Annotation of imach/src/imach.c, revision 1.263
1.263 ! brouard 1: /* $Id: imach.c,v 1.262 2017/04/18 16:48:12 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.263 ! brouard 4: Revision 1.262 2017/04/18 16:48:12 brouard
! 5: *** empty log message ***
! 6:
1.262 brouard 7: Revision 1.261 2017/04/05 10:14:09 brouard
8: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
9:
1.261 brouard 10: Revision 1.260 2017/04/04 17:46:59 brouard
11: Summary: Gnuplot indexations fixed (humm)
12:
1.260 brouard 13: Revision 1.259 2017/04/04 13:01:16 brouard
14: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
15:
1.259 brouard 16: Revision 1.258 2017/04/03 10:17:47 brouard
17: Summary: Version 0.99r12
18:
19: Some cleanings, conformed with updated documentation.
20:
1.258 brouard 21: Revision 1.257 2017/03/29 16:53:30 brouard
22: Summary: Temp
23:
1.257 brouard 24: Revision 1.256 2017/03/27 05:50:23 brouard
25: Summary: Temporary
26:
1.256 brouard 27: Revision 1.255 2017/03/08 16:02:28 brouard
28: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
29:
1.255 brouard 30: Revision 1.254 2017/03/08 07:13:00 brouard
31: Summary: Fixing data parameter line
32:
1.254 brouard 33: Revision 1.253 2016/12/15 11:59:41 brouard
34: Summary: 0.99 in progress
35:
1.253 brouard 36: Revision 1.252 2016/09/15 21:15:37 brouard
37: *** empty log message ***
38:
1.252 brouard 39: Revision 1.251 2016/09/15 15:01:13 brouard
40: Summary: not working
41:
1.251 brouard 42: Revision 1.250 2016/09/08 16:07:27 brouard
43: Summary: continue
44:
1.250 brouard 45: Revision 1.249 2016/09/07 17:14:18 brouard
46: Summary: Starting values from frequencies
47:
1.249 brouard 48: Revision 1.248 2016/09/07 14:10:18 brouard
49: *** empty log message ***
50:
1.248 brouard 51: Revision 1.247 2016/09/02 11:11:21 brouard
52: *** empty log message ***
53:
1.247 brouard 54: Revision 1.246 2016/09/02 08:49:22 brouard
55: *** empty log message ***
56:
1.246 brouard 57: Revision 1.245 2016/09/02 07:25:01 brouard
58: *** empty log message ***
59:
1.245 brouard 60: Revision 1.244 2016/09/02 07:17:34 brouard
61: *** empty log message ***
62:
1.244 brouard 63: Revision 1.243 2016/09/02 06:45:35 brouard
64: *** empty log message ***
65:
1.243 brouard 66: Revision 1.242 2016/08/30 15:01:20 brouard
67: Summary: Fixing a lots
68:
1.242 brouard 69: Revision 1.241 2016/08/29 17:17:25 brouard
70: Summary: gnuplot problem in Back projection to fix
71:
1.241 brouard 72: Revision 1.240 2016/08/29 07:53:18 brouard
73: Summary: Better
74:
1.240 brouard 75: Revision 1.239 2016/08/26 15:51:03 brouard
76: Summary: Improvement in Powell output in order to copy and paste
77:
78: Author:
79:
1.239 brouard 80: Revision 1.238 2016/08/26 14:23:35 brouard
81: Summary: Starting tests of 0.99
82:
1.238 brouard 83: Revision 1.237 2016/08/26 09:20:19 brouard
84: Summary: to valgrind
85:
1.237 brouard 86: Revision 1.236 2016/08/25 10:50:18 brouard
87: *** empty log message ***
88:
1.236 brouard 89: Revision 1.235 2016/08/25 06:59:23 brouard
90: *** empty log message ***
91:
1.235 brouard 92: Revision 1.234 2016/08/23 16:51:20 brouard
93: *** empty log message ***
94:
1.234 brouard 95: Revision 1.233 2016/08/23 07:40:50 brouard
96: Summary: not working
97:
1.233 brouard 98: Revision 1.232 2016/08/22 14:20:21 brouard
99: Summary: not working
100:
1.232 brouard 101: Revision 1.231 2016/08/22 07:17:15 brouard
102: Summary: not working
103:
1.231 brouard 104: Revision 1.230 2016/08/22 06:55:53 brouard
105: Summary: Not working
106:
1.230 brouard 107: Revision 1.229 2016/07/23 09:45:53 brouard
108: Summary: Completing for func too
109:
