Annotation of imach/src/imach.c, revision 1.260
1.260 ! brouard 1: /* $Id: imach.c,v 1.259 2017/04/04 13:01:16 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.260 ! brouard 4: Revision 1.259 2017/04/04 13:01:16 brouard
! 5: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
! 6:
1.259 brouard 7: Revision 1.258 2017/04/03 10:17:47 brouard
8: Summary: Version 0.99r12
9:
10: Some cleanings, conformed with updated documentation.
11:
1.258 brouard 12: Revision 1.257 2017/03/29 16:53:30 brouard
13: Summary: Temp
14:
1.257 brouard 15: Revision 1.256 2017/03/27 05:50:23 brouard
16: Summary: Temporary
17:
1.256 brouard 18: Revision 1.255 2017/03/08 16:02:28 brouard
19: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
20:
1.255 brouard 21: Revision 1.254 2017/03/08 07:13:00 brouard
22: Summary: Fixing data parameter line
23:
1.254 brouard 24: Revision 1.253 2016/12/15 11:59:41 brouard
25: Summary: 0.99 in progress
26:
1.253 brouard 27: Revision 1.252 2016/09/15 21:15:37 brouard
28: *** empty log message ***
29:
1.252 brouard 30: Revision 1.251 2016/09/15 15:01:13 brouard
31: Summary: not working
32:
1.251 brouard 33: Revision 1.250 2016/09/08 16:07:27 brouard
34: Summary: continue
35:
1.250 brouard 36: Revision 1.249 2016/09/07 17:14:18 brouard
37: Summary: Starting values from frequencies
38:
1.249 brouard 39: Revision 1.248 2016/09/07 14:10:18 brouard
40: *** empty log message ***
41:
1.248 brouard 42: Revision 1.247 2016/09/02 11:11:21 brouard
43: *** empty log message ***
44:
1.247 brouard 45: Revision 1.246 2016/09/02 08:49:22 brouard
46: *** empty log message ***
47:
1.246 brouard 48: Revision 1.245 2016/09/02 07:25:01 brouard
49: *** empty log message ***
50:
1.245 brouard 51: Revision 1.244 2016/09/02 07:17:34 brouard
52: *** empty log message ***
53:
1.244 brouard 54: Revision 1.243 2016/09/02 06:45:35 brouard
55: *** empty log message ***
56:
1.243 brouard 57: Revision 1.242 2016/08/30 15:01:20 brouard
58: Summary: Fixing a lots
59:
1.242 brouard 60: Revision 1.241 2016/08/29 17:17:25 brouard
61: Summary: gnuplot problem in Back projection to fix
62:
1.241 brouard 63: Revision 1.240 2016/08/29 07:53:18 brouard
64: Summary: Better
65:
1.240 brouard 66: Revision 1.239 2016/08/26 15:51:03 brouard
67: Summary: Improvement in Powell output in order to copy and paste
68:
69: Author:
70:
1.239 brouard 71: Revision 1.238 2016/08/26 14:23:35 brouard
72: Summary: Starting tests of 0.99
73:
1.238 brouard 74: Revision 1.237 2016/08/26 09:20:19 brouard
75: Summary: to valgrind
76:
1.237 brouard 77: Revision 1.236 2016/08/25 10:50:18 brouard
78: *** empty log message ***
79:
1.236 brouard 80: Revision 1.235 2016/08/25 06:59:23 brouard
81: *** empty log message ***
82:
1.235 brouard 83: Revision 1.234 2016/08/23 16:51:20 brouard
84: *** empty log message ***
85:
1.234 brouard 86: Revision 1.233 2016/08/23 07:40:50 brouard
87: Summary: not working
88:
1.233 brouard 89: Revision 1.232 2016/08/22 14:20:21 brouard
90: Summary: not working
91:
1.232 brouard 92: Revision 1.231 2016/08/22 07:17:15 brouard
93: Summary: not working
94:
1.231 brouard 95: Revision 1.230 2016/08/22 06:55:53 brouard
96: Summary: Not working
97:
1.230 brouard 98: Revision 1.229 2016/07/23 09:45:53 brouard
99: Summary: Completing for func too
100:
1.229 brouard 101: Revision 1.228 2016/07/22 17:45:30 brouard
102: Summary: Fixing some arrays, still debugging
103:
1.227 brouard 104: Revision 1.226 2016/07/12 18:42:34 brouard
105: Summary: temp
106:
1.226 brouard 107: Revision 1.225 2016/07/12 08:40:03 brouard
108: Summary: saving but not running
109:
1.225 brouard 110: Revision 1.224 2016/07/01 13:16:01 brouard
111: Summary: Fixes
112:
1.224 brouard 113: Revision 1.223 2016/02/19 09:23:35 brouard
114: Summary: temporary
115:
1.223 brouard 116: Revision 1.222 2016/02/17 08:14:50 brouard
117: Summary: Probably last 0.98 stable version 0.98r6
118:
1.222 brouard 119: Revision 1.221 2016/02/15 23:35:36 brouard
120: Summary: minor bug
121:
1.220 brouard 122: Revision 1.219 2016/02/15 00:48:12 brouard
123: *** empty log message ***
124:
1.219 brouard 125: Revision 1.218 2016/02/12 11:29:23 brouard
126: Summary: 0.99 Back projections
127:
1.218 brouard 128: Revision 1.217 2015/12/23 17:18:31 brouard
129: Summary: Experimental backcast
130:
1.217 brouard 131: Revision 1.216 2015/12/18 17:32:11 brouard
132: Summary: 0.98r4 Warning and status=-2
133:
134: Version 0.98r4 is now:
135: - displaying an error when status is -1, date of interview unknown and date of death known;
136: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
137: Older changes concerning s=-2, dating from 2005 have been supersed.
138:
1.216 brouard 139: Revision 1.215 2015/12/16 08:52:24 brouard
140: Summary: 0.98r4 working
141:
1.215 brouard 142: Revision 1.214 2015/12/16 06:57:54 brouard
143: Summary: temporary not working
144:
1.214 brouard 145: Revision 1.213 2015/12/11 18:22:17 brouard
146: Summary: 0.98r4
147:
1.213 brouard 148: Revision 1.212 2015/11/21 12:47:24 brouard
149: Summary: minor typo
150:
1.212 brouard 151: Revision 1.211 2015/11/21 12:41:11 brouard
152: Summary: 0.98r3 with some graph of projected cross-sectional
153:
154: Author: Nicolas Brouard
155:
1.211 brouard 156: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 157: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 158: Summary: Adding ftolpl parameter
159: Author: N Brouard
160:
161: We had difficulties to get smoothed confidence intervals. It was due
162: to the period prevalence which wasn't computed accurately. The inner
163: parameter ftolpl is now an outer parameter of the .imach parameter
164: file after estepm. If ftolpl is small 1.e-4 and estepm too,
165: computation are long.
166:
1.209 brouard 167: Revision 1.208 2015/11/17 14:31:57 brouard
168: Summary: temporary
169:
1.208 brouard 170: Revision 1.207 2015/10/27 17:36:57 brouard
171: *** empty log message ***
172:
1.207 brouard 173: Revision 1.206 2015/10/24 07:14:11 brouard
174: *** empty log message ***
175:
1.206 brouard 176: Revision 1.205 2015/10/23 15:50:53 brouard
177: Summary: 0.98r3 some clarification for graphs on likelihood contributions
178:
1.205 brouard 179: Revision 1.204 2015/10/01 16:20:26 brouard
180: Summary: Some new graphs of contribution to likelihood
181:
1.204 brouard 182: Revision 1.203 2015/09/30 17:45:14 brouard
183: Summary: looking at better estimation of the hessian
184:
185: Also a better criteria for convergence to the period prevalence And
186: therefore adding the number of years needed to converge. (The
187: prevalence in any alive state shold sum to one
188:
1.203 brouard 189: Revision 1.202 2015/09/22 19:45:16 brouard
190: Summary: Adding some overall graph on contribution to likelihood. Might change
191:
1.202 brouard 192: Revision 1.201 2015/09/15 17:34:58 brouard
193: Summary: 0.98r0
194:
195: - Some new graphs like suvival functions
196: - Some bugs fixed like model=1+age+V2.
197:
1.201 brouard 198: Revision 1.200 2015/09/09 16:53:55 brouard
199: Summary: Big bug thanks to Flavia
200:
201: Even model=1+age+V2. did not work anymore
202:
1.200 brouard 203: Revision 1.199 2015/09/07 14:09:23 brouard
204: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
205:
1.199 brouard 206: Revision 1.198 2015/09/03 07:14:39 brouard
207: Summary: 0.98q5 Flavia
208:
1.198 brouard 209: Revision 1.197 2015/09/01 18:24:39 brouard
210: *** empty log message ***
211:
1.197 brouard 212: Revision 1.196 2015/08/18 23:17:52 brouard
213: Summary: 0.98q5
214:
1.196 brouard 215: Revision 1.195 2015/08/18 16:28:39 brouard
216: Summary: Adding a hack for testing purpose
217:
218: After reading the title, ftol and model lines, if the comment line has
219: a q, starting with #q, the answer at the end of the run is quit. It
220: permits to run test files in batch with ctest. The former workaround was
221: $ echo q | imach foo.imach
222:
1.195 brouard 223: Revision 1.194 2015/08/18 13:32:00 brouard
224: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
225:
1.194 brouard 226: Revision 1.193 2015/08/04 07:17:42 brouard
227: Summary: 0.98q4
228:
1.193 brouard 229: Revision 1.192 2015/07/16 16:49:02 brouard
230: Summary: Fixing some outputs
231:
1.192 brouard 232: Revision 1.191 2015/07/14 10:00:33 brouard
233: Summary: Some fixes
234:
1.191 brouard 235: Revision 1.190 2015/05/05 08:51:13 brouard
236: Summary: Adding digits in output parameters (7 digits instead of 6)
237:
238: Fix 1+age+.
239:
1.190 brouard 240: Revision 1.189 2015/04/30 14:45:16 brouard
241: Summary: 0.98q2
242:
1.189 brouard 243: Revision 1.188 2015/04/30 08:27:53 brouard
244: *** empty log message ***
245:
1.188 brouard 246: Revision 1.187 2015/04/29 09:11:15 brouard
247: *** empty log message ***
248:
1.187 brouard 249: Revision 1.186 2015/04/23 12:01:52 brouard
250: Summary: V1*age is working now, version 0.98q1
251:
252: Some codes had been disabled in order to simplify and Vn*age was
253: working in the optimization phase, ie, giving correct MLE parameters,
254: but, as usual, outputs were not correct and program core dumped.
255:
1.186 brouard 256: Revision 1.185 2015/03/11 13:26:42 brouard
257: Summary: Inclusion of compile and links command line for Intel Compiler
258:
1.185 brouard 259: Revision 1.184 2015/03/11 11:52:39 brouard
260: Summary: Back from Windows 8. Intel Compiler
261:
1.184 brouard 262: Revision 1.183 2015/03/10 20:34:32 brouard
263: Summary: 0.98q0, trying with directest, mnbrak fixed
264:
265: We use directest instead of original Powell test; probably no
266: incidence on the results, but better justifications;
267: We fixed Numerical Recipes mnbrak routine which was wrong and gave
268: wrong results.
269:
1.183 brouard 270: Revision 1.182 2015/02/12 08:19:57 brouard
271: Summary: Trying to keep directest which seems simpler and more general
272: Author: Nicolas Brouard
273:
1.182 brouard 274: Revision 1.181 2015/02/11 23:22:24 brouard
275: Summary: Comments on Powell added
276:
277: Author:
278:
1.181 brouard 279: Revision 1.180 2015/02/11 17:33:45 brouard
280: Summary: Finishing move from main to function (hpijx and prevalence_limit)
281:
1.180 brouard 282: Revision 1.179 2015/01/04 09:57:06 brouard
283: Summary: back to OS/X
284:
1.179 brouard 285: Revision 1.178 2015/01/04 09:35:48 brouard
286: *** empty log message ***
287:
1.178 brouard 288: Revision 1.177 2015/01/03 18:40:56 brouard
289: Summary: Still testing ilc32 on OSX
290:
1.177 brouard 291: Revision 1.176 2015/01/03 16:45:04 brouard
292: *** empty log message ***
293:
1.176 brouard 294: Revision 1.175 2015/01/03 16:33:42 brouard
295: *** empty log message ***
296:
1.175 brouard 297: Revision 1.174 2015/01/03 16:15:49 brouard
298: Summary: Still in cross-compilation
299:
1.174 brouard 300: Revision 1.173 2015/01/03 12:06:26 brouard
301: Summary: trying to detect cross-compilation
302:
1.173 brouard 303: Revision 1.172 2014/12/27 12:07:47 brouard
304: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
305:
1.172 brouard 306: Revision 1.171 2014/12/23 13:26:59 brouard
307: Summary: Back from Visual C
308:
309: Still problem with utsname.h on Windows
310:
1.171 brouard 311: Revision 1.170 2014/12/23 11:17:12 brouard
312: Summary: Cleaning some \%% back to %%
313:
314: The escape was mandatory for a specific compiler (which one?), but too many warnings.
315:
1.170 brouard 316: Revision 1.169 2014/12/22 23:08:31 brouard
317: Summary: 0.98p
318:
319: Outputs some informations on compiler used, OS etc. Testing on different platforms.
320:
1.169 brouard 321: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 322: Summary: update
1.169 brouard 323:
1.168 brouard 324: Revision 1.167 2014/12/22 13:50:56 brouard
325: Summary: Testing uname and compiler version and if compiled 32 or 64
326:
327: Testing on Linux 64
328:
1.167 brouard 329: Revision 1.166 2014/12/22 11:40:47 brouard
330: *** empty log message ***
331:
1.166 brouard 332: Revision 1.165 2014/12/16 11:20:36 brouard
333: Summary: After compiling on Visual C
334:
335: * imach.c (Module): Merging 1.61 to 1.162
336:
1.165 brouard 337: Revision 1.164 2014/12/16 10:52:11 brouard
338: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
339:
340: * imach.c (Module): Merging 1.61 to 1.162
341:
1.164 brouard 342: Revision 1.163 2014/12/16 10:30:11 brouard
343: * imach.c (Module): Merging 1.61 to 1.162
344:
1.163 brouard 345: Revision 1.162 2014/09/25 11:43:39 brouard
346: Summary: temporary backup 0.99!
347:
1.162 brouard 348: Revision 1.1 2014/09/16 11:06:58 brouard
349: Summary: With some code (wrong) for nlopt
350:
351: Author:
352:
353: Revision 1.161 2014/09/15 20:41:41 brouard
354: Summary: Problem with macro SQR on Intel compiler
355:
1.161 brouard 356: Revision 1.160 2014/09/02 09:24:05 brouard
357: *** empty log message ***
358:
1.160 brouard 359: Revision 1.159 2014/09/01 10:34:10 brouard
360: Summary: WIN32
361: Author: Brouard
362:
1.159 brouard 363: Revision 1.158 2014/08/27 17:11:51 brouard
364: *** empty log message ***
365:
1.158 brouard 366: Revision 1.157 2014/08/27 16:26:55 brouard
367: Summary: Preparing windows Visual studio version
368: Author: Brouard
369:
370: In order to compile on Visual studio, time.h is now correct and time_t
371: and tm struct should be used. difftime should be used but sometimes I
372: just make the differences in raw time format (time(&now).
373: Trying to suppress #ifdef LINUX
374: Add xdg-open for __linux in order to open default browser.
375:
1.157 brouard 376: Revision 1.156 2014/08/25 20:10:10 brouard
377: *** empty log message ***
378:
1.156 brouard 379: Revision 1.155 2014/08/25 18:32:34 brouard
380: Summary: New compile, minor changes
381: Author: Brouard
382:
1.155 brouard 383: Revision 1.154 2014/06/20 17:32:08 brouard
384: Summary: Outputs now all graphs of convergence to period prevalence
385:
1.154 brouard 386: Revision 1.153 2014/06/20 16:45:46 brouard
387: Summary: If 3 live state, convergence to period prevalence on same graph
388: Author: Brouard
389:
1.153 brouard 390: Revision 1.152 2014/06/18 17:54:09 brouard
391: Summary: open browser, use gnuplot on same dir than imach if not found in the path
392:
1.152 brouard 393: Revision 1.151 2014/06/18 16:43:30 brouard
394: *** empty log message ***
395:
1.151 brouard 396: Revision 1.150 2014/06/18 16:42:35 brouard
397: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
398: Author: brouard
399:
1.150 brouard 400: Revision 1.149 2014/06/18 15:51:14 brouard
401: Summary: Some fixes in parameter files errors
402: Author: Nicolas Brouard
403:
1.149 brouard 404: Revision 1.148 2014/06/17 17:38:48 brouard
405: Summary: Nothing new
406: Author: Brouard
407:
408: Just a new packaging for OS/X version 0.98nS
409:
1.148 brouard 410: Revision 1.147 2014/06/16 10:33:11 brouard
411: *** empty log message ***
412:
1.147 brouard 413: Revision 1.146 2014/06/16 10:20:28 brouard
414: Summary: Merge
415: Author: Brouard
416:
417: Merge, before building revised version.
418:
1.146 brouard 419: Revision 1.145 2014/06/10 21:23:15 brouard
420: Summary: Debugging with valgrind
421: Author: Nicolas Brouard
422:
423: Lot of changes in order to output the results with some covariates
424: After the Edimburgh REVES conference 2014, it seems mandatory to
425: improve the code.
426: No more memory valgrind error but a lot has to be done in order to
427: continue the work of splitting the code into subroutines.
428: Also, decodemodel has been improved. Tricode is still not
429: optimal. nbcode should be improved. Documentation has been added in
430: the source code.
431:
1.144 brouard 432: Revision 1.143 2014/01/26 09:45:38 brouard
433: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
434:
435: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
436: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
437:
1.143 brouard 438: Revision 1.142 2014/01/26 03:57:36 brouard
439: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
440:
441: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
442:
1.142 brouard 443: Revision 1.141 2014/01/26 02:42:01 brouard
444: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
445:
1.141 brouard 446: Revision 1.140 2011/09/02 10:37:54 brouard
447: Summary: times.h is ok with mingw32 now.
448:
1.140 brouard 449: Revision 1.139 2010/06/14 07:50:17 brouard
450: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
451: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
452:
1.139 brouard 453: Revision 1.138 2010/04/30 18:19:40 brouard
454: *** empty log message ***
455:
1.138 brouard 456: Revision 1.137 2010/04/29 18:11:38 brouard
457: (Module): Checking covariates for more complex models
458: than V1+V2. A lot of change to be done. Unstable.
459:
1.137 brouard 460: Revision 1.136 2010/04/26 20:30:53 brouard
461: (Module): merging some libgsl code. Fixing computation
462: of likelione (using inter/intrapolation if mle = 0) in order to
463: get same likelihood as if mle=1.
464: Some cleaning of code and comments added.
465:
1.136 brouard 466: Revision 1.135 2009/10/29 15:33:14 brouard
467: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
468:
1.135 brouard 469: Revision 1.134 2009/10/29 13:18:53 brouard
470: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
471:
1.134 brouard 472: Revision 1.133 2009/07/06 10:21:25 brouard
473: just nforces
474:
1.133 brouard 475: Revision 1.132 2009/07/06 08:22:05 brouard
476: Many tings
477:
1.132 brouard 478: Revision 1.131 2009/06/20 16:22:47 brouard
479: Some dimensions resccaled
480:
1.131 brouard 481: Revision 1.130 2009/05/26 06:44:34 brouard
482: (Module): Max Covariate is now set to 20 instead of 8. A
483: lot of cleaning with variables initialized to 0. Trying to make
484: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
485:
1.130 brouard 486: Revision 1.129 2007/08/31 13:49:27 lievre
487: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
488:
1.129 lievre 489: Revision 1.128 2006/06/30 13:02:05 brouard
490: (Module): Clarifications on computing e.j
491:
1.128 brouard 492: Revision 1.127 2006/04/28 18:11:50 brouard
493: (Module): Yes the sum of survivors was wrong since
494: imach-114 because nhstepm was no more computed in the age
495: loop. Now we define nhstepma in the age loop.
496: (Module): In order to speed up (in case of numerous covariates) we
497: compute health expectancies (without variances) in a first step
498: and then all the health expectancies with variances or standard
499: deviation (needs data from the Hessian matrices) which slows the
500: computation.
501: In the future we should be able to stop the program is only health
502: expectancies and graph are needed without standard deviations.
503:
1.127 brouard 504: Revision 1.126 2006/04/28 17:23:28 brouard
505: (Module): Yes the sum of survivors was wrong since
506: imach-114 because nhstepm was no more computed in the age
507: loop. Now we define nhstepma in the age loop.
508: Version 0.98h
509:
1.126 brouard 510: Revision 1.125 2006/04/04 15:20:31 lievre
511: Errors in calculation of health expectancies. Age was not initialized.
512: Forecasting file added.
513:
514: Revision 1.124 2006/03/22 17:13:53 lievre
515: Parameters are printed with %lf instead of %f (more numbers after the comma).
516: The log-likelihood is printed in the log file
517:
518: Revision 1.123 2006/03/20 10:52:43 brouard
519: * imach.c (Module): <title> changed, corresponds to .htm file
520: name. <head> headers where missing.
521:
522: * imach.c (Module): Weights can have a decimal point as for
523: English (a comma might work with a correct LC_NUMERIC environment,
524: otherwise the weight is truncated).
525: Modification of warning when the covariates values are not 0 or
526: 1.
527: Version 0.98g
528:
529: Revision 1.122 2006/03/20 09:45:41 brouard
530: (Module): Weights can have a decimal point as for
531: English (a comma might work with a correct LC_NUMERIC environment,
532: otherwise the weight is truncated).
533: Modification of warning when the covariates values are not 0 or
534: 1.
535: Version 0.98g
536:
537: Revision 1.121 2006/03/16 17:45:01 lievre
538: * imach.c (Module): Comments concerning covariates added
539:
540: * imach.c (Module): refinements in the computation of lli if
541: status=-2 in order to have more reliable computation if stepm is
542: not 1 month. Version 0.98f
543:
544: Revision 1.120 2006/03/16 15:10:38 lievre
545: (Module): refinements in the computation of lli if
546: status=-2 in order to have more reliable computation if stepm is
547: not 1 month. Version 0.98f
548:
549: Revision 1.119 2006/03/15 17:42:26 brouard
550: (Module): Bug if status = -2, the loglikelihood was
551: computed as likelihood omitting the logarithm. Version O.98e
552:
553: Revision 1.118 2006/03/14 18:20:07 brouard
554: (Module): varevsij Comments added explaining the second
555: table of variances if popbased=1 .
556: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
557: (Module): Function pstamp added
558: (Module): Version 0.98d
559:
560: Revision 1.117 2006/03/14 17:16:22 brouard
561: (Module): varevsij Comments added explaining the second
562: table of variances if popbased=1 .
563: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
564: (Module): Function pstamp added
565: (Module): Version 0.98d
566:
567: Revision 1.116 2006/03/06 10:29:27 brouard
568: (Module): Variance-covariance wrong links and
569: varian-covariance of ej. is needed (Saito).
570:
571: Revision 1.115 2006/02/27 12:17:45 brouard
572: (Module): One freematrix added in mlikeli! 0.98c
573:
574: Revision 1.114 2006/02/26 12:57:58 brouard
575: (Module): Some improvements in processing parameter
576: filename with strsep.
577:
578: Revision 1.113 2006/02/24 14:20:24 brouard
579: (Module): Memory leaks checks with valgrind and:
580: datafile was not closed, some imatrix were not freed and on matrix
581: allocation too.
582:
583: Revision 1.112 2006/01/30 09:55:26 brouard
584: (Module): Back to gnuplot.exe instead of wgnuplot.exe
585:
586: Revision 1.111 2006/01/25 20:38:18 brouard
587: (Module): Lots of cleaning and bugs added (Gompertz)
588: (Module): Comments can be added in data file. Missing date values
589: can be a simple dot '.'.
590:
591: Revision 1.110 2006/01/25 00:51:50 brouard
592: (Module): Lots of cleaning and bugs added (Gompertz)
593:
594: Revision 1.109 2006/01/24 19:37:15 brouard
595: (Module): Comments (lines starting with a #) are allowed in data.
596:
597: Revision 1.108 2006/01/19 18:05:42 lievre
598: Gnuplot problem appeared...
599: To be fixed
600:
601: Revision 1.107 2006/01/19 16:20:37 brouard
602: Test existence of gnuplot in imach path
603:
604: Revision 1.106 2006/01/19 13:24:36 brouard
605: Some cleaning and links added in html output
606:
607: Revision 1.105 2006/01/05 20:23:19 lievre
608: *** empty log message ***
609:
610: Revision 1.104 2005/09/30 16:11:43 lievre
611: (Module): sump fixed, loop imx fixed, and simplifications.
612: (Module): If the status is missing at the last wave but we know
613: that the person is alive, then we can code his/her status as -2
614: (instead of missing=-1 in earlier versions) and his/her
615: contributions to the likelihood is 1 - Prob of dying from last
616: health status (= 1-p13= p11+p12 in the easiest case of somebody in
617: the healthy state at last known wave). Version is 0.98
618:
619: Revision 1.103 2005/09/30 15:54:49 lievre
620: (Module): sump fixed, loop imx fixed, and simplifications.
621:
622: Revision 1.102 2004/09/15 17:31:30 brouard
623: Add the possibility to read data file including tab characters.
624:
625: Revision 1.101 2004/09/15 10:38:38 brouard
626: Fix on curr_time
627:
628: Revision 1.100 2004/07/12 18:29:06 brouard
629: Add version for Mac OS X. Just define UNIX in Makefile
630:
631: Revision 1.99 2004/06/05 08:57:40 brouard
632: *** empty log message ***
633:
634: Revision 1.98 2004/05/16 15:05:56 brouard
635: New version 0.97 . First attempt to estimate force of mortality
636: directly from the data i.e. without the need of knowing the health
637: state at each age, but using a Gompertz model: log u =a + b*age .
638: This is the basic analysis of mortality and should be done before any
639: other analysis, in order to test if the mortality estimated from the
640: cross-longitudinal survey is different from the mortality estimated
641: from other sources like vital statistic data.
642:
643: The same imach parameter file can be used but the option for mle should be -3.
644:
1.133 brouard 645: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 646: former routines in order to include the new code within the former code.
647:
648: The output is very simple: only an estimate of the intercept and of
649: the slope with 95% confident intervals.
650:
651: Current limitations:
652: A) Even if you enter covariates, i.e. with the
653: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
654: B) There is no computation of Life Expectancy nor Life Table.
655:
656: Revision 1.97 2004/02/20 13:25:42 lievre
657: Version 0.96d. Population forecasting command line is (temporarily)
658: suppressed.
659:
660: Revision 1.96 2003/07/15 15:38:55 brouard
661: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
662: rewritten within the same printf. Workaround: many printfs.
663:
664: Revision 1.95 2003/07/08 07:54:34 brouard
665: * imach.c (Repository):
666: (Repository): Using imachwizard code to output a more meaningful covariance
667: matrix (cov(a12,c31) instead of numbers.
668:
669: Revision 1.94 2003/06/27 13:00:02 brouard
670: Just cleaning
671:
672: Revision 1.93 2003/06/25 16:33:55 brouard
673: (Module): On windows (cygwin) function asctime_r doesn't
674: exist so I changed back to asctime which exists.
675: (Module): Version 0.96b
676:
677: Revision 1.92 2003/06/25 16:30:45 brouard
678: (Module): On windows (cygwin) function asctime_r doesn't
679: exist so I changed back to asctime which exists.
680:
681: Revision 1.91 2003/06/25 15:30:29 brouard
682: * imach.c (Repository): Duplicated warning errors corrected.
683: (Repository): Elapsed time after each iteration is now output. It
684: helps to forecast when convergence will be reached. Elapsed time
685: is stamped in powell. We created a new html file for the graphs
686: concerning matrix of covariance. It has extension -cov.htm.
687:
688: Revision 1.90 2003/06/24 12:34:15 brouard
689: (Module): Some bugs corrected for windows. Also, when
690: mle=-1 a template is output in file "or"mypar.txt with the design
691: of the covariance matrix to be input.
692:
693: Revision 1.89 2003/06/24 12:30:52 brouard
694: (Module): Some bugs corrected for windows. Also, when
695: mle=-1 a template is output in file "or"mypar.txt with the design
696: of the covariance matrix to be input.
697:
698: Revision 1.88 2003/06/23 17:54:56 brouard
699: * 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.
700:
701: Revision 1.87 2003/06/18 12:26:01 brouard
702: Version 0.96
703:
704: Revision 1.86 2003/06/17 20:04:08 brouard
705: (Module): Change position of html and gnuplot routines and added
706: routine fileappend.
707:
708: Revision 1.85 2003/06/17 13:12:43 brouard
709: * imach.c (Repository): Check when date of death was earlier that
710: current date of interview. It may happen when the death was just
711: prior to the death. In this case, dh was negative and likelihood
712: was wrong (infinity). We still send an "Error" but patch by
713: assuming that the date of death was just one stepm after the
714: interview.
715: (Repository): Because some people have very long ID (first column)
716: we changed int to long in num[] and we added a new lvector for
717: memory allocation. But we also truncated to 8 characters (left
718: truncation)
719: (Repository): No more line truncation errors.
720:
721: Revision 1.84 2003/06/13 21:44:43 brouard
722: * imach.c (Repository): Replace "freqsummary" at a correct
723: place. It differs from routine "prevalence" which may be called
724: many times. Probs is memory consuming and must be used with
725: parcimony.
726: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
727:
728: Revision 1.83 2003/06/10 13:39:11 lievre
729: *** empty log message ***
730:
731: Revision 1.82 2003/06/05 15:57:20 brouard
732: Add log in imach.c and fullversion number is now printed.
733:
734: */
735: /*
736: Interpolated Markov Chain
737:
738: Short summary of the programme:
739:
1.227 brouard 740: This program computes Healthy Life Expectancies or State-specific
741: (if states aren't health statuses) Expectancies from
742: cross-longitudinal data. Cross-longitudinal data consist in:
743:
744: -1- a first survey ("cross") where individuals from different ages
745: are interviewed on their health status or degree of disability (in
746: the case of a health survey which is our main interest)
747:
748: -2- at least a second wave of interviews ("longitudinal") which
749: measure each change (if any) in individual health status. Health
750: expectancies are computed from the time spent in each health state
751: according to a model. More health states you consider, more time is
752: necessary to reach the Maximum Likelihood of the parameters involved
753: in the model. The simplest model is the multinomial logistic model
754: where pij is the probability to be observed in state j at the second
755: wave conditional to be observed in state i at the first
756: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
757: etc , where 'age' is age and 'sex' is a covariate. If you want to
758: have a more complex model than "constant and age", you should modify
759: the program where the markup *Covariates have to be included here
760: again* invites you to do it. More covariates you add, slower the
1.126 brouard 761: convergence.
762:
763: The advantage of this computer programme, compared to a simple
764: multinomial logistic model, is clear when the delay between waves is not
765: identical for each individual. Also, if a individual missed an
766: intermediate interview, the information is lost, but taken into
767: account using an interpolation or extrapolation.
768:
769: hPijx is the probability to be observed in state i at age x+h
770: conditional to the observed state i at age x. The delay 'h' can be
771: split into an exact number (nh*stepm) of unobserved intermediate
772: states. This elementary transition (by month, quarter,
773: semester or year) is modelled as a multinomial logistic. The hPx
774: matrix is simply the matrix product of nh*stepm elementary matrices
775: and the contribution of each individual to the likelihood is simply
776: hPijx.
777:
778: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 779: of the life expectancies. It also computes the period (stable) prevalence.
780:
781: Back prevalence and projections:
1.227 brouard 782:
783: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
784: double agemaxpar, double ftolpl, int *ncvyearp, double
785: dateprev1,double dateprev2, int firstpass, int lastpass, int
786: mobilavproj)
787:
788: Computes the back prevalence limit for any combination of
789: covariate values k at any age between ageminpar and agemaxpar and
790: returns it in **bprlim. In the loops,
791:
792: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
793: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
794:
795: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 796: Computes for any combination of covariates k and any age between bage and fage
797: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
798: oldm=oldms;savm=savms;
1.227 brouard 799:
800: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 801: Computes the transition matrix starting at age 'age' over
802: 'nhstepm*hstepm*stepm' months (i.e. until
803: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 804: nhstepm*hstepm matrices.
805:
806: Returns p3mat[i][j][h] after calling
807: p3mat[i][j][h]=matprod2(newm,
808: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
809: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
810: oldm);
1.226 brouard 811:
812: Important routines
813:
814: - func (or funcone), computes logit (pij) distinguishing
815: o fixed variables (single or product dummies or quantitative);
816: o varying variables by:
817: (1) wave (single, product dummies, quantitative),
818: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
819: % fixed dummy (treated) or quantitative (not done because time-consuming);
820: % varying dummy (not done) or quantitative (not done);
821: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
822: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
823: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
824: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
825: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 826:
1.226 brouard 827:
828:
1.133 brouard 829: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
830: Institut national d'études démographiques, Paris.
1.126 brouard 831: This software have been partly granted by Euro-REVES, a concerted action
832: from the European Union.
833: It is copyrighted identically to a GNU software product, ie programme and
834: software can be distributed freely for non commercial use. Latest version
835: can be accessed at http://euroreves.ined.fr/imach .
836:
837: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
838: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
839:
840: **********************************************************************/
841: /*
842: main
843: read parameterfile
844: read datafile
845: concatwav
846: freqsummary
847: if (mle >= 1)
848: mlikeli
849: print results files
850: if mle==1
851: computes hessian
852: read end of parameter file: agemin, agemax, bage, fage, estepm
853: begin-prev-date,...
854: open gnuplot file
855: open html file
1.145 brouard 856: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
857: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
858: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
859: freexexit2 possible for memory heap.
860:
861: h Pij x | pij_nom ficrestpij
862: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
863: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
864: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
865:
866: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
867: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
868: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
869: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
870: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
871:
1.126 brouard 872: forecasting if prevfcast==1 prevforecast call prevalence()
873: health expectancies
874: Variance-covariance of DFLE
875: prevalence()
876: movingaverage()
877: varevsij()
878: if popbased==1 varevsij(,popbased)
879: total life expectancies
880: Variance of period (stable) prevalence
881: end
882: */
883:
1.187 brouard 884: /* #define DEBUG */
885: /* #define DEBUGBRENT */
1.203 brouard 886: /* #define DEBUGLINMIN */
887: /* #define DEBUGHESS */
888: #define DEBUGHESSIJ
1.224 brouard 889: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 890: #define POWELL /* Instead of NLOPT */
1.224 brouard 891: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 892: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
893: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 894:
895: #include <math.h>
896: #include <stdio.h>
897: #include <stdlib.h>
898: #include <string.h>
1.226 brouard 899: #include <ctype.h>
1.159 brouard 900:
901: #ifdef _WIN32
902: #include <io.h>
1.172 brouard 903: #include <windows.h>
904: #include <tchar.h>
1.159 brouard 905: #else
1.126 brouard 906: #include <unistd.h>
1.159 brouard 907: #endif
1.126 brouard 908:
909: #include <limits.h>
910: #include <sys/types.h>
1.171 brouard 911:
912: #if defined(__GNUC__)
913: #include <sys/utsname.h> /* Doesn't work on Windows */
914: #endif
915:
1.126 brouard 916: #include <sys/stat.h>
917: #include <errno.h>
1.159 brouard 918: /* extern int errno; */
1.126 brouard 919:
1.157 brouard 920: /* #ifdef LINUX */
921: /* #include <time.h> */
922: /* #include "timeval.h" */
923: /* #else */
924: /* #include <sys/time.h> */
925: /* #endif */
926:
1.126 brouard 927: #include <time.h>
928:
1.136 brouard 929: #ifdef GSL
930: #include <gsl/gsl_errno.h>
931: #include <gsl/gsl_multimin.h>
932: #endif
933:
1.167 brouard 934:
1.162 brouard 935: #ifdef NLOPT
936: #include <nlopt.h>
937: typedef struct {
938: double (* function)(double [] );
939: } myfunc_data ;
940: #endif
941:
1.126 brouard 942: /* #include <libintl.h> */
943: /* #define _(String) gettext (String) */
944:
1.251 brouard 945: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 946:
947: #define GNUPLOTPROGRAM "gnuplot"
948: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
949: #define FILENAMELENGTH 132
950:
951: #define GLOCK_ERROR_NOPATH -1 /* empty path */
952: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
953:
1.144 brouard 954: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
955: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 956:
957: #define NINTERVMAX 8
1.144 brouard 958: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
959: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
960: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 961: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 962: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
963: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 964: #define MAXN 20000
1.144 brouard 965: #define YEARM 12. /**< Number of months per year */
1.218 brouard 966: /* #define AGESUP 130 */
967: #define AGESUP 150
968: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 969: #define AGEBASE 40
1.194 brouard 970: #define AGEOVERFLOW 1.e20
1.164 brouard 971: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 972: #ifdef _WIN32
973: #define DIRSEPARATOR '\\'
974: #define CHARSEPARATOR "\\"
975: #define ODIRSEPARATOR '/'
976: #else
1.126 brouard 977: #define DIRSEPARATOR '/'
978: #define CHARSEPARATOR "/"
979: #define ODIRSEPARATOR '\\'
980: #endif
981:
1.260 ! brouard 982: /* $Id: imach.c,v 1.259 2017/04/04 13:01:16 brouard Exp $ */
1.126 brouard 983: /* $State: Exp $ */
1.196 brouard 984: #include "version.h"
985: char version[]=__IMACH_VERSION__;
1.224 brouard 986: 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";
1.260 ! brouard 987: char fullversion[]="$Revision: 1.259 $ $Date: 2017/04/04 13:01:16 $";
1.126 brouard 988: char strstart[80];
989: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 990: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 991: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 992: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
993: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
994: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 995: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
996: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 997: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
998: int cptcovprodnoage=0; /**< Number of covariate products without age */
999: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1000: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1001: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1002: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1003: int nsd=0; /**< Total number of single dummy variables (output) */
1004: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1005: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1006: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1007: int ntveff=0; /**< ntveff number of effective time varying variables */
1008: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1009: int cptcov=0; /* Working variable */
1.218 brouard 1010: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1011: int npar=NPARMAX;
1012: int nlstate=2; /* Number of live states */
1013: int ndeath=1; /* Number of dead states */
1.130 brouard 1014: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1015: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1016: int popbased=0;
1017:
1018: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1019: int maxwav=0; /* Maxim number of waves */
1020: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1021: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1022: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1023: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1024: int mle=1, weightopt=0;
1.126 brouard 1025: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1026: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1027: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1028: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1029: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1030: int selected(int kvar); /* Is covariate kvar selected for printing results */
1031:
1.130 brouard 1032: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1033: double **matprod2(); /* test */
1.126 brouard 1034: double **oldm, **newm, **savm; /* Working pointers to matrices */
1035: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1036: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1037:
1.136 brouard 1038: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1039: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1040: FILE *ficlog, *ficrespow;
1.130 brouard 1041: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1042: double fretone; /* Only one call to likelihood */
1.130 brouard 1043: long ipmx=0; /* Number of contributions */
1.126 brouard 1044: double sw; /* Sum of weights */
1045: char filerespow[FILENAMELENGTH];
1046: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1047: FILE *ficresilk;
1048: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1049: FILE *ficresprobmorprev;
1050: FILE *fichtm, *fichtmcov; /* Html File */
1051: FILE *ficreseij;
1052: char filerese[FILENAMELENGTH];
1053: FILE *ficresstdeij;
1054: char fileresstde[FILENAMELENGTH];
1055: FILE *ficrescveij;
1056: char filerescve[FILENAMELENGTH];
1057: FILE *ficresvij;
1058: char fileresv[FILENAMELENGTH];
1059: FILE *ficresvpl;
1060: char fileresvpl[FILENAMELENGTH];
1061: char title[MAXLINE];
1.234 brouard 1062: char model[MAXLINE]; /**< The model line */
1.217 brouard 1063: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1064: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1065: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1066: char command[FILENAMELENGTH];
1067: int outcmd=0;
1068:
1.217 brouard 1069: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1070: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1071: char filelog[FILENAMELENGTH]; /* Log file */
1072: char filerest[FILENAMELENGTH];
1073: char fileregp[FILENAMELENGTH];
1074: char popfile[FILENAMELENGTH];
1075:
1076: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1077:
1.157 brouard 1078: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1079: /* struct timezone tzp; */
1080: /* extern int gettimeofday(); */
1081: struct tm tml, *gmtime(), *localtime();
1082:
1083: extern time_t time();
1084:
1085: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1086: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1087: struct tm tm;
1088:
1.126 brouard 1089: char strcurr[80], strfor[80];
1090:
1091: char *endptr;
1092: long lval;
1093: double dval;
1094:
1095: #define NR_END 1
1096: #define FREE_ARG char*
1097: #define FTOL 1.0e-10
1098:
1099: #define NRANSI
1.240 brouard 1100: #define ITMAX 200
1101: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1102:
1103: #define TOL 2.0e-4
1104:
1105: #define CGOLD 0.3819660
1106: #define ZEPS 1.0e-10
1107: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1108:
1109: #define GOLD 1.618034
1110: #define GLIMIT 100.0
1111: #define TINY 1.0e-20
1112:
1113: static double maxarg1,maxarg2;
1114: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1115: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1116:
1117: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1118: #define rint(a) floor(a+0.5)
1.166 brouard 1119: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1120: #define mytinydouble 1.0e-16
1.166 brouard 1121: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1122: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1123: /* static double dsqrarg; */
1124: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1125: static double sqrarg;
1126: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1127: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1128: int agegomp= AGEGOMP;
1129:
1130: int imx;
1131: int stepm=1;
1132: /* Stepm, step in month: minimum step interpolation*/
1133:
1134: int estepm;
1135: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1136:
1137: int m,nb;
1138: long *num;
1.197 brouard 1139: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1140: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1141: covariate for which somebody answered excluding
1142: undefined. Usually 2: 0 and 1. */
1143: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1144: covariate for which somebody answered including
1145: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1146: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1147: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1148: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1149: double *ageexmed,*agecens;
1150: double dateintmean=0;
1151:
1152: double *weight;
1153: int **s; /* Status */
1.141 brouard 1154: double *agedc;
1.145 brouard 1155: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1156: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1157: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1158: double **coqvar; /* Fixed quantitative covariate iqv */
1159: double ***cotvar; /* Time varying covariate itv */
1160: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1161: double idx;
1162: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1163: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1164: /*k 1 2 3 4 5 6 7 8 9 */
1165: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1166: /* Tndvar[k] 1 2 3 4 5 */
1167: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1168: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1169: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1170: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1171: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1172: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1173: /* Tprod[i]=k 4 7 */
1174: /* Tage[i]=k 5 8 */
1175: /* */
1176: /* Type */
1177: /* V 1 2 3 4 5 */
1178: /* F F V V V */
1179: /* D Q D D Q */
1180: /* */
1181: int *TvarsD;
1182: int *TvarsDind;
1183: int *TvarsQ;
1184: int *TvarsQind;
1185:
1.235 brouard 1186: #define MAXRESULTLINES 10
1187: int nresult=0;
1.258 brouard 1188: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1189: int TKresult[MAXRESULTLINES];
1.237 brouard 1190: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1191: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1192: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1193: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1194: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1195: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1196:
1.234 brouard 1197: /* 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 1198: 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 */
1199: 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 */
1200: 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 */
1201: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1202: 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 */
1203: 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 1204: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1205: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1206: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1207: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1208: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1209: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1210: 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 */
1211: 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 */
1212:
1.230 brouard 1213: int *Tvarsel; /**< Selected covariates for output */
1214: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1215: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1216: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1217: 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 1218: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1219: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1220: int *Tage;
1.227 brouard 1221: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1222: 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 1223: 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*/
1224: 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 1225: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1226: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1227: int **Tvard;
1228: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1229: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1230: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1231: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1232: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1233: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1234: double *lsurv, *lpop, *tpop;
1235:
1.231 brouard 1236: #define FD 1; /* Fixed dummy covariate */
1237: #define FQ 2; /* Fixed quantitative covariate */
1238: #define FP 3; /* Fixed product covariate */
1239: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1240: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1241: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1242: #define VD 10; /* Varying dummy covariate */
1243: #define VQ 11; /* Varying quantitative covariate */
1244: #define VP 12; /* Varying product covariate */
1245: #define VPDD 13; /* Varying product dummy*dummy covariate */
1246: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1247: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1248: #define APFD 16; /* Age product * fixed dummy covariate */
1249: #define APFQ 17; /* Age product * fixed quantitative covariate */
1250: #define APVD 18; /* Age product * varying dummy covariate */
1251: #define APVQ 19; /* Age product * varying quantitative covariate */
1252:
1253: #define FTYPE 1; /* Fixed covariate */
1254: #define VTYPE 2; /* Varying covariate (loop in wave) */
1255: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1256:
1257: struct kmodel{
1258: int maintype; /* main type */
1259: int subtype; /* subtype */
1260: };
1261: struct kmodel modell[NCOVMAX];
1262:
1.143 brouard 1263: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1264: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1265:
1266: /**************** split *************************/
1267: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1268: {
1269: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1270: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1271: */
1272: char *ss; /* pointer */
1.186 brouard 1273: int l1=0, l2=0; /* length counters */
1.126 brouard 1274:
1275: l1 = strlen(path ); /* length of path */
1276: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1277: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1278: if ( ss == NULL ) { /* no directory, so determine current directory */
1279: strcpy( name, path ); /* we got the fullname name because no directory */
1280: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1281: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1282: /* get current working directory */
1283: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1284: #ifdef WIN32
1285: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1286: #else
1287: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1288: #endif
1.126 brouard 1289: return( GLOCK_ERROR_GETCWD );
1290: }
1291: /* got dirc from getcwd*/
1292: printf(" DIRC = %s \n",dirc);
1.205 brouard 1293: } else { /* strip directory from path */
1.126 brouard 1294: ss++; /* after this, the filename */
1295: l2 = strlen( ss ); /* length of filename */
1296: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1297: strcpy( name, ss ); /* save file name */
1298: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1299: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1300: printf(" DIRC2 = %s \n",dirc);
1301: }
1302: /* We add a separator at the end of dirc if not exists */
1303: l1 = strlen( dirc ); /* length of directory */
1304: if( dirc[l1-1] != DIRSEPARATOR ){
1305: dirc[l1] = DIRSEPARATOR;
1306: dirc[l1+1] = 0;
1307: printf(" DIRC3 = %s \n",dirc);
1308: }
1309: ss = strrchr( name, '.' ); /* find last / */
1310: if (ss >0){
1311: ss++;
1312: strcpy(ext,ss); /* save extension */
1313: l1= strlen( name);
1314: l2= strlen(ss)+1;
1315: strncpy( finame, name, l1-l2);
1316: finame[l1-l2]= 0;
1317: }
1318:
1319: return( 0 ); /* we're done */
1320: }
1321:
1322:
1323: /******************************************/
1324:
1325: void replace_back_to_slash(char *s, char*t)
1326: {
1327: int i;
1328: int lg=0;
1329: i=0;
1330: lg=strlen(t);
1331: for(i=0; i<= lg; i++) {
1332: (s[i] = t[i]);
1333: if (t[i]== '\\') s[i]='/';
1334: }
1335: }
1336:
1.132 brouard 1337: char *trimbb(char *out, char *in)
1.137 brouard 1338: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1339: char *s;
1340: s=out;
1341: while (*in != '\0'){
1.137 brouard 1342: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1343: in++;
1344: }
1345: *out++ = *in++;
1346: }
1347: *out='\0';
1348: return s;
1349: }
1350:
1.187 brouard 1351: /* char *substrchaine(char *out, char *in, char *chain) */
1352: /* { */
1353: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1354: /* char *s, *t; */
1355: /* t=in;s=out; */
1356: /* while ((*in != *chain) && (*in != '\0')){ */
1357: /* *out++ = *in++; */
1358: /* } */
1359:
1360: /* /\* *in matches *chain *\/ */
1361: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1362: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1363: /* } */
1364: /* in--; chain--; */
1365: /* while ( (*in != '\0')){ */
1366: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1367: /* *out++ = *in++; */
1368: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1369: /* } */
1370: /* *out='\0'; */
1371: /* out=s; */
1372: /* return out; */
1373: /* } */
1374: char *substrchaine(char *out, char *in, char *chain)
1375: {
1376: /* Substract chain 'chain' from 'in', return and output 'out' */
1377: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1378:
1379: char *strloc;
1380:
1381: strcpy (out, in);
1382: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1383: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1384: if(strloc != NULL){
1385: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1386: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1387: /* strcpy (strloc, strloc +strlen(chain));*/
1388: }
1389: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1390: return out;
1391: }
1392:
1393:
1.145 brouard 1394: char *cutl(char *blocc, char *alocc, char *in, char occ)
1395: {
1.187 brouard 1396: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1397: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1398: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1399: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1400: */
1.160 brouard 1401: char *s, *t;
1.145 brouard 1402: t=in;s=in;
1403: while ((*in != occ) && (*in != '\0')){
1404: *alocc++ = *in++;
1405: }
1406: if( *in == occ){
1407: *(alocc)='\0';
1408: s=++in;
1409: }
1410:
1411: if (s == t) {/* occ not found */
1412: *(alocc-(in-s))='\0';
1413: in=s;
1414: }
1415: while ( *in != '\0'){
1416: *blocc++ = *in++;
1417: }
1418:
1419: *blocc='\0';
1420: return t;
1421: }
1.137 brouard 1422: char *cutv(char *blocc, char *alocc, char *in, char occ)
1423: {
1.187 brouard 1424: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1425: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1426: gives blocc="abcdef2ghi" and alocc="j".
