]=9999;
1.216 brouard 8851: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8852: s[m][i]=-1;
1.136 brouard 8853: }
8854: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8855: *nberr = *nberr + 1;
1.218 brouard 8856: if(firstone == 0){
8857: firstone=1;
8858: 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);
8859: }
8860: 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 8861: s[m][i]=-1;
8862: }
8863: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8864: (*nberr)++;
1.136 brouard 8865: 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]);
8866: 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]);
8867: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8868: }
8869: }
8870: }
8871:
8872: for (i=1; i<=imx; i++) {
8873: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8874: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8875: 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 8876: if (s[m][i] >= nlstate+1) {
1.169 brouard 8877: if(agedc[i]>0){
8878: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8879: agev[m][i]=agedc[i];
1.214 brouard 8880: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8881: }else {
1.136 brouard 8882: if ((int)andc[i]!=9999){
8883: nbwarn++;
8884: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8885: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8886: agev[m][i]=-1;
8887: }
8888: }
1.169 brouard 8889: } /* agedc > 0 */
1.214 brouard 8890: } /* end if */
1.136 brouard 8891: else if(s[m][i] !=9){ /* Standard case, age in fractional
8892: years but with the precision of a month */
8893: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8894: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8895: agev[m][i]=1;
8896: else if(agev[m][i] < *agemin){
8897: *agemin=agev[m][i];
8898: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8899: }
8900: else if(agev[m][i] >*agemax){
8901: *agemax=agev[m][i];
1.156 brouard 8902: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8903: }
8904: /*agev[m][i]=anint[m][i]-annais[i];*/
8905: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8906: } /* en if 9*/
1.136 brouard 8907: else { /* =9 */
1.214 brouard 8908: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8909: agev[m][i]=1;
8910: s[m][i]=-1;
8911: }
8912: }
1.214 brouard 8913: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8914: agev[m][i]=1;
1.214 brouard 8915: else{
8916: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8917: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8918: agev[m][i]=0;
8919: }
8920: } /* End for lastpass */
8921: }
1.136 brouard 8922:
8923: for (i=1; i<=imx; i++) {
8924: for(m=firstpass; (m<=lastpass); m++){
8925: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8926: (*nberr)++;
1.136 brouard 8927: 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);
8928: 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);
8929: return 1;
8930: }
8931: }
8932: }
8933:
8934: /*for (i=1; i<=imx; i++){
8935: for (m=firstpass; (m<lastpass); m++){
8936: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8937: }
8938:
8939: }*/
8940:
8941:
1.139 brouard 8942: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
8943: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 8944:
8945: return (0);
1.164 brouard 8946: /* endread:*/
1.136 brouard 8947: printf("Exiting calandcheckages: ");
8948: return (1);
8949: }
8950:
1.172 brouard 8951: #if defined(_MSC_VER)
8952: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8953: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
8954: //#include "stdafx.h"
8955: //#include <stdio.h>
8956: //#include <tchar.h>
8957: //#include <windows.h>
8958: //#include <iostream>
8959: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
8960:
8961: LPFN_ISWOW64PROCESS fnIsWow64Process;
8962:
8963: BOOL IsWow64()
8964: {
8965: BOOL bIsWow64 = FALSE;
8966:
8967: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
8968: // (HANDLE, PBOOL);
8969:
8970: //LPFN_ISWOW64PROCESS fnIsWow64Process;
8971:
8972: HMODULE module = GetModuleHandle(_T("kernel32"));
8973: const char funcName[] = "IsWow64Process";
8974: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
8975: GetProcAddress(module, funcName);
8976:
8977: if (NULL != fnIsWow64Process)
8978: {
8979: if (!fnIsWow64Process(GetCurrentProcess(),
8980: &bIsWow64))
8981: //throw std::exception("Unknown error");
8982: printf("Unknown error\n");
8983: }
8984: return bIsWow64 != FALSE;
8985: }
8986: #endif
1.177 brouard 8987:
1.191 brouard 8988: void syscompilerinfo(int logged)
1.167 brouard 8989: {
8990: /* #include "syscompilerinfo.h"*/
1.185 brouard 8991: /* command line Intel compiler 32bit windows, XP compatible:*/
8992: /* /GS /W3 /Gy
8993: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
8994: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
8995: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 8996: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
8997: */
8998: /* 64 bits */
1.185 brouard 8999: /*
9000: /GS /W3 /Gy
9001: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9002: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9003: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9004: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9005: /* Optimization are useless and O3 is slower than O2 */
9006: /*
9007: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9008: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9009: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9010: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9011: */
1.186 brouard 9012: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9013: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9014: /PDB:"visual studio
9015: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9016: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9017: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9018: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9019: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9020: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9021: uiAccess='false'"
9022: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9023: /NOLOGO /TLBID:1
9024: */
1.177 brouard 9025: #if defined __INTEL_COMPILER
1.178 brouard 9026: #if defined(__GNUC__)
9027: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9028: #endif
1.177 brouard 9029: #elif defined(__GNUC__)
1.179 brouard 9030: #ifndef __APPLE__
1.174 brouard 9031: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9032: #endif
1.177 brouard 9033: struct utsname sysInfo;
1.178 brouard 9034: int cross = CROSS;
9035: if (cross){
9036: printf("Cross-");
1.191 brouard 9037: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9038: }
1.174 brouard 9039: #endif
9040:
1.171 brouard 9041: #include <stdint.h>
1.178 brouard 9042:
1.191 brouard 9043: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9044: #if defined(__clang__)
1.191 brouard 9045: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9046: #endif
9047: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9048: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9049: #endif
9050: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9051: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9052: #endif
9053: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9054: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9055: #endif
9056: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9057: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9058: #endif
9059: #if defined(_MSC_VER)
1.191 brouard 9060: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9061: #endif
9062: #if defined(__PGI)
1.191 brouard 9063: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9064: #endif
9065: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9066: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9067: #endif
1.191 brouard 9068: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9069:
1.167 brouard 9070: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9071: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9072: // Windows (x64 and x86)
1.191 brouard 9073: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9074: #elif __unix__ // all unices, not all compilers
9075: // Unix
1.191 brouard 9076: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9077: #elif __linux__
9078: // linux
1.191 brouard 9079: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9080: #elif __APPLE__
1.174 brouard 9081: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9082: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9083: #endif
9084:
9085: /* __MINGW32__ */
9086: /* __CYGWIN__ */
9087: /* __MINGW64__ */
9088: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9089: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9090: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9091: /* _WIN64 // Defined for applications for Win64. */
9092: /* _M_X64 // Defined for compilations that target x64 processors. */
9093: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9094:
1.167 brouard 9095: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9096: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9097: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9098: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9099: #else
1.191 brouard 9100: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9101: #endif
9102:
1.169 brouard 9103: #if defined(__GNUC__)
9104: # if defined(__GNUC_PATCHLEVEL__)
9105: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9106: + __GNUC_MINOR__ * 100 \
9107: + __GNUC_PATCHLEVEL__)
9108: # else
9109: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9110: + __GNUC_MINOR__ * 100)
9111: # endif
1.174 brouard 9112: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9113: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9114:
9115: if (uname(&sysInfo) != -1) {
9116: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9117: 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 9118: }
9119: else
9120: perror("uname() error");
1.179 brouard 9121: //#ifndef __INTEL_COMPILER
9122: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9123: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9124: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9125: #endif
1.169 brouard 9126: #endif
1.172 brouard 9127:
9128: // void main()
9129: // {
1.169 brouard 9130: #if defined(_MSC_VER)
1.174 brouard 9131: if (IsWow64()){
1.191 brouard 9132: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9133: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9134: }
9135: else{
1.191 brouard 9136: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9137: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9138: }
1.172 brouard 9139: // printf("\nPress Enter to continue...");
9140: // getchar();
9141: // }
9142:
1.169 brouard 9143: #endif
9144:
1.167 brouard 9145:
1.219 brouard 9146: }
1.136 brouard 9147:
1.219 brouard 9148: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9149: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9150: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9151: /* double ftolpl = 1.e-10; */
1.180 brouard 9152: double age, agebase, agelim;
1.203 brouard 9153: double tot;
1.180 brouard 9154:
1.202 brouard 9155: strcpy(filerespl,"PL_");
9156: strcat(filerespl,fileresu);
9157: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9158: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9159: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9160: }
1.227 brouard 9161: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9162: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9163: pstamp(ficrespl);
1.203 brouard 9164: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9165: fprintf(ficrespl,"#Age ");
9166: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9167: fprintf(ficrespl,"\n");
1.180 brouard 9168:
1.219 brouard 9169: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9170:
1.219 brouard 9171: agebase=ageminpar;
9172: agelim=agemaxpar;
1.180 brouard 9173:
1.227 brouard 9174: /* i1=pow(2,ncoveff); */
1.234 brouard 9175: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9176: if (cptcovn < 1){i1=1;}
1.180 brouard 9177:
1.238 brouard 9178: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9179: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9180: if(TKresult[nres]!= k)
9181: continue;
1.235 brouard 9182:
1.238 brouard 9183: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9184: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9185: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9186: /* k=k+1; */
9187: /* to clean */
9188: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9189: fprintf(ficrespl,"#******");
9190: printf("#******");
9191: fprintf(ficlog,"#******");
9192: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9193: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9194: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9195: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9196: }