1.229 brouard 110: Revision 1.228 2016/07/22 17:45:30 brouard
111: Summary: Fixing some arrays, still debugging
112:
1.227 brouard 113: Revision 1.226 2016/07/12 18:42:34 brouard
114: Summary: temp
115:
1.226 brouard 116: Revision 1.225 2016/07/12 08:40:03 brouard
117: Summary: saving but not running
118:
1.225 brouard 119: Revision 1.224 2016/07/01 13:16:01 brouard
120: Summary: Fixes
121:
1.224 brouard 122: Revision 1.223 2016/02/19 09:23:35 brouard
123: Summary: temporary
124:
1.223 brouard 125: Revision 1.222 2016/02/17 08:14:50 brouard
126: Summary: Probably last 0.98 stable version 0.98r6
127:
1.222 brouard 128: Revision 1.221 2016/02/15 23:35:36 brouard
129: Summary: minor bug
130:
1.220 brouard 131: Revision 1.219 2016/02/15 00:48:12 brouard
132: *** empty log message ***
133:
1.219 brouard 134: Revision 1.218 2016/02/12 11:29:23 brouard
135: Summary: 0.99 Back projections
136:
1.218 brouard 137: Revision 1.217 2015/12/23 17:18:31 brouard
138: Summary: Experimental backcast
139:
1.217 brouard 140: Revision 1.216 2015/12/18 17:32:11 brouard
141: Summary: 0.98r4 Warning and status=-2
142:
143: Version 0.98r4 is now:
144: - displaying an error when status is -1, date of interview unknown and date of death known;
145: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
146: Older changes concerning s=-2, dating from 2005 have been supersed.
147:
1.216 brouard 148: Revision 1.215 2015/12/16 08:52:24 brouard
149: Summary: 0.98r4 working
150:
1.215 brouard 151: Revision 1.214 2015/12/16 06:57:54 brouard
152: Summary: temporary not working
153:
1.214 brouard 154: Revision 1.213 2015/12/11 18:22:17 brouard
155: Summary: 0.98r4
156:
1.213 brouard 157: Revision 1.212 2015/11/21 12:47:24 brouard
158: Summary: minor typo
159:
1.212 brouard 160: Revision 1.211 2015/11/21 12:41:11 brouard
161: Summary: 0.98r3 with some graph of projected cross-sectional
162:
163: Author: Nicolas Brouard
164:
1.211 brouard 165: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 166: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 167: Summary: Adding ftolpl parameter
168: Author: N Brouard
169:
170: We had difficulties to get smoothed confidence intervals. It was due
171: to the period prevalence which wasn't computed accurately. The inner
172: parameter ftolpl is now an outer parameter of the .imach parameter
173: file after estepm. If ftolpl is small 1.e-4 and estepm too,
174: computation are long.
175:
1.209 brouard 176: Revision 1.208 2015/11/17 14:31:57 brouard
177: Summary: temporary
178:
1.208 brouard 179: Revision 1.207 2015/10/27 17:36:57 brouard
180: *** empty log message ***
181:
1.207 brouard 182: Revision 1.206 2015/10/24 07:14:11 brouard
183: *** empty log message ***
184:
1.206 brouard 185: Revision 1.205 2015/10/23 15:50:53 brouard
186: Summary: 0.98r3 some clarification for graphs on likelihood contributions
187:
1.205 brouard 188: Revision 1.204 2015/10/01 16:20:26 brouard
189: Summary: Some new graphs of contribution to likelihood
190:
1.204 brouard 191: Revision 1.203 2015/09/30 17:45:14 brouard
192: Summary: looking at better estimation of the hessian
193:
194: Also a better criteria for convergence to the period prevalence And
195: therefore adding the number of years needed to converge. (The
196: prevalence in any alive state shold sum to one
197:
1.203 brouard 198: Revision 1.202 2015/09/22 19:45:16 brouard
199: Summary: Adding some overall graph on contribution to likelihood. Might change
200:
1.202 brouard 201: Revision 1.201 2015/09/15 17:34:58 brouard
202: Summary: 0.98r0
203:
204: - Some new graphs like suvival functions
205: - Some bugs fixed like model=1+age+V2.