1427: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1428: */
1429: char *s, *t;
1430: t=in;s=in;
1431: while (*in != '\0'){
1432: while( *in == occ){
1433: *blocc++ = *in++;
1434: s=in;
1435: }
1436: *blocc++ = *in++;
1437: }
1438: if (s == t) /* occ not found */
1439: *(blocc-(in-s))='\0';
1440: else
1441: *(blocc-(in-s)-1)='\0';
1442: in=s;
1443: while ( *in != '\0'){
1444: *alocc++ = *in++;
1445: }
1446:
1447: *alocc='\0';
1448: return s;
1449: }
1450:
1.126 brouard 1451: int nbocc(char *s, char occ)
1452: {
1453: int i,j=0;
1454: int lg=20;
1455: i=0;
1456: lg=strlen(s);
1457: for(i=0; i<= lg; i++) {
1.234 brouard 1458: if (s[i] == occ ) j++;
1.126 brouard 1459: }
1460: return j;
1461: }
1462:
1.137 brouard 1463: /* void cutv(char *u,char *v, char*t, char occ) */
1464: /* { */
1465: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1466: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1467: /* gives u="abcdef2ghi" and v="j" *\/ */
1468: /* int i,lg,j,p=0; */
1469: /* i=0; */
1470: /* lg=strlen(t); */
1471: /* for(j=0; j<=lg-1; j++) { */
1472: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1473: /* } */
1.126 brouard 1474:
1.137 brouard 1475: /* for(j=0; j<p; j++) { */
1476: /* (u[j] = t[j]); */
1477: /* } */
1478: /* u[p]='\0'; */
1.126 brouard 1479:
1.137 brouard 1480: /* for(j=0; j<= lg; j++) { */
1481: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1482: /* } */
1483: /* } */
1.126 brouard 1484:
1.160 brouard 1485: #ifdef _WIN32
1486: char * strsep(char **pp, const char *delim)
1487: {
1488: char *p, *q;
1489:
1490: if ((p = *pp) == NULL)
1491: return 0;
1492: if ((q = strpbrk (p, delim)) != NULL)
1493: {
1494: *pp = q + 1;
1495: *q = '\0';
1496: }
1497: else
1498: *pp = 0;
1499: return p;
1500: }
1501: #endif
1502:
1.126 brouard 1503: /********************** nrerror ********************/
1504:
1505: void nrerror(char error_text[])
1506: {
1507: fprintf(stderr,"ERREUR ...\n");
1508: fprintf(stderr,"%s\n",error_text);
1509: exit(EXIT_FAILURE);
1510: }
1511: /*********************** vector *******************/
1512: double *vector(int nl, int nh)
1513: {
1514: double *v;
1515: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1516: if (!v) nrerror("allocation failure in vector");
1517: return v-nl+NR_END;
1518: }
1519:
1520: /************************ free vector ******************/
1521: void free_vector(double*v, int nl, int nh)
1522: {
1523: free((FREE_ARG)(v+nl-NR_END));
1524: }
1525:
1526: /************************ivector *******************************/
1527: int *ivector(long nl,long nh)
1528: {
1529: int *v;
1530: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1531: if (!v) nrerror("allocation failure in ivector");
1532: return v-nl+NR_END;
1533: }
1534:
1535: /******************free ivector **************************/
1536: void free_ivector(int *v, long nl, long nh)
1537: {
1538: free((FREE_ARG)(v+nl-NR_END));
1539: }
1540:
1541: /************************lvector *******************************/
1542: long *lvector(long nl,long nh)
1543: {
1544: long *v;
1545: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1546: if (!v) nrerror("allocation failure in ivector");
1547: return v-nl+NR_END;
1548: }
1549:
1550: /******************free lvector **************************/
1551: void free_lvector(long *v, long nl, long nh)
1552: {
1553: free((FREE_ARG)(v+nl-NR_END));
1554: }
1555:
1556: /******************* imatrix *******************************/
1557: int **imatrix(long nrl, long nrh, long ncl, long nch)
1558: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1559: {
1560: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1561: int **m;
1562:
1563: /* allocate pointers to rows */
1564: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1565: if (!m) nrerror("allocation failure 1 in matrix()");
1566: m += NR_END;
1567: m -= nrl;
1568:
1569:
1570: /* allocate rows and set pointers to them */
1571: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1572: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1573: m[nrl] += NR_END;
1574: m[nrl] -= ncl;
1575:
1576: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1577:
1578: /* return pointer to array of pointers to rows */
1579: return m;
1580: }
1581:
1582: /****************** free_imatrix *************************/
1583: void free_imatrix(m,nrl,nrh,ncl,nch)
1584: int **m;
1585: long nch,ncl,nrh,nrl;
1586: /* free an int matrix allocated by imatrix() */
1587: {
1588: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1589: free((FREE_ARG) (m+nrl-NR_END));
1590: }
1591:
1592: /******************* matrix *******************************/
1593: double **matrix(long nrl, long nrh, long ncl, long nch)
1594: {
1595: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1596: double **m;
1597:
1598: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1599: if (!m) nrerror("allocation failure 1 in matrix()");
1600: m += NR_END;
1601: m -= nrl;
1602:
1603: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1604: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1605: m[nrl] += NR_END;
1606: m[nrl] -= ncl;
1607:
1608: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1609: return m;
1.145 brouard 1610: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1611: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1612: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1613: */
1614: }
1615:
1616: /*************************free matrix ************************/
1617: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1618: {
1619: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1620: free((FREE_ARG)(m+nrl-NR_END));
1621: }
1622:
1623: /******************* ma3x *******************************/
1624: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1625: {
1626: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1627: double ***m;
1628:
1629: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1630: if (!m) nrerror("allocation failure 1 in matrix()");
1631: m += NR_END;
1632: m -= nrl;
1633:
1634: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1635: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1636: m[nrl] += NR_END;
1637: m[nrl] -= ncl;
1638:
1639: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1640:
1641: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1642: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1643: m[nrl][ncl] += NR_END;
1644: m[nrl][ncl] -= nll;
1645: for (j=ncl+1; j<=nch; j++)
1646: m[nrl][j]=m[nrl][j-1]+nlay;
1647:
1648: for (i=nrl+1; i<=nrh; i++) {
1649: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1650: for (j=ncl+1; j<=nch; j++)
1651: m[i][j]=m[i][j-1]+nlay;
1652: }
1653: return m;
1654: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1655: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1656: */
1657: }
1658:
1659: /*************************free ma3x ************************/
1660: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1661: {
1662: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1663: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1664: free((FREE_ARG)(m+nrl-NR_END));
1665: }
1666:
1667: /*************** function subdirf ***********/
1668: char *subdirf(char fileres[])
1669: {
1670: /* Caution optionfilefiname is hidden */
1671: strcpy(tmpout,optionfilefiname);
1672: strcat(tmpout,"/"); /* Add to the right */
1673: strcat(tmpout,fileres);
1674: return tmpout;
1675: }
1676:
1677: /*************** function subdirf2 ***********/
1678: char *subdirf2(char fileres[], char *preop)
1679: {
1680:
1681: /* Caution optionfilefiname is hidden */
1682: strcpy(tmpout,optionfilefiname);
1683: strcat(tmpout,"/");
1684: strcat(tmpout,preop);
1685: strcat(tmpout,fileres);
1686: return tmpout;
1687: }
1688:
1689: /*************** function subdirf3 ***********/
1690: char *subdirf3(char fileres[], char *preop, char *preop2)
1691: {
1692:
1693: /* Caution optionfilefiname is hidden */
1694: strcpy(tmpout,optionfilefiname);
1695: strcat(tmpout,"/");
1696: strcat(tmpout,preop);
1697: strcat(tmpout,preop2);
1698: strcat(tmpout,fileres);
1699: return tmpout;
1700: }
1.213 brouard 1701:
1702: /*************** function subdirfext ***********/
1703: char *subdirfext(char fileres[], char *preop, char *postop)
1704: {
1705:
1706: strcpy(tmpout,preop);
1707: strcat(tmpout,fileres);
1708: strcat(tmpout,postop);
1709: return tmpout;
1710: }
1.126 brouard 1711:
1.213 brouard 1712: /*************** function subdirfext3 ***********/
1713: char *subdirfext3(char fileres[], char *preop, char *postop)
1714: {
1715:
1716: /* Caution optionfilefiname is hidden */
1717: strcpy(tmpout,optionfilefiname);
1718: strcat(tmpout,"/");
1719: strcat(tmpout,preop);
1720: strcat(tmpout,fileres);
1721: strcat(tmpout,postop);
1722: return tmpout;
1723: }
1724:
1.162 brouard 1725: char *asc_diff_time(long time_sec, char ascdiff[])
1726: {
1727: long sec_left, days, hours, minutes;
1728: days = (time_sec) / (60*60*24);
1729: sec_left = (time_sec) % (60*60*24);
1730: hours = (sec_left) / (60*60) ;
1731: sec_left = (sec_left) %(60*60);
1732: minutes = (sec_left) /60;
1733: sec_left = (sec_left) % (60);
1734: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1735: return ascdiff;
1736: }
1737:
1.126 brouard 1738: /***************** f1dim *************************/
1739: extern int ncom;
1740: extern double *pcom,*xicom;
1741: extern double (*nrfunc)(double []);
1742:
1743: double f1dim(double x)
1744: {
1745: int j;
1746: double f;
1747: double *xt;
1748:
1749: xt=vector(1,ncom);
1750: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1751: f=(*nrfunc)(xt);
1752: free_vector(xt,1,ncom);
1753: return f;
1754: }
1755:
1756: /*****************brent *************************/
1757: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1758: {
1759: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1760: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1761: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1762: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1763: * returned function value.
1764: */
1.126 brouard 1765: int iter;
1766: double a,b,d,etemp;
1.159 brouard 1767: double fu=0,fv,fw,fx;
1.164 brouard 1768: double ftemp=0.;
1.126 brouard 1769: double p,q,r,tol1,tol2,u,v,w,x,xm;
1770: double e=0.0;
1771:
1772: a=(ax < cx ? ax : cx);
1773: b=(ax > cx ? ax : cx);
1774: x=w=v=bx;
1775: fw=fv=fx=(*f)(x);
1776: for (iter=1;iter<=ITMAX;iter++) {
1777: xm=0.5*(a+b);
1778: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1779: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1780: printf(".");fflush(stdout);
1781: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1782: #ifdef DEBUGBRENT
1.126 brouard 1783: 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);
1784: 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);
1785: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1786: #endif
1787: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1788: *xmin=x;
1789: return fx;
1790: }
1791: ftemp=fu;
1792: if (fabs(e) > tol1) {
1793: r=(x-w)*(fx-fv);
1794: q=(x-v)*(fx-fw);
1795: p=(x-v)*q-(x-w)*r;
1796: q=2.0*(q-r);
1797: if (q > 0.0) p = -p;
1798: q=fabs(q);
1799: etemp=e;
1800: e=d;
1801: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1802: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1803: else {
1.224 brouard 1804: d=p/q;
1805: u=x+d;
1806: if (u-a < tol2 || b-u < tol2)
1807: d=SIGN(tol1,xm-x);
1.126 brouard 1808: }
1809: } else {
1810: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1811: }
1812: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1813: fu=(*f)(u);
1814: if (fu <= fx) {
1815: if (u >= x) a=x; else b=x;
1816: SHFT(v,w,x,u)
1.183 brouard 1817: SHFT(fv,fw,fx,fu)
1818: } else {
1819: if (u < x) a=u; else b=u;
1820: if (fu <= fw || w == x) {
1.224 brouard 1821: v=w;
1822: w=u;
1823: fv=fw;
1824: fw=fu;
1.183 brouard 1825: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1826: v=u;
1827: fv=fu;
1.183 brouard 1828: }
1829: }
1.126 brouard 1830: }
1831: nrerror("Too many iterations in brent");
1832: *xmin=x;
1833: return fx;
1834: }
1835:
1836: /****************** mnbrak ***********************/
1837:
1838: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1839: double (*func)(double))
1.183 brouard 1840: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1841: the downhill direction (defined by the function as evaluated at the initial points) and returns
1842: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1843: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1844: */
1.126 brouard 1845: double ulim,u,r,q, dum;
1846: double fu;
1.187 brouard 1847:
1848: double scale=10.;
1849: int iterscale=0;
1850:
1851: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1852: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1853:
1854:
1855: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1856: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1857: /* *bx = *ax - (*ax - *bx)/scale; */
1858: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1859: /* } */
1860:
1.126 brouard 1861: if (*fb > *fa) {
1862: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1863: SHFT(dum,*fb,*fa,dum)
1864: }
1.126 brouard 1865: *cx=(*bx)+GOLD*(*bx-*ax);
1866: *fc=(*func)(*cx);
1.183 brouard 1867: #ifdef DEBUG
1.224 brouard 1868: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1869: 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 1870: #endif
1.224 brouard 1871: 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 1872: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1873: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1874: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1875: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1876: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1877: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1878: fu=(*func)(u);
1.163 brouard 1879: #ifdef DEBUG
1880: /* f(x)=A(x-u)**2+f(u) */
1881: double A, fparabu;
1882: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1883: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1884: 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);
1885: 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 1886: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1887: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1888: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1889: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1890: #endif
1.184 brouard 1891: #ifdef MNBRAKORIGINAL
1.183 brouard 1892: #else
1.191 brouard 1893: /* if (fu > *fc) { */
1894: /* #ifdef DEBUG */
1895: /* printf("mnbrak4 fu > fc \n"); */
1896: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1897: /* #endif */
1898: /* /\* 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 *\\/ *\/ */
1899: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1900: /* dum=u; /\* Shifting c and u *\/ */
1901: /* u = *cx; */
1902: /* *cx = dum; */
1903: /* dum = fu; */
1904: /* fu = *fc; */
1905: /* *fc =dum; */
1906: /* } else { /\* end *\/ */
1907: /* #ifdef DEBUG */
1908: /* printf("mnbrak3 fu < fc \n"); */
1909: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1910: /* #endif */
1911: /* dum=u; /\* Shifting c and u *\/ */
1912: /* u = *cx; */
1913: /* *cx = dum; */
1914: /* dum = fu; */
1915: /* fu = *fc; */
1916: /* *fc =dum; */
1917: /* } */
1.224 brouard 1918: #ifdef DEBUGMNBRAK
1919: double A, fparabu;
1920: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1921: fparabu= *fa - A*(*ax-u)*(*ax-u);
1922: 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);
1923: 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 1924: #endif
1.191 brouard 1925: dum=u; /* Shifting c and u */
1926: u = *cx;
1927: *cx = dum;
1928: dum = fu;
1929: fu = *fc;
1930: *fc =dum;
1.183 brouard 1931: #endif
1.162 brouard 1932: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1933: #ifdef DEBUG
1.224 brouard 1934: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1935: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1936: #endif
1.126 brouard 1937: fu=(*func)(u);
1938: if (fu < *fc) {
1.183 brouard 1939: #ifdef DEBUG
1.224 brouard 1940: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1941: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1942: #endif
1943: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1944: SHFT(*fb,*fc,fu,(*func)(u))
1945: #ifdef DEBUG
1946: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1947: #endif
1948: }
1.162 brouard 1949: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1950: #ifdef DEBUG
1.224 brouard 1951: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1952: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1953: #endif
1.126 brouard 1954: u=ulim;
1955: fu=(*func)(u);
1.183 brouard 1956: } else { /* u could be left to b (if r > q parabola has a maximum) */
1957: #ifdef DEBUG
1.224 brouard 1958: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1959: 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 1960: #endif
1.126 brouard 1961: u=(*cx)+GOLD*(*cx-*bx);
1962: fu=(*func)(u);
1.224 brouard 1963: #ifdef DEBUG
1964: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1965: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1966: #endif
1.183 brouard 1967: } /* end tests */
1.126 brouard 1968: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1969: SHFT(*fa,*fb,*fc,fu)
1970: #ifdef DEBUG
1.224 brouard 1971: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1972: 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 1973: #endif
1974: } /* 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 1975: }
1976:
1977: /*************** linmin ************************/
1.162 brouard 1978: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1979: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1980: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1981: the value of func at the returned location p . This is actually all accomplished by calling the
1982: routines mnbrak and brent .*/
1.126 brouard 1983: int ncom;
1984: double *pcom,*xicom;
1985: double (*nrfunc)(double []);
1986:
1.224 brouard 1987: #ifdef LINMINORIGINAL
1.126 brouard 1988: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1989: #else
1990: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1991: #endif
1.126 brouard 1992: {
1993: double brent(double ax, double bx, double cx,
1994: double (*f)(double), double tol, double *xmin);
1995: double f1dim(double x);
1996: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1997: double *fc, double (*func)(double));
1998: int j;
1999: double xx,xmin,bx,ax;
2000: double fx,fb,fa;
1.187 brouard 2001:
1.203 brouard 2002: #ifdef LINMINORIGINAL
2003: #else
2004: double scale=10., axs, xxs; /* Scale added for infinity */
2005: #endif
2006:
1.126 brouard 2007: ncom=n;
2008: pcom=vector(1,n);
2009: xicom=vector(1,n);
2010: nrfunc=func;
2011: for (j=1;j<=n;j++) {
2012: pcom[j]=p[j];
1.202 brouard 2013: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2014: }
1.187 brouard 2015:
1.203 brouard 2016: #ifdef LINMINORIGINAL
2017: xx=1.;
2018: #else
2019: axs=0.0;
2020: xxs=1.;
2021: do{
2022: xx= xxs;
2023: #endif
1.187 brouard 2024: ax=0.;
2025: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2026: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2027: /* 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)) */
2028: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2029: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2030: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2031: /* 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 2032: #ifdef LINMINORIGINAL
2033: #else
2034: if (fx != fx){
1.224 brouard 2035: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2036: printf("|");
2037: fprintf(ficlog,"|");
1.203 brouard 2038: #ifdef DEBUGLINMIN
1.224 brouard 2039: 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 2040: #endif
2041: }
1.224 brouard 2042: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2043: #endif
2044:
1.191 brouard 2045: #ifdef DEBUGLINMIN
2046: 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 2047: 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 2048: #endif
1.224 brouard 2049: #ifdef LINMINORIGINAL
2050: #else
2051: if(fb == fx){ /* Flat function in the direction */
2052: xmin=xx;
2053: *flat=1;
2054: }else{
2055: *flat=0;
2056: #endif
2057: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2058: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2059: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2060: /* fmin = f(p[j] + xmin * xi[j]) */
2061: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2062: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2063: #ifdef DEBUG
1.224 brouard 2064: 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);
2065: 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);
2066: #endif
2067: #ifdef LINMINORIGINAL
2068: #else
2069: }
1.126 brouard 2070: #endif
1.191 brouard 2071: #ifdef DEBUGLINMIN
2072: printf("linmin end ");
1.202 brouard 2073: fprintf(ficlog,"linmin end ");
1.191 brouard 2074: #endif
1.126 brouard 2075: for (j=1;j<=n;j++) {
1.203 brouard 2076: #ifdef LINMINORIGINAL
2077: xi[j] *= xmin;
2078: #else
2079: #ifdef DEBUGLINMIN
2080: if(xxs <1.0)
2081: printf(" before xi[%d]=%12.8f", j,xi[j]);
2082: #endif
2083: 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) */
2084: #ifdef DEBUGLINMIN
2085: if(xxs <1.0)
2086: 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 );
2087: #endif
2088: #endif
1.187 brouard 2089: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2090: }
1.191 brouard 2091: #ifdef DEBUGLINMIN
1.203 brouard 2092: printf("\n");
1.191 brouard 2093: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2094: 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 2095: for (j=1;j<=n;j++) {
1.202 brouard 2096: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2097: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2098: if(j % ncovmodel == 0){
1.191 brouard 2099: printf("\n");
1.202 brouard 2100: fprintf(ficlog,"\n");
2101: }
1.191 brouard 2102: }
1.203 brouard 2103: #else
1.191 brouard 2104: #endif
1.126 brouard 2105: free_vector(xicom,1,n);
2106: free_vector(pcom,1,n);
2107: }
2108:
2109:
2110: /*************** powell ************************/
1.162 brouard 2111: /*
2112: Minimization of a function func of n variables. Input consists of an initial starting point
2113: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2114: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2115: such that failure to decrease by more than this amount on one iteration signals doneness. On
2116: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2117: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2118: */
1.224 brouard 2119: #ifdef LINMINORIGINAL
2120: #else
2121: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2122: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2123: #endif
1.126 brouard 2124: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2125: double (*func)(double []))
2126: {
1.224 brouard 2127: #ifdef LINMINORIGINAL
2128: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2129: double (*func)(double []));
1.224 brouard 2130: #else
1.241 brouard 2131: void linmin(double p[], double xi[], int n, double *fret,
2132: double (*func)(double []),int *flat);
1.224 brouard 2133: #endif
1.239 brouard 2134: int i,ibig,j,jk,k;
1.126 brouard 2135: double del,t,*pt,*ptt,*xit;
1.181 brouard 2136: double directest;
1.126 brouard 2137: double fp,fptt;
2138: double *xits;
2139: int niterf, itmp;
1.224 brouard 2140: #ifdef LINMINORIGINAL
2141: #else
2142:
2143: flatdir=ivector(1,n);
2144: for (j=1;j<=n;j++) flatdir[j]=0;
2145: #endif
1.126 brouard 2146:
2147: pt=vector(1,n);
2148: ptt=vector(1,n);
2149: xit=vector(1,n);
2150: xits=vector(1,n);
2151: *fret=(*func)(p);
2152: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2153: rcurr_time = time(NULL);
1.126 brouard 2154: for (*iter=1;;++(*iter)) {
1.187 brouard 2155: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2156: ibig=0;
2157: del=0.0;
1.157 brouard 2158: rlast_time=rcurr_time;
2159: /* (void) gettimeofday(&curr_time,&tzp); */
2160: rcurr_time = time(NULL);
2161: curr_time = *localtime(&rcurr_time);
2162: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2163: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2164: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2165: for (i=1;i<=n;i++) {
1.126 brouard 2166: fprintf(ficrespow," %.12lf", p[i]);
2167: }
1.239 brouard 2168: fprintf(ficrespow,"\n");fflush(ficrespow);
2169: printf("\n#model= 1 + age ");
2170: fprintf(ficlog,"\n#model= 1 + age ");
2171: if(nagesqr==1){
1.241 brouard 2172: printf(" + age*age ");
2173: fprintf(ficlog," + age*age ");
1.239 brouard 2174: }
2175: for(j=1;j <=ncovmodel-2;j++){
2176: if(Typevar[j]==0) {
2177: printf(" + V%d ",Tvar[j]);
2178: fprintf(ficlog," + V%d ",Tvar[j]);
2179: }else if(Typevar[j]==1) {
2180: printf(" + V%d*age ",Tvar[j]);
2181: fprintf(ficlog," + V%d*age ",Tvar[j]);
2182: }else if(Typevar[j]==2) {
2183: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2184: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2185: }
2186: }
1.126 brouard 2187: printf("\n");
1.239 brouard 2188: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2189: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2190: fprintf(ficlog,"\n");
1.239 brouard 2191: for(i=1,jk=1; i <=nlstate; i++){
2192: for(k=1; k <=(nlstate+ndeath); k++){
2193: if (k != i) {
2194: printf("%d%d ",i,k);
2195: fprintf(ficlog,"%d%d ",i,k);
2196: for(j=1; j <=ncovmodel; j++){
2197: printf("%12.7f ",p[jk]);
2198: fprintf(ficlog,"%12.7f ",p[jk]);
2199: jk++;
2200: }
2201: printf("\n");
2202: fprintf(ficlog,"\n");
2203: }
2204: }
2205: }
1.241 brouard 2206: if(*iter <=3 && *iter >1){
1.157 brouard 2207: tml = *localtime(&rcurr_time);
2208: strcpy(strcurr,asctime(&tml));
2209: rforecast_time=rcurr_time;
1.126 brouard 2210: itmp = strlen(strcurr);
2211: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2212: strcurr[itmp-1]='\0';
1.162 brouard 2213: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2214: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2215: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2216: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2217: forecast_time = *localtime(&rforecast_time);
2218: strcpy(strfor,asctime(&forecast_time));
2219: itmp = strlen(strfor);
2220: if(strfor[itmp-1]=='\n')
2221: strfor[itmp-1]='\0';
2222: 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);
2223: 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 2224: }
2225: }
1.187 brouard 2226: for (i=1;i<=n;i++) { /* For each direction i */
2227: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2228: fptt=(*fret);
2229: #ifdef DEBUG
1.203 brouard 2230: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2231: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2232: #endif
1.203 brouard 2233: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2234: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2235: #ifdef LINMINORIGINAL
1.188 brouard 2236: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2237: #else
2238: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2239: flatdir[i]=flat; /* Function is vanishing in that direction i */
2240: #endif
2241: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2242: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2243: /* because that direction will be replaced unless the gain del is small */
2244: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2245: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2246: /* with the new direction. */
2247: del=fabs(fptt-(*fret));
2248: ibig=i;
1.126 brouard 2249: }
2250: #ifdef DEBUG
2251: printf("%d %.12e",i,(*fret));
2252: fprintf(ficlog,"%d %.12e",i,(*fret));
2253: for (j=1;j<=n;j++) {
1.224 brouard 2254: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2255: printf(" x(%d)=%.12e",j,xit[j]);
2256: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2257: }
2258: for(j=1;j<=n;j++) {
1.225 brouard 2259: printf(" p(%d)=%.12e",j,p[j]);
2260: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2261: }
2262: printf("\n");
2263: fprintf(ficlog,"\n");
2264: #endif
1.187 brouard 2265: } /* end loop on each direction i */
2266: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2267: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2268: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2269: for(j=1;j<=n;j++) {
1.225 brouard 2270: if(flatdir[j] >0){
2271: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2272: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2273: }
2274: /* printf("\n"); */
2275: /* fprintf(ficlog,"\n"); */
2276: }
1.243 brouard 2277: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2278: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2279: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2280: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2281: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2282: /* decreased of more than 3.84 */
2283: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2284: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2285: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2286:
1.188 brouard 2287: /* Starting the program with initial values given by a former maximization will simply change */
2288: /* the scales of the directions and the directions, because the are reset to canonical directions */
2289: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2290: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2291: #ifdef DEBUG
2292: int k[2],l;
2293: k[0]=1;
2294: k[1]=-1;
2295: printf("Max: %.12e",(*func)(p));
2296: fprintf(ficlog,"Max: %.12e",(*func)(p));
2297: for (j=1;j<=n;j++) {
2298: printf(" %.12e",p[j]);
2299: fprintf(ficlog," %.12e",p[j]);
2300: }
2301: printf("\n");
2302: fprintf(ficlog,"\n");
2303: for(l=0;l<=1;l++) {
2304: for (j=1;j<=n;j++) {
2305: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2306: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2307: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2308: }
2309: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2310: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2311: }
2312: #endif
2313:
1.224 brouard 2314: #ifdef LINMINORIGINAL
2315: #else
2316: free_ivector(flatdir,1,n);
2317: #endif
1.126 brouard 2318: free_vector(xit,1,n);
2319: free_vector(xits,1,n);
2320: free_vector(ptt,1,n);
2321: free_vector(pt,1,n);
2322: return;
1.192 brouard 2323: } /* enough precision */
1.240 brouard 2324: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2325: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2326: ptt[j]=2.0*p[j]-pt[j];
2327: xit[j]=p[j]-pt[j];
2328: pt[j]=p[j];
2329: }
1.181 brouard 2330: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2331: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2332: if (*iter <=4) {
1.225 brouard 2333: #else
2334: #endif
1.224 brouard 2335: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2336: #else
1.161 brouard 2337: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2338: #endif
1.162 brouard 2339: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2340: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2341: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2342: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2343: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2344: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2345: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2346: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2347: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2348: /* Even if f3 <f1, directest can be negative and t >0 */
2349: /* mu² and del² are equal when f3=f1 */
2350: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2351: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2352: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2353: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2354: #ifdef NRCORIGINAL
2355: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2356: #else
2357: 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 2358: t= t- del*SQR(fp-fptt);
1.183 brouard 2359: #endif
1.202 brouard 2360: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2361: #ifdef DEBUG
1.181 brouard 2362: 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);
2363: 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 2364: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2365: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2366: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2367: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2368: 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);
2369: 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);
2370: #endif
1.183 brouard 2371: #ifdef POWELLORIGINAL
2372: if (t < 0.0) { /* Then we use it for new direction */
2373: #else
1.182 brouard 2374: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2375: 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 2376: 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 2377: 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 2378: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2379: }
1.181 brouard 2380: if (directest < 0.0) { /* Then we use it for new direction */
2381: #endif
1.191 brouard 2382: #ifdef DEBUGLINMIN
1.234 brouard 2383: printf("Before linmin in direction P%d-P0\n",n);
2384: for (j=1;j<=n;j++) {
2385: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2386: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2387: if(j % ncovmodel == 0){
2388: printf("\n");
2389: fprintf(ficlog,"\n");
2390: }
2391: }
1.224 brouard 2392: #endif
2393: #ifdef LINMINORIGINAL
1.234 brouard 2394: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2395: #else
1.234 brouard 2396: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2397: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2398: #endif
1.234 brouard 2399:
1.191 brouard 2400: #ifdef DEBUGLINMIN
1.234 brouard 2401: for (j=1;j<=n;j++) {
2402: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2403: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2404: if(j % ncovmodel == 0){
2405: printf("\n");
2406: fprintf(ficlog,"\n");
2407: }
2408: }
1.224 brouard 2409: #endif
1.234 brouard 2410: for (j=1;j<=n;j++) {
2411: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2412: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2413: }
1.224 brouard 2414: #ifdef LINMINORIGINAL
2415: #else
1.234 brouard 2416: for (j=1, flatd=0;j<=n;j++) {
2417: if(flatdir[j]>0)
2418: flatd++;
2419: }
2420: if(flatd >0){
1.255 brouard 2421: printf("%d flat directions: ",flatd);
2422: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2423: for (j=1;j<=n;j++) {
2424: if(flatdir[j]>0){
2425: printf("%d ",j);
2426: fprintf(ficlog,"%d ",j);
2427: }
2428: }
2429: printf("\n");
2430: fprintf(ficlog,"\n");
2431: }
1.191 brouard 2432: #endif
1.234 brouard 2433: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2434: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2435:
1.126 brouard 2436: #ifdef DEBUG
1.234 brouard 2437: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2438: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2439: for(j=1;j<=n;j++){
2440: printf(" %lf",xit[j]);
2441: fprintf(ficlog," %lf",xit[j]);
2442: }
2443: printf("\n");
2444: fprintf(ficlog,"\n");
1.126 brouard 2445: #endif
1.192 brouard 2446: } /* end of t or directest negative */
1.224 brouard 2447: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2448: #else
1.234 brouard 2449: } /* end if (fptt < fp) */
1.192 brouard 2450: #endif
1.225 brouard 2451: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2452: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2453: #else
1.224 brouard 2454: #endif
1.234 brouard 2455: } /* loop iteration */
1.126 brouard 2456: }
1.234 brouard 2457:
1.126 brouard 2458: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2459:
1.235 brouard 2460: 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 2461: {
1.235 brouard 2462: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2463: (and selected quantitative values in nres)
2464: by left multiplying the unit
1.234 brouard 2465: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2466: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2467: /* Wx is row vector: population in state 1, population in state 2, population dead */
2468: /* or prevalence in state 1, prevalence in state 2, 0 */
2469: /* newm is the matrix after multiplications, its rows are identical at a factor */
2470: /* Initial matrix pimij */
2471: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2472: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2473: /* 0, 0 , 1} */
2474: /*
2475: * and after some iteration: */
2476: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2477: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2478: /* 0, 0 , 1} */
2479: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2480: /* {0.51571254859325999, 0.4842874514067399, */
2481: /* 0.51326036147820708, 0.48673963852179264} */
2482: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2483:
1.126 brouard 2484: int i, ii,j,k;
1.209 brouard 2485: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2486: /* double **matprod2(); */ /* test */
1.218 brouard 2487: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2488: double **newm;
1.209 brouard 2489: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2490: int ncvloop=0;
1.169 brouard 2491:
1.209 brouard 2492: min=vector(1,nlstate);
2493: max=vector(1,nlstate);
2494: meandiff=vector(1,nlstate);
2495:
1.218 brouard 2496: /* Starting with matrix unity */
1.126 brouard 2497: for (ii=1;ii<=nlstate+ndeath;ii++)
2498: for (j=1;j<=nlstate+ndeath;j++){
2499: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2500: }
1.169 brouard 2501:
2502: cov[1]=1.;
2503:
2504: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2505: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2506: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2507: ncvloop++;
1.126 brouard 2508: newm=savm;
2509: /* Covariates have to be included here again */
1.138 brouard 2510: cov[2]=agefin;
1.187 brouard 2511: if(nagesqr==1)
2512: cov[3]= agefin*agefin;;
1.234 brouard 2513: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2514: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2515: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2516: /* 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 2517: }
2518: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2519: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2520: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2521: /* 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 2522: }
1.237 brouard 2523: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2524: if(Dummy[Tvar[Tage[k]]]){
2525: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2526: } else{
1.235 brouard 2527: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2528: }
1.235 brouard 2529: /* 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 2530: }
1.237 brouard 2531: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2532: /* 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 2533: if(Dummy[Tvard[k][1]==0]){
2534: if(Dummy[Tvard[k][2]==0]){
2535: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2536: }else{
2537: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2538: }
2539: }else{
2540: if(Dummy[Tvard[k][2]==0]){
2541: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2542: }else{
2543: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2544: }
2545: }
1.234 brouard 2546: }
1.138 brouard 2547: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2548: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2549: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2550: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2551: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2552: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2553: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2554:
1.126 brouard 2555: savm=oldm;
2556: oldm=newm;
1.209 brouard 2557:
2558: for(j=1; j<=nlstate; j++){
2559: max[j]=0.;
2560: min[j]=1.;
2561: }
2562: for(i=1;i<=nlstate;i++){
2563: sumnew=0;
2564: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2565: for(j=1; j<=nlstate; j++){
2566: prlim[i][j]= newm[i][j]/(1-sumnew);
2567: max[j]=FMAX(max[j],prlim[i][j]);
2568: min[j]=FMIN(min[j],prlim[i][j]);
2569: }
2570: }
2571:
1.126 brouard 2572: maxmax=0.;
1.209 brouard 2573: for(j=1; j<=nlstate; j++){
2574: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2575: maxmax=FMAX(maxmax,meandiff[j]);
2576: /* 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 2577: } /* j loop */
1.203 brouard 2578: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2579: /* 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 2580: if(maxmax < ftolpl){
1.209 brouard 2581: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2582: free_vector(min,1,nlstate);
2583: free_vector(max,1,nlstate);
2584: free_vector(meandiff,1,nlstate);
1.126 brouard 2585: return prlim;
2586: }
1.169 brouard 2587: } /* age loop */
1.208 brouard 2588: /* After some age loop it doesn't converge */
1.209 brouard 2589: 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 2590: 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 2591: /* 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); */
2592: free_vector(min,1,nlstate);
2593: free_vector(max,1,nlstate);
2594: free_vector(meandiff,1,nlstate);
1.208 brouard 2595:
1.169 brouard 2596: return prlim; /* should not reach here */
1.126 brouard 2597: }
2598:
1.217 brouard 2599:
2600: /**** Back Prevalence limit (stable or period prevalence) ****************/
2601:
1.218 brouard 2602: /* 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) */
2603: /* 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 2604: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2605: {
1.218 brouard 2606: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2607: matrix by transitions matrix until convergence is reached with precision ftolpl */
2608: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2609: /* Wx is row vector: population in state 1, population in state 2, population dead */
2610: /* or prevalence in state 1, prevalence in state 2, 0 */
2611: /* newm is the matrix after multiplications, its rows are identical at a factor */
2612: /* Initial matrix pimij */
2613: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2614: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2615: /* 0, 0 , 1} */
2616: /*
2617: * and after some iteration: */
2618: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2619: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2620: /* 0, 0 , 1} */
2621: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2622: /* {0.51571254859325999, 0.4842874514067399, */
2623: /* 0.51326036147820708, 0.48673963852179264} */
2624: /* If we start from prlim again, prlim tends to a constant matrix */
2625:
2626: int i, ii,j,k;
1.247 brouard 2627: int first=0;
1.217 brouard 2628: double *min, *max, *meandiff, maxmax,sumnew=0.;
2629: /* double **matprod2(); */ /* test */
2630: double **out, cov[NCOVMAX+1], **bmij();
2631: double **newm;
1.218 brouard 2632: double **dnewm, **doldm, **dsavm; /* for use */
2633: double **oldm, **savm; /* for use */
2634:
1.217 brouard 2635: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2636: int ncvloop=0;
2637:
2638: min=vector(1,nlstate);
2639: max=vector(1,nlstate);
2640: meandiff=vector(1,nlstate);
2641:
1.218 brouard 2642: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2643: oldm=oldms; savm=savms;
2644:
2645: /* Starting with matrix unity */
2646: for (ii=1;ii<=nlstate+ndeath;ii++)
2647: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2648: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2649: }
2650:
2651: cov[1]=1.;
2652:
2653: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2654: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2655: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2656: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2657: ncvloop++;
1.218 brouard 2658: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2659: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2660: /* Covariates have to be included here again */
2661: cov[2]=agefin;
2662: if(nagesqr==1)
2663: cov[3]= agefin*agefin;;
1.242 brouard 2664: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2665: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2666: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2667: /* 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)); */
2668: }
2669: /* for (k=1; k<=cptcovn;k++) { */
2670: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2671: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2672: /* /\* 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])]); *\/ */
2673: /* } */
2674: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2675: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2676: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2677: /* 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]); */
2678: }
2679: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2680: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2681: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2682: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2683: for (k=1; k<=cptcovage;k++){ /* For product with age */
2684: if(Dummy[Tvar[Tage[k]]]){
2685: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2686: } else{
2687: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2688: }
2689: /* 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]); */
2690: }
2691: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2692: /* 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]); */
2693: if(Dummy[Tvard[k][1]==0]){
2694: if(Dummy[Tvard[k][2]==0]){
2695: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2696: }else{
2697: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2698: }
2699: }else{
2700: if(Dummy[Tvard[k][2]==0]){
2701: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2702: }else{
2703: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2704: }
2705: }
1.217 brouard 2706: }
2707:
2708: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2709: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2710: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2711: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2712: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2713: /* ij should be linked to the correct index of cov */
2714: /* age and covariate values ij are in 'cov', but we need to pass
2715: * ij for the observed prevalence at age and status and covariate
2716: * number: prevacurrent[(int)agefin][ii][ij]
2717: */
2718: /* 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 *\/ */
2719: /* 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 *\/ */
2720: 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 2721: savm=oldm;
2722: oldm=newm;
2723: for(j=1; j<=nlstate; j++){
2724: max[j]=0.;
2725: min[j]=1.;
2726: }
2727: for(j=1; j<=nlstate; j++){
2728: for(i=1;i<=nlstate;i++){
1.234 brouard 2729: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2730: bprlim[i][j]= newm[i][j];
2731: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2732: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2733: }
2734: }
1.218 brouard 2735:
1.217 brouard 2736: maxmax=0.;
2737: for(i=1; i<=nlstate; i++){
2738: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2739: maxmax=FMAX(maxmax,meandiff[i]);
2740: /* 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); */
2741: } /* j loop */
2742: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2743: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2744: if(maxmax < ftolpl){
1.220 brouard 2745: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2746: free_vector(min,1,nlstate);
2747: free_vector(max,1,nlstate);
2748: free_vector(meandiff,1,nlstate);
2749: return bprlim;
2750: }
2751: } /* age loop */
2752: /* After some age loop it doesn't converge */
1.247 brouard 2753: if(first){
2754: first=1;
2755: 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\
2756: 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);
2757: }
2758: 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 2759: 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);
2760: /* 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); */
2761: free_vector(min,1,nlstate);
2762: free_vector(max,1,nlstate);
2763: free_vector(meandiff,1,nlstate);
2764:
2765: return bprlim; /* should not reach here */
2766: }
2767:
1.126 brouard 2768: /*************** transition probabilities ***************/
2769:
2770: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2771: {
1.138 brouard 2772: /* According to parameters values stored in x and the covariate's values stored in cov,
2773: computes the probability to be observed in state j being in state i by appying the
2774: model to the ncovmodel covariates (including constant and age).