9197: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9198: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9199: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9200: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9201: }
9202: fprintf(ficrespl,"******\n");
9203: printf("******\n");
9204: fprintf(ficlog,"******\n");
9205: if(invalidvarcomb[k]){
9206: printf("\nCombination (%d) ignored because no case \n",k);
9207: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9208: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9209: continue;
9210: }
1.219 brouard 9211:
1.238 brouard 9212: fprintf(ficrespl,"#Age ");
9213: for(j=1;j<=cptcoveff;j++) {
9214: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9215: }
9216: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9217: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9218:
1.238 brouard 9219: for (age=agebase; age<=agelim; age++){
9220: /* for (age=agebase; age<=agebase; age++){ */
9221: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9222: fprintf(ficrespl,"%.0f ",age );
9223: for(j=1;j<=cptcoveff;j++)
9224: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9225: tot=0.;
9226: for(i=1; i<=nlstate;i++){
9227: tot += prlim[i][i];
9228: fprintf(ficrespl," %.5f", prlim[i][i]);
9229: }
9230: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9231: } /* Age */
9232: /* was end of cptcod */
9233: } /* cptcov */
9234: } /* nres */
1.219 brouard 9235: return 0;
1.180 brouard 9236: }
9237:
1.218 brouard 9238: 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){
9239: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9240:
9241: /* Computes the back prevalence limit for any combination of covariate values
9242: * at any age between ageminpar and agemaxpar
9243: */
1.235 brouard 9244: int i, j, k, i1, nres=0 ;
1.217 brouard 9245: /* double ftolpl = 1.e-10; */
9246: double age, agebase, agelim;
9247: double tot;
1.218 brouard 9248: /* double ***mobaverage; */
9249: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9250:
9251: strcpy(fileresplb,"PLB_");
9252: strcat(fileresplb,fileresu);
9253: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9254: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9255: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9256: }
9257: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9258: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9259: pstamp(ficresplb);
9260: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9261: fprintf(ficresplb,"#Age ");
9262: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9263: fprintf(ficresplb,"\n");
9264:
1.218 brouard 9265:
9266: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9267:
9268: agebase=ageminpar;
9269: agelim=agemaxpar;
9270:
9271:
1.227 brouard 9272: i1=pow(2,cptcoveff);
1.218 brouard 9273: if (cptcovn < 1){i1=1;}
1.227 brouard 9274:
1.238 brouard 9275: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9276: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9277: if(TKresult[nres]!= k)
9278: continue;
9279: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9280: fprintf(ficresplb,"#******");
9281: printf("#******");
9282: fprintf(ficlog,"#******");
9283: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9284: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9285: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9286: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9287: }
9288: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9289: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9290: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9291: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9292: }
9293: fprintf(ficresplb,"******\n");
9294: printf("******\n");
9295: fprintf(ficlog,"******\n");
9296: if(invalidvarcomb[k]){
9297: printf("\nCombination (%d) ignored because no cases \n",k);
9298: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9299: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9300: continue;
9301: }
1.218 brouard 9302:
1.238 brouard 9303: fprintf(ficresplb,"#Age ");
9304: for(j=1;j<=cptcoveff;j++) {
9305: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9306: }
9307: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9308: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9309:
9310:
1.238 brouard 9311: for (age=agebase; age<=agelim; age++){
9312: /* for (age=agebase; age<=agebase; age++){ */
9313: if(mobilavproj > 0){
9314: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9315: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9316: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9317: }else if (mobilavproj == 0){
9318: 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);
9319: 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);
9320: exit(1);
9321: }else{
9322: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9323: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9324: }
9325: fprintf(ficresplb,"%.0f ",age );
9326: for(j=1;j<=cptcoveff;j++)
9327: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9328: tot=0.;
9329: for(i=1; i<=nlstate;i++){
9330: tot += bprlim[i][i];
9331: fprintf(ficresplb," %.5f", bprlim[i][i]);
9332: }
9333: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9334: } /* Age */
9335: /* was end of cptcod */
9336: } /* end of any combination */
9337: } /* end of nres */
1.218 brouard 9338: /* hBijx(p, bage, fage); */
9339: /* fclose(ficrespijb); */
9340:
9341: return 0;
1.217 brouard 9342: }
1.218 brouard 9343:
1.180 brouard 9344: int hPijx(double *p, int bage, int fage){
9345: /*------------- h Pij x at various ages ------------*/
9346:
9347: int stepsize;
9348: int agelim;
9349: int hstepm;
9350: int nhstepm;
1.235 brouard 9351: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9352:
9353: double agedeb;
9354: double ***p3mat;
9355:
1.201 brouard 9356: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9357: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9358: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9359: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9360: }
9361: printf("Computing pij: result on file '%s' \n", filerespij);
9362: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9363:
9364: stepsize=(int) (stepm+YEARM-1)/YEARM;
9365: /*if (stepm<=24) stepsize=2;*/
9366:
9367: agelim=AGESUP;
9368: hstepm=stepsize*YEARM; /* Every year of age */
9369: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9370:
1.180 brouard 9371: /* hstepm=1; aff par mois*/
9372: pstamp(ficrespij);
9373: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9374: i1= pow(2,cptcoveff);
1.218 brouard 9375: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9376: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9377: /* k=k+1; */
1.235 brouard 9378: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9379: for(k=1; k<=i1;k++){
9380: if(TKresult[nres]!= k)
9381: continue;
1.183 brouard 9382: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9383: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9384: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9385: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9386: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9387: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9388: }
1.183 brouard 9389: fprintf(ficrespij,"******\n");
9390:
9391: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9392: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9393: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9394:
9395: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9396:
1.183 brouard 9397: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9398: oldm=oldms;savm=savms;
1.235 brouard 9399: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9400: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9401: for(i=1; i<=nlstate;i++)
9402: for(j=1; j<=nlstate+ndeath;j++)
9403: fprintf(ficrespij," %1d-%1d",i,j);
9404: fprintf(ficrespij,"\n");
9405: for (h=0; h<=nhstepm; h++){
9406: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9407: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9408: for(i=1; i<=nlstate;i++)
9409: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9410: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9411: fprintf(ficrespij,"\n");
9412: }
1.183 brouard 9413: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9414: fprintf(ficrespij,"\n");
9415: }
1.180 brouard 9416: /*}*/
9417: }
1.218 brouard 9418: return 0;
1.180 brouard 9419: }
1.218 brouard 9420:
9421: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9422: /*------------- h Bij x at various ages ------------*/
9423:
9424: int stepsize;
1.218 brouard 9425: /* int agelim; */
9426: int ageminl;
1.217 brouard 9427: int hstepm;
9428: int nhstepm;
1.238 brouard 9429: int h, i, i1, j, k, nres;
1.218 brouard 9430:
1.217 brouard 9431: double agedeb;
9432: double ***p3mat;
1.218 brouard 9433:
9434: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9435: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9436: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9437: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9438: }
9439: printf("Computing pij back: result on file '%s' \n", filerespijb);
9440: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9441:
9442: stepsize=(int) (stepm+YEARM-1)/YEARM;
9443: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9444:
1.218 brouard 9445: /* agelim=AGESUP; */
9446: ageminl=30;
9447: hstepm=stepsize*YEARM; /* Every year of age */
9448: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9449:
9450: /* hstepm=1; aff par mois*/
9451: pstamp(ficrespijb);
9452: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9453: i1= pow(2,cptcoveff);
1.218 brouard 9454: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9455: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9456: /* k=k+1; */
1.238 brouard 9457: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9458: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9459: if(TKresult[nres]!= k)
9460: continue;
9461: fprintf(ficrespijb,"\n#****** ");
9462: for(j=1;j<=cptcoveff;j++)
9463: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9464: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9465: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9466: }
9467: fprintf(ficrespijb,"******\n");
9468: if(invalidvarcomb[k]){
9469: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9470: continue;
9471: }
9472:
9473: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9474: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9475: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9476: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9477: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9478:
9479: /* nhstepm=nhstepm*YEARM; aff par mois*/
9480:
9481: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9482: /* oldm=oldms;savm=savms; */
9483: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9484: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9485: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9486: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9487: for(i=1; i<=nlstate;i++)
9488: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9489: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9490: fprintf(ficrespijb,"\n");
1.238 brouard 9491: for (h=0; h<=nhstepm; h++){
9492: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9493: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9494: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9495: for(i=1; i<=nlstate;i++)
9496: for(j=1; j<=nlstate+ndeath;j++)
9497: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9498: fprintf(ficrespijb,"\n");
9499: }
9500: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9501: fprintf(ficrespijb,"\n");
9502: } /* end age deb */
9503: } /* end combination */
9504: } /* end nres */
1.218 brouard 9505: return 0;
9506: } /* hBijx */
1.217 brouard 9507:
1.180 brouard 9508:
1.136 brouard 9509: /***********************************************/
9510: /**************** Main Program *****************/
9511: /***********************************************/
9512:
9513: int main(int argc, char *argv[])
9514: {
9515: #ifdef GSL
9516: const gsl_multimin_fminimizer_type *T;
9517: size_t iteri = 0, it;
9518: int rval = GSL_CONTINUE;
9519: int status = GSL_SUCCESS;
9520: double ssval;
9521: #endif
9522: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9523: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9524: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9525: int jj, ll, li, lj, lk;
1.136 brouard 9526: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9527: int num_filled;
1.136 brouard 9528: int itimes;
9529: int NDIM=2;
9530: int vpopbased=0;
1.235 brouard 9531: int nres=0;
1.136 brouard 9532:
1.164 brouard 9533: char ca[32], cb[32];
1.136 brouard 9534: /* FILE *fichtm; *//* Html File */
9535: /* FILE *ficgp;*/ /*Gnuplot File */
9536: struct stat info;