206:
1.201 brouard 207: Revision 1.200 2015/09/09 16:53:55 brouard
208: Summary: Big bug thanks to Flavia
209:
210: Even model=1+age+V2. did not work anymore
211:
1.200 brouard 212: Revision 1.199 2015/09/07 14:09:23 brouard
213: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
214:
1.199 brouard 215: Revision 1.198 2015/09/03 07:14:39 brouard
216: Summary: 0.98q5 Flavia
217:
1.198 brouard 218: Revision 1.197 2015/09/01 18:24:39 brouard
219: *** empty log message ***
220:
1.197 brouard 221: Revision 1.196 2015/08/18 23:17:52 brouard
222: Summary: 0.98q5
223:
1.196 brouard 224: Revision 1.195 2015/08/18 16:28:39 brouard
225: Summary: Adding a hack for testing purpose
226:
227: After reading the title, ftol and model lines, if the comment line has
228: a q, starting with #q, the answer at the end of the run is quit. It
229: permits to run test files in batch with ctest. The former workaround was
230: $ echo q | imach foo.imach
231:
1.195 brouard 232: Revision 1.194 2015/08/18 13:32:00 brouard
233: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
234:
1.194 brouard 235: Revision 1.193 2015/08/04 07:17:42 brouard
236: Summary: 0.98q4
237:
1.193 brouard 238: Revision 1.192 2015/07/16 16:49:02 brouard
239: Summary: Fixing some outputs
240:
1.192 brouard 241: Revision 1.191 2015/07/14 10:00:33 brouard
242: Summary: Some fixes
243:
1.191 brouard 244: Revision 1.190 2015/05/05 08:51:13 brouard
245: Summary: Adding digits in output parameters (7 digits instead of 6)
246:
247: Fix 1+age+.
248:
1.190 brouard 249: Revision 1.189 2015/04/30 14:45:16 brouard
250: Summary: 0.98q2
251:
1.189 brouard 252: Revision 1.188 2015/04/30 08:27:53 brouard
253: *** empty log message ***
254:
1.188 brouard 255: Revision 1.187 2015/04/29 09:11:15 brouard
256: *** empty log message ***
257:
1.187 brouard 258: Revision 1.186 2015/04/23 12:01:52 brouard
259: Summary: V1*age is working now, version 0.98q1
260:
261: Some codes had been disabled in order to simplify and Vn*age was
262: working in the optimization phase, ie, giving correct MLE parameters,
263: but, as usual, outputs were not correct and program core dumped.
264:
1.186 brouard 265: Revision 1.185 2015/03/11 13:26:42 brouard
266: Summary: Inclusion of compile and links command line for Intel Compiler
267:
1.185 brouard 268: Revision 1.184 2015/03/11 11:52:39 brouard
269: Summary: Back from Windows 8. Intel Compiler
270:
1.184 brouard 271: Revision 1.183 2015/03/10 20:34:32 brouard
272: Summary: 0.98q0, trying with directest, mnbrak fixed
273:
274: We use directest instead of original Powell test; probably no
275: incidence on the results, but better justifications;
276: We fixed Numerical Recipes mnbrak routine which was wrong and gave
277: wrong results.