2775: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2776: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2777: ncth covariate in the global vector x is given by the formula:
2778: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2779: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2780: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2781: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2782: Outputs ps[i][j] the probability to be observed in j being in j according to
2783: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2784: */
2785: double s1, lnpijopii;
1.126 brouard 2786: /*double t34;*/
1.164 brouard 2787: int i,j, nc, ii, jj;
1.126 brouard 2788:
1.223 brouard 2789: for(i=1; i<= nlstate; i++){
2790: for(j=1; j<i;j++){
2791: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2792: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2793: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2794: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2795: }
2796: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2797: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2798: }
2799: for(j=i+1; j<=nlstate+ndeath;j++){
2800: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2801: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2802: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2803: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2804: }
2805: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2806: }
2807: }
1.218 brouard 2808:
1.223 brouard 2809: for(i=1; i<= nlstate; i++){
2810: s1=0;
2811: for(j=1; j<i; j++){
2812: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2813: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2814: }
2815: for(j=i+1; j<=nlstate+ndeath; j++){
2816: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2817: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2818: }
2819: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2820: ps[i][i]=1./(s1+1.);
2821: /* Computing other pijs */
2822: for(j=1; j<i; j++)
2823: ps[i][j]= exp(ps[i][j])*ps[i][i];
2824: for(j=i+1; j<=nlstate+ndeath; j++)
2825: ps[i][j]= exp(ps[i][j])*ps[i][i];
2826: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2827: } /* end i */
1.218 brouard 2828:
1.223 brouard 2829: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2830: for(jj=1; jj<= nlstate+ndeath; jj++){
2831: ps[ii][jj]=0;
2832: ps[ii][ii]=1;
2833: }
2834: }
1.218 brouard 2835:
2836:
1.223 brouard 2837: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2838: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2839: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2840: /* } */
2841: /* printf("\n "); */
2842: /* } */
2843: /* printf("\n ");printf("%lf ",cov[2]);*/
2844: /*
2845: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2846: goto end;*/
1.223 brouard 2847: return ps;
1.126 brouard 2848: }
2849:
1.218 brouard 2850: /*************** backward transition probabilities ***************/
2851:
2852: /* 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 ) */
2853: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2854: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2855: {
1.222 brouard 2856: /* Computes the backward probability at age agefin and covariate ij
2857: * and returns in **ps as well as **bmij.
2858: */
1.218 brouard 2859: int i, ii, j,k;
1.222 brouard 2860:
2861: double **out, **pmij();
2862: double sumnew=0.;
1.218 brouard 2863: double agefin;
1.222 brouard 2864:
2865: double **dnewm, **dsavm, **doldm;
2866: double **bbmij;
2867:
1.218 brouard 2868: doldm=ddoldms; /* global pointers */
1.222 brouard 2869: dnewm=ddnewms;
2870: dsavm=ddsavms;
2871:
2872: agefin=cov[2];
2873: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2874: the observed prevalence (with this covariate ij) */
2875: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2876: /* We do have the matrix Px in savm and we need pij */
2877: for (j=1;j<=nlstate+ndeath;j++){
2878: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2879: for (ii=1;ii<=nlstate;ii++){
2880: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2881: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2882: for (ii=1;ii<=nlstate+ndeath;ii++){
2883: if(sumnew >= 1.e-10){
2884: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2885: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2886: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2887: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2888: /* }else */
2889: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2890: }else{
1.242 brouard 2891: ;
2892: /* 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 2893: }
2894: } /*End ii */
2895: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2896: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2897: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2898: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2899: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2900: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2901: /* left Product of this matrix by diag matrix of prevalences (savm) */
2902: for (j=1;j<=nlstate+ndeath;j++){
2903: for (ii=1;ii<=nlstate+ndeath;ii++){
2904: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2905: }
2906: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2907: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2908: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2909: /* end bmij */
2910: return ps;
1.218 brouard 2911: }
1.217 brouard 2912: /*************** transition probabilities ***************/
2913:
1.218 brouard 2914: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2915: {
2916: /* According to parameters values stored in x and the covariate's values stored in cov,
2917: computes the probability to be observed in state j being in state i by appying the
2918: model to the ncovmodel covariates (including constant and age).
2919: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2920: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2921: ncth covariate in the global vector x is given by the formula:
2922: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2923: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2924: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2925: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2926: Outputs ps[i][j] the probability to be observed in j being in j according to
2927: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2928: */
2929: double s1, lnpijopii;
2930: /*double t34;*/
2931: int i,j, nc, ii, jj;
2932:
1.234 brouard 2933: for(i=1; i<= nlstate; i++){
2934: for(j=1; j<i;j++){
2935: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2936: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2937: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2938: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2939: }
2940: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2941: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2942: }
2943: for(j=i+1; j<=nlstate+ndeath;j++){
2944: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2945: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2946: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2947: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2948: }
2949: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2950: }
2951: }
2952:
2953: for(i=1; i<= nlstate; i++){
2954: s1=0;
2955: for(j=1; j<i; j++){
2956: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2957: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2958: }
2959: for(j=i+1; j<=nlstate+ndeath; j++){
2960: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2961: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2962: }
2963: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2964: ps[i][i]=1./(s1+1.);
2965: /* Computing other pijs */
2966: for(j=1; j<i; j++)
2967: ps[i][j]= exp(ps[i][j])*ps[i][i];
2968: for(j=i+1; j<=nlstate+ndeath; j++)
2969: ps[i][j]= exp(ps[i][j])*ps[i][i];
2970: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2971: } /* end i */
2972:
2973: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2974: for(jj=1; jj<= nlstate+ndeath; jj++){
2975: ps[ii][jj]=0;
2976: ps[ii][ii]=1;
2977: }
2978: }
2979: /* Added for backcast */ /* Transposed matrix too */
2980: for(jj=1; jj<= nlstate+ndeath; jj++){
2981: s1=0.;
2982: for(ii=1; ii<= nlstate+ndeath; ii++){
2983: s1+=ps[ii][jj];
2984: }
2985: for(ii=1; ii<= nlstate; ii++){
2986: ps[ii][jj]=ps[ii][jj]/s1;
2987: }
2988: }
2989: /* Transposition */
2990: for(jj=1; jj<= nlstate+ndeath; jj++){
2991: for(ii=jj; ii<= nlstate+ndeath; ii++){
2992: s1=ps[ii][jj];
2993: ps[ii][jj]=ps[jj][ii];
2994: ps[jj][ii]=s1;
2995: }
2996: }
2997: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2998: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2999: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3000: /* } */
3001: /* printf("\n "); */
3002: /* } */
3003: /* printf("\n ");printf("%lf ",cov[2]);*/
3004: /*
3005: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3006: goto end;*/
3007: return ps;
1.217 brouard 3008: }
3009:
3010:
1.126 brouard 3011: /**************** Product of 2 matrices ******************/
3012:
1.145 brouard 3013: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3014: {
3015: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3016: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3017: /* in, b, out are matrice of pointers which should have been initialized
3018: before: only the contents of out is modified. The function returns
3019: a pointer to pointers identical to out */
1.145 brouard 3020: int i, j, k;
1.126 brouard 3021: for(i=nrl; i<= nrh; i++)
1.145 brouard 3022: for(k=ncolol; k<=ncoloh; k++){
3023: out[i][k]=0.;
3024: for(j=ncl; j<=nch; j++)
3025: out[i][k] +=in[i][j]*b[j][k];
3026: }
1.126 brouard 3027: return out;
3028: }
3029:
3030:
3031: /************* Higher Matrix Product ***************/
3032:
1.235 brouard 3033: 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 3034: {
1.218 brouard 3035: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3036: 'nhstepm*hstepm*stepm' months (i.e. until
3037: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3038: nhstepm*hstepm matrices.
3039: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3040: (typically every 2 years instead of every month which is too big
3041: for the memory).
3042: Model is determined by parameters x and covariates have to be
3043: included manually here.
3044:
3045: */
3046:
3047: int i, j, d, h, k;
1.131 brouard 3048: double **out, cov[NCOVMAX+1];
1.126 brouard 3049: double **newm;
1.187 brouard 3050: double agexact;
1.214 brouard 3051: double agebegin, ageend;
1.126 brouard 3052:
3053: /* Hstepm could be zero and should return the unit matrix */
3054: for (i=1;i<=nlstate+ndeath;i++)
3055: for (j=1;j<=nlstate+ndeath;j++){
3056: oldm[i][j]=(i==j ? 1.0 : 0.0);
3057: po[i][j][0]=(i==j ? 1.0 : 0.0);
3058: }
3059: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3060: for(h=1; h <=nhstepm; h++){
3061: for(d=1; d <=hstepm; d++){
3062: newm=savm;
3063: /* Covariates have to be included here again */
3064: cov[1]=1.;
1.214 brouard 3065: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3066: cov[2]=agexact;
3067: if(nagesqr==1)
1.227 brouard 3068: cov[3]= agexact*agexact;
1.235 brouard 3069: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3070: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3071: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3072: /* 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)); */
3073: }
3074: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3075: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3076: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3077: /* 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]); */
3078: }
3079: for (k=1; k<=cptcovage;k++){
3080: if(Dummy[Tvar[Tage[k]]]){
3081: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3082: } else{
3083: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3084: }
3085: /* 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]); */
3086: }
3087: for (k=1; k<=cptcovprod;k++){ /* */
3088: /* 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]); */
3089: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3090: }
3091: /* for (k=1; k<=cptcovn;k++) */
3092: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3093: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3094: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3095: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3096: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3097:
3098:
1.126 brouard 3099: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3100: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3101: /* right multiplication of oldm by the current matrix */
1.126 brouard 3102: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3103: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3104: /* if((int)age == 70){ */
3105: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3106: /* for(i=1; i<=nlstate+ndeath; i++) { */
3107: /* printf("%d pmmij ",i); */
3108: /* for(j=1;j<=nlstate+ndeath;j++) { */
3109: /* printf("%f ",pmmij[i][j]); */
3110: /* } */
3111: /* printf(" oldm "); */
3112: /* for(j=1;j<=nlstate+ndeath;j++) { */
3113: /* printf("%f ",oldm[i][j]); */
3114: /* } */
3115: /* printf("\n"); */
3116: /* } */
3117: /* } */
1.126 brouard 3118: savm=oldm;
3119: oldm=newm;
3120: }
3121: for(i=1; i<=nlstate+ndeath; i++)
3122: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3123: po[i][j][h]=newm[i][j];
3124: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3125: }
1.128 brouard 3126: /*printf("h=%d ",h);*/
1.126 brouard 3127: } /* end h */
1.218 brouard 3128: /* printf("\n H=%d \n",h); */
1.126 brouard 3129: return po;
3130: }
3131:
1.217 brouard 3132: /************* Higher Back Matrix Product ***************/
1.218 brouard 3133: /* 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 3134: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3135: {
1.218 brouard 3136: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3137: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3138: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3139: nhstepm*hstepm matrices.
3140: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3141: (typically every 2 years instead of every month which is too big
1.217 brouard 3142: for the memory).
1.218 brouard 3143: Model is determined by parameters x and covariates have to be
3144: included manually here.
1.217 brouard 3145:
1.222 brouard 3146: */
1.217 brouard 3147:
3148: int i, j, d, h, k;
3149: double **out, cov[NCOVMAX+1];
3150: double **newm;
3151: double agexact;
3152: double agebegin, ageend;
1.222 brouard 3153: double **oldm, **savm;
1.217 brouard 3154:
1.222 brouard 3155: oldm=oldms;savm=savms;
1.217 brouard 3156: /* Hstepm could be zero and should return the unit matrix */
3157: for (i=1;i<=nlstate+ndeath;i++)
3158: for (j=1;j<=nlstate+ndeath;j++){
3159: oldm[i][j]=(i==j ? 1.0 : 0.0);
3160: po[i][j][0]=(i==j ? 1.0 : 0.0);
3161: }
3162: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3163: for(h=1; h <=nhstepm; h++){
3164: for(d=1; d <=hstepm; d++){
3165: newm=savm;
3166: /* Covariates have to be included here again */
3167: cov[1]=1.;
3168: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3169: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3170: cov[2]=agexact;
3171: if(nagesqr==1)
1.222 brouard 3172: cov[3]= agexact*agexact;
1.218 brouard 3173: for (k=1; k<=cptcovn;k++)
1.222 brouard 3174: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3175: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3176: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3177: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3178: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3179: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3180: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3181: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3182: /* 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 3183:
3184:
1.217 brouard 3185: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3186: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3187: /* Careful transposed matrix */
1.222 brouard 3188: /* age is in cov[2] */
1.218 brouard 3189: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3190: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3191: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3192: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3193: /* if((int)age == 70){ */
3194: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3195: /* for(i=1; i<=nlstate+ndeath; i++) { */
3196: /* printf("%d pmmij ",i); */
3197: /* for(j=1;j<=nlstate+ndeath;j++) { */
3198: /* printf("%f ",pmmij[i][j]); */
3199: /* } */
3200: /* printf(" oldm "); */
3201: /* for(j=1;j<=nlstate+ndeath;j++) { */
3202: /* printf("%f ",oldm[i][j]); */
3203: /* } */
3204: /* printf("\n"); */
3205: /* } */
3206: /* } */
3207: savm=oldm;
3208: oldm=newm;
3209: }
3210: for(i=1; i<=nlstate+ndeath; i++)
3211: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3212: po[i][j][h]=newm[i][j];
3213: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3214: }
3215: /*printf("h=%d ",h);*/
3216: } /* end h */
1.222 brouard 3217: /* printf("\n H=%d \n",h); */
1.217 brouard 3218: return po;
3219: }
3220:
3221:
1.162 brouard 3222: #ifdef NLOPT
3223: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3224: double fret;
3225: double *xt;
3226: int j;
3227: myfunc_data *d2 = (myfunc_data *) pd;
3228: /* xt = (p1-1); */
3229: xt=vector(1,n);
3230: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3231:
3232: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3233: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3234: printf("Function = %.12lf ",fret);
3235: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3236: printf("\n");
3237: free_vector(xt,1,n);
3238: return fret;
3239: }
3240: #endif
1.126 brouard 3241:
3242: /*************** log-likelihood *************/
3243: double func( double *x)
3244: {
1.226 brouard 3245: int i, ii, j, k, mi, d, kk;
3246: int ioffset=0;
3247: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3248: double **out;
3249: double lli; /* Individual log likelihood */
3250: int s1, s2;
1.228 brouard 3251: 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 3252: double bbh, survp;
3253: long ipmx;
3254: double agexact;
3255: /*extern weight */
3256: /* We are differentiating ll according to initial status */
3257: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3258: /*for(i=1;i<imx;i++)
3259: printf(" %d\n",s[4][i]);
3260: */
1.162 brouard 3261:
1.226 brouard 3262: ++countcallfunc;
1.162 brouard 3263:
1.226 brouard 3264: cov[1]=1.;
1.126 brouard 3265:
1.226 brouard 3266: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3267: ioffset=0;
1.226 brouard 3268: if(mle==1){
3269: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3270: /* Computes the values of the ncovmodel covariates of the model
3271: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3272: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3273: to be observed in j being in i according to the model.
3274: */
1.243 brouard 3275: ioffset=2+nagesqr ;
1.233 brouard 3276: /* Fixed */
1.234 brouard 3277: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3278: 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)*/
3279: }
1.226 brouard 3280: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3281: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3282: has been calculated etc */
3283: /* For an individual i, wav[i] gives the number of effective waves */
3284: /* We compute the contribution to Likelihood of each effective transition
3285: mw[mi][i] is real wave of the mi th effectve wave */
3286: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3287: s2=s[mw[mi+1][i]][i];
3288: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3289: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3290: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3291: */
3292: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3293: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3294: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3295: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3296: }
3297: for (ii=1;ii<=nlstate+ndeath;ii++)
3298: for (j=1;j<=nlstate+ndeath;j++){
3299: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3300: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3301: }
3302: for(d=0; d<dh[mi][i]; d++){
3303: newm=savm;
3304: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3305: cov[2]=agexact;
3306: if(nagesqr==1)
3307: cov[3]= agexact*agexact; /* Should be changed here */
3308: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3309: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3310: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3311: else
3312: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3313: }
3314: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3315: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3316: savm=oldm;
3317: oldm=newm;
3318: } /* end mult */
3319:
3320: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3321: /* But now since version 0.9 we anticipate for bias at large stepm.
3322: * If stepm is larger than one month (smallest stepm) and if the exact delay
3323: * (in months) between two waves is not a multiple of stepm, we rounded to
3324: * the nearest (and in case of equal distance, to the lowest) interval but now
3325: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3326: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3327: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3328: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3329: * -stepm/2 to stepm/2 .
3330: * For stepm=1 the results are the same as for previous versions of Imach.
3331: * For stepm > 1 the results are less biased than in previous versions.
3332: */
1.234 brouard 3333: s1=s[mw[mi][i]][i];
3334: s2=s[mw[mi+1][i]][i];
3335: bbh=(double)bh[mi][i]/(double)stepm;
3336: /* bias bh is positive if real duration
3337: * is higher than the multiple of stepm and negative otherwise.
3338: */
3339: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3340: if( s2 > nlstate){
3341: /* i.e. if s2 is a death state and if the date of death is known
3342: then the contribution to the likelihood is the probability to
3343: die between last step unit time and current step unit time,
3344: which is also equal to probability to die before dh
3345: minus probability to die before dh-stepm .
3346: In version up to 0.92 likelihood was computed
3347: as if date of death was unknown. Death was treated as any other
3348: health state: the date of the interview describes the actual state
3349: and not the date of a change in health state. The former idea was
3350: to consider that at each interview the state was recorded
3351: (healthy, disable or death) and IMaCh was corrected; but when we
3352: introduced the exact date of death then we should have modified
3353: the contribution of an exact death to the likelihood. This new
3354: contribution is smaller and very dependent of the step unit
3355: stepm. It is no more the probability to die between last interview
3356: and month of death but the probability to survive from last
3357: interview up to one month before death multiplied by the
3358: probability to die within a month. Thanks to Chris
3359: Jackson for correcting this bug. Former versions increased
3360: mortality artificially. The bad side is that we add another loop
3361: which slows down the processing. The difference can be up to 10%
3362: lower mortality.
3363: */
3364: /* If, at the beginning of the maximization mostly, the
3365: cumulative probability or probability to be dead is
3366: constant (ie = 1) over time d, the difference is equal to
3367: 0. out[s1][3] = savm[s1][3]: probability, being at state
3368: s1 at precedent wave, to be dead a month before current
3369: wave is equal to probability, being at state s1 at
3370: precedent wave, to be dead at mont of the current
3371: wave. Then the observed probability (that this person died)
3372: is null according to current estimated parameter. In fact,
3373: it should be very low but not zero otherwise the log go to
3374: infinity.
3375: */
1.183 brouard 3376: /* #ifdef INFINITYORIGINAL */
3377: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3378: /* #else */
3379: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3380: /* lli=log(mytinydouble); */
3381: /* else */
3382: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3383: /* #endif */
1.226 brouard 3384: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3385:
1.226 brouard 3386: } else if ( s2==-1 ) { /* alive */
3387: for (j=1,survp=0. ; j<=nlstate; j++)
3388: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3389: /*survp += out[s1][j]; */
3390: lli= log(survp);
3391: }
3392: else if (s2==-4) {
3393: for (j=3,survp=0. ; j<=nlstate; j++)
3394: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3395: lli= log(survp);
3396: }
3397: else if (s2==-5) {
3398: for (j=1,survp=0. ; j<=2; j++)
3399: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3400: lli= log(survp);
3401: }
3402: else{
3403: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3404: /* 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 */
3405: }
3406: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3407: /*if(lli ==000.0)*/
3408: /*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); */
3409: ipmx +=1;
3410: sw += weight[i];
3411: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3412: /* if (lli < log(mytinydouble)){ */
3413: /* 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); */
3414: /* 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]); */
3415: /* } */
3416: } /* end of wave */
3417: } /* end of individual */
3418: } else if(mle==2){
3419: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3420: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3421: for(mi=1; mi<= wav[i]-1; mi++){
3422: for (ii=1;ii<=nlstate+ndeath;ii++)
3423: for (j=1;j<=nlstate+ndeath;j++){
3424: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3425: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3426: }
3427: for(d=0; d<=dh[mi][i]; d++){
3428: newm=savm;
3429: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3430: cov[2]=agexact;
3431: if(nagesqr==1)
3432: cov[3]= agexact*agexact;
3433: for (kk=1; kk<=cptcovage;kk++) {
3434: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3435: }
3436: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3437: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3438: savm=oldm;
3439: oldm=newm;
3440: } /* end mult */
3441:
3442: s1=s[mw[mi][i]][i];
3443: s2=s[mw[mi+1][i]][i];
3444: bbh=(double)bh[mi][i]/(double)stepm;
3445: 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 */
3446: ipmx +=1;
3447: sw += weight[i];
3448: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3449: } /* end of wave */
3450: } /* end of individual */
3451: } else if(mle==3){ /* exponential inter-extrapolation */
3452: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3453: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3454: for(mi=1; mi<= wav[i]-1; mi++){
3455: for (ii=1;ii<=nlstate+ndeath;ii++)
3456: for (j=1;j<=nlstate+ndeath;j++){
3457: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3458: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3459: }
3460: for(d=0; d<dh[mi][i]; d++){
3461: newm=savm;
3462: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3463: cov[2]=agexact;
3464: if(nagesqr==1)
3465: cov[3]= agexact*agexact;
3466: for (kk=1; kk<=cptcovage;kk++) {
3467: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3468: }
3469: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3470: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3471: savm=oldm;
3472: oldm=newm;
3473: } /* end mult */
3474:
3475: s1=s[mw[mi][i]][i];
3476: s2=s[mw[mi+1][i]][i];
3477: bbh=(double)bh[mi][i]/(double)stepm;
3478: 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 */
3479: ipmx +=1;
3480: sw += weight[i];
3481: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3482: } /* end of wave */
3483: } /* end of individual */
3484: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3485: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3486: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3487: for(mi=1; mi<= wav[i]-1; mi++){
3488: for (ii=1;ii<=nlstate+ndeath;ii++)
3489: for (j=1;j<=nlstate+ndeath;j++){
3490: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3491: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3492: }
3493: for(d=0; d<dh[mi][i]; d++){
3494: newm=savm;
3495: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3496: cov[2]=agexact;
3497: if(nagesqr==1)
3498: cov[3]= agexact*agexact;
3499: for (kk=1; kk<=cptcovage;kk++) {
3500: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3501: }
1.126 brouard 3502:
1.226 brouard 3503: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3504: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3505: savm=oldm;
3506: oldm=newm;
3507: } /* end mult */
3508:
3509: s1=s[mw[mi][i]][i];
3510: s2=s[mw[mi+1][i]][i];
3511: if( s2 > nlstate){
3512: lli=log(out[s1][s2] - savm[s1][s2]);
3513: } else if ( s2==-1 ) { /* alive */
3514: for (j=1,survp=0. ; j<=nlstate; j++)
3515: survp += out[s1][j];
3516: lli= log(survp);
3517: }else{
3518: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3519: }
3520: ipmx +=1;
3521: sw += weight[i];
3522: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3523: /* 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 3524: } /* end of wave */
3525: } /* end of individual */
3526: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3527: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3528: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3529: for(mi=1; mi<= wav[i]-1; mi++){
3530: for (ii=1;ii<=nlstate+ndeath;ii++)
3531: for (j=1;j<=nlstate+ndeath;j++){
3532: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3533: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3534: }
3535: for(d=0; d<dh[mi][i]; d++){
3536: newm=savm;
3537: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3538: cov[2]=agexact;
3539: if(nagesqr==1)
3540: cov[3]= agexact*agexact;
3541: for (kk=1; kk<=cptcovage;kk++) {
3542: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3543: }
1.126 brouard 3544:
1.226 brouard 3545: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3546: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3547: savm=oldm;
3548: oldm=newm;
3549: } /* end mult */
3550:
3551: s1=s[mw[mi][i]][i];
3552: s2=s[mw[mi+1][i]][i];
3553: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3554: ipmx +=1;
3555: sw += weight[i];
3556: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3557: /*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]);*/
3558: } /* end of wave */
3559: } /* end of individual */
3560: } /* End of if */
3561: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3562: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3563: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3564: return -l;
1.126 brouard 3565: }
3566:
3567: /*************** log-likelihood *************/
3568: double funcone( double *x)
3569: {
1.228 brouard 3570: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3571: int i, ii, j, k, mi, d, kk;
1.228 brouard 3572: int ioffset=0;
1.131 brouard 3573: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3574: double **out;
3575: double lli; /* Individual log likelihood */
3576: double llt;
3577: int s1, s2;
1.228 brouard 3578: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3579:
1.126 brouard 3580: double bbh, survp;
1.187 brouard 3581: double agexact;
1.214 brouard 3582: double agebegin, ageend;
1.126 brouard 3583: /*extern weight */
3584: /* We are differentiating ll according to initial status */
3585: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3586: /*for(i=1;i<imx;i++)
3587: printf(" %d\n",s[4][i]);
3588: */
3589: cov[1]=1.;
3590:
3591: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3592: ioffset=0;
3593: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3594: /* ioffset=2+nagesqr+cptcovage; */
3595: ioffset=2+nagesqr;
1.232 brouard 3596: /* Fixed */
1.224 brouard 3597: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3598: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3599: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3600: 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)*/
3601: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3602: /* cov[2+6]=covar[Tvar[6]][i]; */
3603: /* cov[2+6]=covar[2][i]; V2 */
3604: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3605: /* cov[2+7]=covar[Tvar[7]][i]; */
3606: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3607: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3608: /* cov[2+9]=covar[Tvar[9]][i]; */
3609: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3610: }
1.232 brouard 3611: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3612: /* 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?)*\/ */
3613: /* } */
1.231 brouard 3614: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3615: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3616: /* } */
1.225 brouard 3617:
1.233 brouard 3618:
3619: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3620: /* Wave varying (but not age varying) */
3621: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3622: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3623: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3624: }
1.232 brouard 3625: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3626: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3627: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3628: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3629: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3630: /* 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 3631: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3632: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3633: /* /\* 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]); *\/ */
3634: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3635: /* } */
1.126 brouard 3636: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3637: for (j=1;j<=nlstate+ndeath;j++){
3638: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3639: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3640: }
1.214 brouard 3641:
3642: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3643: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3644: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3645: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3646: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3647: and mw[mi+1][i]. dh depends on stepm.*/
3648: newm=savm;
1.247 brouard 3649: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3650: cov[2]=agexact;
3651: if(nagesqr==1)
3652: cov[3]= agexact*agexact;
3653: for (kk=1; kk<=cptcovage;kk++) {
3654: if(!FixedV[Tvar[Tage[kk]]])
3655: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3656: else
3657: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3658: }
3659: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3660: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3661: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3662: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3663: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3664: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3665: savm=oldm;
3666: oldm=newm;
1.126 brouard 3667: } /* end mult */
3668:
3669: s1=s[mw[mi][i]][i];
3670: s2=s[mw[mi+1][i]][i];
1.217 brouard 3671: /* if(s2==-1){ */
3672: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3673: /* /\* exit(1); *\/ */
3674: /* } */
1.126 brouard 3675: bbh=(double)bh[mi][i]/(double)stepm;
3676: /* bias is positive if real duration
3677: * is higher than the multiple of stepm and negative otherwise.
3678: */
3679: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3680: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3681: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3682: for (j=1,survp=0. ; j<=nlstate; j++)
3683: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3684: lli= log(survp);
1.126 brouard 3685: }else if (mle==1){
1.242 brouard 3686: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3687: } else if(mle==2){
1.242 brouard 3688: 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 3689: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3690: 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 3691: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3692: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3693: } else{ /* mle=0 back to 1 */
1.242 brouard 3694: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3695: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3696: } /* End of if */
3697: ipmx +=1;
3698: sw += weight[i];
3699: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3700: /*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 3701: if(globpr){
1.246 brouard 3702: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3703: %11.6f %11.6f %11.6f ", \
1.242 brouard 3704: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3705: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3706: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3707: llt +=ll[k]*gipmx/gsw;
3708: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3709: }
3710: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3711: }
1.232 brouard 3712: } /* end of wave */
3713: } /* end of individual */
3714: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3715: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3716: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3717: if(globpr==0){ /* First time we count the contributions and weights */
3718: gipmx=ipmx;
3719: gsw=sw;
3720: }
3721: return -l;
1.126 brouard 3722: }
3723:
3724:
3725: /*************** function likelione ***********/
3726: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3727: {
3728: /* This routine should help understanding what is done with
3729: the selection of individuals/waves and
3730: to check the exact contribution to the likelihood.
3731: Plotting could be done.
3732: */
3733: int k;
3734:
3735: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3736: strcpy(fileresilk,"ILK_");
1.202 brouard 3737: strcat(fileresilk,fileresu);
1.126 brouard 3738: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3739: printf("Problem with resultfile: %s\n", fileresilk);
3740: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3741: }
1.214 brouard 3742: 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");
3743: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3744: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3745: for(k=1; k<=nlstate; k++)
3746: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3747: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3748: }
3749:
3750: *fretone=(*funcone)(p);
3751: if(*globpri !=0){
3752: fclose(ficresilk);
1.205 brouard 3753: if (mle ==0)
3754: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3755: else if(mle >=1)
3756: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3757: 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 3758:
1.208 brouard 3759:
3760: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3761: 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 3762: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3763: }
1.207 brouard 3764: 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 3765: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3766: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3767: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3768: fflush(fichtm);
1.205 brouard 3769: }
1.126 brouard 3770: return;
3771: }
3772:
3773:
3774: /*********** Maximum Likelihood Estimation ***************/
3775:
3776: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3777: {
1.165 brouard 3778: int i,j, iter=0;
1.126 brouard 3779: double **xi;
3780: double fret;
3781: double fretone; /* Only one call to likelihood */
3782: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3783:
3784: #ifdef NLOPT
3785: int creturn;
3786: nlopt_opt opt;
3787: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3788: double *lb;
3789: double minf; /* the minimum objective value, upon return */
3790: double * p1; /* Shifted parameters from 0 instead of 1 */
3791: myfunc_data dinst, *d = &dinst;
3792: #endif
3793:
3794:
1.126 brouard 3795: xi=matrix(1,npar,1,npar);
3796: for (i=1;i<=npar;i++)
3797: for (j=1;j<=npar;j++)
3798: xi[i][j]=(i==j ? 1.0 : 0.0);
3799: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3800: strcpy(filerespow,"POW_");
1.126 brouard 3801: strcat(filerespow,fileres);
3802: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3803: printf("Problem with resultfile: %s\n", filerespow);
3804: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3805: }
3806: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3807: for (i=1;i<=nlstate;i++)
3808: for(j=1;j<=nlstate+ndeath;j++)
3809: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3810: fprintf(ficrespow,"\n");
1.162 brouard 3811: #ifdef POWELL
1.126 brouard 3812: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3813: #endif
1.126 brouard 3814:
1.162 brouard 3815: #ifdef NLOPT
3816: #ifdef NEWUOA
3817: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3818: #else
3819: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3820: #endif
3821: lb=vector(0,npar-1);
3822: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3823: nlopt_set_lower_bounds(opt, lb);
3824: nlopt_set_initial_step1(opt, 0.1);
3825:
3826: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3827: d->function = func;
3828: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3829: nlopt_set_min_objective(opt, myfunc, d);
3830: nlopt_set_xtol_rel(opt, ftol);
3831: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3832: printf("nlopt failed! %d\n",creturn);
3833: }
3834: else {
3835: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3836: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3837: iter=1; /* not equal */
3838: }
3839: nlopt_destroy(opt);
3840: #endif
1.126 brouard 3841: free_matrix(xi,1,npar,1,npar);
3842: fclose(ficrespow);
1.203 brouard 3843: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3844: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3845: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3846:
3847: }
3848:
3849: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3850: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3851: {
3852: double **a,**y,*x,pd;
1.203 brouard 3853: /* double **hess; */
1.164 brouard 3854: int i, j;
1.126 brouard 3855: int *indx;
3856:
3857: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3858: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3859: void lubksb(double **a, int npar, int *indx, double b[]) ;
3860: void ludcmp(double **a, int npar, int *indx, double *d) ;
3861: double gompertz(double p[]);
1.203 brouard 3862: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3863:
3864: printf("\nCalculation of the hessian matrix. Wait...\n");
3865: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3866: for (i=1;i<=npar;i++){
1.203 brouard 3867: printf("%d-",i);fflush(stdout);
3868: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3869:
3870: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3871:
3872: /* printf(" %f ",p[i]);
3873: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3874: }
3875:
3876: for (i=1;i<=npar;i++) {
3877: for (j=1;j<=npar;j++) {
3878: if (j>i) {
1.203 brouard 3879: printf(".%d-%d",i,j);fflush(stdout);
3880: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3881: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3882:
3883: hess[j][i]=hess[i][j];
3884: /*printf(" %lf ",hess[i][j]);*/
3885: }
3886: }
3887: }
3888: printf("\n");
3889: fprintf(ficlog,"\n");
3890:
3891: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3892: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3893:
3894: a=matrix(1,npar,1,npar);
3895: y=matrix(1,npar,1,npar);
3896: x=vector(1,npar);
3897: indx=ivector(1,npar);
3898: for (i=1;i<=npar;i++)
3899: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3900: ludcmp(a,npar,indx,&pd);
3901:
3902: for (j=1;j<=npar;j++) {
3903: for (i=1;i<=npar;i++) x[i]=0;
3904: x[j]=1;
3905: lubksb(a,npar,indx,x);
3906: for (i=1;i<=npar;i++){
3907: matcov[i][j]=x[i];
3908: }
3909: }
3910:
3911: printf("\n#Hessian matrix#\n");
3912: fprintf(ficlog,"\n#Hessian matrix#\n");
3913: for (i=1;i<=npar;i++) {
3914: for (j=1;j<=npar;j++) {
1.203 brouard 3915: printf("%.6e ",hess[i][j]);
3916: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3917: }
3918: printf("\n");
3919: fprintf(ficlog,"\n");
3920: }
3921:
1.203 brouard 3922: /* printf("\n#Covariance matrix#\n"); */
3923: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3924: /* for (i=1;i<=npar;i++) { */
3925: /* for (j=1;j<=npar;j++) { */
3926: /* printf("%.6e ",matcov[i][j]); */
3927: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3928: /* } */
3929: /* printf("\n"); */
3930: /* fprintf(ficlog,"\n"); */
3931: /* } */
3932:
1.126 brouard 3933: /* Recompute Inverse */
1.203 brouard 3934: /* for (i=1;i<=npar;i++) */
3935: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3936: /* ludcmp(a,npar,indx,&pd); */
3937:
3938: /* printf("\n#Hessian matrix recomputed#\n"); */
3939:
3940: /* for (j=1;j<=npar;j++) { */
3941: /* for (i=1;i<=npar;i++) x[i]=0; */
3942: /* x[j]=1; */
3943: /* lubksb(a,npar,indx,x); */
3944: /* for (i=1;i<=npar;i++){ */
3945: /* y[i][j]=x[i]; */
3946: /* printf("%.3e ",y[i][j]); */
3947: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3948: /* } */
3949: /* printf("\n"); */
3950: /* fprintf(ficlog,"\n"); */
3951: /* } */
3952:
3953: /* Verifying the inverse matrix */
3954: #ifdef DEBUGHESS
3955: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3956:
1.203 brouard 3957: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3958: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3959:
3960: for (j=1;j<=npar;j++) {
3961: for (i=1;i<=npar;i++){
1.203 brouard 3962: printf("%.2f ",y[i][j]);
3963: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3964: }
3965: printf("\n");
3966: fprintf(ficlog,"\n");
3967: }
1.203 brouard 3968: #endif
1.126 brouard 3969:
3970: free_matrix(a,1,npar,1,npar);
3971: free_matrix(y,1,npar,1,npar);
3972: free_vector(x,1,npar);
3973: free_ivector(indx,1,npar);
1.203 brouard 3974: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3975:
3976:
3977: }
3978:
3979: /*************** hessian matrix ****************/
3980: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3981: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3982: int i;
3983: int l=1, lmax=20;
1.203 brouard 3984: double k1,k2, res, fx;
1.132 brouard 3985: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3986: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3987: int k=0,kmax=10;
3988: double l1;
3989:
3990: fx=func(x);
3991: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3992: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3993: l1=pow(10,l);
3994: delts=delt;
3995: for(k=1 ; k <kmax; k=k+1){
3996: delt = delta*(l1*k);
3997: p2[theta]=x[theta] +delt;
1.145 brouard 3998: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3999: p2[theta]=x[theta]-delt;
4000: k2=func(p2)-fx;
4001: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4002: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4003:
1.203 brouard 4004: #ifdef DEBUGHESSII
1.126 brouard 4005: 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);
4006: 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);
4007: #endif
4008: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4009: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4010: k=kmax;
4011: }
4012: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4013: k=kmax; l=lmax*10;
1.126 brouard 4014: }
4015: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4016: delts=delt;
4017: }
1.203 brouard 4018: } /* End loop k */
1.126 brouard 4019: }
4020: delti[theta]=delts;
4021: return res;
4022:
4023: }
4024:
1.203 brouard 4025: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4026: {
4027: int i;
1.164 brouard 4028: int l=1, lmax=20;
1.126 brouard 4029: double k1,k2,k3,k4,res,fx;
1.132 brouard 4030: double p2[MAXPARM+1];
1.203 brouard 4031: int k, kmax=1;
4032: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4033:
4034: int firstime=0;
1.203 brouard 4035:
1.126 brouard 4036: fx=func(x);
1.203 brouard 4037: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4038: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4039: p2[thetai]=x[thetai]+delti[thetai]*k;
4040: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4041: k1=func(p2)-fx;
4042:
1.203 brouard 4043: p2[thetai]=x[thetai]+delti[thetai]*k;
4044: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4045: k2=func(p2)-fx;
4046:
1.203 brouard 4047: p2[thetai]=x[thetai]-delti[thetai]*k;
4048: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4049: k3=func(p2)-fx;
4050:
1.203 brouard 4051: p2[thetai]=x[thetai]-delti[thetai]*k;
4052: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4053: k4=func(p2)-fx;
1.203 brouard 4054: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4055: if(k1*k2*k3*k4 <0.){
1.208 brouard 4056: firstime=1;
1.203 brouard 4057: kmax=kmax+10;
1.208 brouard 4058: }
4059: if(kmax >=10 || firstime ==1){
1.246 brouard 4060: 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);
4061: 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 4062: 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);
4063: 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);
4064: }
4065: #ifdef DEBUGHESSIJ
4066: v1=hess[thetai][thetai];
4067: v2=hess[thetaj][thetaj];
4068: cv12=res;
4069: /* Computing eigen value of Hessian matrix */
4070: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4071: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4072: if ((lc2 <0) || (lc1 <0) ){
4073: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4074: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4075: 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);
4076: 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);
4077: }
1.126 brouard 4078: #endif
4079: }
4080: return res;
4081: }
4082:
1.203 brouard 4083: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4084: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4085: /* { */
4086: /* int i; */
4087: /* int l=1, lmax=20; */
4088: /* double k1,k2,k3,k4,res,fx; */
4089: /* double p2[MAXPARM+1]; */
4090: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4091: /* int k=0,kmax=10; */
4092: /* double l1; */
4093:
4094: /* fx=func(x); */
4095: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4096: /* l1=pow(10,l); */
4097: /* delts=delt; */
4098: /* for(k=1 ; k <kmax; k=k+1){ */
4099: /* delt = delti*(l1*k); */
4100: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4101: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4102: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4103: /* k1=func(p2)-fx; */
4104:
4105: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4106: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4107: /* k2=func(p2)-fx; */
4108:
4109: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4110: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4111: /* k3=func(p2)-fx; */
4112:
4113: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4114: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4115: /* k4=func(p2)-fx; */
4116: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4117: /* #ifdef DEBUGHESSIJ */
4118: /* 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); */
4119: /* 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); */
4120: /* #endif */
4121: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4122: /* k=kmax; */
4123: /* } */
4124: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4125: /* k=kmax; l=lmax*10; */
4126: /* } */
4127: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4128: /* delts=delt; */
4129: /* } */
4130: /* } /\* End loop k *\/ */
4131: /* } */
4132: /* delti[theta]=delts; */
4133: /* return res; */
4134: /* } */
4135:
4136:
1.126 brouard 4137: /************** Inverse of matrix **************/
4138: void ludcmp(double **a, int n, int *indx, double *d)
4139: {
4140: int i,imax,j,k;
4141: double big,dum,sum,temp;
4142: double *vv;
4143:
4144: vv=vector(1,n);
4145: *d=1.0;
4146: for (i=1;i<=n;i++) {
4147: big=0.0;
4148: for (j=1;j<=n;j++)
4149: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4150: if (big == 0.0){
4151: printf(" Singular Hessian matrix at row %d:\n",i);
4152: for (j=1;j<=n;j++) {
4153: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4154: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4155: }
4156: fflush(ficlog);
4157: fclose(ficlog);
4158: nrerror("Singular matrix in routine ludcmp");
4159: }
1.126 brouard 4160: vv[i]=1.0/big;
4161: }
4162: for (j=1;j<=n;j++) {
4163: for (i=1;i<j;i++) {
4164: sum=a[i][j];
4165: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4166: a[i][j]=sum;
4167: }
4168: big=0.0;
4169: for (i=j;i<=n;i++) {
4170: sum=a[i][j];
4171: for (k=1;k<j;k++)
4172: sum -= a[i][k]*a[k][j];
4173: a[i][j]=sum;
4174: if ( (dum=vv[i]*fabs(sum)) >= big) {
4175: big=dum;
4176: imax=i;
4177: }
4178: }
4179: if (j != imax) {
4180: for (k=1;k<=n;k++) {
4181: dum=a[imax][k];
4182: a[imax][k]=a[j][k];
4183: a[j][k]=dum;
4184: }
4185: *d = -(*d);
4186: vv[imax]=vv[j];
4187: }
4188: indx[j]=imax;
4189: if (a[j][j] == 0.0) a[j][j]=TINY;
4190: if (j != n) {
4191: dum=1.0/(a[j][j]);
4192: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4193: }
4194: }
4195: free_vector(vv,1,n); /* Doesn't work */
4196: ;
4197: }
4198:
4199: void lubksb(double **a, int n, int *indx, double b[])
4200: {
4201: int i,ii=0,ip,j;
4202: double sum;
4203:
4204: for (i=1;i<=n;i++) {
4205: ip=indx[i];
4206: sum=b[ip];
4207: b[ip]=b[i];
4208: if (ii)
4209: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4210: else if (sum) ii=i;
4211: b[i]=sum;
4212: }
4213: for (i=n;i>=1;i--) {
4214: sum=b[i];
4215: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4216: b[i]=sum/a[i][i];
4217: }
4218: }
4219:
4220: void pstamp(FILE *fichier)
4221: {
1.196 brouard 4222: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4223: }
4224:
1.253 brouard 4225: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4226:
4227: /* y=a+bx regression */
4228: double sumx = 0.0; /* sum of x */
4229: double sumx2 = 0.0; /* sum of x**2 */
4230: double sumxy = 0.0; /* sum of x * y */
4231: double sumy = 0.0; /* sum of y */
4232: double sumy2 = 0.0; /* sum of y**2 */
4233: double sume2; /* sum of square or residuals */
4234: double yhat;
4235:
4236: double denom=0;
4237: int i;
4238: int ne=*no;
4239:
4240: for ( i=ifi, ne=0;i<=ila;i++) {
4241: if(!isfinite(x[i]) || !isfinite(y[i])){
4242: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4243: continue;
4244: }
4245: ne=ne+1;
4246: sumx += x[i];
4247: sumx2 += x[i]*x[i];
4248: sumxy += x[i] * y[i];
4249: sumy += y[i];
4250: sumy2 += y[i]*y[i];
4251: denom = (ne * sumx2 - sumx*sumx);
4252: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4253: }
4254:
4255: denom = (ne * sumx2 - sumx*sumx);
4256: if (denom == 0) {
4257: // vertical, slope m is infinity
4258: *b = INFINITY;
4259: *a = 0;
4260: if (r) *r = 0;
4261: return 1;
4262: }
4263:
4264: *b = (ne * sumxy - sumx * sumy) / denom;
4265: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4266: if (r!=NULL) {
4267: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4268: sqrt((sumx2 - sumx*sumx/ne) *
4269: (sumy2 - sumy*sumy/ne));
4270: }
4271: *no=ne;
4272: for ( i=ifi, ne=0;i<=ila;i++) {
4273: if(!isfinite(x[i]) || !isfinite(y[i])){
4274: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4275: continue;
4276: }
4277: ne=ne+1;
4278: yhat = y[i] - *a -*b* x[i];
4279: sume2 += yhat * yhat ;
4280:
4281: denom = (ne * sumx2 - sumx*sumx);
4282: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4283: }
4284: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4285: *sa= *sb * sqrt(sumx2/ne);
4286:
4287: return 0;
4288: }
4289:
1.126 brouard 4290: /************ Frequencies ********************/
1.251 brouard 4291: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4292: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4293: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4294: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4295:
1.253 brouard 4296: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4297: int iind=0, iage=0;
4298: int mi; /* Effective wave */
4299: int first;
4300: double ***freq; /* Frequencies */
1.253 brouard 4301: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4302: int no;
1.226 brouard 4303: double *meanq;
4304: double **meanqt;
4305: double *pp, **prop, *posprop, *pospropt;
4306: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4307: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4308: double agebegin, ageend;
4309:
4310: pp=vector(1,nlstate);
1.251 brouard 4311: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4312: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4313: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4314: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4315: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4316: meanqt=matrix(1,lastpass,1,nqtveff);
4317: strcpy(fileresp,"P_");
4318: strcat(fileresp,fileresu);
4319: /*strcat(fileresphtm,fileresu);*/
4320: if((ficresp=fopen(fileresp,"w"))==NULL) {
4321: printf("Problem with prevalence resultfile: %s\n", fileresp);
4322: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4323: exit(0);
4324: }
1.240 brouard 4325:
1.226 brouard 4326: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4327: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4328: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4329: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4330: fflush(ficlog);
4331: exit(70);
4332: }
4333: else{
4334: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4335: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4336: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4337: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4338: }
1.237 brouard 4339: 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 4340:
1.226 brouard 4341: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4342: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4343: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4344: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4345: fflush(ficlog);
4346: exit(70);
1.240 brouard 4347: } else{
1.226 brouard 4348: 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 4349: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4350: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4351: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4352: }
1.240 brouard 4353: 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);
4354:
1.253 brouard 4355: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4356: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4357: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4358: j1=0;
1.126 brouard 4359:
1.227 brouard 4360: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4361: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4362: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4363:
4364:
1.226 brouard 4365: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4366: reference=low_education V1=0,V2=0
4367: med_educ V1=1 V2=0,
4368: high_educ V1=0 V2=1
4369: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4370: */
1.249 brouard 4371: dateintsum=0;
4372: k2cpt=0;
4373:
1.253 brouard 4374: if(cptcoveff == 0 )
4375: nl=1; /* Constant model only */
4376: else
4377: nl=2;
4378: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4379: if(nj==1)
4380: j=0; /* First pass for the constant */
4381: else
4382: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4383: first=1;
4384: 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 */
4385: posproptt=0.;
4386: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4387: scanf("%d", i);*/
4388: for (i=-5; i<=nlstate+ndeath; i++)
4389: for (jk=-5; jk<=nlstate+ndeath; jk++)
4390: for(m=iagemin; m <= iagemax+3; m++)
4391: freq[i][jk][m]=0;
4392:
4393: for (i=1; i<=nlstate; i++) {
1.240 brouard 4394: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4395: prop[i][m]=0;
4396: posprop[i]=0;
4397: pospropt[i]=0;
4398: }
4399: /* for (z1=1; z1<= nqfveff; z1++) { */
4400: /* meanq[z1]+=0.; */
4401: /* for(m=1;m<=lastpass;m++){ */
4402: /* meanqt[m][z1]=0.; */
4403: /* } */
4404: /* } */
4405:
4406: /* dateintsum=0; */
4407: /* k2cpt=0; */
4408:
4409: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4410: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4411: bool=1;
4412: if(j !=0){
4413: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4414: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4415: /* for (z1=1; z1<= nqfveff; z1++) { */
4416: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4417: /* } */
4418: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4419: /* if(Tvaraff[z1] ==-20){ */
4420: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4421: /* }else if(Tvaraff[z1] ==-10){ */
4422: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4423: /* }else */
4424: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4425: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4426: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4427: /* 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",
4428: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4429: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4430: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4431: } /* Onlyf fixed */
4432: } /* end z1 */
4433: } /* cptcovn > 0 */
4434: } /* end any */
4435: }/* end j==0 */
4436: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4437: /* for(m=firstpass; m<=lastpass; m++){ */
4438: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4439: m=mw[mi][iind];
4440: if(j!=0){
4441: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4442: for (z1=1; z1<=cptcoveff; z1++) {
4443: if( Fixed[Tmodelind[z1]]==1){
4444: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4445: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4446: value is -1, we don't select. It differs from the
4447: constant and age model which counts them. */
4448: bool=0; /* not selected */
4449: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4450: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4451: bool=0;
4452: }
4453: }
4454: }
4455: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4456: } /* end j==0 */
4457: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4458: if(bool==1){
4459: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4460: and mw[mi+1][iind]. dh depends on stepm. */
4461: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4462: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4463: if(m >=firstpass && m <=lastpass){
4464: k2=anint[m][iind]+(mint[m][iind]/12.);
4465: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4466: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4467: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4468: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4469: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4470: if (m<lastpass) {
4471: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4472: /* 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]); */
4473: if(s[m][iind]==-1)
4474: 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.));
4475: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4476: /* if((int)agev[m][iind] == 55) */
4477: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4478: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4479: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 4480: }
1.251 brouard 4481: } /* end if between passes */
4482: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4483: dateintsum=dateintsum+k2; /* on all covariates ?*/
4484: k2cpt++;
4485: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4486: }
1.251 brouard 4487: }else{
4488: bool=1;
4489: }/* end bool 2 */
4490: } /* end m */
4491: } /* end bool */
4492: } /* end iind = 1 to imx */
4493: /* prop[s][age] is feeded for any initial and valid live state as well as
4494: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4495:
4496:
4497: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4498: pstamp(ficresp);
4499: if (cptcoveff>0 && j!=0){
4500: printf( "\n#********** Variable ");
4501: fprintf(ficresp, "\n#********** Variable ");
4502: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4503: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4504: fprintf(ficlog, "\n#********** Variable ");
4505: for (z1=1; z1<=cptcoveff; z1++){
4506: if(!FixedV[Tvaraff[z1]]){
4507: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4508: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4509: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4510: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4511: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4512: }else{
1.251 brouard 4513: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4514: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4515: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4516: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4517: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4518: }
4519: }
4520: printf( "**********\n#");
4521: fprintf(ficresp, "**********\n#");
4522: fprintf(ficresphtm, "**********</h3>\n");
4523: fprintf(ficresphtmfr, "**********</h3>\n");
4524: fprintf(ficlog, "**********\n");
4525: }
4526: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4527: for(i=1; i<=nlstate;i++) {
4528: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4529: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4530: }
4531: fprintf(ficresp, "\n");
4532: fprintf(ficresphtm, "\n");
4533:
4534: /* Header of frequency table by age */
4535: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4536: fprintf(ficresphtmfr,"<th>Age</th> ");
4537: for(jk=-1; jk <=nlstate+ndeath; jk++){
4538: for(m=-1; m <=nlstate+ndeath; m++){
4539: if(jk!=0 && m!=0)
4540: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4541: }
1.226 brouard 4542: }
1.251 brouard 4543: fprintf(ficresphtmfr, "\n");
4544:
4545: /* For each age */
4546: for(iage=iagemin; iage <= iagemax+3; iage++){
4547: fprintf(ficresphtm,"<tr>");
4548: if(iage==iagemax+1){
4549: fprintf(ficlog,"1");
4550: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4551: }else if(iage==iagemax+2){
4552: fprintf(ficlog,"0");
4553: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4554: }else if(iage==iagemax+3){
4555: fprintf(ficlog,"Total");
4556: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4557: }else{
1.240 brouard 4558: if(first==1){
1.251 brouard 4559: first=0;
4560: printf("See log file for details...\n");
4561: }
4562: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4563: fprintf(ficlog,"Age %d", iage);
4564: }
4565: for(jk=1; jk <=nlstate ; jk++){
4566: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4567: pp[jk] += freq[jk][m][iage];
4568: }
4569: for(jk=1; jk <=nlstate ; jk++){
4570: for(m=-1, pos=0; m <=0 ; m++)
4571: pos += freq[jk][m][iage];
4572: if(pp[jk]>=1.e-10){
4573: if(first==1){
4574: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4575: }
4576: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4577: }else{
4578: if(first==1)
4579: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4580: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4581: }
4582: }
4583:
1.251 brouard 4584: for(jk=1; jk <=nlstate ; jk++){
4585: /* posprop[jk]=0; */
4586: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4587: pp[jk] += freq[jk][m][iage];
4588: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4589:
4590: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4591: pos += pp[jk]; /* pos is the total number of transitions until this age */
4592: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4593: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4594: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4595: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4596: }
1.251 brouard 4597: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4598: if(pos>=1.e-5){
1.251 brouard 4599: if(first==1)
4600: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4601: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4602: }else{
4603: if(first==1)
4604: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4605: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4606: }
4607: if( iage <= iagemax){
4608: if(pos>=1.e-5){
4609: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4610: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4611: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4612: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4613: }
4614: else{
4615: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4616: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4617: }
1.240 brouard 4618: }
1.251 brouard 4619: pospropt[jk] +=posprop[jk];
4620: } /* end loop jk */
4621: /* pospropt=0.; */
4622: for(jk=-1; jk <=nlstate+ndeath; jk++){
4623: for(m=-1; m <=nlstate+ndeath; m++){
4624: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4625: if(first==1){
4626: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4627: }
1.253 brouard 4628: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4629: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4630: }
4631: if(jk!=0 && m!=0)
4632: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4633: }
1.251 brouard 4634: } /* end loop jk */
4635: posproptt=0.;
4636: for(jk=1; jk <=nlstate; jk++){
4637: posproptt += pospropt[jk];
4638: }
4639: fprintf(ficresphtmfr,"</tr>\n ");
4640: if(iage <= iagemax){
4641: fprintf(ficresp,"\n");
4642: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4643: }
1.251 brouard 4644: if(first==1)
4645: printf("Others in log...\n");
4646: fprintf(ficlog,"\n");
4647: } /* end loop age iage */
4648: fprintf(ficresphtm,"<tr><th>Tot</th>");
4649: for(jk=1; jk <=nlstate ; jk++){
4650: if(posproptt < 1.e-5){
4651: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4652: }else{
4653: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4654: }
1.226 brouard 4655: }
1.251 brouard 4656: fprintf(ficresphtm,"</tr>\n");
4657: fprintf(ficresphtm,"</table>\n");
4658: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4659: if(posproptt < 1.e-5){
1.251 brouard 4660: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4661: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 ! brouard 4662: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
! 4663: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4664: invalidvarcomb[j1]=1;
1.226 brouard 4665: }else{
1.251 brouard 4666: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4667: invalidvarcomb[j1]=0;
1.226 brouard 4668: }
1.251 brouard 4669: fprintf(ficresphtmfr,"</table>\n");
4670: fprintf(ficlog,"\n");
4671: if(j!=0){
4672: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4673: for(i=1,jk=1; i <=nlstate; i++){
4674: for(k=1; k <=(nlstate+ndeath); k++){
4675: if (k != i) {
4676: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4677: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4678: if(j1==1){ /* All dummy covariates to zero */
4679: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4680: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4681: printf("%d%d ",i,k);
4682: fprintf(ficlog,"%d%d ",i,k);
4683: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
4684: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4685: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4686: }
1.253 brouard 4687: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4688: for(iage=iagemin; iage <= iagemax+3; iage++){
4689: x[iage]= (double)iage;
4690: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4691: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4692: }
4693: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4694: pstart[jk]=b;
4695: pstart[jk-1]=a;
1.252 brouard 4696: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
4697: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
4698: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.251 brouard 4699: pstart[jk]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4700: printf("%d%d ",i,k);
4701: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4702: printf("jk=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",jk,i,k,jk,p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
4703: }else{ /* Other cases, like quantitative fixed or varying covariates */
4704: ;
4705: }
4706: /* printf("%12.7f )", param[i][jj][k]); */
4707: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4708: jk++;
4709: } /* end jj */
4710: } /* end k!= i */
4711: } /* end k */
4712: } /* end i, jk */
4713: } /* end j !=0 */
4714: } /* end selected combination of covariate j1 */
4715: if(j==0){ /* We can estimate starting values from the occurences in each case */
4716: printf("#Freqsummary: Starting values for the constants:\n");
4717: fprintf(ficlog,"\n");
4718: for(i=1,jk=1; i <=nlstate; i++){
4719: for(k=1; k <=(nlstate+ndeath); k++){
4720: if (k != i) {
4721: printf("%d%d ",i,k);
4722: fprintf(ficlog,"%d%d ",i,k);
4723: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4724: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4725: if(jj==1){ /* Age has to be done */
4726: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4727: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4728: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[jk],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4729: }
4730: /* printf("%12.7f )", param[i][jj][k]); */
4731: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4732: jk++;
1.250 brouard 4733: }
1.251 brouard 4734: printf("\n");
4735: fprintf(ficlog,"\n");
1.250 brouard 4736: }
4737: }
4738: }
1.251 brouard 4739: printf("#Freqsummary\n");
4740: fprintf(ficlog,"\n");
4741: for(jk=-1; jk <=nlstate+ndeath; jk++){
4742: for(m=-1; m <=nlstate+ndeath; m++){
4743: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4744: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4745: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4746: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4747: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4748: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4749: /* } */
4750: }
4751: } /* end loop jk */
4752:
4753: printf("\n");
4754: fprintf(ficlog,"\n");
4755: } /* end j=0 */
1.249 brouard 4756: } /* end j */
1.252 brouard 4757:
1.253 brouard 4758: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4759: for(i=1, jk=1; i <=nlstate; i++){
4760: for(j=1; j <=nlstate+ndeath; j++){
4761: if(j!=i){
4762: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4763: printf("%1d%1d",i,j);
4764: fprintf(ficparo,"%1d%1d",i,j);
4765: for(k=1; k<=ncovmodel;k++){
4766: /* printf(" %lf",param[i][j][k]); */
4767: /* fprintf(ficparo," %lf",param[i][j][k]); */
4768: p[jk]=pstart[jk];
4769: printf(" %f ",pstart[jk]);
4770: fprintf(ficparo," %f ",pstart[jk]);
4771: jk++;
4772: }
4773: printf("\n");
4774: fprintf(ficparo,"\n");
4775: }
4776: }
4777: }
4778: } /* end mle=-2 */
1.226 brouard 4779: dateintmean=dateintsum/k2cpt;
1.240 brouard 4780:
1.226 brouard 4781: fclose(ficresp);
4782: fclose(ficresphtm);
4783: fclose(ficresphtmfr);
4784: free_vector(meanq,1,nqfveff);
4785: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4786: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4787: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4788: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4789: free_vector(pospropt,1,nlstate);
4790: free_vector(posprop,1,nlstate);
1.251 brouard 4791: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4792: free_vector(pp,1,nlstate);
4793: /* End of freqsummary */
4794: }
1.126 brouard 4795:
4796: /************ Prevalence ********************/
1.227 brouard 4797: 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)
4798: {
4799: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4800: in each health status at the date of interview (if between dateprev1 and dateprev2).
4801: We still use firstpass and lastpass as another selection.
4802: */
1.126 brouard 4803:
1.227 brouard 4804: int i, m, jk, j1, bool, z1,j, iv;
4805: int mi; /* Effective wave */
4806: int iage;
4807: double agebegin, ageend;
4808:
4809: double **prop;
4810: double posprop;
4811: double y2; /* in fractional years */
4812: int iagemin, iagemax;
4813: int first; /** to stop verbosity which is redirected to log file */
4814:
4815: iagemin= (int) agemin;
4816: iagemax= (int) agemax;
4817: /*pp=vector(1,nlstate);*/
1.251 brouard 4818: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4819: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4820: j1=0;
1.222 brouard 4821:
1.227 brouard 4822: /*j=cptcoveff;*/
4823: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4824:
1.227 brouard 4825: first=1;
4826: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4827: for (i=1; i<=nlstate; i++)
1.251 brouard 4828: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4829: prop[i][iage]=0.0;
4830: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4831: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4832: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4833:
4834: for (i=1; i<=imx; i++) { /* Each individual */
4835: bool=1;
4836: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4837: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4838: m=mw[mi][i];
4839: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4840: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4841: for (z1=1; z1<=cptcoveff; z1++){
4842: if( Fixed[Tmodelind[z1]]==1){
4843: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4844: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4845: bool=0;
4846: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4847: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4848: bool=0;
4849: }
4850: }
4851: if(bool==1){ /* Otherwise we skip that wave/person */
4852: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4853: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4854: if(m >=firstpass && m <=lastpass){
4855: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4856: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4857: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4858: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4859: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4860: 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);
4861: exit(1);
4862: }
4863: if (s[m][i]>0 && s[m][i]<=nlstate) {
4864: /*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]]);*/
4865: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4866: prop[s[m][i]][iagemax+3] += weight[i];
4867: } /* end valid statuses */
4868: } /* end selection of dates */
4869: } /* end selection of waves */
4870: } /* end bool */
4871: } /* end wave */
4872: } /* end individual */
4873: for(i=iagemin; i <= iagemax+3; i++){
4874: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4875: posprop += prop[jk][i];
4876: }
4877:
4878: for(jk=1; jk <=nlstate ; jk++){
4879: if( i <= iagemax){
4880: if(posprop>=1.e-5){
4881: probs[i][jk][j1]= prop[jk][i]/posprop;
4882: } else{
4883: if(first==1){
4884: first=0;
4885: 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]);
4886: }
4887: }
4888: }
4889: }/* end jk */
4890: }/* end i */
1.222 brouard 4891: /*} *//* end i1 */
1.227 brouard 4892: } /* end j1 */
1.222 brouard 4893:
1.227 brouard 4894: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4895: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4896: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4897: } /* End of prevalence */
1.126 brouard 4898:
4899: /************* Waves Concatenation ***************/
4900:
4901: 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)
4902: {
4903: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4904: Death is a valid wave (if date is known).
4905: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4906: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4907: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4908: */
1.126 brouard 4909:
1.224 brouard 4910: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4911: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4912: double sum=0., jmean=0.;*/
1.224 brouard 4913: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4914: int j, k=0,jk, ju, jl;
4915: double sum=0.;
4916: first=0;
1.214 brouard 4917: firstwo=0;
1.217 brouard 4918: firsthree=0;
1.218 brouard 4919: firstfour=0;
1.164 brouard 4920: jmin=100000;
1.126 brouard 4921: jmax=-1;
4922: jmean=0.;
1.224 brouard 4923:
4924: /* Treating live states */
1.214 brouard 4925: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4926: mi=0; /* First valid wave */
1.227 brouard 4927: mli=0; /* Last valid wave */
1.126 brouard 4928: m=firstpass;
1.214 brouard 4929: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4930: 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 */
4931: mli=m-1;/* mw[++mi][i]=m-1; */
4932: }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 */
4933: mw[++mi][i]=m;
4934: mli=m;
1.224 brouard 4935: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4936: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4937: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4938: }
1.227 brouard 4939: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4940: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4941: break;
1.224 brouard 4942: #else
1.227 brouard 4943: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4944: if(firsthree == 0){
4945: 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);
4946: firsthree=1;
4947: }
4948: 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);
4949: mw[++mi][i]=m;
4950: mli=m;
4951: }
4952: if(s[m][i]==-2){ /* Vital status is really unknown */
4953: nbwarn++;
4954: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4955: 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);
4956: 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);
4957: }
4958: break;
4959: }
4960: break;
1.224 brouard 4961: #endif
1.227 brouard 4962: }/* End m >= lastpass */
1.126 brouard 4963: }/* end while */
1.224 brouard 4964:
1.227 brouard 4965: /* 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 4966: /* After last pass */
1.224 brouard 4967: /* Treating death states */
1.214 brouard 4968: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4969: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4970: /* } */
1.126 brouard 4971: mi++; /* Death is another wave */
4972: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4973: /* Only death is a correct wave */
1.126 brouard 4974: mw[mi][i]=m;
1.257 brouard 4975: } /* else not in a death state */
1.224 brouard 4976: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 4977: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 4978: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4979: 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 */
4980: nbwarn++;
4981: if(firstfiv==0){
4982: 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 );
4983: firstfiv=1;
4984: }else{
4985: 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 );
4986: }
4987: }else{ /* Death occured afer last wave potential bias */
4988: nberr++;
4989: if(firstwo==0){
1.257 brouard 4990: 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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\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 );
1.227 brouard 4991: firstwo=1;
4992: }
1.257 brouard 4993: 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. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.227 brouard 4994: }
1.257 brouard 4995: }else{ /* if date of interview is unknown */
1.227 brouard 4996: /* death is known but not confirmed by death status at any wave */
4997: if(firstfour==0){
4998: 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 );
4999: firstfour=1;
5000: }
5001: 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 5002: }
1.224 brouard 5003: } /* end if date of death is known */
5004: #endif
5005: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5006: /* wav[i]=mw[mi][i]; */
1.126 brouard 5007: if(mi==0){
5008: nbwarn++;
5009: if(first==0){
1.227 brouard 5010: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5011: first=1;
1.126 brouard 5012: }
5013: if(first==1){
1.227 brouard 5014: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5015: }
5016: } /* end mi==0 */
5017: } /* End individuals */
1.214 brouard 5018: /* wav and mw are no more changed */
1.223 brouard 5019:
1.214 brouard 5020:
1.126 brouard 5021: for(i=1; i<=imx; i++){
5022: for(mi=1; mi<wav[i];mi++){
5023: if (stepm <=0)
1.227 brouard 5024: dh[mi][i]=1;
1.126 brouard 5025: else{
1.260 ! brouard 5026: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5027: if (agedc[i] < 2*AGESUP) {
5028: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5029: if(j==0) j=1; /* Survives at least one month after exam */
5030: else if(j<0){
5031: nberr++;
5032: 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]);
5033: j=1; /* Temporary Dangerous patch */
5034: 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);
5035: 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]);
5036: 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);
5037: }
5038: k=k+1;
5039: if (j >= jmax){
5040: jmax=j;
5041: ijmax=i;
5042: }
5043: if (j <= jmin){
5044: jmin=j;
5045: ijmin=i;
5046: }
5047: sum=sum+j;
5048: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5049: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5050: }
5051: }
5052: else{
5053: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5054: /* 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 5055:
1.227 brouard 5056: k=k+1;
5057: if (j >= jmax) {
5058: jmax=j;
5059: ijmax=i;
5060: }
5061: else if (j <= jmin){
5062: jmin=j;
5063: ijmin=i;
5064: }
5065: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5066: /*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]);*/
5067: if(j<0){
5068: nberr++;
5069: 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]);
5070: 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]);
5071: }
5072: sum=sum+j;
5073: }
5074: jk= j/stepm;
5075: jl= j -jk*stepm;
5076: ju= j -(jk+1)*stepm;
5077: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5078: if(jl==0){
5079: dh[mi][i]=jk;
5080: bh[mi][i]=0;
5081: }else{ /* We want a negative bias in order to only have interpolation ie
5082: * to avoid the price of an extra matrix product in likelihood */
5083: dh[mi][i]=jk+1;
5084: bh[mi][i]=ju;
5085: }
5086: }else{
5087: if(jl <= -ju){
5088: dh[mi][i]=jk;
5089: bh[mi][i]=jl; /* bias is positive if real duration
5090: * is higher than the multiple of stepm and negative otherwise.
5091: */
5092: }
5093: else{
5094: dh[mi][i]=jk+1;
5095: bh[mi][i]=ju;
5096: }
5097: if(dh[mi][i]==0){
5098: dh[mi][i]=1; /* At least one step */
5099: bh[mi][i]=ju; /* At least one step */
5100: /* 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);*/
5101: }
5102: } /* end if mle */
1.126 brouard 5103: }
5104: } /* end wave */
5105: }
5106: jmean=sum/k;
5107: 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 5108: 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 5109: }
1.126 brouard 5110:
5111: /*********** Tricode ****************************/
1.220 brouard 5112: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5113: {
5114: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5115: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5116: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5117: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5118: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5119: */
1.130 brouard 5120:
1.242 brouard 5121: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5122: int modmaxcovj=0; /* Modality max of covariates j */
5123: int cptcode=0; /* Modality max of covariates j */
5124: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5125:
5126:
1.242 brouard 5127: /* cptcoveff=0; */
5128: /* *cptcov=0; */
1.126 brouard 5129:
1.242 brouard 5130: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5131:
1.242 brouard 5132: /* Loop on covariates without age and products and no quantitative variable */
5133: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5134: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5135: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5136: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5137: switch(Fixed[k]) {
5138: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5139: 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*/
5140: ij=(int)(covar[Tvar[k]][i]);
5141: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5142: * If product of Vn*Vm, still boolean *:
5143: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5144: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5145: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5146: modality of the nth covariate of individual i. */
5147: if (ij > modmaxcovj)
5148: modmaxcovj=ij;
5149: else if (ij < modmincovj)
5150: modmincovj=ij;
5151: if ((ij < -1) && (ij > NCOVMAX)){
5152: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5153: exit(1);
5154: }else
5155: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5156: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5157: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5158: /* getting the maximum value of the modality of the covariate
5159: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5160: female ies 1, then modmaxcovj=1.
5161: */
5162: } /* end for loop on individuals i */
5163: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5164: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5165: cptcode=modmaxcovj;
5166: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5167: /*for (i=0; i<=cptcode; i++) {*/
5168: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5169: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5170: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5171: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5172: if( j != -1){
5173: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5174: covariate for which somebody answered excluding
5175: undefined. Usually 2: 0 and 1. */
5176: }
5177: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5178: covariate for which somebody answered including
5179: undefined. Usually 3: -1, 0 and 1. */
5180: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5181: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5182: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5183:
1.242 brouard 5184: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5185: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5186: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5187: /* modmincovj=3; modmaxcovj = 7; */
5188: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5189: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5190: /* defining two dummy variables: variables V1_1 and V1_2.*/
5191: /* nbcode[Tvar[j]][ij]=k; */
5192: /* nbcode[Tvar[j]][1]=0; */
5193: /* nbcode[Tvar[j]][2]=1; */
5194: /* nbcode[Tvar[j]][3]=2; */
5195: /* To be continued (not working yet). */
5196: ij=0; /* ij is similar to i but can jump over null modalities */
5197: 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*/
5198: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5199: break;
5200: }
5201: ij++;
5202: 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*/
5203: cptcode = ij; /* New max modality for covar j */
5204: } /* end of loop on modality i=-1 to 1 or more */
5205: break;
5206: case 1: /* Testing on varying covariate, could be simple and
5207: * should look at waves or product of fixed *
5208: * varying. No time to test -1, assuming 0 and 1 only */
5209: ij=0;
5210: for(i=0; i<=1;i++){
5211: nbcode[Tvar[k]][++ij]=i;
5212: }
5213: break;
5214: default:
5215: break;
5216: } /* end switch */
5217: } /* end dummy test */
5218:
5219: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5220: /* /\*recode from 0 *\/ */
5221: /* k is a modality. If we have model=V1+V1*sex */
5222: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5223: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5224: /* } */
5225: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5226: /* if (ij > ncodemax[j]) { */
5227: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5228: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5229: /* break; */
5230: /* } */
5231: /* } /\* end of loop on modality k *\/ */
5232: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5233:
5234: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5235: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5236: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5237: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5238: 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 */
5239: 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 */
5240: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5241: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5242:
5243: ij=0;
5244: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5245: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5246: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5247: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5248: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5249: /* If product not in single variable we don't print results */
5250: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5251: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5252: 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*/
5253: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5254: 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 */
5255: if(Fixed[k]!=0)
5256: anyvaryingduminmodel=1;
5257: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5258: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5259: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5260: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5261: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5262: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5263: }
5264: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5265: /* ij--; */
5266: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5267: *cptcov=ij; /*Number of total real effective covariates: effective
5268: * because they can be excluded from the model and real
5269: * if in the model but excluded because missing values, but how to get k from ij?*/
5270: for(j=ij+1; j<= cptcovt; j++){
5271: Tvaraff[j]=0;
5272: Tmodelind[j]=0;
5273: }
5274: for(j=ntveff+1; j<= cptcovt; j++){
5275: TmodelInvind[j]=0;
5276: }
5277: /* To be sorted */
5278: ;
5279: }
1.126 brouard 5280:
1.145 brouard 5281:
1.126 brouard 5282: /*********** Health Expectancies ****************/
5283:
1.235 brouard 5284: 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 5285:
5286: {
5287: /* Health expectancies, no variances */
1.164 brouard 5288: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5289: int nhstepma, nstepma; /* Decreasing with age */
5290: double age, agelim, hf;
5291: double ***p3mat;
5292: double eip;
5293:
1.238 brouard 5294: /* pstamp(ficreseij); */
1.126 brouard 5295: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5296: fprintf(ficreseij,"# Age");
5297: for(i=1; i<=nlstate;i++){
5298: for(j=1; j<=nlstate;j++){
5299: fprintf(ficreseij," e%1d%1d ",i,j);
5300: }
5301: fprintf(ficreseij," e%1d. ",i);
5302: }
5303: fprintf(ficreseij,"\n");
5304:
5305:
5306: if(estepm < stepm){
5307: printf ("Problem %d lower than %d\n",estepm, stepm);
5308: }
5309: else hstepm=estepm;
5310: /* We compute the life expectancy from trapezoids spaced every estepm months
5311: * This is mainly to measure the difference between two models: for example
5312: * if stepm=24 months pijx are given only every 2 years and by summing them
5313: * we are calculating an estimate of the Life Expectancy assuming a linear
5314: * progression in between and thus overestimating or underestimating according
5315: * to the curvature of the survival function. If, for the same date, we
5316: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5317: * to compare the new estimate of Life expectancy with the same linear
5318: * hypothesis. A more precise result, taking into account a more precise
5319: * curvature will be obtained if estepm is as small as stepm. */
5320:
5321: /* For example we decided to compute the life expectancy with the smallest unit */
5322: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5323: nhstepm is the number of hstepm from age to agelim
5324: nstepm is the number of stepm from age to agelin.
5325: Look at hpijx to understand the reason of that which relies in memory size
5326: and note for a fixed period like estepm months */
5327: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5328: survival function given by stepm (the optimization length). Unfortunately it
5329: means that if the survival funtion is printed only each two years of age and if
5330: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5331: results. So we changed our mind and took the option of the best precision.
5332: */
5333: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5334:
5335: agelim=AGESUP;
5336: /* If stepm=6 months */
5337: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5338: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5339:
5340: /* nhstepm age range expressed in number of stepm */
5341: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5342: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5343: /* if (stepm >= YEARM) hstepm=1;*/
5344: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5345: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5346:
5347: for (age=bage; age<=fage; age ++){
5348: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5349: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5350: /* if (stepm >= YEARM) hstepm=1;*/
5351: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5352:
5353: /* If stepm=6 months */
5354: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5355: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5356:
1.235 brouard 5357: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5358:
5359: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5360:
5361: printf("%d|",(int)age);fflush(stdout);
5362: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5363:
5364: /* Computing expectancies */
5365: for(i=1; i<=nlstate;i++)
5366: for(j=1; j<=nlstate;j++)
5367: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5368: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5369:
5370: /* 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]);*/
5371:
5372: }
5373:
5374: fprintf(ficreseij,"%3.0f",age );
5375: for(i=1; i<=nlstate;i++){
5376: eip=0;
5377: for(j=1; j<=nlstate;j++){
5378: eip +=eij[i][j][(int)age];
5379: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5380: }
5381: fprintf(ficreseij,"%9.4f", eip );
5382: }
5383: fprintf(ficreseij,"\n");
5384:
5385: }
5386: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5387: printf("\n");
5388: fprintf(ficlog,"\n");
5389:
5390: }
5391:
1.235 brouard 5392: 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 5393:
5394: {
5395: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5396: to initial status i, ei. .
1.126 brouard 5397: */
5398: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5399: int nhstepma, nstepma; /* Decreasing with age */
5400: double age, agelim, hf;
5401: double ***p3matp, ***p3matm, ***varhe;
5402: double **dnewm,**doldm;
5403: double *xp, *xm;
5404: double **gp, **gm;
5405: double ***gradg, ***trgradg;
5406: int theta;
5407:
5408: double eip, vip;
5409:
5410: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5411: xp=vector(1,npar);
5412: xm=vector(1,npar);
5413: dnewm=matrix(1,nlstate*nlstate,1,npar);
5414: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5415:
5416: pstamp(ficresstdeij);
5417: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5418: fprintf(ficresstdeij,"# Age");
5419: for(i=1; i<=nlstate;i++){
5420: for(j=1; j<=nlstate;j++)
5421: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5422: fprintf(ficresstdeij," e%1d. ",i);
5423: }
5424: fprintf(ficresstdeij,"\n");
5425:
5426: pstamp(ficrescveij);
5427: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5428: fprintf(ficrescveij,"# Age");
5429: for(i=1; i<=nlstate;i++)
5430: for(j=1; j<=nlstate;j++){
5431: cptj= (j-1)*nlstate+i;
5432: for(i2=1; i2<=nlstate;i2++)
5433: for(j2=1; j2<=nlstate;j2++){
5434: cptj2= (j2-1)*nlstate+i2;
5435: if(cptj2 <= cptj)
5436: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5437: }
5438: }
5439: fprintf(ficrescveij,"\n");
5440:
5441: if(estepm < stepm){
5442: printf ("Problem %d lower than %d\n",estepm, stepm);
5443: }
5444: else hstepm=estepm;
5445: /* We compute the life expectancy from trapezoids spaced every estepm months
5446: * This is mainly to measure the difference between two models: for example
5447: * if stepm=24 months pijx are given only every 2 years and by summing them
5448: * we are calculating an estimate of the Life Expectancy assuming a linear
5449: * progression in between and thus overestimating or underestimating according
5450: * to the curvature of the survival function. If, for the same date, we
5451: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5452: * to compare the new estimate of Life expectancy with the same linear
5453: * hypothesis. A more precise result, taking into account a more precise
5454: * curvature will be obtained if estepm is as small as stepm. */
5455:
5456: /* For example we decided to compute the life expectancy with the smallest unit */
5457: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5458: nhstepm is the number of hstepm from age to agelim
5459: nstepm is the number of stepm from age to agelin.
5460: Look at hpijx to understand the reason of that which relies in memory size
5461: and note for a fixed period like estepm months */
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 only each 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:
5470: /* If stepm=6 months */
5471: /* nhstepm age range expressed in number of stepm */
5472: agelim=AGESUP;
5473: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5474: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5475: /* if (stepm >= YEARM) hstepm=1;*/
5476: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5477:
5478: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5479: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5480: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5481: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5482: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5483: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5484:
5485: for (age=bage; age<=fage; age ++){
5486: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5487: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5488: /* if (stepm >= YEARM) hstepm=1;*/
5489: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5490:
1.126 brouard 5491: /* If stepm=6 months */
5492: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5493: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5494:
5495: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5496:
1.126 brouard 5497: /* Computing Variances of health expectancies */
5498: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5499: decrease memory allocation */
5500: for(theta=1; theta <=npar; theta++){
5501: for(i=1; i<=npar; i++){
1.222 brouard 5502: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5503: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5504: }
1.235 brouard 5505: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5506: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5507:
1.126 brouard 5508: for(j=1; j<= nlstate; j++){
1.222 brouard 5509: for(i=1; i<=nlstate; i++){
5510: for(h=0; h<=nhstepm-1; h++){
5511: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5512: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5513: }
5514: }
1.126 brouard 5515: }
1.218 brouard 5516:
1.126 brouard 5517: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5518: for(h=0; h<=nhstepm-1; h++){
5519: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5520: }
1.126 brouard 5521: }/* End theta */
5522:
5523:
5524: for(h=0; h<=nhstepm-1; h++)
5525: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5526: for(theta=1; theta <=npar; theta++)
5527: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5528:
1.218 brouard 5529:
1.222 brouard 5530: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5531: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5532: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5533:
1.222 brouard 5534: printf("%d|",(int)age);fflush(stdout);
5535: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5536: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5537: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5538: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5539: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5540: for(ij=1;ij<=nlstate*nlstate;ij++)
5541: for(ji=1;ji<=nlstate*nlstate;ji++)
5542: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5543: }
5544: }
1.218 brouard 5545:
1.126 brouard 5546: /* Computing expectancies */
1.235 brouard 5547: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5548: for(i=1; i<=nlstate;i++)
5549: for(j=1; j<=nlstate;j++)
1.222 brouard 5550: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5551: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5552:
1.222 brouard 5553: /* 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 5554:
1.222 brouard 5555: }
1.218 brouard 5556:
1.126 brouard 5557: fprintf(ficresstdeij,"%3.0f",age );
5558: for(i=1; i<=nlstate;i++){
5559: eip=0.;
5560: vip=0.;
5561: for(j=1; j<=nlstate;j++){
1.222 brouard 5562: eip += eij[i][j][(int)age];
5563: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5564: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5565: 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 5566: }
5567: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5568: }
5569: fprintf(ficresstdeij,"\n");
1.218 brouard 5570:
1.126 brouard 5571: fprintf(ficrescveij,"%3.0f",age );
5572: for(i=1; i<=nlstate;i++)
5573: for(j=1; j<=nlstate;j++){
1.222 brouard 5574: cptj= (j-1)*nlstate+i;
5575: for(i2=1; i2<=nlstate;i2++)
5576: for(j2=1; j2<=nlstate;j2++){
5577: cptj2= (j2-1)*nlstate+i2;
5578: if(cptj2 <= cptj)
5579: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5580: }
1.126 brouard 5581: }
5582: fprintf(ficrescveij,"\n");
1.218 brouard 5583:
1.126 brouard 5584: }
5585: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5586: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5587: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5588: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5589: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5590: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5591: printf("\n");
5592: fprintf(ficlog,"\n");
1.218 brouard 5593:
1.126 brouard 5594: free_vector(xm,1,npar);
5595: free_vector(xp,1,npar);
5596: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5597: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5598: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5599: }
1.218 brouard 5600:
1.126 brouard 5601: /************ Variance ******************/
1.235 brouard 5602: 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 5603: {
5604: /* Variance of health expectancies */
5605: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5606: /* double **newm;*/
5607: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5608:
5609: /* int movingaverage(); */
5610: double **dnewm,**doldm;
5611: double **dnewmp,**doldmp;
5612: int i, j, nhstepm, hstepm, h, nstepm ;
5613: int k;
5614: double *xp;
5615: double **gp, **gm; /* for var eij */
5616: double ***gradg, ***trgradg; /*for var eij */
5617: double **gradgp, **trgradgp; /* for var p point j */
5618: double *gpp, *gmp; /* for var p point j */
5619: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5620: double ***p3mat;
5621: double age,agelim, hf;
5622: /* double ***mobaverage; */
5623: int theta;
5624: char digit[4];
5625: char digitp[25];
5626:
5627: char fileresprobmorprev[FILENAMELENGTH];
5628:
5629: if(popbased==1){
5630: if(mobilav!=0)
5631: strcpy(digitp,"-POPULBASED-MOBILAV_");
5632: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5633: }
5634: else
5635: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5636:
1.218 brouard 5637: /* if (mobilav!=0) { */
5638: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5639: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5640: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5641: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5642: /* } */
5643: /* } */
5644:
5645: strcpy(fileresprobmorprev,"PRMORPREV-");
5646: sprintf(digit,"%-d",ij);
5647: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5648: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5649: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5650: strcat(fileresprobmorprev,fileresu);
5651: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5652: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5653: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5654: }
5655: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5656: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5657: pstamp(ficresprobmorprev);
5658: 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 5659: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5660: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5661: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5662: }
5663: for(j=1;j<=cptcoveff;j++)
5664: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5665: fprintf(ficresprobmorprev,"\n");
5666:
1.218 brouard 5667: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5668: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5669: fprintf(ficresprobmorprev," p.%-d SE",j);
5670: for(i=1; i<=nlstate;i++)
5671: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5672: }
5673: fprintf(ficresprobmorprev,"\n");
5674:
5675: fprintf(ficgp,"\n# Routine varevsij");
5676: fprintf(ficgp,"\nunset title \n");
5677: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5678: 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");
5679: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5680: /* } */
5681: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5682: pstamp(ficresvij);
5683: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5684: if(popbased==1)
5685: 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);
5686: else
5687: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5688: fprintf(ficresvij,"# Age");
5689: for(i=1; i<=nlstate;i++)
5690: for(j=1; j<=nlstate;j++)
5691: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5692: fprintf(ficresvij,"\n");
5693:
5694: xp=vector(1,npar);
5695: dnewm=matrix(1,nlstate,1,npar);
5696: doldm=matrix(1,nlstate,1,nlstate);
5697: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5698: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5699:
5700: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5701: gpp=vector(nlstate+1,nlstate+ndeath);
5702: gmp=vector(nlstate+1,nlstate+ndeath);
5703: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5704:
1.218 brouard 5705: if(estepm < stepm){
5706: printf ("Problem %d lower than %d\n",estepm, stepm);
5707: }
5708: else hstepm=estepm;
5709: /* For example we decided to compute the life expectancy with the smallest unit */
5710: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5711: nhstepm is the number of hstepm from age to agelim
5712: nstepm is the number of stepm from age to agelim.
5713: Look at function hpijx to understand why because of memory size limitations,
5714: we decided (b) to get a life expectancy respecting the most precise curvature of the
5715: survival function given by stepm (the optimization length). Unfortunately it
5716: means that if the survival funtion is printed every two years of age and if
5717: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5718: results. So we changed our mind and took the option of the best precision.