1.191 brouard 9537: double agedeb=0.;
1.194 brouard 9538:
9539: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9540: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9541:
1.165 brouard 9542: double fret;
1.191 brouard 9543: double dum=0.; /* Dummy variable */
1.136 brouard 9544: double ***p3mat;
1.218 brouard 9545: /* double ***mobaverage; */
1.164 brouard 9546:
9547: char line[MAXLINE];
1.197 brouard 9548: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9549:
1.234 brouard 9550: char modeltemp[MAXLINE];
1.230 brouard 9551: char resultline[MAXLINE];
9552:
1.136 brouard 9553: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9554: char *tok, *val; /* pathtot */
1.136 brouard 9555: int firstobs=1, lastobs=10;
1.195 brouard 9556: int c, h , cpt, c2;
1.191 brouard 9557: int jl=0;
9558: int i1, j1, jk, stepsize=0;
1.194 brouard 9559: int count=0;
9560:
1.164 brouard 9561: int *tab;
1.136 brouard 9562: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9563: int backcast=0;
1.136 brouard 9564: int mobilav=0,popforecast=0;
1.191 brouard 9565: int hstepm=0, nhstepm=0;
1.136 brouard 9566: int agemortsup;
9567: float sumlpop=0.;
9568: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9569: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9570:
1.191 brouard 9571: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9572: double ftolpl=FTOL;
9573: double **prlim;
1.217 brouard 9574: double **bprlim;
1.136 brouard 9575: double ***param; /* Matrix of parameters */
9576: double *p;
9577: double **matcov; /* Matrix of covariance */
1.203 brouard 9578: double **hess; /* Hessian matrix */
1.136 brouard 9579: double ***delti3; /* Scale */
9580: double *delti; /* Scale */
9581: double ***eij, ***vareij;
9582: double **varpl; /* Variances of prevalence limits by age */
9583: double *epj, vepp;
1.164 brouard 9584:
1.136 brouard 9585: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9586: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9587:
1.136 brouard 9588: double **ximort;
1.145 brouard 9589: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9590: int *dcwave;
9591:
1.164 brouard 9592: char z[1]="c";
1.136 brouard 9593:
9594: /*char *strt;*/
9595: char strtend[80];
1.126 brouard 9596:
1.164 brouard 9597:
1.126 brouard 9598: /* setlocale (LC_ALL, ""); */
9599: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9600: /* textdomain (PACKAGE); */
9601: /* setlocale (LC_CTYPE, ""); */
9602: /* setlocale (LC_MESSAGES, ""); */
9603:
9604: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9605: rstart_time = time(NULL);
9606: /* (void) gettimeofday(&start_time,&tzp);*/
9607: start_time = *localtime(&rstart_time);
1.126 brouard 9608: curr_time=start_time;
1.157 brouard 9609: /*tml = *localtime(&start_time.tm_sec);*/
9610: /* strcpy(strstart,asctime(&tml)); */
9611: strcpy(strstart,asctime(&start_time));
1.126 brouard 9612:
9613: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9614: /* tp.tm_sec = tp.tm_sec +86400; */
9615: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9616: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9617: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9618: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9619: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9620: /* strt=asctime(&tmg); */
9621: /* printf("Time(after) =%s",strstart); */
9622: /* (void) time (&time_value);
9623: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9624: * tm = *localtime(&time_value);
9625: * strstart=asctime(&tm);
9626: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9627: */
9628:
9629: nberr=0; /* Number of errors and warnings */
9630: nbwarn=0;
1.184 brouard 9631: #ifdef WIN32
9632: _getcwd(pathcd, size);
9633: #else
1.126 brouard 9634: getcwd(pathcd, size);
1.184 brouard 9635: #endif
1.191 brouard 9636: syscompilerinfo(0);
1.196 brouard 9637: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9638: if(argc <=1){
9639: printf("\nEnter the parameter file name: ");
1.205 brouard 9640: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9641: printf("ERROR Empty parameter file name\n");
9642: goto end;
9643: }
1.126 brouard 9644: i=strlen(pathr);
9645: if(pathr[i-1]=='\n')
9646: pathr[i-1]='\0';
1.156 brouard 9647: i=strlen(pathr);
1.205 brouard 9648: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9649: pathr[i-1]='\0';
1.205 brouard 9650: }
9651: i=strlen(pathr);
9652: if( i==0 ){
9653: printf("ERROR Empty parameter file name\n");
9654: goto end;
9655: }
9656: for (tok = pathr; tok != NULL; ){
1.126 brouard 9657: printf("Pathr |%s|\n",pathr);
9658: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9659: printf("val= |%s| pathr=%s\n",val,pathr);
9660: strcpy (pathtot, val);
9661: if(pathr[0] == '\0') break; /* Dirty */
9662: }
9663: }
9664: else{
9665: strcpy(pathtot,argv[1]);
9666: }
9667: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9668: /*cygwin_split_path(pathtot,path,optionfile);
9669: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9670: /* cutv(path,optionfile,pathtot,'\\');*/
9671:
9672: /* Split argv[0], imach program to get pathimach */
9673: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9674: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9675: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9676: /* strcpy(pathimach,argv[0]); */
9677: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9678: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9679: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9680: #ifdef WIN32
9681: _chdir(path); /* Can be a relative path */
9682: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9683: #else
1.126 brouard 9684: chdir(path); /* Can be a relative path */
1.184 brouard 9685: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9686: #endif
9687: printf("Current directory %s!\n",pathcd);
1.126 brouard 9688: strcpy(command,"mkdir ");
9689: strcat(command,optionfilefiname);
9690: if((outcmd=system(command)) != 0){
1.169 brouard 9691: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9692: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9693: /* fclose(ficlog); */
9694: /* exit(1); */
9695: }
9696: /* if((imk=mkdir(optionfilefiname))<0){ */
9697: /* perror("mkdir"); */
9698: /* } */
9699:
9700: /*-------- arguments in the command line --------*/
9701:
1.186 brouard 9702: /* Main Log file */
1.126 brouard 9703: strcat(filelog, optionfilefiname);
9704: strcat(filelog,".log"); /* */
9705: if((ficlog=fopen(filelog,"w"))==NULL) {
9706: printf("Problem with logfile %s\n",filelog);
9707: goto end;
9708: }
9709: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9710: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9711: fprintf(ficlog,"\nEnter the parameter file name: \n");
9712: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9713: path=%s \n\
9714: optionfile=%s\n\
9715: optionfilext=%s\n\
1.156 brouard 9716: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9717:
1.197 brouard 9718: syscompilerinfo(1);
1.167 brouard 9719:
1.126 brouard 9720: printf("Local time (at start):%s",strstart);
9721: fprintf(ficlog,"Local time (at start): %s",strstart);
9722: fflush(ficlog);
9723: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9724: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9725:
9726: /* */
9727: strcpy(fileres,"r");
9728: strcat(fileres, optionfilefiname);
1.201 brouard 9729: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9730: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9731: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9732:
1.186 brouard 9733: /* Main ---------arguments file --------*/
1.126 brouard 9734:
9735: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9736: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9737: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9738: fflush(ficlog);
1.149 brouard 9739: /* goto end; */
9740: exit(70);
1.126 brouard 9741: }
9742:
9743:
9744:
9745: strcpy(filereso,"o");
1.201 brouard 9746: strcat(filereso,fileresu);
1.126 brouard 9747: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9748: printf("Problem with Output resultfile: %s\n", filereso);
9749: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9750: fflush(ficlog);
9751: goto end;
9752: }
9753:
9754: /* Reads comments: lines beginning with '#' */
9755: numlinepar=0;
1.197 brouard 9756:
9757: /* First parameter line */
9758: while(fgets(line, MAXLINE, ficpar)) {
9759: /* If line starts with a # it is a comment */
9760: if (line[0] == '#') {
9761: numlinepar++;
9762: fputs(line,stdout);
9763: fputs(line,ficparo);
9764: fputs(line,ficlog);
9765: continue;
9766: }else
9767: break;
9768: }
9769: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9770: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9771: if (num_filled != 5) {
9772: printf("Should be 5 parameters\n");
9773: }
1.126 brouard 9774: numlinepar++;
1.197 brouard 9775: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9776: }
9777: /* Second parameter line */
9778: while(fgets(line, MAXLINE, ficpar)) {
9779: /* If line starts with a # it is a comment */
9780: if (line[0] == '#') {
9781: numlinepar++;
9782: fputs(line,stdout);
9783: fputs(line,ficparo);
9784: fputs(line,ficlog);
9785: continue;
9786: }else
9787: break;
9788: }
1.223 brouard 9789: 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", \
9790: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9791: if (num_filled != 11) {
9792: 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 9793: printf("but line=%s\n",line);
1.197 brouard 9794: }
1.223 brouard 9795: 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 9796: }
1.203 brouard 9797: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9798: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9799: /* Third parameter line */
9800: while(fgets(line, MAXLINE, ficpar)) {
9801: /* If line starts with a # it is a comment */
9802: if (line[0] == '#') {
9803: numlinepar++;
9804: fputs(line,stdout);
9805: fputs(line,ficparo);
9806: fputs(line,ficlog);
9807: continue;
9808: }else
9809: break;
9810: }
1.201 brouard 9811: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9812: if (num_filled == 0)
9813: model[0]='\0';
9814: else if (num_filled != 1){
1.197 brouard 9815: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9816: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9817: model[0]='\0';
9818: goto end;
9819: }
9820: else{
9821: if (model[0]=='+'){
9822: for(i=1; i<=strlen(model);i++)
9823: modeltemp[i-1]=model[i];
1.201 brouard 9824: strcpy(model,modeltemp);
1.197 brouard 9825: }
9826: }
1.199 brouard 9827: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9828: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9829: }
9830: /* 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); */
9831: /* numlinepar=numlinepar+3; /\* In general *\/ */
9832: /* 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 9833: 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);
9834: 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 9835: fflush(ficlog);
1.190 brouard 9836: /* if(model[0]=='#'|| model[0]== '\0'){ */
9837: if(model[0]=='#'){
1.187 brouard 9838: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9839: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9840: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9841: if(mle != -1){
9842: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9843: exit(1);
9844: }
9845: }
1.126 brouard 9846: while((c=getc(ficpar))=='#' && c!= EOF){
9847: ungetc(c,ficpar);
9848: fgets(line, MAXLINE, ficpar);
9849: numlinepar++;
1.195 brouard 9850: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9851: z[0]=line[1];
9852: }
9853: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9854: fputs(line, stdout);
9855: //puts(line);
1.126 brouard 9856: fputs(line,ficparo);
9857: fputs(line,ficlog);
9858: }
9859: ungetc(c,ficpar);
9860:
9861:
1.145 brouard 9862: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9863: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9864: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9865: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9866: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9867: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9868: v1+v2*age+v2*v3 makes cptcovn = 3
9869: */
9870: if (strlen(model)>1)
1.187 brouard 9871: 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 9872: else
1.187 brouard 9873: ncovmodel=2; /* Constant and age */
1.133 brouard 9874: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9875: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9876: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9877: 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);
9878: 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);
9879: fflush(stdout);
9880: fclose (ficlog);
9881: goto end;
9882: }
1.126 brouard 9883: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9884: delti=delti3[1][1];