278:
1.183 brouard 279: Revision 1.182 2015/02/12 08:19:57 brouard
280: Summary: Trying to keep directest which seems simpler and more general
281: Author: Nicolas Brouard
282:
1.182 brouard 283: Revision 1.181 2015/02/11 23:22:24 brouard
284: Summary: Comments on Powell added
285:
286: Author:
287:
1.181 brouard 288: Revision 1.180 2015/02/11 17:33:45 brouard
289: Summary: Finishing move from main to function (hpijx and prevalence_limit)
290:
1.180 brouard 291: Revision 1.179 2015/01/04 09:57:06 brouard
292: Summary: back to OS/X
293:
1.179 brouard 294: Revision 1.178 2015/01/04 09:35:48 brouard
295: *** empty log message ***
296:
1.178 brouard 297: Revision 1.177 2015/01/03 18:40:56 brouard
298: Summary: Still testing ilc32 on OSX
299:
1.177 brouard 300: Revision 1.176 2015/01/03 16:45:04 brouard
301: *** empty log message ***
302:
1.176 brouard 303: Revision 1.175 2015/01/03 16:33:42 brouard
304: *** empty log message ***
305:
1.175 brouard 306: Revision 1.174 2015/01/03 16:15:49 brouard
307: Summary: Still in cross-compilation
308:
1.174 brouard 309: Revision 1.173 2015/01/03 12:06:26 brouard
310: Summary: trying to detect cross-compilation
311:
1.173 brouard 312: Revision 1.172 2014/12/27 12:07:47 brouard
313: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
314:
1.172 brouard 315: Revision 1.171 2014/12/23 13:26:59 brouard
316: Summary: Back from Visual C
317:
318: Still problem with utsname.h on Windows
319:
1.171 brouard 320: Revision 1.170 2014/12/23 11:17:12 brouard
321: Summary: Cleaning some \%% back to %%
322:
323: The escape was mandatory for a specific compiler (which one?), but too many warnings.
324:
1.170 brouard 325: Revision 1.169 2014/12/22 23:08:31 brouard
326: Summary: 0.98p
327:
328: Outputs some informations on compiler used, OS etc. Testing on different platforms.
329:
1.169 brouard 330: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 331: Summary: update
1.169 brouard 332:
1.168 brouard 333: Revision 1.167 2014/12/22 13:50:56 brouard
334: Summary: Testing uname and compiler version and if compiled 32 or 64
335:
336: Testing on Linux 64
337:
1.167 brouard 338: Revision 1.166 2014/12/22 11:40:47 brouard
339: *** empty log message ***
340:
1.166 brouard 341: Revision 1.165 2014/12/16 11:20:36 brouard
342: Summary: After compiling on Visual C
343:
344: * imach.c (Module): Merging 1.61 to 1.162
345:
1.165 brouard 346: Revision 1.164 2014/12/16 10:52:11 brouard
347: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
348:
349: * imach.c (Module): Merging 1.61 to 1.162
350:
1.164 brouard 351: Revision 1.163 2014/12/16 10:30:11 brouard
352: * imach.c (Module): Merging 1.61 to 1.162
353:
1.163 brouard 354: Revision 1.162 2014/09/25 11:43:39 brouard
355: Summary: temporary backup 0.99!
356:
1.162 brouard 357: Revision 1.1 2014/09/16 11:06:58 brouard
358: Summary: With some code (wrong) for nlopt
359:
360: Author:
361:
362: Revision 1.161 2014/09/15 20:41:41 brouard
363: Summary: Problem with macro SQR on Intel compiler
364:
1.161 brouard 365: Revision 1.160 2014/09/02 09:24:05 brouard
366: *** empty log message ***
367:
1.160 brouard 368: Revision 1.159 2014/09/01 10:34:10 brouard
369: Summary: WIN32
370: Author: Brouard
371:
1.159 brouard 372: Revision 1.158 2014/08/27 17:11:51 brouard
373: *** empty log message ***
374:
1.158 brouard 375: Revision 1.157 2014/08/27 16:26:55 brouard
376: Summary: Preparing windows Visual studio version
377: Author: Brouard
378:
379: In order to compile on Visual studio, time.h is now correct and time_t
380: and tm struct should be used. difftime should be used but sometimes I
381: just make the differences in raw time format (time(&now).