5719: */
5720: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5721: agelim = AGESUP;
5722: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5723: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5724: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5725: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5726: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5727: gp=matrix(0,nhstepm,1,nlstate);
5728: gm=matrix(0,nhstepm,1,nlstate);
5729:
5730:
5731: for(theta=1; theta <=npar; theta++){
5732: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5733: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5734: }
5735:
1.242 brouard 5736: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5737:
5738: if (popbased==1) {
5739: if(mobilav ==0){
5740: for(i=1; i<=nlstate;i++)
5741: prlim[i][i]=probs[(int)age][i][ij];
5742: }else{ /* mobilav */
5743: for(i=1; i<=nlstate;i++)
5744: prlim[i][i]=mobaverage[(int)age][i][ij];
5745: }
5746: }
5747:
1.235 brouard 5748: 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 5749: for(j=1; j<= nlstate; j++){
5750: for(h=0; h<=nhstepm; h++){
5751: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5752: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5753: }
5754: }
5755: /* Next for computing probability of death (h=1 means
5756: computed over hstepm matrices product = hstepm*stepm months)
5757: as a weighted average of prlim.
5758: */
5759: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5760: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5761: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5762: }
5763: /* end probability of death */
5764:
5765: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5766: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5767:
1.242 brouard 5768: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5769:
5770: if (popbased==1) {
5771: if(mobilav ==0){
5772: for(i=1; i<=nlstate;i++)
5773: prlim[i][i]=probs[(int)age][i][ij];
5774: }else{ /* mobilav */
5775: for(i=1; i<=nlstate;i++)
5776: prlim[i][i]=mobaverage[(int)age][i][ij];
5777: }
5778: }
5779:
1.235 brouard 5780: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5781:
5782: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5783: for(h=0; h<=nhstepm; h++){
5784: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5785: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5786: }
5787: }
5788: /* This for computing probability of death (h=1 means
5789: computed over hstepm matrices product = hstepm*stepm months)
5790: as a weighted average of prlim.
5791: */
5792: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5793: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5794: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5795: }
5796: /* end probability of death */
5797:
5798: for(j=1; j<= nlstate; j++) /* vareij */
5799: for(h=0; h<=nhstepm; h++){
5800: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5801: }
5802:
5803: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5804: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5805: }
5806:
5807: } /* End theta */
5808:
5809: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5810:
5811: for(h=0; h<=nhstepm; h++) /* veij */
5812: for(j=1; j<=nlstate;j++)
5813: for(theta=1; theta <=npar; theta++)
5814: trgradg[h][j][theta]=gradg[h][theta][j];
5815:
5816: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5817: for(theta=1; theta <=npar; theta++)
5818: trgradgp[j][theta]=gradgp[theta][j];
5819:
5820:
5821: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5822: for(i=1;i<=nlstate;i++)
5823: for(j=1;j<=nlstate;j++)
5824: vareij[i][j][(int)age] =0.;
5825:
5826: for(h=0;h<=nhstepm;h++){
5827: for(k=0;k<=nhstepm;k++){
5828: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5829: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5830: for(i=1;i<=nlstate;i++)
5831: for(j=1;j<=nlstate;j++)
5832: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5833: }
5834: }
5835:
5836: /* pptj */
5837: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5838: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5839: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5840: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5841: varppt[j][i]=doldmp[j][i];
5842: /* end ppptj */
5843: /* x centered again */
5844:
1.242 brouard 5845: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5846:
5847: if (popbased==1) {
5848: if(mobilav ==0){
5849: for(i=1; i<=nlstate;i++)
5850: prlim[i][i]=probs[(int)age][i][ij];
5851: }else{ /* mobilav */
5852: for(i=1; i<=nlstate;i++)
5853: prlim[i][i]=mobaverage[(int)age][i][ij];
5854: }
5855: }
5856:
5857: /* This for computing probability of death (h=1 means
5858: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5859: as a weighted average of prlim.
5860: */
1.235 brouard 5861: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5862: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5863: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5864: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5865: }
5866: /* end probability of death */
5867:
5868: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5869: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5870: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5871: for(i=1; i<=nlstate;i++){
5872: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5873: }
5874: }
5875: fprintf(ficresprobmorprev,"\n");
5876:
5877: fprintf(ficresvij,"%.0f ",age );
5878: for(i=1; i<=nlstate;i++)
5879: for(j=1; j<=nlstate;j++){
5880: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5881: }
5882: fprintf(ficresvij,"\n");
5883: free_matrix(gp,0,nhstepm,1,nlstate);
5884: free_matrix(gm,0,nhstepm,1,nlstate);
5885: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5886: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5887: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5888: } /* End age */
5889: free_vector(gpp,nlstate+1,nlstate+ndeath);
5890: free_vector(gmp,nlstate+1,nlstate+ndeath);
5891: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5892: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5893: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5894: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5895: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5896: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5897: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5898: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5899: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5900: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5901: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5902: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5903: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5904: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5905: 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);
5906: /* 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 5907: */
1.218 brouard 5908: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5909: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5910:
1.218 brouard 5911: free_vector(xp,1,npar);
5912: free_matrix(doldm,1,nlstate,1,nlstate);
5913: free_matrix(dnewm,1,nlstate,1,npar);
5914: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5915: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5916: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5917: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5918: fclose(ficresprobmorprev);
5919: fflush(ficgp);
5920: fflush(fichtm);
5921: } /* end varevsij */
1.126 brouard 5922:
5923: /************ Variance of prevlim ******************/
1.235 brouard 5924: 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 5925: {
1.205 brouard 5926: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5927: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5928:
1.126 brouard 5929: double **dnewm,**doldm;
5930: int i, j, nhstepm, hstepm;
5931: double *xp;
5932: double *gp, *gm;
5933: double **gradg, **trgradg;
1.208 brouard 5934: double **mgm, **mgp;
1.126 brouard 5935: double age,agelim;
5936: int theta;
5937:
5938: pstamp(ficresvpl);
5939: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5940: fprintf(ficresvpl,"# Age ");
5941: if(nresult >=1)
5942: fprintf(ficresvpl," Result# ");
1.126 brouard 5943: for(i=1; i<=nlstate;i++)
5944: fprintf(ficresvpl," %1d-%1d",i,i);
5945: fprintf(ficresvpl,"\n");
5946:
5947: xp=vector(1,npar);
5948: dnewm=matrix(1,nlstate,1,npar);
5949: doldm=matrix(1,nlstate,1,nlstate);
5950:
5951: hstepm=1*YEARM; /* Every year of age */
5952: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5953: agelim = AGESUP;
5954: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5955: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5956: if (stepm >= YEARM) hstepm=1;
5957: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5958: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5959: mgp=matrix(1,npar,1,nlstate);
5960: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5961: gp=vector(1,nlstate);
5962: gm=vector(1,nlstate);
5963:
5964: for(theta=1; theta <=npar; theta++){
5965: for(i=1; i<=npar; i++){ /* Computes gradient */
5966: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5967: }
1.209 brouard 5968: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5969: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5970: else
1.235 brouard 5971: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5972: for(i=1;i<=nlstate;i++){
1.126 brouard 5973: gp[i] = prlim[i][i];
1.208 brouard 5974: mgp[theta][i] = prlim[i][i];
5975: }
1.126 brouard 5976: for(i=1; i<=npar; i++) /* Computes gradient */
5977: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5978: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5979: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5980: else
1.235 brouard 5981: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5982: for(i=1;i<=nlstate;i++){
1.126 brouard 5983: gm[i] = prlim[i][i];
1.208 brouard 5984: mgm[theta][i] = prlim[i][i];
5985: }
1.126 brouard 5986: for(i=1;i<=nlstate;i++)
5987: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5988: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5989: } /* End theta */
5990:
5991: trgradg =matrix(1,nlstate,1,npar);
5992:
5993: for(j=1; j<=nlstate;j++)
5994: for(theta=1; theta <=npar; theta++)
5995: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5996: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5997: /* printf("\nmgm mgp %d ",(int)age); */
5998: /* for(j=1; j<=nlstate;j++){ */
5999: /* printf(" %d ",j); */
6000: /* for(theta=1; theta <=npar; theta++) */
6001: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6002: /* printf("\n "); */
6003: /* } */
6004: /* } */
6005: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6006: /* printf("\n gradg %d ",(int)age); */
6007: /* for(j=1; j<=nlstate;j++){ */
6008: /* printf("%d ",j); */
6009: /* for(theta=1; theta <=npar; theta++) */
6010: /* printf("%d %lf ",theta,gradg[theta][j]); */
6011: /* printf("\n "); */
6012: /* } */
6013: /* } */
1.126 brouard 6014:
6015: for(i=1;i<=nlstate;i++)
6016: varpl[i][(int)age] =0.;
1.209 brouard 6017: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6018: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6019: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6020: }else{
1.126 brouard 6021: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6022: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6023: }
1.126 brouard 6024: for(i=1;i<=nlstate;i++)
6025: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6026:
6027: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6028: if(nresult >=1)
6029: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6030: for(i=1; i<=nlstate;i++)
6031: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6032: fprintf(ficresvpl,"\n");
6033: free_vector(gp,1,nlstate);
6034: free_vector(gm,1,nlstate);
1.208 brouard 6035: free_matrix(mgm,1,npar,1,nlstate);
6036: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6037: free_matrix(gradg,1,npar,1,nlstate);
6038: free_matrix(trgradg,1,nlstate,1,npar);
6039: } /* End age */
6040:
6041: free_vector(xp,1,npar);
6042: free_matrix(doldm,1,nlstate,1,npar);
6043: free_matrix(dnewm,1,nlstate,1,nlstate);
6044:
6045: }
6046:
6047: /************ Variance of one-step probabilities ******************/
6048: 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 6049: {
6050: int i, j=0, k1, l1, tj;
6051: int k2, l2, j1, z1;
6052: int k=0, l;
6053: int first=1, first1, first2;
6054: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6055: double **dnewm,**doldm;
6056: double *xp;
6057: double *gp, *gm;
6058: double **gradg, **trgradg;
6059: double **mu;
6060: double age, cov[NCOVMAX+1];
6061: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6062: int theta;
6063: char fileresprob[FILENAMELENGTH];
6064: char fileresprobcov[FILENAMELENGTH];
6065: char fileresprobcor[FILENAMELENGTH];
6066: double ***varpij;
6067:
6068: strcpy(fileresprob,"PROB_");
6069: strcat(fileresprob,fileres);
6070: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6071: printf("Problem with resultfile: %s\n", fileresprob);
6072: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6073: }
6074: strcpy(fileresprobcov,"PROBCOV_");
6075: strcat(fileresprobcov,fileresu);
6076: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6077: printf("Problem with resultfile: %s\n", fileresprobcov);
6078: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6079: }
6080: strcpy(fileresprobcor,"PROBCOR_");
6081: strcat(fileresprobcor,fileresu);
6082: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6083: printf("Problem with resultfile: %s\n", fileresprobcor);
6084: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6085: }
6086: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6087: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6088: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6089: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6090: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6091: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6092: pstamp(ficresprob);
6093: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6094: fprintf(ficresprob,"# Age");
6095: pstamp(ficresprobcov);
6096: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6097: fprintf(ficresprobcov,"# Age");
6098: pstamp(ficresprobcor);
6099: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6100: fprintf(ficresprobcor,"# Age");
1.126 brouard 6101:
6102:
1.222 brouard 6103: for(i=1; i<=nlstate;i++)
6104: for(j=1; j<=(nlstate+ndeath);j++){
6105: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6106: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6107: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6108: }
6109: /* fprintf(ficresprob,"\n");
6110: fprintf(ficresprobcov,"\n");
6111: fprintf(ficresprobcor,"\n");
6112: */
6113: xp=vector(1,npar);
6114: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6115: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6116: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6117: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6118: first=1;
6119: fprintf(ficgp,"\n# Routine varprob");
6120: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6121: fprintf(fichtm,"\n");
6122:
6123: 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);
6124: 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);
6125: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6126: and drawn. It helps understanding how is the covariance between two incidences.\
6127: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6128: 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 6129: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6130: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6131: standard deviations wide on each axis. <br>\
6132: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6133: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6134: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6135:
1.222 brouard 6136: cov[1]=1;
6137: /* tj=cptcoveff; */
1.225 brouard 6138: tj = (int) pow(2,cptcoveff);
1.222 brouard 6139: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6140: j1=0;
1.224 brouard 6141: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6142: if (cptcovn>0) {
6143: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6144: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6145: fprintf(ficresprob, "**********\n#\n");
6146: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6147: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6148: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6149:
1.222 brouard 6150: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6151: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6152: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6153:
6154:
1.222 brouard 6155: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6156: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6157: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6158:
1.222 brouard 6159: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6160: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6161: fprintf(ficresprobcor, "**********\n#");
6162: if(invalidvarcomb[j1]){
6163: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6164: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6165: continue;
6166: }
6167: }
6168: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6169: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6170: gp=vector(1,(nlstate)*(nlstate+ndeath));
6171: gm=vector(1,(nlstate)*(nlstate+ndeath));
6172: for (age=bage; age<=fage; age ++){
6173: cov[2]=age;
6174: if(nagesqr==1)
6175: cov[3]= age*age;
6176: for (k=1; k<=cptcovn;k++) {
6177: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6178: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6179: * 1 1 1 1 1
6180: * 2 2 1 1 1
6181: * 3 1 2 1 1
6182: */
6183: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6184: }
6185: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6186: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6187: for (k=1; k<=cptcovprod;k++)
6188: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6189:
6190:
1.222 brouard 6191: for(theta=1; theta <=npar; theta++){
6192: for(i=1; i<=npar; i++)
6193: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6194:
1.222 brouard 6195: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6196:
1.222 brouard 6197: k=0;
6198: for(i=1; i<= (nlstate); i++){
6199: for(j=1; j<=(nlstate+ndeath);j++){
6200: k=k+1;
6201: gp[k]=pmmij[i][j];
6202: }
6203: }
1.220 brouard 6204:
1.222 brouard 6205: for(i=1; i<=npar; i++)
6206: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6207:
1.222 brouard 6208: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6209: k=0;
6210: for(i=1; i<=(nlstate); i++){
6211: for(j=1; j<=(nlstate+ndeath);j++){
6212: k=k+1;
6213: gm[k]=pmmij[i][j];
6214: }
6215: }
1.220 brouard 6216:
1.222 brouard 6217: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6218: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6219: }
1.126 brouard 6220:
1.222 brouard 6221: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6222: for(theta=1; theta <=npar; theta++)
6223: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6224:
1.222 brouard 6225: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6226: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6227:
1.222 brouard 6228: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6229:
1.222 brouard 6230: k=0;
6231: for(i=1; i<=(nlstate); i++){
6232: for(j=1; j<=(nlstate+ndeath);j++){
6233: k=k+1;
6234: mu[k][(int) age]=pmmij[i][j];
6235: }
6236: }
6237: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6238: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6239: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6240:
1.222 brouard 6241: /*printf("\n%d ",(int)age);
6242: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6243: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6244: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6245: }*/
1.220 brouard 6246:
1.222 brouard 6247: fprintf(ficresprob,"\n%d ",(int)age);
6248: fprintf(ficresprobcov,"\n%d ",(int)age);
6249: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6250:
1.222 brouard 6251: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6252: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6253: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6254: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6255: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6256: }
6257: i=0;
6258: for (k=1; k<=(nlstate);k++){
6259: for (l=1; l<=(nlstate+ndeath);l++){
6260: i++;
6261: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6262: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6263: for (j=1; j<=i;j++){
6264: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6265: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6266: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6267: }
6268: }
6269: }/* end of loop for state */
6270: } /* end of loop for age */
6271: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6272: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6273: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6274: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6275:
6276: /* Confidence intervalle of pij */
6277: /*
6278: fprintf(ficgp,"\nunset parametric;unset label");
6279: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6280: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6281: 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);
6282: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6283: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6284: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6285: */
6286:
6287: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6288: first1=1;first2=2;
6289: for (k2=1; k2<=(nlstate);k2++){
6290: for (l2=1; l2<=(nlstate+ndeath);l2++){
6291: if(l2==k2) continue;
6292: j=(k2-1)*(nlstate+ndeath)+l2;
6293: for (k1=1; k1<=(nlstate);k1++){
6294: for (l1=1; l1<=(nlstate+ndeath);l1++){
6295: if(l1==k1) continue;
6296: i=(k1-1)*(nlstate+ndeath)+l1;
6297: if(i<=j) continue;
6298: for (age=bage; age<=fage; age ++){
6299: if ((int)age %5==0){
6300: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6301: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6302: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6303: mu1=mu[i][(int) age]/stepm*YEARM ;
6304: mu2=mu[j][(int) age]/stepm*YEARM;
6305: c12=cv12/sqrt(v1*v2);
6306: /* Computing eigen value of matrix of covariance */
6307: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6308: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6309: if ((lc2 <0) || (lc1 <0) ){
6310: if(first2==1){
6311: first1=0;
6312: 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);
6313: }
6314: 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);
6315: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6316: /* lc2=fabs(lc2); */
6317: }
1.220 brouard 6318:
1.222 brouard 6319: /* Eigen vectors */
6320: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6321: /*v21=sqrt(1.-v11*v11); *//* error */
6322: v21=(lc1-v1)/cv12*v11;
6323: v12=-v21;
6324: v22=v11;
6325: tnalp=v21/v11;
6326: if(first1==1){
6327: first1=0;
6328: 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);
6329: }
6330: 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);
6331: /*printf(fignu*/
6332: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6333: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6334: if(first==1){
6335: first=0;
6336: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6337: fprintf(ficgp,"\nset parametric;unset label");
6338: 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);
6339: fprintf(ficgp,"\nset ter svg size 640, 480");
6340: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6341: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6342: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6343: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6344: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6345: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6346: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6347: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6348: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6349: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6350: 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", \
6351: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6352: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6353: }else{
6354: first=0;
6355: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6356: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6357: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6358: 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", \
6359: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6360: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6361: }/* if first */
6362: } /* age mod 5 */
6363: } /* end loop age */
6364: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6365: first=1;
6366: } /*l12 */
6367: } /* k12 */
6368: } /*l1 */
6369: }/* k1 */
6370: } /* loop on combination of covariates j1 */
6371: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6372: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6373: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6374: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6375: free_vector(xp,1,npar);
6376: fclose(ficresprob);
6377: fclose(ficresprobcov);
6378: fclose(ficresprobcor);
6379: fflush(ficgp);
6380: fflush(fichtmcov);
6381: }
1.126 brouard 6382:
6383:
6384: /******************* Printing html file ***********/
1.201 brouard 6385: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6386: int lastpass, int stepm, int weightopt, char model[],\
6387: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6388: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6389: double jprev1, double mprev1,double anprev1, double dateprev1, \
6390: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6391: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6392:
6393: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6394: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6395: </ul>");
1.237 brouard 6396: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6397: </ul>", model);
1.214 brouard 6398: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6399: 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",
6400: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6401: 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 6402: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6403: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6404: fprintf(fichtm,"\
6405: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6406: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6407: fprintf(fichtm,"\
1.217 brouard 6408: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6409: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6410: fprintf(fichtm,"\
1.126 brouard 6411: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6412: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6413: fprintf(fichtm,"\
1.217 brouard 6414: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6415: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6416: fprintf(fichtm,"\
1.211 brouard 6417: - (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 6418: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6419: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6420: if(prevfcast==1){
6421: fprintf(fichtm,"\
6422: - Prevalence projections by age and states: \
1.201 brouard 6423: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6424: }
1.126 brouard 6425:
1.222 brouard 6426: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6427:
1.225 brouard 6428: m=pow(2,cptcoveff);
1.222 brouard 6429: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6430:
1.222 brouard 6431: jj1=0;
1.237 brouard 6432:
6433: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6434: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6435: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6436: continue;
1.220 brouard 6437:
1.222 brouard 6438: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6439: jj1++;
6440: if (cptcovn > 0) {
6441: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6442: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6443: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6444: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6445: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6446: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6447: }
1.237 brouard 6448: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6449: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6450: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6451: }
6452:
1.230 brouard 6453: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6454: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6455: if(invalidvarcomb[k1]){
6456: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6457: printf("\nCombination (%d) ignored because no cases \n",k1);
6458: continue;
6459: }
6460: }
6461: /* aij, bij */
1.259 brouard 6462: fprintf(fichtm,"<br>- Logit model (yours is: logit(pij)=log(pij/pii)= aij+ bij age+%s) as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
1.241 brouard 6463: <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 6464: /* Pij */
1.241 brouard 6465: 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> \
6466: <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 6467: /* Quasi-incidences */
6468: 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 6469: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6470: 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 6471: 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> \
6472: <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 6473: /* Survival functions (period) in state j */
6474: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6475: 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> \
6476: <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 6477: }
6478: /* State specific survival functions (period) */
6479: for(cpt=1; cpt<=nlstate;cpt++){
6480: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6481: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6482: <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 6483: }
6484: /* Period (stable) prevalence in each health state */
6485: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6486: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability to be in state %d some years earlier, knowing that we will be in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6487: <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 6488: }
6489: if(backcast==1){
6490: /* Period (stable) back prevalence in each health state */
6491: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6492: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability to be in state %d at a younger age, knowing that we will be in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6493: <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 6494: }
1.217 brouard 6495: }
1.222 brouard 6496: if(prevfcast==1){
6497: /* Projection of prevalence up to period (stable) prevalence in each health state */
6498: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6499: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d) 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> \
6500: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6501: }
6502: }
1.220 brouard 6503:
1.222 brouard 6504: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6505: 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> \
6506: <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 6507: }
6508: /* } /\* end i1 *\/ */
6509: }/* End k1 */
6510: fprintf(fichtm,"</ul>");
1.126 brouard 6511:
1.222 brouard 6512: fprintf(fichtm,"\
1.126 brouard 6513: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6514: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6515: - 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 6516: But because parameters are usually highly correlated (a higher incidence of disability \
6517: and a higher incidence of recovery can give very close observed transition) it might \
6518: be very useful to look not only at linear confidence intervals estimated from the \
6519: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6520: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6521: covariance matrix of the one-step probabilities. \
6522: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6523:
1.222 brouard 6524: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6525: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6526: fprintf(fichtm,"\
1.126 brouard 6527: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6528: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6529:
1.222 brouard 6530: fprintf(fichtm,"\
1.126 brouard 6531: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6532: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6533: fprintf(fichtm,"\
1.126 brouard 6534: - 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): \
6535: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6536: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6537: fprintf(fichtm,"\
1.126 brouard 6538: - (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): \
6539: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6540: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6541: fprintf(fichtm,"\
1.128 brouard 6542: - 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 6543: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6544: fprintf(fichtm,"\
1.128 brouard 6545: - 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 6546: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6547: fprintf(fichtm,"\
1.126 brouard 6548: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6549: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6550:
6551: /* if(popforecast==1) fprintf(fichtm,"\n */
6552: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6553: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6554: /* <br>",fileres,fileres,fileres,fileres); */
6555: /* else */
6556: /* 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 6557: fflush(fichtm);
6558: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6559:
1.225 brouard 6560: m=pow(2,cptcoveff);
1.222 brouard 6561: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6562:
1.222 brouard 6563: jj1=0;
1.237 brouard 6564:
1.241 brouard 6565: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6566: for(k1=1; k1<=m;k1++){
1.253 brouard 6567: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6568: continue;
1.222 brouard 6569: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6570: jj1++;
1.126 brouard 6571: if (cptcovn > 0) {
6572: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6573: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6574: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6575: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6576: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6577: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6578: }
6579:
1.126 brouard 6580: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6581:
1.222 brouard 6582: if(invalidvarcomb[k1]){
6583: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6584: continue;
6585: }
1.126 brouard 6586: }
6587: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6588: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6589: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
1.258 brouard 6590: <img src=\"%s_%d-%d-%d.svg\">",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 6591: }
6592: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6593: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6594: true period expectancies (those weighted with period prevalences are also\
6595: drawn in addition to the population based expectancies computed using\
1.241 brouard 6596: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6597: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6598: /* } /\* end i1 *\/ */
6599: }/* End k1 */
1.241 brouard 6600: }/* End nres */
1.222 brouard 6601: fprintf(fichtm,"</ul>");
6602: fflush(fichtm);
1.126 brouard 6603: }
6604:
6605: /******************* Gnuplot file **************/
1.223 brouard 6606: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6607:
6608: char dirfileres[132],optfileres[132];
1.223 brouard 6609: char gplotcondition[132];
1.237 brouard 6610: 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 6611: int lv=0, vlv=0, kl=0;
1.130 brouard 6612: int ng=0;
1.201 brouard 6613: int vpopbased;
1.223 brouard 6614: int ioffset; /* variable offset for columns */
1.235 brouard 6615: int nres=0; /* Index of resultline */
1.219 brouard 6616:
1.126 brouard 6617: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6618: /* printf("Problem with file %s",optionfilegnuplot); */
6619: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6620: /* } */
6621:
6622: /*#ifdef windows */
6623: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6624: /*#endif */
1.225 brouard 6625: m=pow(2,cptcoveff);
1.126 brouard 6626:
1.202 brouard 6627: /* Contribution to likelihood */
6628: /* Plot the probability implied in the likelihood */
1.223 brouard 6629: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6630: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6631: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6632: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6633: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6634: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6635: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6636: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6637: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6638: 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));
6639: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6640: 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));
6641: for (i=1; i<= nlstate ; i ++) {
6642: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6643: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6644: 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);
6645: for (j=2; j<= nlstate+ndeath ; j ++) {
6646: 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);
6647: }
6648: fprintf(ficgp,";\nset out; unset ylabel;\n");
6649: }
6650: /* 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 */
6651: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6652: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6653: fprintf(ficgp,"\nset out;unset log\n");
6654: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6655:
1.126 brouard 6656: strcpy(dirfileres,optionfilefiname);
6657: strcpy(optfileres,"vpl");
1.223 brouard 6658: /* 1eme*/
1.238 brouard 6659: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6660: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6661: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6662: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6663: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6664: continue;
6665: /* We are interested in selected combination by the resultline */
1.246 brouard 6666: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6667: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6668: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6669: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6670: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6671: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6672: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6673: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6674: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6675: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6676: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6677: }
6678: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6679: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6680: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6681: }
1.246 brouard 6682: /* printf("\n#\n"); */
1.238 brouard 6683: fprintf(ficgp,"\n#\n");
6684: if(invalidvarcomb[k1]){
1.260 ! brouard 6685: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 6686: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6687: continue;
6688: }
1.235 brouard 6689:
1.241 brouard 6690: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6691: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.260 ! brouard 6692: 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_"),nres-1,nres-1,nres);
! 6693: /* 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); */
! 6694: /* k1-1 error should be nres-1*/
1.238 brouard 6695: for (i=1; i<= nlstate ; i ++) {
6696: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6697: else fprintf(ficgp," %%*lf (%%*lf)");
6698: }
1.260 ! brouard 6699: 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_"),nres-1,nres-1,nres);
1.238 brouard 6700: for (i=1; i<= nlstate ; i ++) {
6701: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6702: else fprintf(ficgp," %%*lf (%%*lf)");
6703: }
1.260 ! brouard 6704: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6705: for (i=1; i<= nlstate ; i ++) {
6706: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6707: else fprintf(ficgp," %%*lf (%%*lf)");
6708: }
6709: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6710: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6711: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6712: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6713: if(cptcoveff ==0){
1.245 brouard 6714: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6715: }else{
6716: kl=0;
6717: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6718: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6719: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6720: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6721: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6722: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6723: kl++;
1.238 brouard 6724: /* 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 *\/ */
6725: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6726: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6727: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6728: if(k==cptcoveff){
1.245 brouard 6729: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6730: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6731: }else{
6732: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6733: kl++;
6734: }
6735: } /* end covariate */
6736: } /* end if no covariate */
6737: } /* end if backcast */
6738: fprintf(ficgp,"\nset out \n");
6739: } /* nres */
1.201 brouard 6740: } /* k1 */
6741: } /* cpt */
1.235 brouard 6742:
6743:
1.126 brouard 6744: /*2 eme*/
1.238 brouard 6745: for (k1=1; k1<= m ; k1 ++){
6746: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6747: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6748: continue;
6749: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6750: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6751: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6752: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6753: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6754: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6755: vlv= nbcode[Tvaraff[k]][lv];
6756: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6757: }
1.237 brouard 6758: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6759: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6760: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6761: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6762: }
1.211 brouard 6763: fprintf(ficgp,"\n#\n");
1.223 brouard 6764: if(invalidvarcomb[k1]){
6765: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6766: continue;
6767: }
1.219 brouard 6768:
1.241 brouard 6769: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6770: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6771: if(vpopbased==0)
6772: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6773: else
6774: fprintf(ficgp,"\nreplot ");
6775: for (i=1; i<= nlstate+1 ; i ++) {
6776: k=2*i;
6777: 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);
6778: for (j=1; j<= nlstate+1 ; j ++) {
6779: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6780: else fprintf(ficgp," %%*lf (%%*lf)");
6781: }
6782: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6783: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6784: 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);
6785: for (j=1; j<= nlstate+1 ; j ++) {
6786: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6787: else fprintf(ficgp," %%*lf (%%*lf)");
6788: }
6789: fprintf(ficgp,"\" t\"\" w l lt 0,");
6790: 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);
6791: for (j=1; j<= nlstate+1 ; j ++) {
6792: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6793: else fprintf(ficgp," %%*lf (%%*lf)");
6794: }
6795: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6796: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6797: } /* state */
6798: } /* vpopbased */
1.244 brouard 6799: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6800: } /* end nres */
6801: } /* k1 end 2 eme*/
6802:
6803:
6804: /*3eme*/
6805: for (k1=1; k1<= m ; k1 ++){
6806: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6807: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6808: continue;
6809:
6810: for (cpt=1; cpt<= nlstate ; cpt ++) {
6811: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6812: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6813: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6814: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6815: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6816: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6817: vlv= nbcode[Tvaraff[k]][lv];
6818: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6819: }
6820: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6821: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6822: }
6823: fprintf(ficgp,"\n#\n");
6824: if(invalidvarcomb[k1]){
6825: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6826: continue;
6827: }
6828:
6829: /* k=2+nlstate*(2*cpt-2); */
6830: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6831: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6832: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6833: 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 6834: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6835: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6836: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6837: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6838: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6839: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6840:
1.238 brouard 6841: */
6842: for (i=1; i< nlstate ; i ++) {
6843: 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);
6844: /* 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 6845:
1.238 brouard 6846: }
6847: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6848: }
6849: } /* end nres */
6850: } /* end kl 3eme */
1.126 brouard 6851:
1.223 brouard 6852: /* 4eme */
1.201 brouard 6853: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6854: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6855: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6856: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6857: continue;
1.238 brouard 6858: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6859: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6860: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6861: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6862: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6863: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6864: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6865: vlv= nbcode[Tvaraff[k]][lv];
6866: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6867: }
6868: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6869: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6870: }
6871: fprintf(ficgp,"\n#\n");
6872: if(invalidvarcomb[k1]){
6873: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6874: continue;
1.223 brouard 6875: }
1.238 brouard 6876:
1.241 brouard 6877: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6878: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6879: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6880: k=3;
6881: for (i=1; i<= nlstate ; i ++){
6882: if(i==1){
6883: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6884: }else{
6885: fprintf(ficgp,", '' ");
6886: }
6887: l=(nlstate+ndeath)*(i-1)+1;
6888: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6889: for (j=2; j<= nlstate+ndeath ; j ++)
6890: fprintf(ficgp,"+$%d",k+l+j-1);
6891: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6892: } /* nlstate */
6893: fprintf(ficgp,"\nset out\n");
6894: } /* end cpt state*/
6895: } /* end nres */
6896: } /* end covariate k1 */
6897:
1.220 brouard 6898: /* 5eme */
1.201 brouard 6899: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6900: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6901: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6902: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6903: continue;
1.238 brouard 6904: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6905: 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);
6906: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6907: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6908: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6909: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6910: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6911: vlv= nbcode[Tvaraff[k]][lv];
6912: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6913: }
6914: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6915: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6916: }
6917: fprintf(ficgp,"\n#\n");
6918: if(invalidvarcomb[k1]){
6919: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6920: continue;
6921: }
1.227 brouard 6922:
1.241 brouard 6923: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6924: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6925: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6926: k=3;
6927: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6928: if(j==1)
6929: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6930: else
6931: fprintf(ficgp,", '' ");
6932: l=(nlstate+ndeath)*(cpt-1) +j;
6933: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6934: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6935: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6936: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6937: } /* nlstate */
6938: fprintf(ficgp,", '' ");
6939: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6940: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6941: l=(nlstate+ndeath)*(cpt-1) +j;
6942: if(j < nlstate)
6943: fprintf(ficgp,"$%d +",k+l);
6944: else
6945: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6946: }
6947: fprintf(ficgp,"\nset out\n");
6948: } /* end cpt state*/
6949: } /* end covariate */
6950: } /* end nres */
1.227 brouard 6951:
1.220 brouard 6952: /* 6eme */
1.202 brouard 6953: /* CV preval stable (period) for each covariate */
1.237 brouard 6954: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6955: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6956: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6957: continue;
1.255 brouard 6958: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6959:
1.211 brouard 6960: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6961: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6962: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6963: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6964: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6965: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6966: vlv= nbcode[Tvaraff[k]][lv];
6967: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6968: }
1.237 brouard 6969: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6970: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6971: }
1.211 brouard 6972: fprintf(ficgp,"\n#\n");
1.223 brouard 6973: if(invalidvarcomb[k1]){
1.227 brouard 6974: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6975: continue;
1.223 brouard 6976: }
1.227 brouard 6977:
1.241 brouard 6978: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6979: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6980: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6981: k=3; /* Offset */
1.255 brouard 6982: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6983: if(i==1)
6984: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6985: else
6986: fprintf(ficgp,", '' ");
1.255 brouard 6987: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6988: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6989: for (j=2; j<= nlstate ; j ++)
6990: fprintf(ficgp,"+$%d",k+l+j-1);
6991: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6992: } /* nlstate */
1.201 brouard 6993: fprintf(ficgp,"\nset out\n");
1.153 brouard 6994: } /* end cpt state*/
6995: } /* end covariate */
1.227 brouard 6996:
6997:
1.220 brouard 6998: /* 7eme */
1.218 brouard 6999: if(backcast == 1){
1.217 brouard 7000: /* CV back preval stable (period) for each covariate */
1.237 brouard 7001: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7002: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7003: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7004: continue;
1.255 brouard 7005: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
7006: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7007: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7008: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7009: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7010: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7011: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7012: vlv= nbcode[Tvaraff[k]][lv];
7013: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7014: }
1.237 brouard 7015: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7016: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7017: }
1.227 brouard 7018: fprintf(ficgp,"\n#\n");
7019: if(invalidvarcomb[k1]){
7020: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7021: continue;
7022: }
7023:
1.241 brouard 7024: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7025: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7026: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7027: k=3; /* Offset */
1.255 brouard 7028: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7029: if(i==1)
7030: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7031: else
7032: fprintf(ficgp,", '' ");
7033: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7034: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7035: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7036: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
1.255 brouard 7037: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7038: /* for (j=2; j<= nlstate ; j ++) */
7039: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7040: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7041: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7042: } /* nlstate */
7043: fprintf(ficgp,"\nset out\n");
1.218 brouard 7044: } /* end cpt state*/
7045: } /* end covariate */
7046: } /* End if backcast */
7047:
1.223 brouard 7048: /* 8eme */
1.218 brouard 7049: if(prevfcast==1){
7050: /* Projection from cross-sectional to stable (period) for each covariate */
7051:
1.237 brouard 7052: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7053: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7054: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7055: continue;
1.211 brouard 7056: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7057: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7058: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7059: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7060: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7061: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7062: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7063: vlv= nbcode[Tvaraff[k]][lv];
7064: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7065: }
1.237 brouard 7066: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7067: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7068: }
1.227 brouard 7069: fprintf(ficgp,"\n#\n");
7070: if(invalidvarcomb[k1]){
7071: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7072: continue;
7073: }
7074:
7075: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7076: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7077: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7078: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7079: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7080: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7081: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7082: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7083: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7084: if(i==1){
7085: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7086: }else{
7087: fprintf(ficgp,",\\\n '' ");
7088: }
7089: if(cptcoveff ==0){ /* No covariate */
7090: ioffset=2; /* Age is in 2 */
7091: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7092: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7093: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7094: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7095: fprintf(ficgp," u %d:(", ioffset);
7096: if(i==nlstate+1)
7097: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7098: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7099: else
7100: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7101: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7102: }else{ /* more than 2 covariates */
7103: if(cptcoveff ==1){
7104: ioffset=4; /* Age is in 4 */
7105: }else{
7106: ioffset=6; /* Age is in 6 */
7107: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7108: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7109: }
7110: fprintf(ficgp," u %d:(",ioffset);
7111: kl=0;
7112: strcpy(gplotcondition,"(");
7113: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7114: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7115: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7116: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7117: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7118: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7119: kl++;
7120: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7121: kl++;
7122: if(k <cptcoveff && cptcoveff>1)
7123: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7124: }
7125: strcpy(gplotcondition+strlen(gplotcondition),")");
7126: /* 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 *\/ */
7127: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7128: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7129: /* '' 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*/
7130: if(i==nlstate+1){
7131: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7132: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7133: }else{
7134: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7135: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7136: }
7137: } /* end if covariate */
7138: } /* nlstate */
7139: fprintf(ficgp,"\nset out\n");
1.223 brouard 7140: } /* end cpt state*/
7141: } /* end covariate */
7142: } /* End if prevfcast */
1.227 brouard 7143:
7144:
1.238 brouard 7145: /* 9eme writing MLE parameters */
7146: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7147: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7148: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7149: for(k=1; k <=(nlstate+ndeath); k++){
7150: if (k != i) {
1.227 brouard 7151: fprintf(ficgp,"# current state %d\n",k);
7152: for(j=1; j <=ncovmodel; j++){
7153: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7154: jk++;
7155: }
7156: fprintf(ficgp,"\n");
1.126 brouard 7157: }
7158: }
1.223 brouard 7159: }
1.187 brouard 7160: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7161:
1.145 brouard 7162: /*goto avoid;*/
1.238 brouard 7163: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7164: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7165: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7166: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7167: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7168: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7169: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7170: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7171: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7172: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7173: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7174: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7175: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7176: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7177: fprintf(ficgp,"#\n");
1.223 brouard 7178: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7179: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7180: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7181: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7182: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7183: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7184: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7185: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7186: continue;
7187: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7188: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7189: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7190: }
7191: fprintf(ficgp,"\n#\n");
1.241 brouard 7192: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7193: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7194: if (ng==1){
7195: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7196: fprintf(ficgp,"\nunset log y");
7197: }else if (ng==2){
7198: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7199: fprintf(ficgp,"\nset log y");
7200: }else if (ng==3){
7201: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7202: fprintf(ficgp,"\nset log y");
7203: }else
7204: fprintf(ficgp,"\nunset title ");
7205: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7206: i=1;
7207: for(k2=1; k2<=nlstate; k2++) {
7208: k3=i;
7209: for(k=1; k<=(nlstate+ndeath); k++) {
7210: if (k != k2){
7211: switch( ng) {
7212: case 1:
7213: if(nagesqr==0)
7214: fprintf(ficgp," p%d+p%d*x",i,i+1);
7215: else /* nagesqr =1 */
7216: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7217: break;
7218: case 2: /* ng=2 */
7219: if(nagesqr==0)
7220: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7221: else /* nagesqr =1 */
7222: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7223: break;
7224: case 3:
7225: if(nagesqr==0)
7226: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7227: else /* nagesqr =1 */
7228: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7229: break;
7230: }
7231: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7232: ijp=1; /* product no age */
7233: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7234: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7235: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7236: if(j==Tage[ij]) { /* Product by age */
7237: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7238: if(DummyV[j]==0){
1.237 brouard 7239: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7240: }else{ /* quantitative */
7241: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7242: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7243: }
7244: ij++;
7245: }
7246: }else if(j==Tprod[ijp]) { /* */
7247: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7248: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7249: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7250: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7251: /* 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)]); */
7252: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7253: }else{ /* Vn is dummy and Vm is quanti */
7254: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7255: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7256: }
7257: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7258: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7259: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7260: }else{ /* Both quanti */
7261: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7262: }
7263: }
1.238 brouard 7264: ijp++;
1.237 brouard 7265: }
7266: } else{ /* simple covariate */
7267: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7268: if(Dummy[j]==0){
7269: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7270: }else{ /* quantitative */
7271: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7272: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7273: }
1.237 brouard 7274: } /* end simple */
7275: } /* end j */
1.223 brouard 7276: }else{
7277: i=i-ncovmodel;
7278: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7279: fprintf(ficgp," (1.");
7280: }
1.227 brouard 7281:
1.223 brouard 7282: if(ng != 1){
7283: fprintf(ficgp,")/(1");
1.227 brouard 7284:
1.223 brouard 7285: for(k1=1; k1 <=nlstate; k1++){
7286: if(nagesqr==0)
7287: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7288: else /* nagesqr =1 */
7289: 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 7290:
1.223 brouard 7291: ij=1;
7292: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7293: if((j-2)==Tage[ij]) { /* Bug valgrind */
7294: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7295: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7296: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7297: ij++;
7298: }
7299: }
7300: else
1.225 brouard 7301: 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 7302: }
7303: fprintf(ficgp,")");
7304: }
7305: fprintf(ficgp,")");
7306: if(ng ==2)
7307: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7308: else /* ng= 3 */
7309: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7310: }else{ /* end ng <> 1 */
7311: if( k !=k2) /* logit p11 is hard to draw */
7312: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7313: }
7314: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7315: fprintf(ficgp,",");
7316: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7317: fprintf(ficgp,",");
7318: i=i+ncovmodel;
7319: } /* end k */
7320: } /* end k2 */
7321: fprintf(ficgp,"\n set out\n");
7322: } /* end jk */
7323: } /* end ng */
7324: /* avoid: */
7325: fflush(ficgp);
1.126 brouard 7326: } /* end gnuplot */
7327:
7328:
7329: /*************** Moving average **************/
1.219 brouard 7330: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7331: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7332:
1.222 brouard 7333: int i, cpt, cptcod;
7334: int modcovmax =1;
7335: int mobilavrange, mob;
7336: int iage=0;
7337:
7338: double sum=0.;
7339: double age;
7340: double *sumnewp, *sumnewm;
7341: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7342:
7343:
1.225 brouard 7344: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7345: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7346:
7347: sumnewp = vector(1,ncovcombmax);
7348: sumnewm = vector(1,ncovcombmax);
7349: agemingood = vector(1,ncovcombmax);
7350: agemaxgood = vector(1,ncovcombmax);
7351:
7352: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7353: sumnewm[cptcod]=0.;
7354: sumnewp[cptcod]=0.;
7355: agemingood[cptcod]=0;
7356: agemaxgood[cptcod]=0;
7357: }
7358: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7359:
7360: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7361: if(mobilav==1) mobilavrange=5; /* default */
7362: else mobilavrange=mobilav;
7363: for (age=bage; age<=fage; age++)
7364: for (i=1; i<=nlstate;i++)
7365: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7366: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7367: /* We keep the original values on the extreme ages bage, fage and for
7368: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7369: we use a 5 terms etc. until the borders are no more concerned.