9885: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9886: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 ! brouard 9887: /* We could also provide initial parameters values giving by simple logistic regression
! 9888: * only one way, that is without matrix product. We will have nlstate maximizations */
! 9889: /* for(i=1;i<nlstate;i++){ */
! 9890: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
! 9891: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
! 9892: /* } */
1.126 brouard 9893: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9894: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9895: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9896: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9897: fclose (ficparo);
9898: fclose (ficlog);
9899: goto end;
9900: exit(0);
1.220 brouard 9901: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9902: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9903: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9904: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9905: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9906: matcov=matrix(1,npar,1,npar);
1.203 brouard 9907: hess=matrix(1,npar,1,npar);
1.220 brouard 9908: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9909: /* Read guessed parameters */
1.126 brouard 9910: /* Reads comments: lines beginning with '#' */
9911: while((c=getc(ficpar))=='#' && c!= EOF){
9912: ungetc(c,ficpar);
9913: fgets(line, MAXLINE, ficpar);
9914: numlinepar++;
1.141 brouard 9915: fputs(line,stdout);
1.126 brouard 9916: fputs(line,ficparo);
9917: fputs(line,ficlog);
9918: }
9919: ungetc(c,ficpar);
9920:
9921: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9922: for(i=1; i <=nlstate; i++){
1.234 brouard 9923: j=0;
1.126 brouard 9924: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9925: if(jj==i) continue;
9926: j++;
9927: fscanf(ficpar,"%1d%1d",&i1,&j1);
9928: if ((i1 != i) || (j1 != jj)){
9929: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9930: It might be a problem of design; if ncovcol and the model are correct\n \
9931: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9932: exit(1);
9933: }
9934: fprintf(ficparo,"%1d%1d",i1,j1);
9935: if(mle==1)
9936: printf("%1d%1d",i,jj);
9937: fprintf(ficlog,"%1d%1d",i,jj);
9938: for(k=1; k<=ncovmodel;k++){
9939: fscanf(ficpar," %lf",¶m[i][j][k]);
9940: if(mle==1){
9941: printf(" %lf",param[i][j][k]);
9942: fprintf(ficlog," %lf",param[i][j][k]);
9943: }
9944: else
9945: fprintf(ficlog," %lf",param[i][j][k]);
9946: fprintf(ficparo," %lf",param[i][j][k]);
9947: }
9948: fscanf(ficpar,"\n");
9949: numlinepar++;
9950: if(mle==1)
9951: printf("\n");
9952: fprintf(ficlog,"\n");
9953: fprintf(ficparo,"\n");
1.126 brouard 9954: }
9955: }
9956: fflush(ficlog);
1.234 brouard 9957:
1.145 brouard 9958: /* Reads scales values */
1.126 brouard 9959: p=param[1][1];
9960:
9961: /* Reads comments: lines beginning with '#' */
9962: while((c=getc(ficpar))=='#' && c!= EOF){
9963: ungetc(c,ficpar);
9964: fgets(line, MAXLINE, ficpar);
9965: numlinepar++;
1.141 brouard 9966: fputs(line,stdout);
1.126 brouard 9967: fputs(line,ficparo);
9968: fputs(line,ficlog);
9969: }
9970: ungetc(c,ficpar);
9971:
9972: for(i=1; i <=nlstate; i++){
9973: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 9974: fscanf(ficpar,"%1d%1d",&i1,&j1);
9975: if ( (i1-i) * (j1-j) != 0){
9976: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
9977: exit(1);
9978: }
9979: printf("%1d%1d",i,j);
9980: fprintf(ficparo,"%1d%1d",i1,j1);
9981: fprintf(ficlog,"%1d%1d",i1,j1);
9982: for(k=1; k<=ncovmodel;k++){
9983: fscanf(ficpar,"%le",&delti3[i][j][k]);
9984: printf(" %le",delti3[i][j][k]);
9985: fprintf(ficparo," %le",delti3[i][j][k]);
9986: fprintf(ficlog," %le",delti3[i][j][k]);
9987: }
9988: fscanf(ficpar,"\n");
9989: numlinepar++;
9990: printf("\n");
9991: fprintf(ficparo,"\n");
9992: fprintf(ficlog,"\n");
1.126 brouard 9993: }
9994: }
9995: fflush(ficlog);
1.234 brouard 9996:
1.145 brouard 9997: /* Reads covariance matrix */
1.126 brouard 9998: delti=delti3[1][1];
1.220 brouard 9999:
10000:
1.126 brouard 10001: /* 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 10002:
1.126 brouard 10003: /* Reads comments: lines beginning with '#' */
10004: while((c=getc(ficpar))=='#' && c!= EOF){
10005: ungetc(c,ficpar);
10006: fgets(line, MAXLINE, ficpar);
10007: numlinepar++;
1.141 brouard 10008: fputs(line,stdout);
1.126 brouard 10009: fputs(line,ficparo);
10010: fputs(line,ficlog);
10011: }
10012: ungetc(c,ficpar);
1.220 brouard 10013:
1.126 brouard 10014: matcov=matrix(1,npar,1,npar);
1.203 brouard 10015: hess=matrix(1,npar,1,npar);
1.131 brouard 10016: for(i=1; i <=npar; i++)
10017: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10018:
1.194 brouard 10019: /* Scans npar lines */
1.126 brouard 10020: for(i=1; i <=npar; i++){
1.226 brouard 10021: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10022: if(count != 3){
1.226 brouard 10023: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10024: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10025: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10026: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10027: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10028: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10029: exit(1);
1.220 brouard 10030: }else{
1.226 brouard 10031: if(mle==1)
10032: printf("%1d%1d%d",i1,j1,jk);
10033: }
10034: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10035: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10036: for(j=1; j <=i; j++){
1.226 brouard 10037: fscanf(ficpar," %le",&matcov[i][j]);
10038: if(mle==1){
10039: printf(" %.5le",matcov[i][j]);
10040: }
10041: fprintf(ficlog," %.5le",matcov[i][j]);
10042: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10043: }
10044: fscanf(ficpar,"\n");
10045: numlinepar++;
10046: if(mle==1)
1.220 brouard 10047: printf("\n");
1.126 brouard 10048: fprintf(ficlog,"\n");
10049: fprintf(ficparo,"\n");
10050: }
1.194 brouard 10051: /* End of read covariance matrix npar lines */
1.126 brouard 10052: for(i=1; i <=npar; i++)
10053: for(j=i+1;j<=npar;j++)
1.226 brouard 10054: matcov[i][j]=matcov[j][i];
1.126 brouard 10055:
10056: if(mle==1)
10057: printf("\n");
10058: fprintf(ficlog,"\n");
10059:
10060: fflush(ficlog);
10061:
10062: /*-------- Rewriting parameter file ----------*/
10063: strcpy(rfileres,"r"); /* "Rparameterfile */
10064: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10065: strcat(rfileres,"."); /* */
10066: strcat(rfileres,optionfilext); /* Other files have txt extension */
10067: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10068: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10069: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10070: }
10071: fprintf(ficres,"#%s\n",version);
10072: } /* End of mle != -3 */
1.218 brouard 10073:
1.186 brouard 10074: /* Main data
10075: */
1.126 brouard 10076: n= lastobs;
10077: num=lvector(1,n);
10078: moisnais=vector(1,n);
10079: annais=vector(1,n);
10080: moisdc=vector(1,n);
10081: andc=vector(1,n);
1.220 brouard 10082: weight=vector(1,n);
1.126 brouard 10083: agedc=vector(1,n);
10084: cod=ivector(1,n);
1.220 brouard 10085: for(i=1;i<=n;i++){
1.234 brouard 10086: num[i]=0;
10087: moisnais[i]=0;
10088: annais[i]=0;
10089: moisdc[i]=0;
10090: andc[i]=0;
10091: agedc[i]=0;
10092: cod[i]=0;
10093: weight[i]=1.0; /* Equal weights, 1 by default */
10094: }
1.126 brouard 10095: mint=matrix(1,maxwav,1,n);
10096: anint=matrix(1,maxwav,1,n);
1.131 brouard 10097: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10098: tab=ivector(1,NCOVMAX);
1.144 brouard 10099: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10100: 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 10101:
1.136 brouard 10102: /* Reads data from file datafile */
10103: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10104: goto end;
10105:
10106: /* Calculation of the number of parameters from char model */
1.234 brouard 10107: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10108: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10109: k=3 V4 Tvar[k=3]= 4 (from V4)
10110: k=2 V1 Tvar[k=2]= 1 (from V1)
10111: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10112: */
10113:
10114: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10115: TvarsDind=ivector(1,NCOVMAX); /* */
10116: TvarsD=ivector(1,NCOVMAX); /* */
10117: TvarsQind=ivector(1,NCOVMAX); /* */
10118: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10119: TvarF=ivector(1,NCOVMAX); /* */
10120: TvarFind=ivector(1,NCOVMAX); /* */
10121: TvarV=ivector(1,NCOVMAX); /* */
10122: TvarVind=ivector(1,NCOVMAX); /* */
10123: TvarA=ivector(1,NCOVMAX); /* */
10124: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10125: TvarFD=ivector(1,NCOVMAX); /* */
10126: TvarFDind=ivector(1,NCOVMAX); /* */
10127: TvarFQ=ivector(1,NCOVMAX); /* */
10128: TvarFQind=ivector(1,NCOVMAX); /* */
10129: TvarVD=ivector(1,NCOVMAX); /* */
10130: TvarVDind=ivector(1,NCOVMAX); /* */
10131: TvarVQ=ivector(1,NCOVMAX); /* */
10132: TvarVQind=ivector(1,NCOVMAX); /* */
10133:
1.230 brouard 10134: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10135: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10136: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10137: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10138: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10139: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10140: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10141: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10142: */
10143: /* For model-covariate k tells which data-covariate to use but
10144: because this model-covariate is a construction we invent a new column
10145: ncovcol + k1
10146: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10147: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10148: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10149: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10150: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10151: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10152: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10153: */
1.145 brouard 10154: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10155: 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 10156: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10157: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10158: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10159: 4 covariates (3 plus signs)
10160: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10161: */
1.230 brouard 10162: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10163: * individual dummy, fixed or varying:
10164: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10165: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10166: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10167: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10168: * Tmodelind[1]@9={9,0,3,2,}*/
10169: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10170: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10171: * individual quantitative, fixed or varying:
10172: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10173: * 3, 1, 0, 0, 0, 0, 0, 0},
10174: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10175: /* Main decodemodel */
10176:
1.187 brouard 10177:
1.223 brouard 10178: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10179: goto end;
10180:
1.137 brouard 10181: if((double)(lastobs-imx)/(double)imx > 1.10){
10182: nbwarn++;
10183: 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);
10184: 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);
10185: }
1.136 brouard 10186: /* if(mle==1){*/
1.137 brouard 10187: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10188: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10189: }
10190:
10191: /*-calculation of age at interview from date of interview and age at death -*/
10192: agev=matrix(1,maxwav,1,imx);
10193:
10194: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10195: goto end;
10196:
1.126 brouard 10197:
1.136 brouard 10198: agegomp=(int)agemin;
10199: free_vector(moisnais,1,n);
10200: free_vector(annais,1,n);
1.126 brouard 10201: /* free_matrix(mint,1,maxwav,1,n);
10202: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10203: /* free_vector(moisdc,1,n); */
10204: /* free_vector(andc,1,n); */
1.145 brouard 10205: /* */
10206:
1.126 brouard 10207: wav=ivector(1,imx);
1.214 brouard 10208: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10209: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10210: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10211: 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.*/
10212: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10213: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10214:
10215: /* Concatenates waves */
1.214 brouard 10216: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10217: Death is a valid wave (if date is known).