382: Trying to suppress #ifdef LINUX
383: Add xdg-open for __linux in order to open default browser.
384:
1.157 brouard 385: Revision 1.156 2014/08/25 20:10:10 brouard
386: *** empty log message ***
387:
1.156 brouard 388: Revision 1.155 2014/08/25 18:32:34 brouard
389: Summary: New compile, minor changes
390: Author: Brouard
391:
1.155 brouard 392: Revision 1.154 2014/06/20 17:32:08 brouard
393: Summary: Outputs now all graphs of convergence to period prevalence
394:
1.154 brouard 395: Revision 1.153 2014/06/20 16:45:46 brouard
396: Summary: If 3 live state, convergence to period prevalence on same graph
397: Author: Brouard
398:
1.153 brouard 399: Revision 1.152 2014/06/18 17:54:09 brouard
400: Summary: open browser, use gnuplot on same dir than imach if not found in the path
401:
1.152 brouard 402: Revision 1.151 2014/06/18 16:43:30 brouard
403: *** empty log message ***
404:
1.151 brouard 405: Revision 1.150 2014/06/18 16:42:35 brouard
406: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
407: Author: brouard
408:
1.150 brouard 409: Revision 1.149 2014/06/18 15:51:14 brouard
410: Summary: Some fixes in parameter files errors
411: Author: Nicolas Brouard
412:
1.149 brouard 413: Revision 1.148 2014/06/17 17:38:48 brouard
414: Summary: Nothing new
415: Author: Brouard
416:
417: Just a new packaging for OS/X version 0.98nS
418:
1.148 brouard 419: Revision 1.147 2014/06/16 10:33:11 brouard
420: *** empty log message ***
421:
1.147 brouard 422: Revision 1.146 2014/06/16 10:20:28 brouard
423: Summary: Merge
424: Author: Brouard
425:
426: Merge, before building revised version.
427:
1.146 brouard 428: Revision 1.145 2014/06/10 21:23:15 brouard
429: Summary: Debugging with valgrind
430: Author: Nicolas Brouard
431:
432: Lot of changes in order to output the results with some covariates
433: After the Edimburgh REVES conference 2014, it seems mandatory to
434: improve the code.
435: No more memory valgrind error but a lot has to be done in order to
436: continue the work of splitting the code into subroutines.
437: Also, decodemodel has been improved. Tricode is still not
438: optimal. nbcode should be improved. Documentation has been added in
439: the source code.
440:
1.144 brouard 441: Revision 1.143 2014/01/26 09:45:38 brouard
442: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
443:
444: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
445: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
446:
1.143 brouard 447: Revision 1.142 2014/01/26 03:57:36 brouard
448: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
449:
450: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
451:
1.142 brouard 452: Revision 1.141 2014/01/26 02:42:01 brouard
453: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
454:
1.141 brouard 455: Revision 1.140 2011/09/02 10:37:54 brouard
456: Summary: times.h is ok with mingw32 now.
457:
1.140 brouard 458: Revision 1.139 2010/06/14 07:50:17 brouard
459: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
460: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
461:
1.139 brouard 462: Revision 1.138 2010/04/30 18:19:40 brouard
463: *** empty log message ***
464:
1.138 brouard 465: Revision 1.137 2010/04/29 18:11:38 brouard
466: (Module): Checking covariates for more complex models
467: than V1+V2. A lot of change to be done. Unstable.
468:
1.137 brouard 469: Revision 1.136 2010/04/26 20:30:53 brouard
470: (Module): merging some libgsl code. Fixing computation
471: of likelione (using inter/intrapolation if mle = 0) in order to
472: get same likelihood as if mle=1.
473: Some cleaning of code and comments added.