7370: */
7371: for (mob=3;mob <=mobilavrange;mob=mob+2){
7372: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7373: for (i=1; i<=nlstate;i++){
7374: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7375: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7376: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7377: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7378: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7379: }
7380: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7381: }
7382: }
7383: }/* end age */
7384: }/* end mob */
7385: }else
7386: return -1;
7387: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7388: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7389: if(invalidvarcomb[cptcod]){
7390: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7391: continue;
7392: }
1.219 brouard 7393:
1.222 brouard 7394: agemingood[cptcod]=fage-(mob-1)/2;
7395: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7396: sumnewm[cptcod]=0.;
7397: for (i=1; i<=nlstate;i++){
7398: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7399: }
7400: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7401: agemingood[cptcod]=age;
7402: }else{ /* bad */
7403: for (i=1; i<=nlstate;i++){
7404: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7405: } /* i */
7406: } /* end bad */
7407: }/* age */
7408: sum=0.;
7409: for (i=1; i<=nlstate;i++){
7410: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7411: }
7412: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7413: 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);
7414: /* for (i=1; i<=nlstate;i++){ */
7415: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7416: /* } /\* i *\/ */
7417: } /* end bad */
7418: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7419: /* From youngest, finding the oldest wrong */
7420: agemaxgood[cptcod]=bage+(mob-1)/2;
7421: for (age=bage+(mob-1)/2; age<=fage; age++){
7422: sumnewm[cptcod]=0.;
7423: for (i=1; i<=nlstate;i++){
7424: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7425: }
7426: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7427: agemaxgood[cptcod]=age;
7428: }else{ /* bad */
7429: for (i=1; i<=nlstate;i++){
7430: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7431: } /* i */
7432: } /* end bad */
7433: }/* age */
7434: sum=0.;
7435: for (i=1; i<=nlstate;i++){
7436: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7437: }
7438: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7439: 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);
7440: /* for (i=1; i<=nlstate;i++){ */
7441: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7442: /* } /\* i *\/ */
7443: } /* end bad */
7444:
7445: for (age=bage; age<=fage; age++){
1.235 brouard 7446: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7447: sumnewp[cptcod]=0.;
7448: sumnewm[cptcod]=0.;
7449: for (i=1; i<=nlstate;i++){
7450: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7451: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7452: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7453: }
7454: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7455: }
7456: /* printf("\n"); */
7457: /* } */
7458: /* brutal averaging */
7459: for (i=1; i<=nlstate;i++){
7460: for (age=1; age<=bage; age++){
7461: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7462: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7463: }
7464: for (age=fage; age<=AGESUP; age++){
7465: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7466: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7467: }
7468: } /* end i status */
7469: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7470: for (age=1; age<=AGESUP; age++){
7471: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7472: mobaverage[(int)age][i][cptcod]=0.;
7473: }
7474: }
7475: }/* end cptcod */
7476: free_vector(sumnewm,1, ncovcombmax);
7477: free_vector(sumnewp,1, ncovcombmax);
7478: free_vector(agemaxgood,1, ncovcombmax);
7479: free_vector(agemingood,1, ncovcombmax);
7480: return 0;
7481: }/* End movingaverage */
1.218 brouard 7482:
1.126 brouard 7483:
7484: /************** Forecasting ******************/
1.235 brouard 7485: 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 7486: /* proj1, year, month, day of starting projection
7487: agemin, agemax range of age
7488: dateprev1 dateprev2 range of dates during which prevalence is computed
7489: anproj2 year of en of projection (same day and month as proj1).
7490: */
1.235 brouard 7491: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7492: double agec; /* generic age */
7493: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7494: double *popeffectif,*popcount;
7495: double ***p3mat;
1.218 brouard 7496: /* double ***mobaverage; */
1.126 brouard 7497: char fileresf[FILENAMELENGTH];
7498:
7499: agelim=AGESUP;
1.211 brouard 7500: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7501: in each health status at the date of interview (if between dateprev1 and dateprev2).
7502: We still use firstpass and lastpass as another selection.
7503: */
1.214 brouard 7504: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7505: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7506:
1.201 brouard 7507: strcpy(fileresf,"F_");
7508: strcat(fileresf,fileresu);
1.126 brouard 7509: if((ficresf=fopen(fileresf,"w"))==NULL) {
7510: printf("Problem with forecast resultfile: %s\n", fileresf);
7511: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7512: }
1.235 brouard 7513: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7514: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7515:
1.225 brouard 7516: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7517:
7518:
7519: stepsize=(int) (stepm+YEARM-1)/YEARM;
7520: if (stepm<=12) stepsize=1;
7521: if(estepm < stepm){
7522: printf ("Problem %d lower than %d\n",estepm, stepm);
7523: }
7524: else hstepm=estepm;
7525:
7526: hstepm=hstepm/stepm;
7527: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7528: fractional in yp1 */
7529: anprojmean=yp;
7530: yp2=modf((yp1*12),&yp);
7531: mprojmean=yp;
7532: yp1=modf((yp2*30.5),&yp);
7533: jprojmean=yp;
7534: if(jprojmean==0) jprojmean=1;
7535: if(mprojmean==0) jprojmean=1;
7536:
1.227 brouard 7537: i1=pow(2,cptcoveff);
1.126 brouard 7538: if (cptcovn < 1){i1=1;}
7539:
7540: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7541:
7542: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7543:
1.126 brouard 7544: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7545: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7546: for(k=1; k<=i1;k++){
1.253 brouard 7547: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7548: continue;
1.227 brouard 7549: if(invalidvarcomb[k]){
7550: printf("\nCombination (%d) projection ignored because no cases \n",k);
7551: continue;
7552: }
7553: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7554: for(j=1;j<=cptcoveff;j++) {
7555: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7556: }
1.235 brouard 7557: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7558: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7559: }
1.227 brouard 7560: fprintf(ficresf," yearproj age");
7561: for(j=1; j<=nlstate+ndeath;j++){
7562: for(i=1; i<=nlstate;i++)
7563: fprintf(ficresf," p%d%d",i,j);
7564: fprintf(ficresf," wp.%d",j);
7565: }
7566: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7567: fprintf(ficresf,"\n");
7568: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7569: for (agec=fage; agec>=(ageminpar-1); agec--){
7570: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7571: nhstepm = nhstepm/hstepm;
7572: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7573: oldm=oldms;savm=savms;
1.235 brouard 7574: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7575:
7576: for (h=0; h<=nhstepm; h++){
7577: if (h*hstepm/YEARM*stepm ==yearp) {
7578: fprintf(ficresf,"\n");
7579: for(j=1;j<=cptcoveff;j++)
7580: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7581: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7582: }
7583: for(j=1; j<=nlstate+ndeath;j++) {
7584: ppij=0.;
7585: for(i=1; i<=nlstate;i++) {
7586: if (mobilav==1)
7587: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7588: else {
7589: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7590: }
7591: if (h*hstepm/YEARM*stepm== yearp) {
7592: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7593: }
7594: } /* end i */
7595: if (h*hstepm/YEARM*stepm==yearp) {
7596: fprintf(ficresf," %.3f", ppij);
7597: }
7598: }/* end j */
7599: } /* end h */
7600: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7601: } /* end agec */
7602: } /* end yearp */
7603: } /* end k */
1.219 brouard 7604:
1.126 brouard 7605: fclose(ficresf);
1.215 brouard 7606: printf("End of Computing forecasting \n");
7607: fprintf(ficlog,"End of Computing forecasting\n");
7608:
1.126 brouard 7609: }
7610:
1.218 brouard 7611: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7612: /* 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 7613: /* /\* back1, year, month, day of starting backection */
7614: /* agemin, agemax range of age */
7615: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7616: /* anback2 year of en of backection (same day and month as back1). */
7617: /* *\/ */
7618: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7619: /* double agec; /\* generic age *\/ */
7620: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7621: /* double *popeffectif,*popcount; */
7622: /* double ***p3mat; */
7623: /* /\* double ***mobaverage; *\/ */
7624: /* char fileresfb[FILENAMELENGTH]; */
7625:
7626: /* agelim=AGESUP; */
7627: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7628: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7629: /* We still use firstpass and lastpass as another selection. */
7630: /* *\/ */
7631: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7632: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7633: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7634:
7635: /* strcpy(fileresfb,"FB_"); */
7636: /* strcat(fileresfb,fileresu); */
7637: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7638: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7639: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7640: /* } */
7641: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7642: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7643:
1.225 brouard 7644: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7645:
7646: /* /\* if (mobilav!=0) { *\/ */
7647: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7648: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7649: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7650: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7651: /* /\* } *\/ */
7652: /* /\* } *\/ */
7653:
7654: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7655: /* if (stepm<=12) stepsize=1; */
7656: /* if(estepm < stepm){ */
7657: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7658: /* } */
7659: /* else hstepm=estepm; */
7660:
7661: /* hstepm=hstepm/stepm; */
7662: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7663: /* fractional in yp1 *\/ */
7664: /* anprojmean=yp; */
7665: /* yp2=modf((yp1*12),&yp); */
7666: /* mprojmean=yp; */
7667: /* yp1=modf((yp2*30.5),&yp); */
7668: /* jprojmean=yp; */
7669: /* if(jprojmean==0) jprojmean=1; */
7670: /* if(mprojmean==0) jprojmean=1; */
7671:
1.225 brouard 7672: /* i1=cptcoveff; */
1.218 brouard 7673: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7674:
1.218 brouard 7675: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7676:
1.218 brouard 7677: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7678:
7679: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7680: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7681: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7682: /* k=k+1; */
7683: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7684: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7685: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7686: /* } */
7687: /* fprintf(ficresfb," yearbproj age"); */
7688: /* for(j=1; j<=nlstate+ndeath;j++){ */
7689: /* for(i=1; i<=nlstate;i++) */
7690: /* fprintf(ficresfb," p%d%d",i,j); */
7691: /* fprintf(ficresfb," p.%d",j); */
7692: /* } */
7693: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7694: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7695: /* fprintf(ficresfb,"\n"); */
7696: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7697: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7698: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7699: /* nhstepm = nhstepm/hstepm; */
7700: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7701: /* oldm=oldms;savm=savms; */
7702: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7703: /* for (h=0; h<=nhstepm; h++){ */
7704: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7705: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7706: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7707: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7708: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7709: /* } */
7710: /* for(j=1; j<=nlstate+ndeath;j++) { */
7711: /* ppij=0.; */
7712: /* for(i=1; i<=nlstate;i++) { */
7713: /* if (mobilav==1) */
7714: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7715: /* else { */
7716: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7717: /* } */
7718: /* if (h*hstepm/YEARM*stepm== yearp) { */
7719: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7720: /* } */
7721: /* } /\* end i *\/ */
7722: /* if (h*hstepm/YEARM*stepm==yearp) { */
7723: /* fprintf(ficresfb," %.3f", ppij); */
7724: /* } */
7725: /* }/\* end j *\/ */
7726: /* } /\* end h *\/ */
7727: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7728: /* } /\* end agec *\/ */
7729: /* } /\* end yearp *\/ */
7730: /* } /\* end cptcod *\/ */
7731: /* } /\* end cptcov *\/ */
7732:
7733: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7734:
7735: /* fclose(ficresfb); */
7736: /* printf("End of Computing Back forecasting \n"); */
7737: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7738:
1.218 brouard 7739: /* } */
1.217 brouard 7740:
1.126 brouard 7741: /************** Forecasting *****not tested NB*************/
1.227 brouard 7742: /* 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 7743:
1.227 brouard 7744: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7745: /* int *popage; */
7746: /* double calagedatem, agelim, kk1, kk2; */
7747: /* double *popeffectif,*popcount; */
7748: /* double ***p3mat,***tabpop,***tabpopprev; */
7749: /* /\* double ***mobaverage; *\/ */
7750: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7751:
1.227 brouard 7752: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7753: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7754: /* agelim=AGESUP; */
7755: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7756:
1.227 brouard 7757: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7758:
7759:
1.227 brouard 7760: /* strcpy(filerespop,"POP_"); */
7761: /* strcat(filerespop,fileresu); */
7762: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7763: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7764: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7765: /* } */
7766: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7767: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7768:
1.227 brouard 7769: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7770:
1.227 brouard 7771: /* /\* if (mobilav!=0) { *\/ */
7772: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7773: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7774: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7775: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7776: /* /\* } *\/ */
7777: /* /\* } *\/ */
1.126 brouard 7778:
1.227 brouard 7779: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7780: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7781:
1.227 brouard 7782: /* agelim=AGESUP; */
1.126 brouard 7783:
1.227 brouard 7784: /* hstepm=1; */
7785: /* hstepm=hstepm/stepm; */
1.218 brouard 7786:
1.227 brouard 7787: /* if (popforecast==1) { */
7788: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7789: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7790: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7791: /* } */
7792: /* popage=ivector(0,AGESUP); */
7793: /* popeffectif=vector(0,AGESUP); */
7794: /* popcount=vector(0,AGESUP); */
1.126 brouard 7795:
1.227 brouard 7796: /* i=1; */
7797: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7798:
1.227 brouard 7799: /* imx=i; */
7800: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7801: /* } */
1.218 brouard 7802:
1.227 brouard 7803: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7804: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7805: /* k=k+1; */
7806: /* fprintf(ficrespop,"\n#******"); */
7807: /* for(j=1;j<=cptcoveff;j++) { */
7808: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7809: /* } */
7810: /* fprintf(ficrespop,"******\n"); */
7811: /* fprintf(ficrespop,"# Age"); */
7812: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7813: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7814:
1.227 brouard 7815: /* for (cpt=0; cpt<=0;cpt++) { */
7816: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7817:
1.227 brouard 7818: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7819: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7820: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7821:
1.227 brouard 7822: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7823: /* oldm=oldms;savm=savms; */
7824: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7825:
1.227 brouard 7826: /* for (h=0; h<=nhstepm; h++){ */
7827: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7828: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7829: /* } */
7830: /* for(j=1; j<=nlstate+ndeath;j++) { */
7831: /* kk1=0.;kk2=0; */
7832: /* for(i=1; i<=nlstate;i++) { */
7833: /* if (mobilav==1) */
7834: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7835: /* else { */
7836: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7837: /* } */
7838: /* } */
7839: /* if (h==(int)(calagedatem+12*cpt)){ */
7840: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7841: /* /\*fprintf(ficrespop," %.3f", kk1); */
7842: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7843: /* } */
7844: /* } */
7845: /* for(i=1; i<=nlstate;i++){ */
7846: /* kk1=0.; */
7847: /* for(j=1; j<=nlstate;j++){ */
7848: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7849: /* } */
7850: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7851: /* } */
1.218 brouard 7852:
1.227 brouard 7853: /* if (h==(int)(calagedatem+12*cpt)) */
7854: /* for(j=1; j<=nlstate;j++) */
7855: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7856: /* } */
7857: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7858: /* } */
7859: /* } */
1.218 brouard 7860:
1.227 brouard 7861: /* /\******\/ */
1.218 brouard 7862:
1.227 brouard 7863: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7864: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7865: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7866: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7867: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7868:
1.227 brouard 7869: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7870: /* oldm=oldms;savm=savms; */
7871: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7872: /* for (h=0; h<=nhstepm; h++){ */
7873: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7874: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7875: /* } */
7876: /* for(j=1; j<=nlstate+ndeath;j++) { */
7877: /* kk1=0.;kk2=0; */
7878: /* for(i=1; i<=nlstate;i++) { */
7879: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7880: /* } */
7881: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7882: /* } */
7883: /* } */
7884: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7885: /* } */
7886: /* } */
7887: /* } */
7888: /* } */
1.218 brouard 7889:
1.227 brouard 7890: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7891:
1.227 brouard 7892: /* if (popforecast==1) { */
7893: /* free_ivector(popage,0,AGESUP); */
7894: /* free_vector(popeffectif,0,AGESUP); */
7895: /* free_vector(popcount,0,AGESUP); */
7896: /* } */
7897: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7898: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7899: /* fclose(ficrespop); */
7900: /* } /\* End of popforecast *\/ */
1.218 brouard 7901:
1.126 brouard 7902: int fileappend(FILE *fichier, char *optionfich)
7903: {
7904: if((fichier=fopen(optionfich,"a"))==NULL) {
7905: printf("Problem with file: %s\n", optionfich);
7906: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7907: return (0);
7908: }
7909: fflush(fichier);
7910: return (1);
7911: }
7912:
7913:
7914: /**************** function prwizard **********************/
7915: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7916: {
7917:
7918: /* Wizard to print covariance matrix template */
7919:
1.164 brouard 7920: char ca[32], cb[32];
7921: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7922: int numlinepar;
7923:
7924: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7925: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7926: for(i=1; i <=nlstate; i++){
7927: jj=0;
7928: for(j=1; j <=nlstate+ndeath; j++){
7929: if(j==i) continue;
7930: jj++;
7931: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7932: printf("%1d%1d",i,j);
7933: fprintf(ficparo,"%1d%1d",i,j);
7934: for(k=1; k<=ncovmodel;k++){
7935: /* printf(" %lf",param[i][j][k]); */
7936: /* fprintf(ficparo," %lf",param[i][j][k]); */
7937: printf(" 0.");
7938: fprintf(ficparo," 0.");
7939: }
7940: printf("\n");
7941: fprintf(ficparo,"\n");
7942: }
7943: }
7944: printf("# Scales (for hessian or gradient estimation)\n");
7945: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7946: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7947: for(i=1; i <=nlstate; i++){
7948: jj=0;
7949: for(j=1; j <=nlstate+ndeath; j++){
7950: if(j==i) continue;
7951: jj++;
7952: fprintf(ficparo,"%1d%1d",i,j);
7953: printf("%1d%1d",i,j);
7954: fflush(stdout);
7955: for(k=1; k<=ncovmodel;k++){
7956: /* printf(" %le",delti3[i][j][k]); */
7957: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7958: printf(" 0.");
7959: fprintf(ficparo," 0.");
7960: }
7961: numlinepar++;
7962: printf("\n");
7963: fprintf(ficparo,"\n");
7964: }
7965: }
7966: printf("# Covariance matrix\n");
7967: /* # 121 Var(a12)\n\ */
7968: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7969: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7970: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7971: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7972: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7973: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7974: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7975: fflush(stdout);
7976: fprintf(ficparo,"# Covariance matrix\n");
7977: /* # 121 Var(a12)\n\ */
7978: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7979: /* # ...\n\ */
7980: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7981:
7982: for(itimes=1;itimes<=2;itimes++){
7983: jj=0;
7984: for(i=1; i <=nlstate; i++){
7985: for(j=1; j <=nlstate+ndeath; j++){
7986: if(j==i) continue;
7987: for(k=1; k<=ncovmodel;k++){
7988: jj++;
7989: ca[0]= k+'a'-1;ca[1]='\0';
7990: if(itimes==1){
7991: printf("#%1d%1d%d",i,j,k);
7992: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7993: }else{
7994: printf("%1d%1d%d",i,j,k);
7995: fprintf(ficparo,"%1d%1d%d",i,j,k);
7996: /* printf(" %.5le",matcov[i][j]); */
7997: }
7998: ll=0;
7999: for(li=1;li <=nlstate; li++){
8000: for(lj=1;lj <=nlstate+ndeath; lj++){
8001: if(lj==li) continue;
8002: for(lk=1;lk<=ncovmodel;lk++){
8003: ll++;
8004: if(ll<=jj){
8005: cb[0]= lk +'a'-1;cb[1]='\0';
8006: if(ll<jj){
8007: if(itimes==1){
8008: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8009: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8010: }else{
8011: printf(" 0.");
8012: fprintf(ficparo," 0.");
8013: }
8014: }else{
8015: if(itimes==1){
8016: printf(" Var(%s%1d%1d)",ca,i,j);
8017: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8018: }else{
8019: printf(" 0.");
8020: fprintf(ficparo," 0.");
8021: }
8022: }
8023: }
8024: } /* end lk */
8025: } /* end lj */
8026: } /* end li */
8027: printf("\n");
8028: fprintf(ficparo,"\n");
8029: numlinepar++;
8030: } /* end k*/
8031: } /*end j */
8032: } /* end i */
8033: } /* end itimes */
8034:
8035: } /* end of prwizard */
8036: /******************* Gompertz Likelihood ******************************/
8037: double gompertz(double x[])
8038: {
8039: double A,B,L=0.0,sump=0.,num=0.;
8040: int i,n=0; /* n is the size of the sample */
8041:
1.220 brouard 8042: for (i=1;i<=imx ; i++) {
1.126 brouard 8043: sump=sump+weight[i];
8044: /* sump=sump+1;*/
8045: num=num+1;
8046: }
8047:
8048:
8049: /* for (i=0; i<=imx; i++)
8050: 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]);*/
8051:
8052: for (i=1;i<=imx ; i++)
8053: {
8054: if (cens[i] == 1 && wav[i]>1)
8055: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8056:
8057: if (cens[i] == 0 && wav[i]>1)
8058: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8059: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8060:
8061: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8062: if (wav[i] > 1 ) { /* ??? */
8063: L=L+A*weight[i];
8064: /* 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]);*/
8065: }
8066: }
8067:
8068: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8069:
8070: return -2*L*num/sump;
8071: }
8072:
1.136 brouard 8073: #ifdef GSL
8074: /******************* Gompertz_f Likelihood ******************************/
8075: double gompertz_f(const gsl_vector *v, void *params)
8076: {
8077: double A,B,LL=0.0,sump=0.,num=0.;
8078: double *x= (double *) v->data;
8079: int i,n=0; /* n is the size of the sample */
8080:
8081: for (i=0;i<=imx-1 ; i++) {
8082: sump=sump+weight[i];
8083: /* sump=sump+1;*/
8084: num=num+1;
8085: }
8086:
8087:
8088: /* for (i=0; i<=imx; i++)
8089: 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]);*/
8090: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8091: for (i=1;i<=imx ; i++)
8092: {
8093: if (cens[i] == 1 && wav[i]>1)
8094: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8095:
8096: if (cens[i] == 0 && wav[i]>1)
8097: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8098: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8099:
8100: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8101: if (wav[i] > 1 ) { /* ??? */
8102: LL=LL+A*weight[i];
8103: /* 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]);*/
8104: }
8105: }
8106:
8107: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8108: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8109:
8110: return -2*LL*num/sump;
8111: }
8112: #endif
8113:
1.126 brouard 8114: /******************* Printing html file ***********/
1.201 brouard 8115: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8116: int lastpass, int stepm, int weightopt, char model[],\
8117: int imx, double p[],double **matcov,double agemortsup){
8118: int i,k;
8119:
8120: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8121: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8122: for (i=1;i<=2;i++)
8123: 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 8124: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8125: fprintf(fichtm,"</ul>");
8126:
8127: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8128:
8129: 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>");
8130:
8131: for (k=agegomp;k<(agemortsup-2);k++)
8132: 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]);
8133:
8134:
8135: fflush(fichtm);
8136: }
8137:
8138: /******************* Gnuplot file **************/
1.201 brouard 8139: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8140:
8141: char dirfileres[132],optfileres[132];
1.164 brouard 8142:
1.126 brouard 8143: int ng;
8144:
8145:
8146: /*#ifdef windows */
8147: fprintf(ficgp,"cd \"%s\" \n",pathc);
8148: /*#endif */
8149:
8150:
8151: strcpy(dirfileres,optionfilefiname);
8152: strcpy(optfileres,"vpl");
1.199 brouard 8153: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8154: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8155: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8156: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8157: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8158:
8159: }
8160:
1.136 brouard 8161: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8162: {
1.126 brouard 8163:
1.136 brouard 8164: /*-------- data file ----------*/
8165: FILE *fic;
8166: char dummy[]=" ";
1.240 brouard 8167: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8168: int lstra;
1.136 brouard 8169: int linei, month, year,iout;
8170: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8171: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8172: char *stratrunc;
1.223 brouard 8173:
1.240 brouard 8174: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8175: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8176:
1.240 brouard 8177: for(v=1; v <=ncovcol;v++){
8178: DummyV[v]=0;
8179: FixedV[v]=0;
8180: }
8181: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8182: DummyV[v]=1;
8183: FixedV[v]=0;
8184: }
8185: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8186: DummyV[v]=0;
8187: FixedV[v]=1;
8188: }
8189: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8190: DummyV[v]=1;
8191: FixedV[v]=1;
8192: }
8193: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8194: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8195: 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]);
8196: }
1.126 brouard 8197:
1.136 brouard 8198: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8199: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8200: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8201: }
1.126 brouard 8202:
1.136 brouard 8203: i=1;
8204: linei=0;
8205: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8206: linei=linei+1;
8207: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8208: if(line[j] == '\t')
8209: line[j] = ' ';
8210: }
8211: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8212: ;
8213: };
8214: line[j+1]=0; /* Trims blanks at end of line */
8215: if(line[0]=='#'){
8216: fprintf(ficlog,"Comment line\n%s\n",line);
8217: printf("Comment line\n%s\n",line);
8218: continue;
8219: }
8220: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8221: strcpy(line, linetmp);
1.223 brouard 8222:
8223: /* Loops on waves */
8224: for (j=maxwav;j>=1;j--){
8225: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8226: cutv(stra, strb, line, ' ');
8227: if(strb[0]=='.') { /* Missing value */
8228: lval=-1;
8229: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8230: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8231: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8232: 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);
8233: 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);
8234: return 1;
8235: }
8236: }else{
8237: errno=0;
8238: /* what_kind_of_number(strb); */
8239: dval=strtod(strb,&endptr);
8240: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8241: /* if(strb != endptr && *endptr == '\0') */
8242: /* dval=dlval; */
8243: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8244: if( strb[0]=='\0' || (*endptr != '\0')){
8245: 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);
8246: 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);
8247: return 1;
8248: }
8249: cotqvar[j][iv][i]=dval;
8250: cotvar[j][ntv+iv][i]=dval;
8251: }
8252: strcpy(line,stra);
1.223 brouard 8253: }/* end loop ntqv */
1.225 brouard 8254:
1.223 brouard 8255: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8256: cutv(stra, strb, line, ' ');
8257: if(strb[0]=='.') { /* Missing value */
8258: lval=-1;
8259: }else{
8260: errno=0;
8261: lval=strtol(strb,&endptr,10);
8262: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8263: if( strb[0]=='\0' || (*endptr != '\0')){
8264: 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);
8265: 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);
8266: return 1;
8267: }
8268: }
8269: if(lval <-1 || lval >1){
8270: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8271: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8272: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8273: For example, for multinomial values like 1, 2 and 3,\n \
8274: build V1=0 V2=0 for the reference value (1),\n \
8275: V1=1 V2=0 for (2) \n \
1.223 brouard 8276: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8277: output of IMaCh is often meaningless.\n \
1.223 brouard 8278: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8279: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8280: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8281: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8282: For example, for multinomial values like 1, 2 and 3,\n \
8283: build V1=0 V2=0 for the reference value (1),\n \
8284: V1=1 V2=0 for (2) \n \
1.223 brouard 8285: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8286: output of IMaCh is often meaningless.\n \
1.223 brouard 8287: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8288: return 1;
8289: }
8290: cotvar[j][iv][i]=(double)(lval);
8291: strcpy(line,stra);
1.223 brouard 8292: }/* end loop ntv */
1.225 brouard 8293:
1.223 brouard 8294: /* Statuses at wave */
1.137 brouard 8295: cutv(stra, strb, line, ' ');
1.223 brouard 8296: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8297: lval=-1;
1.136 brouard 8298: }else{
1.238 brouard 8299: errno=0;
8300: lval=strtol(strb,&endptr,10);
8301: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8302: if( strb[0]=='\0' || (*endptr != '\0')){
8303: 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);
8304: 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);
8305: return 1;
8306: }
1.136 brouard 8307: }
1.225 brouard 8308:
1.136 brouard 8309: s[j][i]=lval;
1.225 brouard 8310:
1.223 brouard 8311: /* Date of Interview */
1.136 brouard 8312: strcpy(line,stra);
8313: cutv(stra, strb,line,' ');
1.169 brouard 8314: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8315: }
1.169 brouard 8316: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8317: month=99;
8318: year=9999;
1.136 brouard 8319: }else{
1.225 brouard 8320: 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);
8321: 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);
8322: return 1;
1.136 brouard 8323: }
8324: anint[j][i]= (double) year;
8325: mint[j][i]= (double)month;
8326: strcpy(line,stra);
1.223 brouard 8327: } /* End loop on waves */
1.225 brouard 8328:
1.223 brouard 8329: /* Date of death */
1.136 brouard 8330: cutv(stra, strb,line,' ');
1.169 brouard 8331: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8332: }
1.169 brouard 8333: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8334: month=99;
8335: year=9999;
8336: }else{
1.141 brouard 8337: 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 8338: 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);
8339: return 1;
1.136 brouard 8340: }
8341: andc[i]=(double) year;
8342: moisdc[i]=(double) month;
8343: strcpy(line,stra);
8344:
1.223 brouard 8345: /* Date of birth */
1.136 brouard 8346: cutv(stra, strb,line,' ');
1.169 brouard 8347: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8348: }
1.169 brouard 8349: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8350: month=99;
8351: year=9999;
8352: }else{
1.141 brouard 8353: 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);
8354: 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 8355: return 1;
1.136 brouard 8356: }
8357: if (year==9999) {
1.141 brouard 8358: 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);
8359: 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 8360: return 1;
8361:
1.136 brouard 8362: }
8363: annais[i]=(double)(year);
8364: moisnais[i]=(double)(month);
8365: strcpy(line,stra);
1.225 brouard 8366:
1.223 brouard 8367: /* Sample weight */
1.136 brouard 8368: cutv(stra, strb,line,' ');
8369: errno=0;
8370: dval=strtod(strb,&endptr);
8371: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8372: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8373: 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 8374: fflush(ficlog);
8375: return 1;
8376: }
8377: weight[i]=dval;
8378: strcpy(line,stra);
1.225 brouard 8379:
1.223 brouard 8380: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8381: cutv(stra, strb, line, ' ');
8382: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8383: lval=-1;
1.223 brouard 8384: }else{
1.225 brouard 8385: errno=0;
8386: /* what_kind_of_number(strb); */
8387: dval=strtod(strb,&endptr);
8388: /* if(strb != endptr && *endptr == '\0') */
8389: /* dval=dlval; */
8390: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8391: if( strb[0]=='\0' || (*endptr != '\0')){
8392: 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);
8393: 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);
8394: return 1;
8395: }
8396: coqvar[iv][i]=dval;
1.226 brouard 8397: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8398: }
8399: strcpy(line,stra);
8400: }/* end loop nqv */
1.136 brouard 8401:
1.223 brouard 8402: /* Covariate values */
1.136 brouard 8403: for (j=ncovcol;j>=1;j--){
8404: cutv(stra, strb,line,' ');
1.223 brouard 8405: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8406: lval=-1;
1.136 brouard 8407: }else{
1.225 brouard 8408: errno=0;
8409: lval=strtol(strb,&endptr,10);
8410: if( strb[0]=='\0' || (*endptr != '\0')){
8411: 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);
8412: 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);
8413: return 1;
8414: }
1.136 brouard 8415: }
8416: if(lval <-1 || lval >1){
1.225 brouard 8417: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8418: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8419: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8420: For example, for multinomial values like 1, 2 and 3,\n \
8421: build V1=0 V2=0 for the reference value (1),\n \
8422: V1=1 V2=0 for (2) \n \
1.136 brouard 8423: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8424: output of IMaCh is often meaningless.\n \
1.136 brouard 8425: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8426: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8427: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8428: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8429: For example, for multinomial values like 1, 2 and 3,\n \
8430: build V1=0 V2=0 for the reference value (1),\n \
8431: V1=1 V2=0 for (2) \n \
1.136 brouard 8432: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8433: output of IMaCh is often meaningless.\n \
1.136 brouard 8434: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8435: return 1;
1.136 brouard 8436: }
8437: covar[j][i]=(double)(lval);
8438: strcpy(line,stra);
8439: }
8440: lstra=strlen(stra);
1.225 brouard 8441:
1.136 brouard 8442: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8443: stratrunc = &(stra[lstra-9]);
8444: num[i]=atol(stratrunc);
8445: }
8446: else
8447: num[i]=atol(stra);
8448: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8449: 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;}*/
8450:
8451: i=i+1;
8452: } /* End loop reading data */
1.225 brouard 8453:
1.136 brouard 8454: *imax=i-1; /* Number of individuals */
8455: fclose(fic);
1.225 brouard 8456:
1.136 brouard 8457: return (0);
1.164 brouard 8458: /* endread: */
1.225 brouard 8459: printf("Exiting readdata: ");
8460: fclose(fic);
8461: return (1);
1.223 brouard 8462: }
1.126 brouard 8463:
1.234 brouard 8464: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8465: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8466: while (*p2 == ' ')
1.234 brouard 8467: p2++;
8468: /* while ((*p1++ = *p2++) !=0) */
8469: /* ; */
8470: /* do */
8471: /* while (*p2 == ' ') */
8472: /* p2++; */
8473: /* while (*p1++ == *p2++); */
8474: *stri=p2;
1.145 brouard 8475: }
8476:
1.235 brouard 8477: int decoderesult ( char resultline[], int nres)
1.230 brouard 8478: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8479: {
1.235 brouard 8480: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8481: char resultsav[MAXLINE];
1.234 brouard 8482: int resultmodel[MAXLINE];
8483: int modelresult[MAXLINE];
1.230 brouard 8484: char stra[80], strb[80], strc[80], strd[80],stre[80];
8485:
1.234 brouard 8486: removefirstspace(&resultline);
1.233 brouard 8487: printf("decoderesult:%s\n",resultline);
1.230 brouard 8488:
8489: if (strstr(resultline,"v") !=0){
8490: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8491: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8492: return 1;
8493: }
8494: trimbb(resultsav, resultline);
8495: if (strlen(resultsav) >1){
8496: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8497: }
1.253 brouard 8498: if(j == 0){ /* Resultline but no = */
8499: TKresult[nres]=0; /* Combination for the nresult and the model */
8500: return (0);
8501: }
8502:
1.234 brouard 8503: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8504: 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);
8505: 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);
8506: }
8507: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8508: if(nbocc(resultsav,'=') >1){
8509: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8510: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8511: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8512: }else
8513: cutl(strc,strd,resultsav,'=');
1.230 brouard 8514: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8515:
1.230 brouard 8516: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8517: Tvarsel[k]=atoi(strc);
8518: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8519: /* cptcovsel++; */
8520: if (nbocc(stra,'=') >0)
8521: strcpy(resultsav,stra); /* and analyzes it */
8522: }
1.235 brouard 8523: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8524: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8525: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8526: match=0;
1.236 brouard 8527: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8528: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8529: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8530: match=1;
8531: break;
8532: }
8533: }
8534: if(match == 0){
8535: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8536: }
8537: }
8538: }
1.235 brouard 8539: /* Checking for missing or useless values in comparison of current model needs */
8540: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8541: match=0;
1.235 brouard 8542: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8543: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8544: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8545: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8546: ++match;
8547: }
8548: }
8549: }
8550: if(match == 0){
8551: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8552: }else if(match > 1){
8553: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8554: }
8555: }
1.235 brouard 8556:
1.234 brouard 8557: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8558: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8559: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8560: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8561: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8562: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8563: /* 1 0 0 0 */
8564: /* 2 1 0 0 */
8565: /* 3 0 1 0 */
8566: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8567: /* 5 0 0 1 */
8568: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8569: /* 7 0 1 1 */
8570: /* 8 1 1 1 */
1.237 brouard 8571: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8572: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8573: /* V5*age V5 known which value for nres? */
8574: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8575: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8576: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8577: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8578: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8579: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8580: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8581: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8582: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8583: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8584: k4++;;
8585: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8586: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8587: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8588: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8589: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8590: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8591: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8592: k4q++;;
8593: }
8594: }
1.234 brouard 8595:
1.235 brouard 8596: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8597: return (0);
8598: }
1.235 brouard 8599:
1.230 brouard 8600: int decodemodel( char model[], int lastobs)
8601: /**< This routine decodes the model and returns:
1.224 brouard 8602: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8603: * - nagesqr = 1 if age*age in the model, otherwise 0.
8604: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8605: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8606: * - cptcovage number of covariates with age*products =2
8607: * - cptcovs number of simple covariates
8608: * - 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
8609: * which is a new column after the 9 (ncovcol) variables.
8610: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8611: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8612: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8613: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8614: */
1.136 brouard 8615: {
1.238 brouard 8616: int i, j, k, ks, v;
1.227 brouard 8617: int j1, k1, k2, k3, k4;
1.136 brouard 8618: char modelsav[80];
1.145 brouard 8619: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8620: char *strpt;
1.136 brouard 8621:
1.145 brouard 8622: /*removespace(model);*/
1.136 brouard 8623: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8624: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8625: if (strstr(model,"AGE") !=0){
1.192 brouard 8626: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8627: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8628: return 1;
8629: }
1.141 brouard 8630: if (strstr(model,"v") !=0){
8631: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8632: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8633: return 1;
8634: }
1.187 brouard 8635: strcpy(modelsav,model);
8636: if ((strpt=strstr(model,"age*age")) !=0){
8637: printf(" strpt=%s, model=%s\n",strpt, model);
8638: if(strpt != model){
1.234 brouard 8639: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8640: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8641: corresponding column of parameters.\n",model);
1.234 brouard 8642: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8643: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8644: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8645: return 1;
1.225 brouard 8646: }
1.187 brouard 8647: nagesqr=1;
8648: if (strstr(model,"+age*age") !=0)
1.234 brouard 8649: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8650: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8651: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8652: else
1.234 brouard 8653: substrchaine(modelsav, model, "age*age");
1.187 brouard 8654: }else
8655: nagesqr=0;
8656: if (strlen(modelsav) >1){
8657: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8658: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8659: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8660: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8661: * cst, age and age*age
8662: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8663: /* including age products which are counted in cptcovage.
8664: * but the covariates which are products must be treated
8665: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8666: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8667: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8668:
8669:
1.187 brouard 8670: /* Design
8671: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8672: * < ncovcol=8 >
8673: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8674: * k= 1 2 3 4 5 6 7 8
8675: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8676: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8677: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8678: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8679: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8680: * Tage[++cptcovage]=k
8681: * if products, new covar are created after ncovcol with k1
8682: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8683: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8684: * 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
8685: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8686: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8687: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8688: * < ncovcol=8 >
8689: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8690: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8691: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8692: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8693: * p Tprod[1]@2={ 6, 5}
8694: *p Tvard[1][1]@4= {7, 8, 5, 6}
8695: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8696: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8697: *How to reorganize?