10218: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10219: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10220: and mw[mi+1][i]. dh depends on stepm.
10221: */
10222:
1.126 brouard 10223: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.145 brouard 10224: /* */
10225:
1.215 brouard 10226: free_vector(moisdc,1,n);
10227: free_vector(andc,1,n);
10228:
1.126 brouard 10229: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10230: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10231: ncodemax[1]=1;
1.145 brouard 10232: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10233: cptcoveff=0;
1.220 brouard 10234: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10235: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10236: }
10237:
10238: ncovcombmax=pow(2,cptcoveff);
10239: invalidvarcomb=ivector(1, ncovcombmax);
10240: for(i=1;i<ncovcombmax;i++)
10241: invalidvarcomb[i]=0;
10242:
1.211 brouard 10243: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10244: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10245: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10246:
1.200 brouard 10247: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10248: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10249: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10250: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10251: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10252: * (currently 0 or 1) in the data.
10253: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10254: * corresponding modality (h,j).
10255: */
10256:
1.145 brouard 10257: h=0;
10258: /*if (cptcovn > 0) */
1.126 brouard 10259: m=pow(2,cptcoveff);
10260:
1.144 brouard 10261: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10262: * For k=4 covariates, h goes from 1 to m=2**k
10263: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10264: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10265: * h\k 1 2 3 4
1.143 brouard 10266: *______________________________
10267: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10268: * 2 2 1 1 1
10269: * 3 i=2 1 2 1 1
10270: * 4 2 2 1 1
10271: * 5 i=3 1 i=2 1 2 1
10272: * 6 2 1 2 1
10273: * 7 i=4 1 2 2 1
10274: * 8 2 2 2 1
1.197 brouard 10275: * 9 i=5 1 i=3 1 i=2 1 2
10276: * 10 2 1 1 2
10277: * 11 i=6 1 2 1 2
10278: * 12 2 2 1 2
10279: * 13 i=7 1 i=4 1 2 2
10280: * 14 2 1 2 2
10281: * 15 i=8 1 2 2 2
10282: * 16 2 2 2 2
1.143 brouard 10283: */
1.212 brouard 10284: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10285: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10286: * and the value of each covariate?
10287: * V1=1, V2=1, V3=2, V4=1 ?
10288: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10289: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10290: * In order to get the real value in the data, we use nbcode
10291: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10292: * We are keeping this crazy system in order to be able (in the future?)
10293: * to have more than 2 values (0 or 1) for a covariate.
10294: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10295: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10296: * bbbbbbbb
10297: * 76543210
10298: * h-1 00000101 (6-1=5)
1.219 brouard 10299: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10300: * &
10301: * 1 00000001 (1)
1.219 brouard 10302: * 00000000 = 1 & ((h-1) >> (k-1))
10303: * +1= 00000001 =1
1.211 brouard 10304: *
10305: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10306: * h' 1101 =2^3+2^2+0x2^1+2^0
10307: * >>k' 11
10308: * & 00000001
10309: * = 00000001
10310: * +1 = 00000010=2 = codtabm(14,3)
10311: * Reverse h=6 and m=16?
10312: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10313: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10314: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10315: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10316: * V3=decodtabm(14,3,2**4)=2
10317: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10318: *(h-1) >> (j-1) 0011 =13 >> 2
10319: * &1 000000001
10320: * = 000000001
10321: * +1= 000000010 =2
10322: * 2211
10323: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10324: * V3=2
1.220 brouard 10325: * codtabm and decodtabm are identical
1.211 brouard 10326: */
10327:
1.145 brouard 10328:
10329: free_ivector(Ndum,-1,NCOVMAX);
10330:
10331:
1.126 brouard 10332:
1.186 brouard 10333: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10334: strcpy(optionfilegnuplot,optionfilefiname);
10335: if(mle==-3)
1.201 brouard 10336: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10337: strcat(optionfilegnuplot,".gp");
10338:
10339: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10340: printf("Problem with file %s",optionfilegnuplot);
10341: }
10342: else{
1.204 brouard 10343: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10344: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10345: //fprintf(ficgp,"set missing 'NaNq'\n");
10346: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10347: }
10348: /* fclose(ficgp);*/
1.186 brouard 10349:
10350:
10351: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10352:
10353: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10354: if(mle==-3)
1.201 brouard 10355: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10356: strcat(optionfilehtm,".htm");
10357: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10358: printf("Problem with %s \n",optionfilehtm);
10359: exit(0);
1.126 brouard 10360: }
10361:
10362: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10363: strcat(optionfilehtmcov,"-cov.htm");
10364: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10365: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10366: }
10367: else{
10368: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10369: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10370: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10371: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10372: }
10373:
1.213 brouard 10374: 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 10375: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10376: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10377: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10378: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10379: \n\
10380: <hr size=\"2\" color=\"#EC5E5E\">\
10381: <ul><li><h4>Parameter files</h4>\n\
10382: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10383: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10384: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10385: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10386: - Date and time at start: %s</ul>\n",\
10387: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10388: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10389: fileres,fileres,\
10390: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10391: fflush(fichtm);
10392:
10393: strcpy(pathr,path);
10394: strcat(pathr,optionfilefiname);
1.184 brouard 10395: #ifdef WIN32
10396: _chdir(optionfilefiname); /* Move to directory named optionfile */
10397: #else
1.126 brouard 10398: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10399: #endif
10400:
1.126 brouard 10401:
1.220 brouard 10402: /* Calculates basic frequencies. Computes observed prevalence at single age
10403: and for any valid combination of covariates
1.126 brouard 10404: and prints on file fileres'p'. */
1.227 brouard 10405: freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
10406: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10407:
10408: fprintf(fichtm,"\n");
10409: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10410: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10411: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10412: imx,agemin,agemax,jmin,jmax,jmean);
10413: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10414: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10415: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10416: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10417: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10418:
1.126 brouard 10419: /* For Powell, parameters are in a vector p[] starting at p[1]
10420: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10421: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10422:
10423: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10424: /* For mortality only */
1.126 brouard 10425: if (mle==-3){
1.136 brouard 10426: ximort=matrix(1,NDIM,1,NDIM);
1.220 brouard 10427: for(i=1;i<=NDIM;i++)
10428: for(j=1;j<=NDIM;j++)
10429: ximort[i][j]=0.;
1.186 brouard 10430: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10431: cens=ivector(1,n);
10432: ageexmed=vector(1,n);
10433: agecens=vector(1,n);
10434: dcwave=ivector(1,n);
1.223 brouard 10435:
1.126 brouard 10436: for (i=1; i<=imx; i++){
10437: dcwave[i]=-1;
10438: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10439: if (s[m][i]>nlstate) {
10440: dcwave[i]=m;
10441: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10442: break;
10443: }
1.126 brouard 10444: }
1.226 brouard 10445:
1.126 brouard 10446: for (i=1; i<=imx; i++) {
10447: if (wav[i]>0){
1.226 brouard 10448: ageexmed[i]=agev[mw[1][i]][i];
10449: j=wav[i];
10450: agecens[i]=1.;
10451:
10452: if (ageexmed[i]> 1 && wav[i] > 0){
10453: agecens[i]=agev[mw[j][i]][i];
10454: cens[i]= 1;
10455: }else if (ageexmed[i]< 1)
10456: cens[i]= -1;
10457: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10458: cens[i]=0 ;
1.126 brouard 10459: }
10460: else cens[i]=-1;
10461: }
10462:
10463: for (i=1;i<=NDIM;i++) {
10464: for (j=1;j<=NDIM;j++)
1.226 brouard 10465: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10466: }
10467:
1.145 brouard 10468: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10469: /*printf("%lf %lf", p[1], p[2]);*/
10470:
10471:
1.136 brouard 10472: #ifdef GSL
10473: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10474: #else
1.126 brouard 10475: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10476: #endif
1.201 brouard 10477: strcpy(filerespow,"POW-MORT_");
10478: strcat(filerespow,fileresu);
1.126 brouard 10479: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10480: printf("Problem with resultfile: %s\n", filerespow);
10481: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10482: }
1.136 brouard 10483: #ifdef GSL
10484: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10485: #else
1.126 brouard 10486: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10487: #endif
1.126 brouard 10488: /* for (i=1;i<=nlstate;i++)
10489: for(j=1;j<=nlstate+ndeath;j++)
10490: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10491: */
10492: fprintf(ficrespow,"\n");
1.136 brouard 10493: #ifdef GSL
10494: /* gsl starts here */
10495: T = gsl_multimin_fminimizer_nmsimplex;
10496: gsl_multimin_fminimizer *sfm = NULL;
10497: gsl_vector *ss, *x;
10498: gsl_multimin_function minex_func;
10499:
10500: /* Initial vertex size vector */
10501: ss = gsl_vector_alloc (NDIM);
10502:
10503: if (ss == NULL){
10504: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10505: }
10506: /* Set all step sizes to 1 */
10507: gsl_vector_set_all (ss, 0.001);
10508:
10509: /* Starting point */
1.126 brouard 10510:
1.136 brouard 10511: x = gsl_vector_alloc (NDIM);
10512:
10513: if (x == NULL){
10514: gsl_vector_free(ss);
10515: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10516: }
10517:
10518: /* Initialize method and iterate */
10519: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10520: /* gsl_vector_set(x, 0, 0.0268); */
10521: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10522: gsl_vector_set(x, 0, p[1]);
10523: gsl_vector_set(x, 1, p[2]);
10524:
10525: minex_func.f = &gompertz_f;
10526: minex_func.n = NDIM;
10527: minex_func.params = (void *)&p; /* ??? */
10528:
10529: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10530: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10531:
10532: printf("Iterations beginning .....\n\n");
10533: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10534:
10535: iteri=0;
10536: while (rval == GSL_CONTINUE){
10537: iteri++;
10538: status = gsl_multimin_fminimizer_iterate(sfm);
10539:
10540: if (status) printf("error: %s\n", gsl_strerror (status));
10541: fflush(0);
10542:
10543: if (status)
10544: break;
10545:
10546: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10547: ssval = gsl_multimin_fminimizer_size (sfm);
10548:
10549: if (rval == GSL_SUCCESS)
10550: printf ("converged to a local maximum at\n");
10551:
10552: printf("%5d ", iteri);
10553: for (it = 0; it < NDIM; it++){
10554: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10555: }
10556: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10557: }
10558:
10559: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10560:
10561: gsl_vector_free(x); /* initial values */
10562: gsl_vector_free(ss); /* inital step size */
10563: for (it=0; it<NDIM; it++){
10564: p[it+1]=gsl_vector_get(sfm->x,it);
10565: fprintf(ficrespow," %.12lf", p[it]);
10566: }
10567: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10568: #endif
10569: #ifdef POWELL