8698: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8699: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8700: * {2, 1, 4, 8, 5, 6, 3, 7}
8701: * Struct []
8702: */
1.225 brouard 8703:
1.187 brouard 8704: /* This loop fills the array Tvar from the string 'model'.*/
8705: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8706: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8707: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8708: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8709: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8710: /* k=1 Tvar[1]=2 (from V2) */
8711: /* k=5 Tvar[5] */
8712: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8713: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8714: /* } */
1.198 brouard 8715: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8716: /*
8717: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8718: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8719: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8720: }
1.187 brouard 8721: cptcovage=0;
8722: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8723: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8724: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8725: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8726: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8727: /*scanf("%d",i);*/
8728: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8729: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8730: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8731: /* covar is not filled and then is empty */
8732: cptcovprod--;
8733: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8734: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8735: Typevar[k]=1; /* 1 for age product */
8736: cptcovage++; /* Sums the number of covariates which include age as a product */
8737: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8738: /*printf("stre=%s ", stre);*/
8739: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8740: cptcovprod--;
8741: cutl(stre,strb,strc,'V');
8742: Tvar[k]=atoi(stre);
8743: Typevar[k]=1; /* 1 for age product */
8744: cptcovage++;
8745: Tage[cptcovage]=k;
8746: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8747: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8748: cptcovn++;
8749: cptcovprodnoage++;k1++;
8750: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8751: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8752: because this model-covariate is a construction we invent a new column
8753: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8754: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8755: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8756: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8757: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8758: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8759: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8760: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8761: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8762: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8763: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8764: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8765: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8766: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8767: for (i=1; i<=lastobs;i++){
8768: /* Computes the new covariate which is a product of
8769: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8770: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8771: }
8772: } /* End age is not in the model */
8773: } /* End if model includes a product */
8774: else { /* no more sum */
8775: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8776: /* scanf("%d",i);*/
8777: cutl(strd,strc,strb,'V');
8778: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8779: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8780: Tvar[k]=atoi(strd);
8781: Typevar[k]=0; /* 0 for simple covariates */
8782: }
8783: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8784: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8785: scanf("%d",i);*/
1.187 brouard 8786: } /* end of loop + on total covariates */
8787: } /* end if strlen(modelsave == 0) age*age might exist */
8788: } /* end if strlen(model == 0) */
1.136 brouard 8789:
8790: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8791: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8792:
1.136 brouard 8793: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8794: printf("cptcovprod=%d ", cptcovprod);
8795: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8796: scanf("%d ",i);*/
8797:
8798:
1.230 brouard 8799: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8800: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8801: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8802: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8803: k = 1 2 3 4 5 6 7 8 9
8804: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8805: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8806: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8807: Dummy[k] 1 0 0 0 3 1 1 2 3
8808: Tmodelind[combination of covar]=k;
1.225 brouard 8809: */
8810: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8811: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8812: /* 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 8813: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8814: printf("Model=%s\n\
8815: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8816: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8817: 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);
8818: fprintf(ficlog,"Model=%s\n\
8819: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8820: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8821: 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 8822: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8823: 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 */
8824: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8825: Fixed[k]= 0;
8826: Dummy[k]= 0;
1.225 brouard 8827: ncoveff++;
1.232 brouard 8828: ncovf++;
1.234 brouard 8829: nsd++;
8830: modell[k].maintype= FTYPE;
8831: TvarsD[nsd]=Tvar[k];
8832: TvarsDind[nsd]=k;
8833: TvarF[ncovf]=Tvar[k];
8834: TvarFind[ncovf]=k;
8835: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8836: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8837: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8838: Fixed[k]= 0;
8839: Dummy[k]= 0;
8840: ncoveff++;
8841: ncovf++;
8842: modell[k].maintype= FTYPE;
8843: TvarF[ncovf]=Tvar[k];
8844: TvarFind[ncovf]=k;
1.230 brouard 8845: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8846: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8847: }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 8848: Fixed[k]= 0;
8849: Dummy[k]= 1;
1.230 brouard 8850: nqfveff++;
1.234 brouard 8851: modell[k].maintype= FTYPE;
8852: modell[k].subtype= FQ;
8853: nsq++;
8854: TvarsQ[nsq]=Tvar[k];
8855: TvarsQind[nsq]=k;
1.232 brouard 8856: ncovf++;
1.234 brouard 8857: TvarF[ncovf]=Tvar[k];
8858: TvarFind[ncovf]=k;
1.231 brouard 8859: 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 8860: 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 8861: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8862: Fixed[k]= 1;
8863: Dummy[k]= 0;
1.225 brouard 8864: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8865: modell[k].maintype= VTYPE;
8866: modell[k].subtype= VD;
8867: nsd++;
8868: TvarsD[nsd]=Tvar[k];
8869: TvarsDind[nsd]=k;
8870: ncovv++; /* Only simple time varying variables */
8871: TvarV[ncovv]=Tvar[k];
1.242 brouard 8872: 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 8873: 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 */
8874: 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 8875: 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);
8876: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8877: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8878: Fixed[k]= 1;
8879: Dummy[k]= 1;
8880: nqtveff++;
8881: modell[k].maintype= VTYPE;
8882: modell[k].subtype= VQ;
8883: ncovv++; /* Only simple time varying variables */
8884: nsq++;
8885: TvarsQ[nsq]=Tvar[k];
8886: TvarsQind[nsq]=k;
8887: TvarV[ncovv]=Tvar[k];
1.242 brouard 8888: 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 8889: 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 */
8890: 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 8891: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8892: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8893: 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 8894: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8895: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8896: ncova++;
8897: TvarA[ncova]=Tvar[k];
8898: TvarAind[ncova]=k;
1.231 brouard 8899: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8900: Fixed[k]= 2;
8901: Dummy[k]= 2;
8902: modell[k].maintype= ATYPE;
8903: modell[k].subtype= APFD;
8904: /* ncoveff++; */
1.227 brouard 8905: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8906: Fixed[k]= 2;
8907: Dummy[k]= 3;
8908: modell[k].maintype= ATYPE;
8909: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8910: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8911: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8912: Fixed[k]= 3;
8913: Dummy[k]= 2;
8914: modell[k].maintype= ATYPE;
8915: modell[k].subtype= APVD; /* Product age * varying dummy */
8916: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8917: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8918: Fixed[k]= 3;
8919: Dummy[k]= 3;
8920: modell[k].maintype= ATYPE;
8921: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8922: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8923: }
8924: }else if (Typevar[k] == 2) { /* product without age */
8925: k1=Tposprod[k];
8926: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8927: if(Tvard[k1][2] <=ncovcol){
8928: Fixed[k]= 1;
8929: Dummy[k]= 0;
8930: modell[k].maintype= FTYPE;
8931: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8932: ncovf++; /* Fixed variables without age */
8933: TvarF[ncovf]=Tvar[k];
8934: TvarFind[ncovf]=k;
8935: }else if(Tvard[k1][2] <=ncovcol+nqv){
8936: Fixed[k]= 0; /* or 2 ?*/
8937: Dummy[k]= 1;
8938: modell[k].maintype= FTYPE;
8939: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8940: ncovf++; /* Varying variables without age */
8941: TvarF[ncovf]=Tvar[k];
8942: TvarFind[ncovf]=k;
8943: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8944: Fixed[k]= 1;
8945: Dummy[k]= 0;
8946: modell[k].maintype= VTYPE;
8947: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8948: ncovv++; /* Varying variables without age */
8949: TvarV[ncovv]=Tvar[k];
8950: TvarVind[ncovv]=k;
8951: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8952: Fixed[k]= 1;
8953: Dummy[k]= 1;
8954: modell[k].maintype= VTYPE;
8955: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8956: ncovv++; /* Varying variables without age */
8957: TvarV[ncovv]=Tvar[k];
8958: TvarVind[ncovv]=k;
8959: }
1.227 brouard 8960: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8961: if(Tvard[k1][2] <=ncovcol){
8962: Fixed[k]= 0; /* or 2 ?*/
8963: Dummy[k]= 1;
8964: modell[k].maintype= FTYPE;
8965: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8966: ncovf++; /* Fixed variables without age */
8967: TvarF[ncovf]=Tvar[k];
8968: TvarFind[ncovf]=k;
8969: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8970: Fixed[k]= 1;
8971: Dummy[k]= 1;
8972: modell[k].maintype= VTYPE;
8973: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8974: ncovv++; /* Varying variables without age */
8975: TvarV[ncovv]=Tvar[k];
8976: TvarVind[ncovv]=k;
8977: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8978: Fixed[k]= 1;
8979: Dummy[k]= 1;
8980: modell[k].maintype= VTYPE;
8981: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8982: ncovv++; /* Varying variables without age */
8983: TvarV[ncovv]=Tvar[k];
8984: TvarVind[ncovv]=k;
8985: ncovv++; /* Varying variables without age */
8986: TvarV[ncovv]=Tvar[k];
8987: TvarVind[ncovv]=k;
8988: }
1.227 brouard 8989: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8990: if(Tvard[k1][2] <=ncovcol){
8991: Fixed[k]= 1;
8992: Dummy[k]= 1;
8993: modell[k].maintype= VTYPE;
8994: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8995: ncovv++; /* Varying variables without age */
8996: TvarV[ncovv]=Tvar[k];
8997: TvarVind[ncovv]=k;
8998: }else if(Tvard[k1][2] <=ncovcol+nqv){
8999: Fixed[k]= 1;
9000: Dummy[k]= 1;
9001: modell[k].maintype= VTYPE;
9002: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9003: ncovv++; /* Varying variables without age */
9004: TvarV[ncovv]=Tvar[k];
9005: TvarVind[ncovv]=k;
9006: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9007: Fixed[k]= 1;
9008: Dummy[k]= 0;
9009: modell[k].maintype= VTYPE;
9010: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9011: ncovv++; /* Varying variables without age */
9012: TvarV[ncovv]=Tvar[k];
9013: TvarVind[ncovv]=k;
9014: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9015: Fixed[k]= 1;
9016: Dummy[k]= 1;
9017: modell[k].maintype= VTYPE;
9018: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9019: ncovv++; /* Varying variables without age */
9020: TvarV[ncovv]=Tvar[k];
9021: TvarVind[ncovv]=k;
9022: }
1.227 brouard 9023: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9024: if(Tvard[k1][2] <=ncovcol){
9025: Fixed[k]= 1;
9026: Dummy[k]= 1;
9027: modell[k].maintype= VTYPE;
9028: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9029: ncovv++; /* Varying variables without age */
9030: TvarV[ncovv]=Tvar[k];
9031: TvarVind[ncovv]=k;
9032: }else if(Tvard[k1][2] <=ncovcol+nqv){
9033: Fixed[k]= 1;
9034: Dummy[k]= 1;
9035: modell[k].maintype= VTYPE;
9036: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9037: ncovv++; /* Varying variables without age */
9038: TvarV[ncovv]=Tvar[k];
9039: TvarVind[ncovv]=k;
9040: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9041: Fixed[k]= 1;
9042: Dummy[k]= 1;
9043: modell[k].maintype= VTYPE;
9044: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9045: ncovv++; /* Varying variables without age */
9046: TvarV[ncovv]=Tvar[k];
9047: TvarVind[ncovv]=k;
9048: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9049: Fixed[k]= 1;
9050: Dummy[k]= 1;
9051: modell[k].maintype= VTYPE;
9052: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9053: ncovv++; /* Varying variables without age */
9054: TvarV[ncovv]=Tvar[k];
9055: TvarVind[ncovv]=k;
9056: }
1.227 brouard 9057: }else{
1.240 brouard 9058: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9059: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9060: } /*end k1*/
1.225 brouard 9061: }else{
1.226 brouard 9062: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9063: 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 9064: }
1.227 brouard 9065: 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 9066: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9067: 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]);
9068: }
9069: /* Searching for doublons in the model */
9070: for(k1=1; k1<= cptcovt;k1++){
9071: for(k2=1; k2 <k1;k2++){
9072: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9073: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9074: if(Tvar[k1]==Tvar[k2]){
9075: 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]]);
9076: 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);
9077: return(1);
9078: }
9079: }else if (Typevar[k1] ==2){
9080: k3=Tposprod[k1];
9081: k4=Tposprod[k2];
9082: 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])) ){
9083: 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]]);
9084: 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);
9085: return(1);
9086: }
9087: }
1.227 brouard 9088: }
9089: }
1.225 brouard 9090: }
9091: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9092: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9093: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9094: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9095: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9096: /*endread:*/
1.225 brouard 9097: printf("Exiting decodemodel: ");
9098: return (1);
1.136 brouard 9099: }
9100:
1.169 brouard 9101: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9102: {/* Check ages at death */
1.136 brouard 9103: int i, m;
1.218 brouard 9104: int firstone=0;
9105:
1.136 brouard 9106: for (i=1; i<=imx; i++) {
9107: for(m=2; (m<= maxwav); m++) {
9108: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9109: anint[m][i]=9999;
1.216 brouard 9110: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9111: s[m][i]=-1;
1.136 brouard 9112: }
9113: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 ! brouard 9114: *nberr = *nberr + 1;
1.218 brouard 9115: if(firstone == 0){
9116: firstone=1;
1.260 ! brouard 9117: printf("Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 9118: }
1.260 ! brouard 9119: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
! 9120: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9121: }
9122: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9123: (*nberr)++;
1.259 brouard 9124: printf("Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
9125: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
9126: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9127: }
9128: }
9129: }
9130:
9131: for (i=1; i<=imx; i++) {
9132: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9133: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9134: 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 9135: if (s[m][i] >= nlstate+1) {
1.169 brouard 9136: if(agedc[i]>0){
9137: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9138: agev[m][i]=agedc[i];
1.214 brouard 9139: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9140: }else {
1.136 brouard 9141: if ((int)andc[i]!=9999){
9142: nbwarn++;
9143: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9144: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9145: agev[m][i]=-1;
9146: }
9147: }
1.169 brouard 9148: } /* agedc > 0 */
1.214 brouard 9149: } /* end if */
1.136 brouard 9150: else if(s[m][i] !=9){ /* Standard case, age in fractional
9151: years but with the precision of a month */
9152: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9153: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9154: agev[m][i]=1;
9155: else if(agev[m][i] < *agemin){
9156: *agemin=agev[m][i];
9157: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9158: }
9159: else if(agev[m][i] >*agemax){
9160: *agemax=agev[m][i];
1.156 brouard 9161: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9162: }
9163: /*agev[m][i]=anint[m][i]-annais[i];*/
9164: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9165: } /* en if 9*/
1.136 brouard 9166: else { /* =9 */
1.214 brouard 9167: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9168: agev[m][i]=1;
9169: s[m][i]=-1;
9170: }
9171: }
1.214 brouard 9172: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9173: agev[m][i]=1;
1.214 brouard 9174: else{
9175: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9176: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9177: agev[m][i]=0;
9178: }
9179: } /* End for lastpass */
9180: }
1.136 brouard 9181:
9182: for (i=1; i<=imx; i++) {
9183: for(m=firstpass; (m<=lastpass); m++){
9184: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9185: (*nberr)++;
1.136 brouard 9186: 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);
9187: 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);
9188: return 1;
9189: }
9190: }
9191: }
9192:
9193: /*for (i=1; i<=imx; i++){
9194: for (m=firstpass; (m<lastpass); m++){
9195: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9196: }
9197:
9198: }*/
9199:
9200:
1.139 brouard 9201: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9202: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9203:
9204: return (0);
1.164 brouard 9205: /* endread:*/
1.136 brouard 9206: printf("Exiting calandcheckages: ");
9207: return (1);
9208: }
9209:
1.172 brouard 9210: #if defined(_MSC_VER)
9211: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9212: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9213: //#include "stdafx.h"
9214: //#include <stdio.h>
9215: //#include <tchar.h>
9216: //#include <windows.h>
9217: //#include <iostream>
9218: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9219:
9220: LPFN_ISWOW64PROCESS fnIsWow64Process;
9221:
9222: BOOL IsWow64()
9223: {
9224: BOOL bIsWow64 = FALSE;
9225:
9226: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9227: // (HANDLE, PBOOL);
9228:
9229: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9230:
9231: HMODULE module = GetModuleHandle(_T("kernel32"));
9232: const char funcName[] = "IsWow64Process";
9233: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9234: GetProcAddress(module, funcName);
9235:
9236: if (NULL != fnIsWow64Process)
9237: {
9238: if (!fnIsWow64Process(GetCurrentProcess(),
9239: &bIsWow64))
9240: //throw std::exception("Unknown error");
9241: printf("Unknown error\n");
9242: }
9243: return bIsWow64 != FALSE;
9244: }
9245: #endif
1.177 brouard 9246:
1.191 brouard 9247: void syscompilerinfo(int logged)
1.167 brouard 9248: {
9249: /* #include "syscompilerinfo.h"*/
1.185 brouard 9250: /* command line Intel compiler 32bit windows, XP compatible:*/
9251: /* /GS /W3 /Gy
9252: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9253: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9254: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9255: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9256: */
9257: /* 64 bits */
1.185 brouard 9258: /*
9259: /GS /W3 /Gy
9260: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9261: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9262: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9263: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9264: /* Optimization are useless and O3 is slower than O2 */
9265: /*
9266: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9267: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9268: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9269: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9270: */
1.186 brouard 9271: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9272: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9273: /PDB:"visual studio
9274: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9275: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9276: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9277: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9278: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9279: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9280: uiAccess='false'"
9281: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9282: /NOLOGO /TLBID:1
9283: */
1.177 brouard 9284: #if defined __INTEL_COMPILER
1.178 brouard 9285: #if defined(__GNUC__)
9286: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9287: #endif
1.177 brouard 9288: #elif defined(__GNUC__)
1.179 brouard 9289: #ifndef __APPLE__
1.174 brouard 9290: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9291: #endif
1.177 brouard 9292: struct utsname sysInfo;
1.178 brouard 9293: int cross = CROSS;
9294: if (cross){
9295: printf("Cross-");
1.191 brouard 9296: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9297: }
1.174 brouard 9298: #endif
9299:
1.171 brouard 9300: #include <stdint.h>
1.178 brouard 9301:
1.191 brouard 9302: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9303: #if defined(__clang__)
1.191 brouard 9304: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9305: #endif
9306: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9307: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9308: #endif
9309: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9310: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9311: #endif
9312: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9313: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9314: #endif
9315: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9316: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9317: #endif
9318: #if defined(_MSC_VER)
1.191 brouard 9319: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9320: #endif
9321: #if defined(__PGI)
1.191 brouard 9322: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9323: #endif
9324: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9325: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9326: #endif
1.191 brouard 9327: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9328:
1.167 brouard 9329: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9330: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9331: // Windows (x64 and x86)
1.191 brouard 9332: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9333: #elif __unix__ // all unices, not all compilers
9334: // Unix
1.191 brouard 9335: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9336: #elif __linux__
9337: // linux
1.191 brouard 9338: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9339: #elif __APPLE__
1.174 brouard 9340: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9341: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9342: #endif
9343:
9344: /* __MINGW32__ */
9345: /* __CYGWIN__ */
9346: /* __MINGW64__ */
9347: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9348: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9349: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9350: /* _WIN64 // Defined for applications for Win64. */
9351: /* _M_X64 // Defined for compilations that target x64 processors. */
9352: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9353:
1.167 brouard 9354: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9355: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9356: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9357: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9358: #else
1.191 brouard 9359: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9360: #endif
9361:
1.169 brouard 9362: #if defined(__GNUC__)
9363: # if defined(__GNUC_PATCHLEVEL__)
9364: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9365: + __GNUC_MINOR__ * 100 \
9366: + __GNUC_PATCHLEVEL__)
9367: # else
9368: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9369: + __GNUC_MINOR__ * 100)
9370: # endif
1.174 brouard 9371: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9372: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9373:
9374: if (uname(&sysInfo) != -1) {
9375: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9376: 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 9377: }
9378: else
9379: perror("uname() error");
1.179 brouard 9380: //#ifndef __INTEL_COMPILER
9381: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9382: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9383: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9384: #endif
1.169 brouard 9385: #endif
1.172 brouard 9386:
9387: // void main()
9388: // {
1.169 brouard 9389: #if defined(_MSC_VER)
1.174 brouard 9390: if (IsWow64()){
1.191 brouard 9391: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9392: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9393: }
9394: else{
1.191 brouard 9395: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9396: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9397: }
1.172 brouard 9398: // printf("\nPress Enter to continue...");
9399: // getchar();
9400: // }
9401:
1.169 brouard 9402: #endif
9403:
1.167 brouard 9404:
1.219 brouard 9405: }
1.136 brouard 9406:
1.219 brouard 9407: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9408: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9409: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9410: /* double ftolpl = 1.e-10; */
1.180 brouard 9411: double age, agebase, agelim;
1.203 brouard 9412: double tot;
1.180 brouard 9413:
1.202 brouard 9414: strcpy(filerespl,"PL_");
9415: strcat(filerespl,fileresu);
9416: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9417: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9418: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9419: }
1.227 brouard 9420: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9421: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9422: pstamp(ficrespl);
1.203 brouard 9423: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9424: fprintf(ficrespl,"#Age ");
9425: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9426: fprintf(ficrespl,"\n");
1.180 brouard 9427:
1.219 brouard 9428: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9429:
1.219 brouard 9430: agebase=ageminpar;
9431: agelim=agemaxpar;
1.180 brouard 9432:
1.227 brouard 9433: /* i1=pow(2,ncoveff); */
1.234 brouard 9434: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9435: if (cptcovn < 1){i1=1;}
1.180 brouard 9436:
1.238 brouard 9437: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9438: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9439: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9440: continue;
1.235 brouard 9441:
1.238 brouard 9442: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9443: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9444: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9445: /* k=k+1; */
9446: /* to clean */
9447: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9448: fprintf(ficrespl,"#******");
9449: printf("#******");
9450: fprintf(ficlog,"#******");
9451: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9452: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9453: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9454: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9455: }
9456: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9457: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9458: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9459: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9460: }
9461: fprintf(ficrespl,"******\n");
9462: printf("******\n");
9463: fprintf(ficlog,"******\n");
9464: if(invalidvarcomb[k]){
9465: printf("\nCombination (%d) ignored because no case \n",k);
9466: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9467: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9468: continue;
9469: }
1.219 brouard 9470:
1.238 brouard 9471: fprintf(ficrespl,"#Age ");
9472: for(j=1;j<=cptcoveff;j++) {
9473: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9474: }
9475: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9476: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9477:
1.238 brouard 9478: for (age=agebase; age<=agelim; age++){
9479: /* for (age=agebase; age<=agebase; age++){ */
9480: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9481: fprintf(ficrespl,"%.0f ",age );
9482: for(j=1;j<=cptcoveff;j++)
9483: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9484: tot=0.;
9485: for(i=1; i<=nlstate;i++){
9486: tot += prlim[i][i];
9487: fprintf(ficrespl," %.5f", prlim[i][i]);
9488: }
9489: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9490: } /* Age */
9491: /* was end of cptcod */
9492: } /* cptcov */
9493: } /* nres */
1.219 brouard 9494: return 0;
1.180 brouard 9495: }
9496:
1.218 brouard 9497: 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){
9498: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9499:
9500: /* Computes the back prevalence limit for any combination of covariate values
9501: * at any age between ageminpar and agemaxpar
9502: */
1.235 brouard 9503: int i, j, k, i1, nres=0 ;
1.217 brouard 9504: /* double ftolpl = 1.e-10; */
9505: double age, agebase, agelim;
9506: double tot;
1.218 brouard 9507: /* double ***mobaverage; */
9508: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9509:
9510: strcpy(fileresplb,"PLB_");
9511: strcat(fileresplb,fileresu);
9512: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9513: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9514: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9515: }
9516: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9517: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9518: pstamp(ficresplb);
9519: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9520: fprintf(ficresplb,"#Age ");
9521: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9522: fprintf(ficresplb,"\n");
9523:
1.218 brouard 9524:
9525: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9526:
9527: agebase=ageminpar;
9528: agelim=agemaxpar;
9529:
9530:
1.227 brouard 9531: i1=pow(2,cptcoveff);
1.218 brouard 9532: if (cptcovn < 1){i1=1;}
1.227 brouard 9533:
1.238 brouard 9534: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9535: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9536: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9537: continue;
9538: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9539: fprintf(ficresplb,"#******");
9540: printf("#******");
9541: fprintf(ficlog,"#******");
9542: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9543: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9544: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9545: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9546: }
9547: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9548: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9549: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9550: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9551: }
9552: fprintf(ficresplb,"******\n");
9553: printf("******\n");
9554: fprintf(ficlog,"******\n");
9555: if(invalidvarcomb[k]){
9556: printf("\nCombination (%d) ignored because no cases \n",k);
9557: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9558: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9559: continue;
9560: }
1.218 brouard 9561:
1.238 brouard 9562: fprintf(ficresplb,"#Age ");
9563: for(j=1;j<=cptcoveff;j++) {
9564: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9565: }
9566: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9567: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9568:
9569:
1.238 brouard 9570: for (age=agebase; age<=agelim; age++){
9571: /* for (age=agebase; age<=agebase; age++){ */
9572: if(mobilavproj > 0){
9573: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9574: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9575: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9576: }else if (mobilavproj == 0){
9577: 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);
9578: 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);
9579: exit(1);
9580: }else{
9581: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9582: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9583: }
9584: fprintf(ficresplb,"%.0f ",age );
9585: for(j=1;j<=cptcoveff;j++)
9586: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9587: tot=0.;
9588: for(i=1; i<=nlstate;i++){
9589: tot += bprlim[i][i];
9590: fprintf(ficresplb," %.5f", bprlim[i][i]);
9591: }
9592: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9593: } /* Age */
9594: /* was end of cptcod */
1.255 brouard 9595: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9596: } /* end of any combination */
9597: } /* end of nres */
1.218 brouard 9598: /* hBijx(p, bage, fage); */
9599: /* fclose(ficrespijb); */
9600:
9601: return 0;
1.217 brouard 9602: }
1.218 brouard 9603:
1.180 brouard 9604: int hPijx(double *p, int bage, int fage){
9605: /*------------- h Pij x at various ages ------------*/
9606:
9607: int stepsize;
9608: int agelim;
9609: int hstepm;
9610: int nhstepm;
1.235 brouard 9611: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9612:
9613: double agedeb;
9614: double ***p3mat;
9615:
1.201 brouard 9616: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9617: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9618: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9619: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9620: }
9621: printf("Computing pij: result on file '%s' \n", filerespij);
9622: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9623:
9624: stepsize=(int) (stepm+YEARM-1)/YEARM;
9625: /*if (stepm<=24) stepsize=2;*/
9626:
9627: agelim=AGESUP;
9628: hstepm=stepsize*YEARM; /* Every year of age */
9629: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9630:
1.180 brouard 9631: /* hstepm=1; aff par mois*/
9632: pstamp(ficrespij);
9633: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9634: i1= pow(2,cptcoveff);
1.218 brouard 9635: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9636: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9637: /* k=k+1; */
1.235 brouard 9638: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9639: for(k=1; k<=i1;k++){
1.253 brouard 9640: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9641: continue;
1.183 brouard 9642: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9643: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9644: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9645: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9646: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9647: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9648: }
1.183 brouard 9649: fprintf(ficrespij,"******\n");
9650:
9651: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9652: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9653: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9654:
9655: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9656:
1.183 brouard 9657: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9658: oldm=oldms;savm=savms;
1.235 brouard 9659: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9660: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9661: for(i=1; i<=nlstate;i++)
9662: for(j=1; j<=nlstate+ndeath;j++)
9663: fprintf(ficrespij," %1d-%1d",i,j);
9664: fprintf(ficrespij,"\n");
9665: for (h=0; h<=nhstepm; h++){
9666: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9667: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9668: for(i=1; i<=nlstate;i++)
9669: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9670: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9671: fprintf(ficrespij,"\n");
9672: }
1.183 brouard 9673: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9674: fprintf(ficrespij,"\n");
9675: }
1.180 brouard 9676: /*}*/
9677: }
1.218 brouard 9678: return 0;
1.180 brouard 9679: }
1.218 brouard 9680:
9681: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9682: /*------------- h Bij x at various ages ------------*/
9683:
9684: int stepsize;
1.218 brouard 9685: /* int agelim; */
9686: int ageminl;
1.217 brouard 9687: int hstepm;
9688: int nhstepm;
1.238 brouard 9689: int h, i, i1, j, k, nres;
1.218 brouard 9690:
1.217 brouard 9691: double agedeb;
9692: double ***p3mat;
1.218 brouard 9693:
9694: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9695: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9696: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9697: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9698: }
9699: printf("Computing pij back: result on file '%s' \n", filerespijb);
9700: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9701:
9702: stepsize=(int) (stepm+YEARM-1)/YEARM;
9703: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9704:
1.218 brouard 9705: /* agelim=AGESUP; */
9706: ageminl=30;
9707: hstepm=stepsize*YEARM; /* Every year of age */
9708: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9709:
9710: /* hstepm=1; aff par mois*/
9711: pstamp(ficrespijb);
1.255 brouard 9712: fprintf(ficrespijb,"#****** h Bij x Back probability to be in state i at age x-h being in j at x: B1j+B2j+...=1 ");
1.227 brouard 9713: i1= pow(2,cptcoveff);
1.218 brouard 9714: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9715: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9716: /* k=k+1; */
1.238 brouard 9717: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9718: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9719: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9720: continue;
9721: fprintf(ficrespijb,"\n#****** ");
9722: for(j=1;j<=cptcoveff;j++)
9723: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9724: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9725: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9726: }
9727: fprintf(ficrespijb,"******\n");
9728: if(invalidvarcomb[k]){
9729: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9730: continue;
9731: }
9732:
9733: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9734: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9735: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9736: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9737: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9738:
9739: /* nhstepm=nhstepm*YEARM; aff par mois*/
9740:
9741: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9742: /* oldm=oldms;savm=savms; */
9743: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9744: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9745: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9746: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9747: for(i=1; i<=nlstate;i++)
9748: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9749: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9750: fprintf(ficrespijb,"\n");
1.238 brouard 9751: for (h=0; h<=nhstepm; h++){
9752: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9753: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9754: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9755: for(i=1; i<=nlstate;i++)
9756: for(j=1; j<=nlstate+ndeath;j++)
9757: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9758: fprintf(ficrespijb,"\n");
9759: }
9760: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9761: fprintf(ficrespijb,"\n");
9762: } /* end age deb */
9763: } /* end combination */
9764: } /* end nres */
1.218 brouard 9765: return 0;
9766: } /* hBijx */
1.217 brouard 9767:
1.180 brouard 9768:
1.136 brouard 9769: /***********************************************/
9770: /**************** Main Program *****************/
9771: /***********************************************/
9772:
9773: int main(int argc, char *argv[])
9774: {
9775: #ifdef GSL
9776: const gsl_multimin_fminimizer_type *T;
9777: size_t iteri = 0, it;
9778: int rval = GSL_CONTINUE;
9779: int status = GSL_SUCCESS;
9780: double ssval;
9781: #endif
9782: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9783: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9784: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9785: int jj, ll, li, lj, lk;
1.136 brouard 9786: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9787: int num_filled;
1.136 brouard 9788: int itimes;
9789: int NDIM=2;
9790: int vpopbased=0;
1.235 brouard 9791: int nres=0;
1.258 brouard 9792: int endishere=0;
1.136 brouard 9793:
1.164 brouard 9794: char ca[32], cb[32];
1.136 brouard 9795: /* FILE *fichtm; *//* Html File */
9796: /* FILE *ficgp;*/ /*Gnuplot File */
9797: struct stat info;
1.191 brouard 9798: double agedeb=0.;
1.194 brouard 9799:
9800: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9801: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9802:
1.165 brouard 9803: double fret;
1.191 brouard 9804: double dum=0.; /* Dummy variable */
1.136 brouard 9805: double ***p3mat;
1.218 brouard 9806: /* double ***mobaverage; */
1.164 brouard 9807:
9808: char line[MAXLINE];
1.197 brouard 9809: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9810:
1.234 brouard 9811: char modeltemp[MAXLINE];
1.230 brouard 9812: char resultline[MAXLINE];
9813:
1.136 brouard 9814: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9815: char *tok, *val; /* pathtot */
1.136 brouard 9816: int firstobs=1, lastobs=10;
1.195 brouard 9817: int c, h , cpt, c2;
1.191 brouard 9818: int jl=0;
9819: int i1, j1, jk, stepsize=0;
1.194 brouard 9820: int count=0;
9821:
1.164 brouard 9822: int *tab;
1.136 brouard 9823: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9824: int backcast=0;
1.136 brouard 9825: int mobilav=0,popforecast=0;
1.191 brouard 9826: int hstepm=0, nhstepm=0;
1.136 brouard 9827: int agemortsup;
9828: float sumlpop=0.;
9829: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9830: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9831:
1.191 brouard 9832: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9833: double ftolpl=FTOL;
9834: double **prlim;
1.217 brouard 9835: double **bprlim;
1.136 brouard 9836: double ***param; /* Matrix of parameters */
1.251 brouard 9837: double ***paramstart; /* Matrix of starting parameter values */
9838: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9839: double **matcov; /* Matrix of covariance */
1.203 brouard 9840: double **hess; /* Hessian matrix */
1.136 brouard 9841: double ***delti3; /* Scale */
9842: double *delti; /* Scale */
9843: double ***eij, ***vareij;
9844: double **varpl; /* Variances of prevalence limits by age */
9845: double *epj, vepp;
1.164 brouard 9846:
1.136 brouard 9847: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9848: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9849:
1.136 brouard 9850: double **ximort;
1.145 brouard 9851: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9852: int *dcwave;
9853:
1.164 brouard 9854: char z[1]="c";
1.136 brouard 9855:
9856: /*char *strt;*/
9857: char strtend[80];
1.126 brouard 9858:
1.164 brouard 9859:
1.126 brouard 9860: /* setlocale (LC_ALL, ""); */
9861: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9862: /* textdomain (PACKAGE); */
9863: /* setlocale (LC_CTYPE, ""); */
9864: /* setlocale (LC_MESSAGES, ""); */
9865:
9866: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9867: rstart_time = time(NULL);
9868: /* (void) gettimeofday(&start_time,&tzp);*/
9869: start_time = *localtime(&rstart_time);
1.126 brouard 9870: curr_time=start_time;
1.157 brouard 9871: /*tml = *localtime(&start_time.tm_sec);*/
9872: /* strcpy(strstart,asctime(&tml)); */
9873: strcpy(strstart,asctime(&start_time));
1.126 brouard 9874:
9875: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9876: /* tp.tm_sec = tp.tm_sec +86400; */
9877: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9878: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9879: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9880: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9881: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9882: /* strt=asctime(&tmg); */
9883: /* printf("Time(after) =%s",strstart); */
9884: /* (void) time (&time_value);
9885: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9886: * tm = *localtime(&time_value);
9887: * strstart=asctime(&tm);
9888: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9889: */
9890:
9891: nberr=0; /* Number of errors and warnings */
9892: nbwarn=0;
1.184 brouard 9893: #ifdef WIN32
9894: _getcwd(pathcd, size);
9895: #else
1.126 brouard 9896: getcwd(pathcd, size);
1.184 brouard 9897: #endif
1.191 brouard 9898: syscompilerinfo(0);
1.196 brouard 9899: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9900: if(argc <=1){
9901: printf("\nEnter the parameter file name: ");
1.205 brouard 9902: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9903: printf("ERROR Empty parameter file name\n");
9904: goto end;
9905: }
1.126 brouard 9906: i=strlen(pathr);
9907: if(pathr[i-1]=='\n')
9908: pathr[i-1]='\0';
1.156 brouard 9909: i=strlen(pathr);
1.205 brouard 9910: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9911: pathr[i-1]='\0';
1.205 brouard 9912: }
9913: i=strlen(pathr);
9914: if( i==0 ){
9915: printf("ERROR Empty parameter file name\n");
9916: goto end;
9917: }
9918: for (tok = pathr; tok != NULL; ){
1.126 brouard 9919: printf("Pathr |%s|\n",pathr);
9920: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9921: printf("val= |%s| pathr=%s\n",val,pathr);
9922: strcpy (pathtot, val);
9923: if(pathr[0] == '\0') break; /* Dirty */
9924: }
9925: }
9926: else{
9927: strcpy(pathtot,argv[1]);
9928: }
9929: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9930: /*cygwin_split_path(pathtot,path,optionfile);
9931: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9932: /* cutv(path,optionfile,pathtot,'\\');*/
9933:
9934: /* Split argv[0], imach program to get pathimach */
9935: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9936: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9937: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9938: /* strcpy(pathimach,argv[0]); */
9939: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9940: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9941: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9942: #ifdef WIN32
9943: _chdir(path); /* Can be a relative path */
9944: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9945: #else
1.126 brouard 9946: chdir(path); /* Can be a relative path */
1.184 brouard 9947: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9948: #endif
9949: printf("Current directory %s!\n",pathcd);
1.126 brouard 9950: strcpy(command,"mkdir ");
9951: strcat(command,optionfilefiname);
9952: if((outcmd=system(command)) != 0){
1.169 brouard 9953: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9954: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9955: /* fclose(ficlog); */
9956: /* exit(1); */
9957: }
9958: /* if((imk=mkdir(optionfilefiname))<0){ */
9959: /* perror("mkdir"); */
9960: /* } */
9961:
9962: /*-------- arguments in the command line --------*/
9963:
1.186 brouard 9964: /* Main Log file */
1.126 brouard 9965: strcat(filelog, optionfilefiname);
9966: strcat(filelog,".log"); /* */
9967: if((ficlog=fopen(filelog,"w"))==NULL) {
9968: printf("Problem with logfile %s\n",filelog);
9969: goto end;
9970: }
9971: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9972: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9973: fprintf(ficlog,"\nEnter the parameter file name: \n");
9974: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9975: path=%s \n\
9976: optionfile=%s\n\
9977: optionfilext=%s\n\
1.156 brouard 9978: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9979:
1.197 brouard 9980: syscompilerinfo(1);
1.167 brouard 9981:
1.126 brouard 9982: printf("Local time (at start):%s",strstart);
9983: fprintf(ficlog,"Local time (at start): %s",strstart);
9984: fflush(ficlog);
9985: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9986: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9987:
9988: /* */
9989: strcpy(fileres,"r");
9990: strcat(fileres, optionfilefiname);
1.201 brouard 9991: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9992: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9993: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9994:
1.186 brouard 9995: /* Main ---------arguments file --------*/
1.126 brouard 9996:
9997: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9998: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9999: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10000: fflush(ficlog);
1.149 brouard 10001: /* goto end; */
10002: exit(70);
1.126 brouard 10003: }
10004:
10005:
10006:
10007: strcpy(filereso,"o");
1.201 brouard 10008: strcat(filereso,fileresu);
1.126 brouard 10009: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10010: printf("Problem with Output resultfile: %s\n", filereso);
10011: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10012: fflush(ficlog);
10013: goto end;
10014: }
10015:
10016: /* Reads comments: lines beginning with '#' */
10017: numlinepar=0;
1.197 brouard 10018:
10019: /* First parameter line */
10020: while(fgets(line, MAXLINE, ficpar)) {
10021: /* If line starts with a # it is a comment */
10022: if (line[0] == '#') {
10023: numlinepar++;
10024: fputs(line,stdout);
10025: fputs(line,ficparo);
10026: fputs(line,ficlog);
10027: continue;
10028: }else
10029: break;
10030: }
10031: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10032: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10033: if (num_filled != 5) {
10034: printf("Should be 5 parameters\n");
10035: }
1.126 brouard 10036: numlinepar++;
1.197 brouard 10037: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10038: }
10039: /* Second parameter line */
10040: while(fgets(line, MAXLINE, ficpar)) {
10041: /* If line starts with a # it is a comment */
10042: if (line[0] == '#') {
10043: numlinepar++;
10044: fputs(line,stdout);
10045: fputs(line,ficparo);
10046: fputs(line,ficlog);
10047: continue;
10048: }else
10049: break;
10050: }
1.223 brouard 10051: 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", \
10052: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10053: if (num_filled != 11) {
10054: 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 10055: printf("but line=%s\n",line);
1.197 brouard 10056: }
1.223 brouard 10057: 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 10058: }
1.203 brouard 10059: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10060: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10061: /* Third parameter line */
10062: while(fgets(line, MAXLINE, ficpar)) {
10063: /* If line starts with a # it is a comment */
10064: if (line[0] == '#') {
10065: numlinepar++;
10066: fputs(line,stdout);
10067: fputs(line,ficparo);
10068: fputs(line,ficlog);
10069: continue;
10070: }else
10071: break;
10072: }
1.201 brouard 10073: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10074: if (num_filled == 0)
10075: model[0]='\0';
10076: else if (num_filled != 1){
1.197 brouard 10077: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10078: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10079: model[0]='\0';
10080: goto end;
10081: }
10082: else{
10083: if (model[0]=='+'){
10084: for(i=1; i<=strlen(model);i++)
10085: modeltemp[i-1]=model[i];
1.201 brouard 10086: strcpy(model,modeltemp);
1.197 brouard 10087: }
10088: }
1.199 brouard 10089: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10090: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10091: }
10092: /* 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); */
10093: /* numlinepar=numlinepar+3; /\* In general *\/ */
10094: /* 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 10095: 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);
10096: 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 10097: fflush(ficlog);
1.190 brouard 10098: /* if(model[0]=='#'|| model[0]== '\0'){ */
10099: if(model[0]=='#'){
1.187 brouard 10100: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10101: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10102: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10103: if(mle != -1){
10104: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10105: exit(1);
10106: }
10107: }
1.126 brouard 10108: while((c=getc(ficpar))=='#' && c!= EOF){
10109: ungetc(c,ficpar);
10110: fgets(line, MAXLINE, ficpar);
10111: numlinepar++;
1.195 brouard 10112: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10113: z[0]=line[1];
10114: }
10115: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10116: fputs(line, stdout);
10117: //puts(line);
1.126 brouard 10118: fputs(line,ficparo);
10119: fputs(line,ficlog);
10120: }
10121: ungetc(c,ficpar);
10122:
10123:
1.145 brouard 10124: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10125: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10126: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10127: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10128: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10129: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10130: v1+v2*age+v2*v3 makes cptcovn = 3
10131: */
10132: if (strlen(model)>1)
1.187 brouard 10133: 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 10134: else
1.187 brouard 10135: ncovmodel=2; /* Constant and age */
1.133 brouard 10136: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10137: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10138: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10139: 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);
10140: 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);
10141: fflush(stdout);
10142: fclose (ficlog);
10143: goto end;
10144: }
1.126 brouard 10145: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10146: delti=delti3[1][1];
10147: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10148: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10149: /* We could also provide initial parameters values giving by simple logistic regression
10150: * only one way, that is without matrix product. We will have nlstate maximizations */
10151: /* for(i=1;i<nlstate;i++){ */
10152: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10153: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10154: /* } */
1.126 brouard 10155: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10156: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10157: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10158: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10159: fclose (ficparo);
10160: fclose (ficlog);
10161: goto end;
10162: exit(0);
1.220 brouard 10163: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10164: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10165: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10166: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10167: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10168: matcov=matrix(1,npar,1,npar);
1.203 brouard 10169: hess=matrix(1,npar,1,npar);
1.220 brouard 10170: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10171: /* Read guessed parameters */
1.126 brouard 10172: /* Reads comments: lines beginning with '#' */
10173: while((c=getc(ficpar))=='#' && c!= EOF){
10174: ungetc(c,ficpar);
10175: fgets(line, MAXLINE, ficpar);
10176: numlinepar++;
1.141 brouard 10177: fputs(line,stdout);
1.126 brouard 10178: fputs(line,ficparo);
10179: fputs(line,ficlog);
10180: }
10181: ungetc(c,ficpar);
10182:
10183: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10184: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10185: for(i=1; i <=nlstate; i++){
1.234 brouard 10186: j=0;
1.126 brouard 10187: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10188: if(jj==i) continue;
10189: j++;
10190: fscanf(ficpar,"%1d%1d",&i1,&j1);
10191: if ((i1 != i) || (j1 != jj)){
10192: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10193: It might be a problem of design; if ncovcol and the model are correct\n \
10194: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10195: exit(1);
10196: }
10197: fprintf(ficparo,"%1d%1d",i1,j1);
10198: if(mle==1)
10199: printf("%1d%1d",i,jj);
10200: fprintf(ficlog,"%1d%1d",i,jj);
10201: for(k=1; k<=ncovmodel;k++){
10202: fscanf(ficpar," %lf",¶m[i][j][k]);
10203: if(mle==1){
10204: printf(" %lf",param[i][j][k]);
10205: fprintf(ficlog," %lf",param[i][j][k]);
10206: }
10207: else
10208: fprintf(ficlog," %lf",param[i][j][k]);
10209: fprintf(ficparo," %lf",param[i][j][k]);
10210: }
10211: fscanf(ficpar,"\n");
10212: numlinepar++;
10213: if(mle==1)
10214: printf("\n");
10215: fprintf(ficlog,"\n");
10216: fprintf(ficparo,"\n");
1.126 brouard 10217: }
10218: }
10219: fflush(ficlog);
1.234 brouard 10220:
1.251 brouard 10221: /* Reads parameters values */
1.126 brouard 10222: p=param[1][1];
1.251 brouard 10223: pstart=paramstart[1][1];
1.126 brouard 10224:
10225: /* Reads comments: lines beginning with '#' */
10226: while((c=getc(ficpar))=='#' && c!= EOF){
10227: ungetc(c,ficpar);
10228: fgets(line, MAXLINE, ficpar);
10229: numlinepar++;
1.141 brouard 10230: fputs(line,stdout);
1.126 brouard 10231: fputs(line,ficparo);
10232: fputs(line,ficlog);
10233: }
10234: ungetc(c,ficpar);
10235:
10236: for(i=1; i <=nlstate; i++){
10237: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10238: fscanf(ficpar,"%1d%1d",&i1,&j1);
10239: if ( (i1-i) * (j1-j) != 0){
10240: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10241: exit(1);
10242: }
10243: printf("%1d%1d",i,j);
10244: fprintf(ficparo,"%1d%1d",i1,j1);
10245: fprintf(ficlog,"%1d%1d",i1,j1);
10246: for(k=1; k<=ncovmodel;k++){
10247: fscanf(ficpar,"%le",&delti3[i][j][k]);
10248: printf(" %le",delti3[i][j][k]);
10249: fprintf(ficparo," %le",delti3[i][j][k]);
10250: fprintf(ficlog," %le",delti3[i][j][k]);
10251: }
10252: fscanf(ficpar,"\n");
10253: numlinepar++;
10254: printf("\n");
10255: fprintf(ficparo,"\n");
10256: fprintf(ficlog,"\n");
1.126 brouard 10257: }
10258: }
10259: fflush(ficlog);
1.234 brouard 10260:
1.145 brouard 10261: /* Reads covariance matrix */
1.126 brouard 10262: delti=delti3[1][1];
1.220 brouard 10263:
10264:
1.126 brouard 10265: /* 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 10266:
1.126 brouard 10267: /* Reads comments: lines beginning with '#' */
10268: while((c=getc(ficpar))=='#' && c!= EOF){
10269: ungetc(c,ficpar);
10270: fgets(line, MAXLINE, ficpar);
10271: numlinepar++;
1.141 brouard 10272: fputs(line,stdout);
1.126 brouard 10273: fputs(line,ficparo);
10274: fputs(line,ficlog);
10275: }
10276: ungetc(c,ficpar);
1.220 brouard 10277:
1.126 brouard 10278: matcov=matrix(1,npar,1,npar);
1.203 brouard 10279: hess=matrix(1,npar,1,npar);
1.131 brouard 10280: for(i=1; i <=npar; i++)
10281: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10282:
1.194 brouard 10283: /* Scans npar lines */
1.126 brouard 10284: for(i=1; i <=npar; i++){
1.226 brouard 10285: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10286: if(count != 3){
1.226 brouard 10287: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10288: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10289: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10290: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10291: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10292: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10293: exit(1);
1.220 brouard 10294: }else{
1.226 brouard 10295: if(mle==1)
10296: printf("%1d%1d%d",i1,j1,jk);
10297: }
10298: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10299: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10300: for(j=1; j <=i; j++){
1.226 brouard 10301: fscanf(ficpar," %le",&matcov[i][j]);
10302: if(mle==1){
10303: printf(" %.5le",matcov[i][j]);
10304: }
10305: fprintf(ficlog," %.5le",matcov[i][j]);
10306: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10307: }
10308: fscanf(ficpar,"\n");
10309: numlinepar++;
10310: if(mle==1)
1.220 brouard 10311: printf("\n");
1.126 brouard 10312: fprintf(ficlog,"\n");
10313: fprintf(ficparo,"\n");
10314: }
1.194 brouard 10315: /* End of read covariance matrix npar lines */
1.126 brouard 10316: for(i=1; i <=npar; i++)
10317: for(j=i+1;j<=npar;j++)
1.226 brouard 10318: matcov[i][j]=matcov[j][i];
1.126 brouard 10319:
10320: if(mle==1)
10321: printf("\n");
10322: fprintf(ficlog,"\n");
10323:
10324: fflush(ficlog);
10325:
10326: /*-------- Rewriting parameter file ----------*/
10327: strcpy(rfileres,"r"); /* "Rparameterfile */
10328: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10329: strcat(rfileres,"."); /* */
10330: strcat(rfileres,optionfilext); /* Other files have txt extension */
10331: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10332: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10333: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10334: }
10335: fprintf(ficres,"#%s\n",version);
10336: } /* End of mle != -3 */
1.218 brouard 10337:
1.186 brouard 10338: /* Main data
10339: */
1.126 brouard 10340: n= lastobs;
10341: num=lvector(1,n);
10342: moisnais=vector(1,n);
10343: annais=vector(1,n);
10344: moisdc=vector(1,n);
10345: andc=vector(1,n);
1.220 brouard 10346: weight=vector(1,n);
1.126 brouard 10347: agedc=vector(1,n);
10348: cod=ivector(1,n);
1.220 brouard 10349: for(i=1;i<=n;i++){
1.234 brouard 10350: num[i]=0;
10351: moisnais[i]=0;
10352: annais[i]=0;
10353: moisdc[i]=0;
10354: andc[i]=0;
10355: agedc[i]=0;
10356: cod[i]=0;
10357: weight[i]=1.0; /* Equal weights, 1 by default */
10358: }
1.126 brouard 10359: mint=matrix(1,maxwav,1,n);
10360: anint=matrix(1,maxwav,1,n);
1.131 brouard 10361: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10362: tab=ivector(1,NCOVMAX);
1.144 brouard 10363: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10364: 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 10365:
1.136 brouard 10366: /* Reads data from file datafile */
10367: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10368: goto end;
10369:
10370: /* Calculation of the number of parameters from char model */
1.234 brouard 10371: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10372: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10373: k=3 V4 Tvar[k=3]= 4 (from V4)
10374: k=2 V1 Tvar[k=2]= 1 (from V1)
10375: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10376: */
10377:
10378: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10379: TvarsDind=ivector(1,NCOVMAX); /* */
10380: TvarsD=ivector(1,NCOVMAX); /* */
10381: TvarsQind=ivector(1,NCOVMAX); /* */
10382: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10383: TvarF=ivector(1,NCOVMAX); /* */
10384: TvarFind=ivector(1,NCOVMAX); /* */
10385: TvarV=ivector(1,NCOVMAX); /* */
10386: TvarVind=ivector(1,NCOVMAX); /* */
10387: TvarA=ivector(1,NCOVMAX); /* */
10388: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10389: TvarFD=ivector(1,NCOVMAX); /* */
10390: TvarFDind=ivector(1,NCOVMAX); /* */
10391: TvarFQ=ivector(1,NCOVMAX); /* */
10392: TvarFQind=ivector(1,NCOVMAX); /* */
10393: TvarVD=ivector(1,NCOVMAX); /* */
10394: TvarVDind=ivector(1,NCOVMAX); /* */
10395: TvarVQ=ivector(1,NCOVMAX); /* */
10396: TvarVQind=ivector(1,NCOVMAX); /* */
10397:
1.230 brouard 10398: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10399: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10400: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10401: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10402: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10403: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10404: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10405: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10406: */
10407: /* For model-covariate k tells which data-covariate to use but
10408: because this model-covariate is a construction we invent a new column
10409: ncovcol + k1
10410: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10411: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10412: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10413: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10414: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10415: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10416: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10417: */
1.145 brouard 10418: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10419: 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 10420: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10421: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10422: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10423: 4 covariates (3 plus signs)
10424: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10425: */
1.230 brouard 10426: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10427: * individual dummy, fixed or varying:
10428: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10429: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10430: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10431: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10432: * Tmodelind[1]@9={9,0,3,2,}*/
10433: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10434: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10435: * individual quantitative, fixed or varying:
10436: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10437: * 3, 1, 0, 0, 0, 0, 0, 0},
10438: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10439: /* Main decodemodel */
10440:
1.187 brouard 10441:
1.223 brouard 10442: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10443: goto end;
10444:
1.137 brouard 10445: if((double)(lastobs-imx)/(double)imx > 1.10){
10446: nbwarn++;
10447: 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);
10448: 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);
10449: }
1.136 brouard 10450: /* if(mle==1){*/
1.137 brouard 10451: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10452: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10453: }
10454:
10455: /*-calculation of age at interview from date of interview and age at death -*/
10456: agev=matrix(1,maxwav,1,imx);
10457:
10458: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10459: goto end;
10460:
1.126 brouard 10461:
1.136 brouard 10462: agegomp=(int)agemin;
10463: free_vector(moisnais,1,n);
10464: free_vector(annais,1,n);
1.126 brouard 10465: /* free_matrix(mint,1,maxwav,1,n);
10466: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10467: /* free_vector(moisdc,1,n); */
10468: /* free_vector(andc,1,n); */
1.145 brouard 10469: /* */
10470:
1.126 brouard 10471: wav=ivector(1,imx);
1.214 brouard 10472: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10473: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10474: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10475: 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.*/
10476: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10477: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10478:
10479: /* Concatenates waves */
1.214 brouard 10480: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10481: Death is a valid wave (if date is known).