10570: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10571: #endif
1.126 brouard 10572: fclose(ficrespow);
10573:
1.203 brouard 10574: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10575:
10576: for(i=1; i <=NDIM; i++)
10577: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10578: matcov[i][j]=matcov[j][i];
1.126 brouard 10579:
10580: printf("\nCovariance matrix\n ");
1.203 brouard 10581: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10582: for(i=1; i <=NDIM; i++) {
10583: for(j=1;j<=NDIM;j++){
1.220 brouard 10584: printf("%f ",matcov[i][j]);
10585: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10586: }
1.203 brouard 10587: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10588: }
10589:
10590: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10591: for (i=1;i<=NDIM;i++) {
1.126 brouard 10592: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10593: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10594: }
1.126 brouard 10595: lsurv=vector(1,AGESUP);
10596: lpop=vector(1,AGESUP);
10597: tpop=vector(1,AGESUP);
10598: lsurv[agegomp]=100000;
10599:
10600: for (k=agegomp;k<=AGESUP;k++) {
10601: agemortsup=k;
10602: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10603: }
10604:
10605: for (k=agegomp;k<agemortsup;k++)
10606: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10607:
10608: for (k=agegomp;k<agemortsup;k++){
10609: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10610: sumlpop=sumlpop+lpop[k];
10611: }
10612:
10613: tpop[agegomp]=sumlpop;
10614: for (k=agegomp;k<(agemortsup-3);k++){
10615: /* tpop[k+1]=2;*/
10616: tpop[k+1]=tpop[k]-lpop[k];
10617: }
10618:
10619:
10620: printf("\nAge lx qx dx Lx Tx e(x)\n");
10621: for (k=agegomp;k<(agemortsup-2);k++)
10622: 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]);
10623:
10624:
10625: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10626: ageminpar=50;
10627: agemaxpar=100;
1.194 brouard 10628: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10629: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10630: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10631: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10632: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10633: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10634: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10635: }else{
10636: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10637: 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 10638: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10639: }
1.201 brouard 10640: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10641: stepm, weightopt,\
10642: model,imx,p,matcov,agemortsup);
10643:
10644: free_vector(lsurv,1,AGESUP);
10645: free_vector(lpop,1,AGESUP);
10646: free_vector(tpop,1,AGESUP);
1.220 brouard 10647: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10648: free_ivector(cens,1,n);
10649: free_vector(agecens,1,n);
10650: free_ivector(dcwave,1,n);
1.220 brouard 10651: #ifdef GSL
1.136 brouard 10652: #endif
1.186 brouard 10653: } /* Endof if mle==-3 mortality only */
1.205 brouard 10654: /* Standard */
10655: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10656: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10657: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10658: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10659: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10660: for (k=1; k<=npar;k++)
10661: printf(" %d %8.5f",k,p[k]);
10662: printf("\n");
1.205 brouard 10663: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10664: /* mlikeli uses func not funcone */
1.247 ! brouard 10665: /* for(i=1;i<nlstate;i++){ */
! 10666: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
! 10667: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
! 10668: /* } */
1.205 brouard 10669: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10670: }
10671: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10672: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10673: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10674: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10675: }
10676: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10677: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10678: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10679: for (k=1; k<=npar;k++)
10680: printf(" %d %8.5f",k,p[k]);
10681: printf("\n");
10682:
10683: /*--------- results files --------------*/
1.224 brouard 10684: 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 10685:
10686:
10687: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10688: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10689: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10690: for(i=1,jk=1; i <=nlstate; i++){
10691: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10692: if (k != i) {
10693: printf("%d%d ",i,k);
10694: fprintf(ficlog,"%d%d ",i,k);
10695: fprintf(ficres,"%1d%1d ",i,k);
10696: for(j=1; j <=ncovmodel; j++){
10697: printf("%12.7f ",p[jk]);
10698: fprintf(ficlog,"%12.7f ",p[jk]);
10699: fprintf(ficres,"%12.7f ",p[jk]);
10700: jk++;
10701: }
10702: printf("\n");
10703: fprintf(ficlog,"\n");
10704: fprintf(ficres,"\n");
10705: }
1.126 brouard 10706: }
10707: }
1.203 brouard 10708: if(mle != 0){
10709: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10710: ftolhess=ftol; /* Usually correct */
1.203 brouard 10711: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10712: 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");
10713: 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");
10714: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10715: for(k=1; k <=(nlstate+ndeath); k++){
10716: if (k != i) {
10717: printf("%d%d ",i,k);
10718: fprintf(ficlog,"%d%d ",i,k);
10719: for(j=1; j <=ncovmodel; j++){
10720: 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]));
10721: 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]));
10722: jk++;
10723: }
10724: printf("\n");
10725: fprintf(ficlog,"\n");
10726: }
10727: }
1.193 brouard 10728: }
1.203 brouard 10729: } /* end of hesscov and Wald tests */
1.225 brouard 10730:
1.203 brouard 10731: /* */
1.126 brouard 10732: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10733: printf("# Scales (for hessian or gradient estimation)\n");
10734: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10735: for(i=1,jk=1; i <=nlstate; i++){
10736: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10737: if (j!=i) {
10738: fprintf(ficres,"%1d%1d",i,j);
10739: printf("%1d%1d",i,j);
10740: fprintf(ficlog,"%1d%1d",i,j);
10741: for(k=1; k<=ncovmodel;k++){
10742: printf(" %.5e",delti[jk]);
10743: fprintf(ficlog," %.5e",delti[jk]);
10744: fprintf(ficres," %.5e",delti[jk]);
10745: jk++;
10746: }
10747: printf("\n");
10748: fprintf(ficlog,"\n");
10749: fprintf(ficres,"\n");
10750: }
1.126 brouard 10751: }
10752: }
10753:
10754: 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 10755: if(mle >= 1) /* To big for the screen */
1.126 brouard 10756: 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");
10757: 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");
10758: /* # 121 Var(a12)\n\ */
10759: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10760: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10761: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10762: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10763: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10764: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10765: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10766:
10767:
10768: /* Just to have a covariance matrix which will be more understandable
10769: even is we still don't want to manage dictionary of variables
10770: */
10771: for(itimes=1;itimes<=2;itimes++){
10772: jj=0;
10773: for(i=1; i <=nlstate; i++){
1.225 brouard 10774: for(j=1; j <=nlstate+ndeath; j++){
10775: if(j==i) continue;
10776: for(k=1; k<=ncovmodel;k++){
10777: jj++;
10778: ca[0]= k+'a'-1;ca[1]='\0';
10779: if(itimes==1){
10780: if(mle>=1)
10781: printf("#%1d%1d%d",i,j,k);
10782: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10783: fprintf(ficres,"#%1d%1d%d",i,j,k);
10784: }else{
10785: if(mle>=1)
10786: printf("%1d%1d%d",i,j,k);
10787: fprintf(ficlog,"%1d%1d%d",i,j,k);
10788: fprintf(ficres,"%1d%1d%d",i,j,k);
10789: }
10790: ll=0;
10791: for(li=1;li <=nlstate; li++){
10792: for(lj=1;lj <=nlstate+ndeath; lj++){
10793: if(lj==li) continue;
10794: for(lk=1;lk<=ncovmodel;lk++){
10795: ll++;
10796: if(ll<=jj){
10797: cb[0]= lk +'a'-1;cb[1]='\0';
10798: if(ll<jj){
10799: if(itimes==1){
10800: if(mle>=1)
10801: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10802: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10803: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10804: }else{
10805: if(mle>=1)
10806: printf(" %.5e",matcov[jj][ll]);
10807: fprintf(ficlog," %.5e",matcov[jj][ll]);
10808: fprintf(ficres," %.5e",matcov[jj][ll]);
10809: }
10810: }else{
10811: if(itimes==1){
10812: if(mle>=1)
10813: printf(" Var(%s%1d%1d)",ca,i,j);
10814: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10815: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10816: }else{
10817: if(mle>=1)
10818: printf(" %.7e",matcov[jj][ll]);
10819: fprintf(ficlog," %.7e",matcov[jj][ll]);
10820: fprintf(ficres," %.7e",matcov[jj][ll]);
10821: }
10822: }
10823: }
10824: } /* end lk */
10825: } /* end lj */
10826: } /* end li */
10827: if(mle>=1)
10828: printf("\n");
10829: fprintf(ficlog,"\n");
10830: fprintf(ficres,"\n");
10831: numlinepar++;
10832: } /* end k*/
10833: } /*end j */
1.126 brouard 10834: } /* end i */
10835: } /* end itimes */
10836:
10837: fflush(ficlog);
10838: fflush(ficres);
1.225 brouard 10839: while(fgets(line, MAXLINE, ficpar)) {
10840: /* If line starts with a # it is a comment */
10841: if (line[0] == '#') {
10842: numlinepar++;
10843: fputs(line,stdout);
10844: fputs(line,ficparo);
10845: fputs(line,ficlog);
10846: continue;
10847: }else
10848: break;
10849: }
10850:
1.209 brouard 10851: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10852: /* ungetc(c,ficpar); */
10853: /* fgets(line, MAXLINE, ficpar); */
10854: /* fputs(line,stdout); */
10855: /* fputs(line,ficparo); */
10856: /* } */
10857: /* ungetc(c,ficpar); */
1.126 brouard 10858:
10859: estepm=0;
1.209 brouard 10860: 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 10861:
10862: if (num_filled != 6) {
10863: 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);
10864: 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);
10865: goto end;
10866: }
10867: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10868: }
10869: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10870: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10871:
1.209 brouard 10872: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10873: if (estepm==0 || estepm < stepm) estepm=stepm;
10874: if (fage <= 2) {
10875: bage = ageminpar;
10876: fage = agemaxpar;
10877: }
10878:
10879: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10880: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10881: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10882:
1.186 brouard 10883: /* Other stuffs, more or less useful */
1.126 brouard 10884: while((c=getc(ficpar))=='#' && c!= EOF){
10885: ungetc(c,ficpar);
10886: fgets(line, MAXLINE, ficpar);
1.141 brouard 10887: fputs(line,stdout);
1.126 brouard 10888: fputs(line,ficparo);
10889: }
10890: ungetc(c,ficpar);
10891:
10892: 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);
10893: 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);
10894: 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);
10895: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10896: 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);
10897:
10898: while((c=getc(ficpar))=='#' && c!= EOF){
10899: ungetc(c,ficpar);
10900: fgets(line, MAXLINE, ficpar);
1.141 brouard 10901: fputs(line,stdout);
1.126 brouard 10902: fputs(line,ficparo);
10903: }
10904: ungetc(c,ficpar);
10905:
10906:
10907: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10908: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10909:
10910: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10911: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10912: fprintf(ficparo,"pop_based=%d\n",popbased);
10913: fprintf(ficres,"pop_based=%d\n",popbased);
10914:
10915: while((c=getc(ficpar))=='#' && c!= EOF){
10916: ungetc(c,ficpar);
10917: fgets(line, MAXLINE, ficpar);
1.141 brouard 10918: fputs(line,stdout);
1.238 brouard 10919: fputs(line,ficres);
1.126 brouard 10920: fputs(line,ficparo);
10921: }
10922: ungetc(c,ficpar);
10923:
10924: 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);
10925: 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);
10926: 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);
10927: 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);
10928: 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);
10929: /* day and month of proj2 are not used but only year anproj2.*/
10930:
1.217 brouard 10931: while((c=getc(ficpar))=='#' && c!= EOF){
10932: ungetc(c,ficpar);
10933: fgets(line, MAXLINE, ficpar);
10934: fputs(line,stdout);
10935: fputs(line,ficparo);
1.238 brouard 10936: fputs(line,ficres);
1.217 brouard 10937: }
10938: ungetc(c,ficpar);
10939:
10940: fscanf(ficpar,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj);
1.223 brouard 10941: 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);
10942: 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);
10943: 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 10944: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 10945:
1.230 brouard 10946: /* Results */
1.235 brouard 10947: nresult=0;
1.230 brouard 10948: while(fgets(line, MAXLINE, ficpar)) {
10949: /* If line starts with a # it is a comment */
10950: if (line[0] == '#') {
10951: numlinepar++;
10952: fputs(line,stdout);
10953: fputs(line,ficparo);
10954: fputs(line,ficlog);
1.238 brouard 10955: fputs(line,ficres);
1.230 brouard 10956: continue;
10957: }else
10958: break;
10959: }
1.240 brouard 10960: if (!feof(ficpar))
1.230 brouard 10961: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 10962: if (num_filled == 0){
1.230 brouard 10963: resultline[0]='\0';
1.240 brouard 10964: break;
10965: } else if (num_filled != 1){
1.230 brouard 10966: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
10967: }
1.235 brouard 10968: nresult++; /* Sum of resultlines */
10969: printf("Result %d: result=%s\n",nresult, resultline);
10970: if(nresult > MAXRESULTLINES){
10971: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10972: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
10973: goto end;
10974: }
10975: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 10976: fprintf(ficparo,"result: %s\n",resultline);
10977: fprintf(ficres,"result: %s\n",resultline);
10978: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 10979: while(fgets(line, MAXLINE, ficpar)) {
10980: /* If line starts with a # it is a comment */
10981: if (line[0] == '#') {
10982: numlinepar++;
10983: fputs(line,stdout);
10984: fputs(line,ficparo);
1.238 brouard 10985: fputs(line,ficres);
1.230 brouard 10986: fputs(line,ficlog);
10987: continue;
10988: }else
10989: break;
10990: }
10991: if (feof(ficpar))
10992: break;
10993: else{ /* Processess output results for this combination of covariate values */
10994: }
1.240 brouard 10995: } /* end while */
1.230 brouard 10996:
10997:
1.126 brouard 10998:
1.230 brouard 10999: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11000: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11001:
11002: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11003: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11004: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11005: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11006: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11007: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11008: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11009: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11010: }else{
1.218 brouard 11011: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11012: }
11013: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11014: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11015: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11016:
1.225 brouard 11017: /*------------ free_vector -------------*/
11018: /* chdir(path); */
1.220 brouard 11019:
1.215 brouard 11020: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11021: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11022: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11023: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11024: free_lvector(num,1,n);
11025: free_vector(agedc,1,n);
11026: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11027: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11028: fclose(ficparo);
11029: fclose(ficres);
1.220 brouard 11030:
11031:
1.186 brouard 11032: /* Other results (useful)*/
1.220 brouard 11033:
11034:
1.126 brouard 11035: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11036: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11037: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11038: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11039: fclose(ficrespl);
11040:
11041: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11042: /*#include "hpijx.h"*/
11043: hPijx(p, bage, fage);
1.145 brouard 11044: fclose(ficrespij);
1.227 brouard 11045:
1.220 brouard 11046: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11047: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11048: k=1;
1.126 brouard 11049: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11050:
1.219 brouard 11051: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11052: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11053: for(i=1;i<=AGESUP;i++)
1.219 brouard 11054: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11055: for(k=1;k<=ncovcombmax;k++)
11056: probs[i][j][k]=0.;
1.219 brouard 11057: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11058: if (mobilav!=0 ||mobilavproj !=0 ) {
11059: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11060: for(i=1;i<=AGESUP;i++)
11061: for(j=1;j<=nlstate;j++)
11062: for(k=1;k<=ncovcombmax;k++)
11063: mobaverages[i][j][k]=0.;
1.219 brouard 11064: mobaverage=mobaverages;
11065: if (mobilav!=0) {
1.235 brouard 11066: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11067: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11068: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11069: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11070: }
1.219 brouard 11071: }
11072: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11073: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11074: else if (mobilavproj !=0) {
1.235 brouard 11075: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11076: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11077: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11078: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11079: }
1.219 brouard 11080: }
11081: }/* end if moving average */
1.227 brouard 11082:
1.126 brouard 11083: /*---------- Forecasting ------------------*/
11084: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11085: if(prevfcast==1){
11086: /* if(stepm ==1){*/
1.225 brouard 11087: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11088: }
1.217 brouard 11089: if(backcast==1){
1.219 brouard 11090: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11091: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11092: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11093:
11094: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11095:
11096: bprlim=matrix(1,nlstate,1,nlstate);
11097: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11098: fclose(ficresplb);
11099:
1.222 brouard 11100: hBijx(p, bage, fage, mobaverage);
11101: fclose(ficrespijb);
1.219 brouard 11102: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11103:
11104: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11105: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11106: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11107: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11108: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11109: }
1.217 brouard 11110:
1.186 brouard 11111:
11112: /* ------ Other prevalence ratios------------ */
1.126 brouard 11113:
1.215 brouard 11114: free_ivector(wav,1,imx);
11115: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11116: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11117: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11118:
11119:
1.127 brouard 11120: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11121:
1.201 brouard 11122: strcpy(filerese,"E_");
11123: strcat(filerese,fileresu);
1.126 brouard 11124: if((ficreseij=fopen(filerese,"w"))==NULL) {
11125: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11126: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11127: }
1.208 brouard 11128: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11129: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11130:
11131: pstamp(ficreseij);
1.219 brouard 11132:
1.235 brouard 11133: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11134: if (cptcovn < 1){i1=1;}
11135:
11136: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11137: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11138: if(TKresult[nres]!= k)
11139: continue;
1.219 brouard 11140: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11141: printf("\n#****** ");
1.225 brouard 11142: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11143: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11144: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11145: }
11146: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11147: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11148: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11149: }
11150: fprintf(ficreseij,"******\n");
1.235 brouard 11151: printf("******\n");
1.219 brouard 11152:
11153: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11154: oldm=oldms;savm=savms;
1.235 brouard 11155: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11156:
1.219 brouard 11157: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11158: }
11159: fclose(ficreseij);
1.208 brouard 11160: printf("done evsij\n");fflush(stdout);
11161: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11162:
1.227 brouard 11163: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11164:
11165:
1.201 brouard 11166: strcpy(filerest,"T_");
11167: strcat(filerest,fileresu);
1.127 brouard 11168: if((ficrest=fopen(filerest,"w"))==NULL) {
11169: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11170: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11171: }
1.208 brouard 11172: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11173: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11174:
1.126 brouard 11175:
1.201 brouard 11176: strcpy(fileresstde,"STDE_");
11177: strcat(fileresstde,fileresu);
1.126 brouard 11178: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11179: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11180: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11181: }
1.227 brouard 11182: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11183: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11184:
1.201 brouard 11185: strcpy(filerescve,"CVE_");
11186: strcat(filerescve,fileresu);
1.126 brouard 11187: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11188: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11189: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11190: }
1.227 brouard 11191: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11192: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11193:
1.201 brouard 11194: strcpy(fileresv,"V_");
11195: strcat(fileresv,fileresu);
1.126 brouard 11196: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11197: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11198: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11199: }
1.227 brouard 11200: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11201: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11202:
1.145 brouard 11203: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11204: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11205:
1.235 brouard 11206: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11207: if (cptcovn < 1){i1=1;}
11208:
11209: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11210: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11211: if(TKresult[nres]!= k)
11212: continue;
1.242 brouard 11213: printf("\n#****** Result for:");
11214: fprintf(ficrest,"\n#****** Result for:");
11215: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11216: for(j=1;j<=cptcoveff;j++){
11217: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11218: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11219: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11220: }
1.235 brouard 11221: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11222: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11223: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11224: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11225: }
1.208 brouard 11226: fprintf(ficrest,"******\n");
1.227 brouard 11227: fprintf(ficlog,"******\n");
11228: printf("******\n");
1.208 brouard 11229:
11230: fprintf(ficresstdeij,"\n#****** ");
11231: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11232: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11233: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11234: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11235: }
1.235 brouard 11236: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11237: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11238: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11239: }
1.208 brouard 11240: fprintf(ficresstdeij,"******\n");
11241: fprintf(ficrescveij,"******\n");
11242:
11243: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11244: /* pstamp(ficresvij); */