10482: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10483: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10484: and mw[mi+1][i]. dh depends on stepm.
10485: */
10486:
1.126 brouard 10487: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10488: /* Concatenates waves */
1.145 brouard 10489:
1.215 brouard 10490: free_vector(moisdc,1,n);
10491: free_vector(andc,1,n);
10492:
1.126 brouard 10493: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10494: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10495: ncodemax[1]=1;
1.145 brouard 10496: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10497: cptcoveff=0;
1.220 brouard 10498: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10499: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10500: }
10501:
10502: ncovcombmax=pow(2,cptcoveff);
10503: invalidvarcomb=ivector(1, ncovcombmax);
10504: for(i=1;i<ncovcombmax;i++)
10505: invalidvarcomb[i]=0;
10506:
1.211 brouard 10507: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10508: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10509: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10510:
1.200 brouard 10511: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10512: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10513: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10514: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10515: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10516: * (currently 0 or 1) in the data.
10517: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10518: * corresponding modality (h,j).
10519: */
10520:
1.145 brouard 10521: h=0;
10522: /*if (cptcovn > 0) */
1.126 brouard 10523: m=pow(2,cptcoveff);
10524:
1.144 brouard 10525: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10526: * For k=4 covariates, h goes from 1 to m=2**k
10527: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10528: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10529: * h\k 1 2 3 4
1.143 brouard 10530: *______________________________
10531: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10532: * 2 2 1 1 1
10533: * 3 i=2 1 2 1 1
10534: * 4 2 2 1 1
10535: * 5 i=3 1 i=2 1 2 1
10536: * 6 2 1 2 1
10537: * 7 i=4 1 2 2 1
10538: * 8 2 2 2 1
1.197 brouard 10539: * 9 i=5 1 i=3 1 i=2 1 2
10540: * 10 2 1 1 2
10541: * 11 i=6 1 2 1 2
10542: * 12 2 2 1 2
10543: * 13 i=7 1 i=4 1 2 2
10544: * 14 2 1 2 2
10545: * 15 i=8 1 2 2 2
10546: * 16 2 2 2 2
1.143 brouard 10547: */
1.212 brouard 10548: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10549: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10550: * and the value of each covariate?
10551: * V1=1, V2=1, V3=2, V4=1 ?
10552: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10553: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10554: * In order to get the real value in the data, we use nbcode
10555: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10556: * We are keeping this crazy system in order to be able (in the future?)
10557: * to have more than 2 values (0 or 1) for a covariate.
10558: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10559: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10560: * bbbbbbbb
10561: * 76543210
10562: * h-1 00000101 (6-1=5)
1.219 brouard 10563: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10564: * &
10565: * 1 00000001 (1)
1.219 brouard 10566: * 00000000 = 1 & ((h-1) >> (k-1))
10567: * +1= 00000001 =1
1.211 brouard 10568: *
10569: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10570: * h' 1101 =2^3+2^2+0x2^1+2^0
10571: * >>k' 11
10572: * & 00000001
10573: * = 00000001
10574: * +1 = 00000010=2 = codtabm(14,3)
10575: * Reverse h=6 and m=16?
10576: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10577: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10578: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10579: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10580: * V3=decodtabm(14,3,2**4)=2
10581: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10582: *(h-1) >> (j-1) 0011 =13 >> 2
10583: * &1 000000001
10584: * = 000000001
10585: * +1= 000000010 =2
10586: * 2211
10587: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10588: * V3=2
1.220 brouard 10589: * codtabm and decodtabm are identical
1.211 brouard 10590: */
10591:
1.145 brouard 10592:
10593: free_ivector(Ndum,-1,NCOVMAX);
10594:
10595:
1.126 brouard 10596:
1.186 brouard 10597: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10598: strcpy(optionfilegnuplot,optionfilefiname);
10599: if(mle==-3)
1.201 brouard 10600: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10601: strcat(optionfilegnuplot,".gp");
10602:
10603: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10604: printf("Problem with file %s",optionfilegnuplot);
10605: }
10606: else{
1.204 brouard 10607: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10608: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10609: //fprintf(ficgp,"set missing 'NaNq'\n");
10610: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10611: }
10612: /* fclose(ficgp);*/
1.186 brouard 10613:
10614:
10615: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10616:
10617: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10618: if(mle==-3)
1.201 brouard 10619: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10620: strcat(optionfilehtm,".htm");
10621: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10622: printf("Problem with %s \n",optionfilehtm);
10623: exit(0);
1.126 brouard 10624: }
10625:
10626: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10627: strcat(optionfilehtmcov,"-cov.htm");
10628: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10629: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10630: }
10631: else{
10632: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10633: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10634: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10635: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10636: }
10637:
1.213 brouard 10638: 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 10639: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10640: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10641: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10642: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10643: \n\
10644: <hr size=\"2\" color=\"#EC5E5E\">\
10645: <ul><li><h4>Parameter files</h4>\n\
10646: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10647: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10648: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10649: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10650: - Date and time at start: %s</ul>\n",\
10651: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10652: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10653: fileres,fileres,\
10654: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10655: fflush(fichtm);
10656:
10657: strcpy(pathr,path);
10658: strcat(pathr,optionfilefiname);
1.184 brouard 10659: #ifdef WIN32
10660: _chdir(optionfilefiname); /* Move to directory named optionfile */
10661: #else
1.126 brouard 10662: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10663: #endif
10664:
1.126 brouard 10665:
1.220 brouard 10666: /* Calculates basic frequencies. Computes observed prevalence at single age
10667: and for any valid combination of covariates
1.126 brouard 10668: and prints on file fileres'p'. */
1.251 brouard 10669: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10670: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10671:
10672: fprintf(fichtm,"\n");
10673: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10674: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10675: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10676: imx,agemin,agemax,jmin,jmax,jmean);
10677: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10678: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10679: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10680: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10681: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10682:
1.126 brouard 10683: /* For Powell, parameters are in a vector p[] starting at p[1]
10684: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10685: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10686:
10687: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10688: /* For mortality only */
1.126 brouard 10689: if (mle==-3){
1.136 brouard 10690: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10691: for(i=1;i<=NDIM;i++)
10692: for(j=1;j<=NDIM;j++)
10693: ximort[i][j]=0.;
1.186 brouard 10694: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10695: cens=ivector(1,n);
10696: ageexmed=vector(1,n);
10697: agecens=vector(1,n);
10698: dcwave=ivector(1,n);
1.223 brouard 10699:
1.126 brouard 10700: for (i=1; i<=imx; i++){
10701: dcwave[i]=-1;
10702: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10703: if (s[m][i]>nlstate) {
10704: dcwave[i]=m;
10705: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10706: break;
10707: }
1.126 brouard 10708: }
1.226 brouard 10709:
1.126 brouard 10710: for (i=1; i<=imx; i++) {
10711: if (wav[i]>0){
1.226 brouard 10712: ageexmed[i]=agev[mw[1][i]][i];
10713: j=wav[i];
10714: agecens[i]=1.;
10715:
10716: if (ageexmed[i]> 1 && wav[i] > 0){
10717: agecens[i]=agev[mw[j][i]][i];
10718: cens[i]= 1;
10719: }else if (ageexmed[i]< 1)
10720: cens[i]= -1;
10721: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10722: cens[i]=0 ;
1.126 brouard 10723: }
10724: else cens[i]=-1;
10725: }
10726:
10727: for (i=1;i<=NDIM;i++) {
10728: for (j=1;j<=NDIM;j++)
1.226 brouard 10729: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10730: }
10731:
1.145 brouard 10732: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10733: /*printf("%lf %lf", p[1], p[2]);*/
10734:
10735:
1.136 brouard 10736: #ifdef GSL
10737: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10738: #else
1.126 brouard 10739: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10740: #endif
1.201 brouard 10741: strcpy(filerespow,"POW-MORT_");
10742: strcat(filerespow,fileresu);
1.126 brouard 10743: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10744: printf("Problem with resultfile: %s\n", filerespow);
10745: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10746: }
1.136 brouard 10747: #ifdef GSL
10748: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10749: #else
1.126 brouard 10750: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10751: #endif
1.126 brouard 10752: /* for (i=1;i<=nlstate;i++)
10753: for(j=1;j<=nlstate+ndeath;j++)
10754: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10755: */
10756: fprintf(ficrespow,"\n");
1.136 brouard 10757: #ifdef GSL
10758: /* gsl starts here */
10759: T = gsl_multimin_fminimizer_nmsimplex;
10760: gsl_multimin_fminimizer *sfm = NULL;
10761: gsl_vector *ss, *x;
10762: gsl_multimin_function minex_func;
10763:
10764: /* Initial vertex size vector */
10765: ss = gsl_vector_alloc (NDIM);
10766:
10767: if (ss == NULL){
10768: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10769: }
10770: /* Set all step sizes to 1 */
10771: gsl_vector_set_all (ss, 0.001);
10772:
10773: /* Starting point */
1.126 brouard 10774:
1.136 brouard 10775: x = gsl_vector_alloc (NDIM);
10776:
10777: if (x == NULL){
10778: gsl_vector_free(ss);
10779: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10780: }
10781:
10782: /* Initialize method and iterate */
10783: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10784: /* gsl_vector_set(x, 0, 0.0268); */
10785: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10786: gsl_vector_set(x, 0, p[1]);
10787: gsl_vector_set(x, 1, p[2]);
10788:
10789: minex_func.f = &gompertz_f;
10790: minex_func.n = NDIM;
10791: minex_func.params = (void *)&p; /* ??? */
10792:
10793: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10794: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10795:
10796: printf("Iterations beginning .....\n\n");
10797: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10798:
10799: iteri=0;
10800: while (rval == GSL_CONTINUE){
10801: iteri++;
10802: status = gsl_multimin_fminimizer_iterate(sfm);
10803:
10804: if (status) printf("error: %s\n", gsl_strerror (status));
10805: fflush(0);
10806:
10807: if (status)
10808: break;
10809:
10810: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10811: ssval = gsl_multimin_fminimizer_size (sfm);
10812:
10813: if (rval == GSL_SUCCESS)
10814: printf ("converged to a local maximum at\n");
10815:
10816: printf("%5d ", iteri);
10817: for (it = 0; it < NDIM; it++){
10818: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10819: }
10820: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10821: }
10822:
10823: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10824:
10825: gsl_vector_free(x); /* initial values */
10826: gsl_vector_free(ss); /* inital step size */
10827: for (it=0; it<NDIM; it++){
10828: p[it+1]=gsl_vector_get(sfm->x,it);
10829: fprintf(ficrespow," %.12lf", p[it]);
10830: }
10831: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10832: #endif
10833: #ifdef POWELL
10834: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10835: #endif
1.126 brouard 10836: fclose(ficrespow);
10837:
1.203 brouard 10838: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10839:
10840: for(i=1; i <=NDIM; i++)
10841: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10842: matcov[i][j]=matcov[j][i];
1.126 brouard 10843:
10844: printf("\nCovariance matrix\n ");
1.203 brouard 10845: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10846: for(i=1; i <=NDIM; i++) {
10847: for(j=1;j<=NDIM;j++){
1.220 brouard 10848: printf("%f ",matcov[i][j]);
10849: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10850: }
1.203 brouard 10851: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10852: }
10853:
10854: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10855: for (i=1;i<=NDIM;i++) {
1.126 brouard 10856: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10857: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10858: }
1.126 brouard 10859: lsurv=vector(1,AGESUP);
10860: lpop=vector(1,AGESUP);
10861: tpop=vector(1,AGESUP);
10862: lsurv[agegomp]=100000;
10863:
10864: for (k=agegomp;k<=AGESUP;k++) {
10865: agemortsup=k;
10866: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10867: }
10868:
10869: for (k=agegomp;k<agemortsup;k++)
10870: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10871:
10872: for (k=agegomp;k<agemortsup;k++){
10873: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10874: sumlpop=sumlpop+lpop[k];
10875: }
10876:
10877: tpop[agegomp]=sumlpop;
10878: for (k=agegomp;k<(agemortsup-3);k++){
10879: /* tpop[k+1]=2;*/
10880: tpop[k+1]=tpop[k]-lpop[k];
10881: }
10882:
10883:
10884: printf("\nAge lx qx dx Lx Tx e(x)\n");
10885: for (k=agegomp;k<(agemortsup-2);k++)
10886: 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]);
10887:
10888:
10889: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10890: ageminpar=50;
10891: agemaxpar=100;
1.194 brouard 10892: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10893: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10894: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10895: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10896: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10897: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10898: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10899: }else{
10900: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10901: 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 10902: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10903: }
1.201 brouard 10904: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10905: stepm, weightopt,\
10906: model,imx,p,matcov,agemortsup);
10907:
10908: free_vector(lsurv,1,AGESUP);
10909: free_vector(lpop,1,AGESUP);
10910: free_vector(tpop,1,AGESUP);
1.220 brouard 10911: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10912: free_ivector(cens,1,n);
10913: free_vector(agecens,1,n);
10914: free_ivector(dcwave,1,n);
1.220 brouard 10915: #ifdef GSL
1.136 brouard 10916: #endif
1.186 brouard 10917: } /* Endof if mle==-3 mortality only */
1.205 brouard 10918: /* Standard */
10919: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10920: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10921: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10922: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10923: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10924: for (k=1; k<=npar;k++)
10925: printf(" %d %8.5f",k,p[k]);
10926: printf("\n");
1.205 brouard 10927: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10928: /* mlikeli uses func not funcone */
1.247 brouard 10929: /* for(i=1;i<nlstate;i++){ */
10930: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10931: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10932: /* } */
1.205 brouard 10933: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10934: }
10935: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10936: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10937: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10938: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10939: }
10940: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10941: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10942: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10943: for (k=1; k<=npar;k++)
10944: printf(" %d %8.5f",k,p[k]);
10945: printf("\n");
10946:
10947: /*--------- results files --------------*/
1.224 brouard 10948: 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 10949:
10950:
10951: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10952: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10953: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10954: for(i=1,jk=1; i <=nlstate; i++){
10955: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10956: if (k != i) {
10957: printf("%d%d ",i,k);
10958: fprintf(ficlog,"%d%d ",i,k);
10959: fprintf(ficres,"%1d%1d ",i,k);
10960: for(j=1; j <=ncovmodel; j++){
10961: printf("%12.7f ",p[jk]);
10962: fprintf(ficlog,"%12.7f ",p[jk]);
10963: fprintf(ficres,"%12.7f ",p[jk]);
10964: jk++;
10965: }
10966: printf("\n");
10967: fprintf(ficlog,"\n");
10968: fprintf(ficres,"\n");
10969: }
1.126 brouard 10970: }
10971: }
1.203 brouard 10972: if(mle != 0){
10973: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10974: ftolhess=ftol; /* Usually correct */
1.203 brouard 10975: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10976: 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");
10977: 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");
10978: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10979: for(k=1; k <=(nlstate+ndeath); k++){
10980: if (k != i) {
10981: printf("%d%d ",i,k);
10982: fprintf(ficlog,"%d%d ",i,k);
10983: for(j=1; j <=ncovmodel; j++){
10984: 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]));
10985: 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]));
10986: jk++;
10987: }
10988: printf("\n");
10989: fprintf(ficlog,"\n");
10990: }
10991: }
1.193 brouard 10992: }
1.203 brouard 10993: } /* end of hesscov and Wald tests */
1.225 brouard 10994:
1.203 brouard 10995: /* */
1.126 brouard 10996: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10997: printf("# Scales (for hessian or gradient estimation)\n");
10998: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10999: for(i=1,jk=1; i <=nlstate; i++){
11000: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11001: if (j!=i) {
11002: fprintf(ficres,"%1d%1d",i,j);
11003: printf("%1d%1d",i,j);
11004: fprintf(ficlog,"%1d%1d",i,j);
11005: for(k=1; k<=ncovmodel;k++){
11006: printf(" %.5e",delti[jk]);
11007: fprintf(ficlog," %.5e",delti[jk]);
11008: fprintf(ficres," %.5e",delti[jk]);
11009: jk++;
11010: }
11011: printf("\n");
11012: fprintf(ficlog,"\n");
11013: fprintf(ficres,"\n");
11014: }
1.126 brouard 11015: }
11016: }
11017:
11018: 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 11019: if(mle >= 1) /* To big for the screen */
1.126 brouard 11020: 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");
11021: 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");
11022: /* # 121 Var(a12)\n\ */
11023: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11024: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11025: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11026: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11027: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11028: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11029: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11030:
11031:
11032: /* Just to have a covariance matrix which will be more understandable
11033: even is we still don't want to manage dictionary of variables
11034: */
11035: for(itimes=1;itimes<=2;itimes++){
11036: jj=0;
11037: for(i=1; i <=nlstate; i++){
1.225 brouard 11038: for(j=1; j <=nlstate+ndeath; j++){
11039: if(j==i) continue;
11040: for(k=1; k<=ncovmodel;k++){
11041: jj++;
11042: ca[0]= k+'a'-1;ca[1]='\0';
11043: if(itimes==1){
11044: if(mle>=1)
11045: printf("#%1d%1d%d",i,j,k);
11046: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11047: fprintf(ficres,"#%1d%1d%d",i,j,k);
11048: }else{
11049: if(mle>=1)
11050: printf("%1d%1d%d",i,j,k);
11051: fprintf(ficlog,"%1d%1d%d",i,j,k);
11052: fprintf(ficres,"%1d%1d%d",i,j,k);
11053: }
11054: ll=0;
11055: for(li=1;li <=nlstate; li++){
11056: for(lj=1;lj <=nlstate+ndeath; lj++){
11057: if(lj==li) continue;
11058: for(lk=1;lk<=ncovmodel;lk++){
11059: ll++;
11060: if(ll<=jj){
11061: cb[0]= lk +'a'-1;cb[1]='\0';
11062: if(ll<jj){
11063: if(itimes==1){
11064: if(mle>=1)
11065: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11066: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11067: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11068: }else{
11069: if(mle>=1)
11070: printf(" %.5e",matcov[jj][ll]);
11071: fprintf(ficlog," %.5e",matcov[jj][ll]);
11072: fprintf(ficres," %.5e",matcov[jj][ll]);
11073: }
11074: }else{
11075: if(itimes==1){
11076: if(mle>=1)
11077: printf(" Var(%s%1d%1d)",ca,i,j);
11078: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11079: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11080: }else{
11081: if(mle>=1)
11082: printf(" %.7e",matcov[jj][ll]);
11083: fprintf(ficlog," %.7e",matcov[jj][ll]);
11084: fprintf(ficres," %.7e",matcov[jj][ll]);
11085: }
11086: }
11087: }
11088: } /* end lk */
11089: } /* end lj */
11090: } /* end li */
11091: if(mle>=1)
11092: printf("\n");
11093: fprintf(ficlog,"\n");
11094: fprintf(ficres,"\n");
11095: numlinepar++;
11096: } /* end k*/
11097: } /*end j */
1.126 brouard 11098: } /* end i */
11099: } /* end itimes */
11100:
11101: fflush(ficlog);
11102: fflush(ficres);
1.225 brouard 11103: while(fgets(line, MAXLINE, ficpar)) {
11104: /* If line starts with a # it is a comment */
11105: if (line[0] == '#') {
11106: numlinepar++;
11107: fputs(line,stdout);
11108: fputs(line,ficparo);
11109: fputs(line,ficlog);
11110: continue;
11111: }else
11112: break;
11113: }
11114:
1.209 brouard 11115: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11116: /* ungetc(c,ficpar); */
11117: /* fgets(line, MAXLINE, ficpar); */
11118: /* fputs(line,stdout); */
11119: /* fputs(line,ficparo); */
11120: /* } */
11121: /* ungetc(c,ficpar); */
1.126 brouard 11122:
11123: estepm=0;
1.209 brouard 11124: 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 11125:
11126: if (num_filled != 6) {
11127: 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);
11128: 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);
11129: goto end;
11130: }
11131: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11132: }
11133: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11134: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11135:
1.209 brouard 11136: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11137: if (estepm==0 || estepm < stepm) estepm=stepm;
11138: if (fage <= 2) {
11139: bage = ageminpar;
11140: fage = agemaxpar;
11141: }
11142:
11143: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11144: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11145: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11146:
1.186 brouard 11147: /* Other stuffs, more or less useful */
1.254 brouard 11148: while(fgets(line, MAXLINE, ficpar)) {
11149: /* If line starts with a # it is a comment */
11150: if (line[0] == '#') {
11151: numlinepar++;
11152: fputs(line,stdout);
11153: fputs(line,ficparo);
11154: fputs(line,ficlog);
11155: continue;
11156: }else
11157: break;
11158: }
11159:
11160: if((num_filled=sscanf(line,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav)) !=EOF){
11161:
11162: if (num_filled != 7) {
11163: printf("Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11164: fprintf(ficlog,"Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11165: goto end;
11166: }
11167: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11168: 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);
11169: 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);
11170: fprintf(ficlog,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
1.126 brouard 11171: }
1.254 brouard 11172:
11173: while(fgets(line, MAXLINE, ficpar)) {
11174: /* If line starts with a # it is a comment */
11175: if (line[0] == '#') {
11176: numlinepar++;
11177: fputs(line,stdout);
11178: fputs(line,ficparo);
11179: fputs(line,ficlog);
11180: continue;
11181: }else
11182: break;
1.126 brouard 11183: }
11184:
11185:
11186: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11187: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11188:
1.254 brouard 11189: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11190: if (num_filled != 1) {
11191: printf("Error: Not 1 (data)parameters in line but %d, for example:pop_based=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11192: fprintf(ficlog,"Error: Not 1 (data)parameters in line but %d, for example: pop_based=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11193: goto end;
11194: }
11195: printf("pop_based=%d\n",popbased);
11196: fprintf(ficlog,"pop_based=%d\n",popbased);
11197: fprintf(ficparo,"pop_based=%d\n",popbased);
11198: fprintf(ficres,"pop_based=%d\n",popbased);
11199: }
11200:
1.258 brouard 11201: /* Results */
11202: nresult=0;
11203: do{
11204: if(!fgets(line, MAXLINE, ficpar)){
11205: endishere=1;
11206: parameterline=14;
11207: }else if (line[0] == '#') {
11208: /* If line starts with a # it is a comment */
1.254 brouard 11209: numlinepar++;
11210: fputs(line,stdout);
11211: fputs(line,ficparo);
11212: fputs(line,ficlog);
11213: continue;
1.258 brouard 11214: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11215: parameterline=11;
11216: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11217: parameterline=12;
11218: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11219: parameterline=13;
11220: else{
11221: parameterline=14;
1.254 brouard 11222: }
1.258 brouard 11223: switch (parameterline){
11224: case 11:
11225: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF){
11226: if (num_filled != 8) {
11227: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11228: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11229: goto end;
11230: }
11231: 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);
11232: 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);
11233: 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);
11234: 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);
11235: /* day and month of proj2 are not used but only year anproj2.*/
11236: }
1.254 brouard 11237: break;
1.258 brouard 11238: case 12:
11239: /*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);*/
11240: if((num_filled=sscanf(line,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF){
11241: if (num_filled != 8) {
11242: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11243: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11244: goto end;
11245: }
11246: printf("backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
11247: 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);
11248: 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);
11249: 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);
11250: /* day and month of proj2 are not used but only year anproj2.*/
11251: }
1.230 brouard 11252: break;
1.258 brouard 11253: case 13:
11254: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11255: if (num_filled == 0){
11256: resultline[0]='\0';
11257: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11258: fprintf(ficlog,"Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11259: break;
11260: } else if (num_filled != 1){
11261: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11262: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11263: }
11264: nresult++; /* Sum of resultlines */
11265: printf("Result %d: result=%s\n",nresult, resultline);
11266: if(nresult > MAXRESULTLINES){
11267: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11268: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11269: goto end;
11270: }
11271: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11272: fprintf(ficparo,"result: %s\n",resultline);
11273: fprintf(ficres,"result: %s\n",resultline);
11274: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11275: break;
1.258 brouard 11276: case 14:
1.259 brouard 11277: if(ncovmodel >2 && nresult==0 ){
11278: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11279: goto end;
11280: }
1.259 brouard 11281: break;
1.258 brouard 11282: default:
11283: nresult=1;
11284: decoderesult(".",nresult ); /* No covariate */
11285: }
11286: } /* End switch parameterline */
11287: }while(endishere==0); /* End do */
1.126 brouard 11288:
1.230 brouard 11289: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11290: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11291:
11292: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11293: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11294: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11295: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11296: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11297: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11298: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11299: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11300: }else{
1.218 brouard 11301: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11302: }
11303: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11304: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11305: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11306:
1.225 brouard 11307: /*------------ free_vector -------------*/
11308: /* chdir(path); */
1.220 brouard 11309:
1.215 brouard 11310: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11311: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11312: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11313: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11314: free_lvector(num,1,n);
11315: free_vector(agedc,1,n);
11316: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11317: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11318: fclose(ficparo);
11319: fclose(ficres);
1.220 brouard 11320:
11321:
1.186 brouard 11322: /* Other results (useful)*/
1.220 brouard 11323:
11324:
1.126 brouard 11325: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11326: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11327: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11328: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11329: fclose(ficrespl);
11330:
11331: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11332: /*#include "hpijx.h"*/
11333: hPijx(p, bage, fage);
1.145 brouard 11334: fclose(ficrespij);
1.227 brouard 11335:
1.220 brouard 11336: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11337: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11338: k=1;
1.126 brouard 11339: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11340:
1.219 brouard 11341: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11342: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11343: for(i=1;i<=AGESUP;i++)
1.219 brouard 11344: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11345: for(k=1;k<=ncovcombmax;k++)
11346: probs[i][j][k]=0.;
1.219 brouard 11347: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11348: if (mobilav!=0 ||mobilavproj !=0 ) {
11349: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11350: for(i=1;i<=AGESUP;i++)
11351: for(j=1;j<=nlstate;j++)
11352: for(k=1;k<=ncovcombmax;k++)
11353: mobaverages[i][j][k]=0.;
1.219 brouard 11354: mobaverage=mobaverages;
11355: if (mobilav!=0) {
1.235 brouard 11356: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11357: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11358: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11359: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11360: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11361: }
1.219 brouard 11362: }
11363: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11364: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11365: else if (mobilavproj !=0) {
1.235 brouard 11366: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11367: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11368: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11369: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11370: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11371: }
1.219 brouard 11372: }
11373: }/* end if moving average */
1.227 brouard 11374:
1.126 brouard 11375: /*---------- Forecasting ------------------*/
11376: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11377: if(prevfcast==1){
11378: /* if(stepm ==1){*/
1.225 brouard 11379: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11380: }
1.217 brouard 11381: if(backcast==1){
1.219 brouard 11382: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11383: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11384: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11385:
11386: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11387:
11388: bprlim=matrix(1,nlstate,1,nlstate);
11389: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11390: fclose(ficresplb);
11391:
1.222 brouard 11392: hBijx(p, bage, fage, mobaverage);
11393: fclose(ficrespijb);
1.219 brouard 11394: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11395:
11396: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11397: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11398: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11399: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11400: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11401: }
1.217 brouard 11402:
1.186 brouard 11403:
11404: /* ------ Other prevalence ratios------------ */
1.126 brouard 11405:
1.215 brouard 11406: free_ivector(wav,1,imx);
11407: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11408: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11409: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11410:
11411:
1.127 brouard 11412: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11413:
1.201 brouard 11414: strcpy(filerese,"E_");
11415: strcat(filerese,fileresu);
1.126 brouard 11416: if((ficreseij=fopen(filerese,"w"))==NULL) {
11417: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11418: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11419: }
1.208 brouard 11420: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11421: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11422:
11423: pstamp(ficreseij);
1.219 brouard 11424:
1.235 brouard 11425: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11426: if (cptcovn < 1){i1=1;}
11427:
11428: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11429: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11430: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11431: continue;
1.219 brouard 11432: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11433: printf("\n#****** ");
1.225 brouard 11434: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11435: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11436: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11437: }
11438: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11439: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11440: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11441: }
11442: fprintf(ficreseij,"******\n");
1.235 brouard 11443: printf("******\n");
1.219 brouard 11444:
11445: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11446: oldm=oldms;savm=savms;
1.235 brouard 11447: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11448:
1.219 brouard 11449: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11450: }
11451: fclose(ficreseij);
1.208 brouard 11452: printf("done evsij\n");fflush(stdout);
11453: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11454:
1.227 brouard 11455: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11456:
11457:
1.201 brouard 11458: strcpy(filerest,"T_");
11459: strcat(filerest,fileresu);
1.127 brouard 11460: if((ficrest=fopen(filerest,"w"))==NULL) {
11461: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11462: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11463: }
1.208 brouard 11464: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11465: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11466:
1.126 brouard 11467:
1.201 brouard 11468: strcpy(fileresstde,"STDE_");
11469: strcat(fileresstde,fileresu);
1.126 brouard 11470: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11471: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11472: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11473: }
1.227 brouard 11474: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11475: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11476:
1.201 brouard 11477: strcpy(filerescve,"CVE_");
11478: strcat(filerescve,fileresu);
1.126 brouard 11479: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11480: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11481: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11482: }
1.227 brouard 11483: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11484: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11485:
1.201 brouard 11486: strcpy(fileresv,"V_");
11487: strcat(fileresv,fileresu);
1.126 brouard 11488: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11489: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11490: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11491: }
1.227 brouard 11492: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11493: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11494:
1.145 brouard 11495: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11496: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11497:
1.235 brouard 11498: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11499: if (cptcovn < 1){i1=1;}
11500:
11501: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11502: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11503: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11504: continue;
1.242 brouard 11505: printf("\n#****** Result for:");
11506: fprintf(ficrest,"\n#****** Result for:");
11507: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11508: for(j=1;j<=cptcoveff;j++){
11509: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11510: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11511: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11512: }
1.235 brouard 11513: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11514: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11515: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11516: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11517: }
1.208 brouard 11518: fprintf(ficrest,"******\n");
1.227 brouard 11519: fprintf(ficlog,"******\n");
11520: printf("******\n");
1.208 brouard 11521:
11522: fprintf(ficresstdeij,"\n#****** ");
11523: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11524: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11525: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11526: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11527: }
1.235 brouard 11528: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11529: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11530: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11531: }
1.208 brouard 11532: fprintf(ficresstdeij,"******\n");
11533: fprintf(ficrescveij,"******\n");
11534:
11535: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11536: /* pstamp(ficresvij); */
1.225 brouard 11537: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11538: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11539: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11540: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11541: }
1.208 brouard 11542: fprintf(ficresvij,"******\n");
11543:
11544: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11545: oldm=oldms;savm=savms;
1.235 brouard 11546: printf(" cvevsij ");
11547: fprintf(ficlog, " cvevsij ");
11548: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11549: printf(" end cvevsij \n ");
11550: fprintf(ficlog, " end cvevsij \n ");
11551:
11552: /*
11553: */
11554: /* goto endfree; */
11555:
11556: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11557: pstamp(ficrest);
11558:
11559:
11560: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11561: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11562: cptcod= 0; /* To be deleted */
11563: printf("varevsij vpopbased=%d \n",vpopbased);
11564: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11565: 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 11566: 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 ");
11567: if(vpopbased==1)
11568: 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);
11569: else
11570: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11571: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11572: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11573: fprintf(ficrest,"\n");
11574: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11575: epj=vector(1,nlstate+1);
11576: printf("Computing age specific period (stable) prevalences in each health state \n");
11577: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11578: for(age=bage; age <=fage ;age++){
1.235 brouard 11579: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11580: if (vpopbased==1) {
11581: if(mobilav ==0){
11582: for(i=1; i<=nlstate;i++)
11583: prlim[i][i]=probs[(int)age][i][k];
11584: }else{ /* mobilav */
11585: for(i=1; i<=nlstate;i++)
11586: prlim[i][i]=mobaverage[(int)age][i][k];
11587: }
11588: }
1.219 brouard 11589:
1.227 brouard 11590: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11591: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11592: /* printf(" age %4.0f ",age); */
11593: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11594: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11595: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11596: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11597: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11598: }
11599: epj[nlstate+1] +=epj[j];
11600: }
11601: /* printf(" age %4.0f \n",age); */
1.219 brouard 11602:
1.227 brouard 11603: for(i=1, vepp=0.;i <=nlstate;i++)
11604: for(j=1;j <=nlstate;j++)
11605: vepp += vareij[i][j][(int)age];
11606: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11607: for(j=1;j <=nlstate;j++){
11608: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11609: }
11610: fprintf(ficrest,"\n");
11611: }
1.208 brouard 11612: } /* End vpopbased */
11613: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11614: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11615: free_vector(epj,1,nlstate+1);
1.235 brouard 11616: printf("done selection\n");fflush(stdout);
11617: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11618:
1.145 brouard 11619: /*}*/
1.235 brouard 11620: } /* End k selection */
1.227 brouard 11621:
11622: printf("done State-specific expectancies\n");fflush(stdout);
11623: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11624:
1.126 brouard 11625: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11626:
1.201 brouard 11627: strcpy(fileresvpl,"VPL_");
11628: strcat(fileresvpl,fileresu);
1.126 brouard 11629: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11630: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11631: exit(0);
11632: }
1.208 brouard 11633: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11634: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11635:
1.145 brouard 11636: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11637: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11638:
1.235 brouard 11639: i1=pow(2,cptcoveff);
11640: if (cptcovn < 1){i1=1;}
11641:
11642: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11643: for(k=1; k<=i1;k++){
1.253 brouard 11644: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11645: continue;
1.227 brouard 11646: fprintf(ficresvpl,"\n#****** ");
11647: printf("\n#****** ");
11648: fprintf(ficlog,"\n#****** ");
11649: for(j=1;j<=cptcoveff;j++) {
11650: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11651: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11652: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11653: }
1.235 brouard 11654: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11655: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11656: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11657: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11658: }
1.227 brouard 11659: fprintf(ficresvpl,"******\n");
11660: printf("******\n");
11661: fprintf(ficlog,"******\n");
11662:
11663: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11664: oldm=oldms;savm=savms;
1.235 brouard 11665: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11666: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11667: /*}*/
1.126 brouard 11668: }
1.227 brouard 11669:
1.126 brouard 11670: fclose(ficresvpl);
1.208 brouard 11671: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11672: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11673:
11674: free_vector(weight,1,n);
11675: free_imatrix(Tvard,1,NCOVMAX,1,2);
11676: free_imatrix(s,1,maxwav+1,1,n);
11677: free_matrix(anint,1,maxwav,1,n);
11678: free_matrix(mint,1,maxwav,1,n);
11679: free_ivector(cod,1,n);
11680: free_ivector(tab,1,NCOVMAX);
11681: fclose(ficresstdeij);
11682: fclose(ficrescveij);
11683: fclose(ficresvij);
11684: fclose(ficrest);
11685: fclose(ficpar);
11686:
11687:
1.126 brouard 11688: /*---------- End : free ----------------*/
1.219 brouard 11689: if (mobilav!=0 ||mobilavproj !=0)
11690: 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 11691: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11692: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11693: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11694: } /* mle==-3 arrives here for freeing */
1.227 brouard 11695: /* endfree:*/
11696: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11697: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11698: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11699: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11700: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11701: free_matrix(coqvar,1,maxwav,1,n);
11702: free_matrix(covar,0,NCOVMAX,1,n);
11703: free_matrix(matcov,1,npar,1,npar);
11704: free_matrix(hess,1,npar,1,npar);
11705: /*free_vector(delti,1,npar);*/
11706: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11707: free_matrix(agev,1,maxwav,1,imx);
11708: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11709:
11710: free_ivector(ncodemax,1,NCOVMAX);
11711: free_ivector(ncodemaxwundef,1,NCOVMAX);
11712: free_ivector(Dummy,-1,NCOVMAX);
11713: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11714: free_ivector(DummyV,1,NCOVMAX);
11715: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11716: free_ivector(Typevar,-1,NCOVMAX);
11717: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11718: free_ivector(TvarsQ,1,NCOVMAX);
11719: free_ivector(TvarsQind,1,NCOVMAX);
11720: free_ivector(TvarsD,1,NCOVMAX);
11721: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11722: free_ivector(TvarFD,1,NCOVMAX);
11723: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11724: free_ivector(TvarF,1,NCOVMAX);
11725: free_ivector(TvarFind,1,NCOVMAX);
11726: free_ivector(TvarV,1,NCOVMAX);
11727: free_ivector(TvarVind,1,NCOVMAX);
11728: free_ivector(TvarA,1,NCOVMAX);
11729: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11730: free_ivector(TvarFQ,1,NCOVMAX);
11731: free_ivector(TvarFQind,1,NCOVMAX);
11732: free_ivector(TvarVD,1,NCOVMAX);
11733: free_ivector(TvarVDind,1,NCOVMAX);
11734: free_ivector(TvarVQ,1,NCOVMAX);
11735: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11736: free_ivector(Tvarsel,1,NCOVMAX);
11737: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11738: free_ivector(Tposprod,1,NCOVMAX);
11739: free_ivector(Tprod,1,NCOVMAX);
11740: free_ivector(Tvaraff,1,NCOVMAX);
11741: free_ivector(invalidvarcomb,1,ncovcombmax);
11742: free_ivector(Tage,1,NCOVMAX);
11743: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11744: free_ivector(TmodelInvind,1,NCOVMAX);
11745: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11746:
11747: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11748: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11749: fflush(fichtm);
11750: fflush(ficgp);
11751:
1.227 brouard 11752:
1.126 brouard 11753: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11754: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11755: 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 11756: }else{
11757: printf("End of Imach\n");
11758: fprintf(ficlog,"End of Imach\n");
11759: }
11760: printf("See log file on %s\n",filelog);
11761: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11762: /*(void) gettimeofday(&end_time,&tzp);*/
11763: rend_time = time(NULL);
11764: end_time = *localtime(&rend_time);
11765: /* tml = *localtime(&end_time.tm_sec); */
11766: strcpy(strtend,asctime(&end_time));
1.126 brouard 11767: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11768: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11769: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11770:
1.157 brouard 11771: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11772: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11773: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11774: /* printf("Total time was %d uSec.\n", total_usecs);*/
11775: /* if(fileappend(fichtm,optionfilehtm)){ */
11776: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11777: fclose(fichtm);
11778: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11779: fclose(fichtmcov);
11780: fclose(ficgp);
11781: fclose(ficlog);
11782: /*------ End -----------*/
1.227 brouard 11783:
11784:
11785: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11786: #ifdef WIN32
1.227 brouard 11787: if (_chdir(pathcd) != 0)
11788: printf("Can't move to directory %s!\n",path);
11789: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11790: #else
1.227 brouard 11791: if(chdir(pathcd) != 0)
11792: printf("Can't move to directory %s!\n", path);
11793: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11794: #endif
1.126 brouard 11795: printf("Current directory %s!\n",pathcd);
11796: /*strcat(plotcmd,CHARSEPARATOR);*/
11797: sprintf(plotcmd,"gnuplot");
1.157 brouard 11798: #ifdef _WIN32
1.126 brouard 11799: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11800: #endif
11801: if(!stat(plotcmd,&info)){
1.158 brouard 11802: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11803: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11804: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11805: }else
11806: strcpy(pplotcmd,plotcmd);
1.157 brouard 11807: #ifdef __unix
1.126 brouard 11808: strcpy(plotcmd,GNUPLOTPROGRAM);
11809: if(!stat(plotcmd,&info)){
1.158 brouard 11810: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11811: }else
11812: strcpy(pplotcmd,plotcmd);
11813: #endif
11814: }else
11815: strcpy(pplotcmd,plotcmd);
11816:
11817: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11818: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11819:
1.126 brouard 11820: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11821: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11822: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11823: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11824: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11825: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11826: }
1.158 brouard 11827: printf(" Successful, please wait...");
1.126 brouard 11828: while (z[0] != 'q') {
11829: /* chdir(path); */
1.154 brouard 11830: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11831: scanf("%s",z);
11832: /* if (z[0] == 'c') system("./imach"); */
11833: if (z[0] == 'e') {
1.158 brouard 11834: #ifdef __APPLE__
1.152 brouard 11835: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11836: #elif __linux
11837: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11838: #else
1.152 brouard 11839: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11840: #endif
11841: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11842: system(pplotcmd);
1.126 brouard 11843: }
11844: else if (z[0] == 'g') system(plotcmd);
11845: else if (z[0] == 'q') exit(0);
11846: }
1.227 brouard 11847: end:
1.126 brouard 11848: while (z[0] != 'q') {
1.195 brouard 11849: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11850: scanf("%s",z);
11851: }
11852: }
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