1.225 brouard 11245: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11246: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11247: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11248: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11249: }
1.208 brouard 11250: fprintf(ficresvij,"******\n");
11251:
11252: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11253: oldm=oldms;savm=savms;
1.235 brouard 11254: printf(" cvevsij ");
11255: fprintf(ficlog, " cvevsij ");
11256: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11257: printf(" end cvevsij \n ");
11258: fprintf(ficlog, " end cvevsij \n ");
11259:
11260: /*
11261: */
11262: /* goto endfree; */
11263:
11264: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11265: pstamp(ficrest);
11266:
11267:
11268: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11269: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11270: cptcod= 0; /* To be deleted */
11271: printf("varevsij vpopbased=%d \n",vpopbased);
11272: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11273: 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 11274: 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 ");
11275: if(vpopbased==1)
11276: 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);
11277: else
11278: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11279: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11280: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11281: fprintf(ficrest,"\n");
11282: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11283: epj=vector(1,nlstate+1);
11284: printf("Computing age specific period (stable) prevalences in each health state \n");
11285: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11286: for(age=bage; age <=fage ;age++){
1.235 brouard 11287: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11288: if (vpopbased==1) {
11289: if(mobilav ==0){
11290: for(i=1; i<=nlstate;i++)
11291: prlim[i][i]=probs[(int)age][i][k];
11292: }else{ /* mobilav */
11293: for(i=1; i<=nlstate;i++)
11294: prlim[i][i]=mobaverage[(int)age][i][k];
11295: }
11296: }
1.219 brouard 11297:
1.227 brouard 11298: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11299: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11300: /* printf(" age %4.0f ",age); */
11301: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11302: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11303: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11304: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11305: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11306: }
11307: epj[nlstate+1] +=epj[j];
11308: }
11309: /* printf(" age %4.0f \n",age); */
1.219 brouard 11310:
1.227 brouard 11311: for(i=1, vepp=0.;i <=nlstate;i++)
11312: for(j=1;j <=nlstate;j++)
11313: vepp += vareij[i][j][(int)age];
11314: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11315: for(j=1;j <=nlstate;j++){
11316: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11317: }
11318: fprintf(ficrest,"\n");
11319: }
1.208 brouard 11320: } /* End vpopbased */
11321: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11322: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11323: free_vector(epj,1,nlstate+1);
1.235 brouard 11324: printf("done selection\n");fflush(stdout);
11325: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11326:
1.145 brouard 11327: /*}*/
1.235 brouard 11328: } /* End k selection */
1.227 brouard 11329:
11330: printf("done State-specific expectancies\n");fflush(stdout);
11331: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11332:
1.126 brouard 11333: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11334:
1.201 brouard 11335: strcpy(fileresvpl,"VPL_");
11336: strcat(fileresvpl,fileresu);
1.126 brouard 11337: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11338: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11339: exit(0);
11340: }
1.208 brouard 11341: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11342: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11343:
1.145 brouard 11344: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11345: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11346:
1.235 brouard 11347: i1=pow(2,cptcoveff);
11348: if (cptcovn < 1){i1=1;}
11349:
11350: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11351: for(k=1; k<=i1;k++){
11352: if(TKresult[nres]!= k)
11353: continue;
1.227 brouard 11354: fprintf(ficresvpl,"\n#****** ");
11355: printf("\n#****** ");
11356: fprintf(ficlog,"\n#****** ");
11357: for(j=1;j<=cptcoveff;j++) {
11358: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11359: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11360: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11361: }
1.235 brouard 11362: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11363: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11364: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11365: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11366: }
1.227 brouard 11367: fprintf(ficresvpl,"******\n");
11368: printf("******\n");
11369: fprintf(ficlog,"******\n");
11370:
11371: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11372: oldm=oldms;savm=savms;
1.235 brouard 11373: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11374: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11375: /*}*/
1.126 brouard 11376: }
1.227 brouard 11377:
1.126 brouard 11378: fclose(ficresvpl);
1.208 brouard 11379: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11380: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11381:
11382: free_vector(weight,1,n);
11383: free_imatrix(Tvard,1,NCOVMAX,1,2);
11384: free_imatrix(s,1,maxwav+1,1,n);
11385: free_matrix(anint,1,maxwav,1,n);
11386: free_matrix(mint,1,maxwav,1,n);
11387: free_ivector(cod,1,n);
11388: free_ivector(tab,1,NCOVMAX);
11389: fclose(ficresstdeij);
11390: fclose(ficrescveij);
11391: fclose(ficresvij);
11392: fclose(ficrest);
11393: fclose(ficpar);
11394:
11395:
1.126 brouard 11396: /*---------- End : free ----------------*/
1.219 brouard 11397: if (mobilav!=0 ||mobilavproj !=0)
11398: 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 11399: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11400: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11401: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11402: } /* mle==-3 arrives here for freeing */
1.227 brouard 11403: /* endfree:*/
11404: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11405: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11406: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11407: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11408: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11409: free_matrix(coqvar,1,maxwav,1,n);
11410: free_matrix(covar,0,NCOVMAX,1,n);
11411: free_matrix(matcov,1,npar,1,npar);
11412: free_matrix(hess,1,npar,1,npar);
11413: /*free_vector(delti,1,npar);*/
11414: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11415: free_matrix(agev,1,maxwav,1,imx);
11416: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11417:
11418: free_ivector(ncodemax,1,NCOVMAX);
11419: free_ivector(ncodemaxwundef,1,NCOVMAX);
11420: free_ivector(Dummy,-1,NCOVMAX);
11421: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11422: free_ivector(DummyV,1,NCOVMAX);
11423: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11424: free_ivector(Typevar,-1,NCOVMAX);
11425: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11426: free_ivector(TvarsQ,1,NCOVMAX);
11427: free_ivector(TvarsQind,1,NCOVMAX);
11428: free_ivector(TvarsD,1,NCOVMAX);
11429: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11430: free_ivector(TvarFD,1,NCOVMAX);
11431: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11432: free_ivector(TvarF,1,NCOVMAX);
11433: free_ivector(TvarFind,1,NCOVMAX);
11434: free_ivector(TvarV,1,NCOVMAX);
11435: free_ivector(TvarVind,1,NCOVMAX);
11436: free_ivector(TvarA,1,NCOVMAX);
11437: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11438: free_ivector(TvarFQ,1,NCOVMAX);
11439: free_ivector(TvarFQind,1,NCOVMAX);
11440: free_ivector(TvarVD,1,NCOVMAX);
11441: free_ivector(TvarVDind,1,NCOVMAX);
11442: free_ivector(TvarVQ,1,NCOVMAX);
11443: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11444: free_ivector(Tvarsel,1,NCOVMAX);
11445: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11446: free_ivector(Tposprod,1,NCOVMAX);
11447: free_ivector(Tprod,1,NCOVMAX);
11448: free_ivector(Tvaraff,1,NCOVMAX);
11449: free_ivector(invalidvarcomb,1,ncovcombmax);
11450: free_ivector(Tage,1,NCOVMAX);
11451: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11452: free_ivector(TmodelInvind,1,NCOVMAX);
11453: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11454:
11455: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11456: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11457: fflush(fichtm);
11458: fflush(ficgp);
11459:
1.227 brouard 11460:
1.126 brouard 11461: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11462: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11463: 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 11464: }else{
11465: printf("End of Imach\n");
11466: fprintf(ficlog,"End of Imach\n");
11467: }
11468: printf("See log file on %s\n",filelog);
11469: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11470: /*(void) gettimeofday(&end_time,&tzp);*/
11471: rend_time = time(NULL);
11472: end_time = *localtime(&rend_time);
11473: /* tml = *localtime(&end_time.tm_sec); */
11474: strcpy(strtend,asctime(&end_time));
1.126 brouard 11475: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11476: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11477: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11478:
1.157 brouard 11479: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11480: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11481: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11482: /* printf("Total time was %d uSec.\n", total_usecs);*/
11483: /* if(fileappend(fichtm,optionfilehtm)){ */
11484: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11485: fclose(fichtm);
11486: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11487: fclose(fichtmcov);
11488: fclose(ficgp);
11489: fclose(ficlog);
11490: /*------ End -----------*/
1.227 brouard 11491:
11492:
11493: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11494: #ifdef WIN32
1.227 brouard 11495: if (_chdir(pathcd) != 0)
11496: printf("Can't move to directory %s!\n",path);
11497: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11498: #else
1.227 brouard 11499: if(chdir(pathcd) != 0)
11500: printf("Can't move to directory %s!\n", path);
11501: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11502: #endif
1.126 brouard 11503: printf("Current directory %s!\n",pathcd);
11504: /*strcat(plotcmd,CHARSEPARATOR);*/
11505: sprintf(plotcmd,"gnuplot");
1.157 brouard 11506: #ifdef _WIN32
1.126 brouard 11507: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11508: #endif
11509: if(!stat(plotcmd,&info)){
1.158 brouard 11510: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11511: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11512: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11513: }else
11514: strcpy(pplotcmd,plotcmd);
1.157 brouard 11515: #ifdef __unix
1.126 brouard 11516: strcpy(plotcmd,GNUPLOTPROGRAM);
11517: if(!stat(plotcmd,&info)){
1.158 brouard 11518: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11519: }else
11520: strcpy(pplotcmd,plotcmd);
11521: #endif
11522: }else
11523: strcpy(pplotcmd,plotcmd);
11524:
11525: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11526: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11527:
1.126 brouard 11528: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11529: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11530: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11531: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11532: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11533: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11534: }
1.158 brouard 11535: printf(" Successful, please wait...");
1.126 brouard 11536: while (z[0] != 'q') {
11537: /* chdir(path); */
1.154 brouard 11538: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11539: scanf("%s",z);
11540: /* if (z[0] == 'c') system("./imach"); */
11541: if (z[0] == 'e') {
1.158 brouard 11542: #ifdef __APPLE__
1.152 brouard 11543: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11544: #elif __linux
11545: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11546: #else
1.152 brouard 11547: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11548: #endif
11549: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11550: system(pplotcmd);
1.126 brouard 11551: }
11552: else if (z[0] == 'g') system(plotcmd);
11553: else if (z[0] == 'q') exit(0);
11554: }
1.227 brouard 11555: end:
1.126 brouard 11556: while (z[0] != 'q') {
1.195 brouard 11557: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11558: scanf("%s",z);
11559: }
11560: }
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