474:
1.136 brouard 475: Revision 1.135 2009/10/29 15:33:14 brouard
476: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
477:
1.135 brouard 478: Revision 1.134 2009/10/29 13:18:53 brouard
479: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
480:
1.134 brouard 481: Revision 1.133 2009/07/06 10:21:25 brouard
482: just nforces
483:
1.133 brouard 484: Revision 1.132 2009/07/06 08:22:05 brouard
485: Many tings
486:
1.132 brouard 487: Revision 1.131 2009/06/20 16:22:47 brouard
488: Some dimensions resccaled
489:
1.131 brouard 490: Revision 1.130 2009/05/26 06:44:34 brouard
491: (Module): Max Covariate is now set to 20 instead of 8. A
492: lot of cleaning with variables initialized to 0. Trying to make
493: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
494:
1.130 brouard 495: Revision 1.129 2007/08/31 13:49:27 lievre
496: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
497:
1.129 lievre 498: Revision 1.128 2006/06/30 13:02:05 brouard
499: (Module): Clarifications on computing e.j
500:
1.128 brouard 501: Revision 1.127 2006/04/28 18:11:50 brouard
502: (Module): Yes the sum of survivors was wrong since
503: imach-114 because nhstepm was no more computed in the age
504: loop. Now we define nhstepma in the age loop.
505: (Module): In order to speed up (in case of numerous covariates) we
506: compute health expectancies (without variances) in a first step
507: and then all the health expectancies with variances or standard
508: deviation (needs data from the Hessian matrices) which slows the
509: computation.
510: In the future we should be able to stop the program is only health
511: expectancies and graph are needed without standard deviations.
512:
1.127 brouard 513: Revision 1.126 2006/04/28 17:23:28 brouard
514: (Module): Yes the sum of survivors was wrong since
515: imach-114 because nhstepm was no more computed in the age
516: loop. Now we define nhstepma in the age loop.
517: Version 0.98h
518:
1.126 brouard 519: Revision 1.125 2006/04/04 15:20:31 lievre
520: Errors in calculation of health expectancies. Age was not initialized.
521: Forecasting file added.
522:
523: Revision 1.124 2006/03/22 17:13:53 lievre
524: Parameters are printed with %lf instead of %f (more numbers after the comma).
525: The log-likelihood is printed in the log file
526:
527: Revision 1.123 2006/03/20 10:52:43 brouard
528: * imach.c (Module): <title> changed, corresponds to .htm file
529: name. <head> headers where missing.
530:
531: * imach.c (Module): Weights can have a decimal point as for
532: English (a comma might work with a correct LC_NUMERIC environment,
533: otherwise the weight is truncated).
534: Modification of warning when the covariates values are not 0 or
535: 1.
536: Version 0.98g
537:
538: Revision 1.122 2006/03/20 09:45:41 brouard
539: (Module): Weights can have a decimal point as for
540: English (a comma might work with a correct LC_NUMERIC environment,
541: otherwise the weight is truncated).
542: Modification of warning when the covariates values are not 0 or
543: 1.
544: Version 0.98g
545:
546: Revision 1.121 2006/03/16 17:45:01 lievre
547: * imach.c (Module): Comments concerning covariates added
548:
549: * imach.c (Module): refinements in the computation of lli if
550: status=-2 in order to have more reliable computation if stepm is
551: not 1 month. Version 0.98f
552:
553: Revision 1.120 2006/03/16 15:10:38 lievre
554: (Module): refinements in the computation of lli if
555: status=-2 in order to have more reliable computation if stepm is
556: not 1 month. Version 0.98f
557:
558: Revision 1.119 2006/03/15 17:42:26 brouard
559: (Module): Bug if status = -2, the loglikelihood was
560: computed as likelihood omitting the logarithm. Version O.98e
561:
562: Revision 1.118 2006/03/14 18:20:07 brouard
563: (Module): varevsij Comments added explaining the second
564: table of variances if popbased=1 .
565: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
566: (Module): Function pstamp added
567: (Module): Version 0.98d
568:
569: Revision 1.117 2006/03/14 17:16:22 brouard
570: (Module): varevsij Comments added explaining the second
571: table of variances if popbased=1 .
572: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
573: (Module): Function pstamp added
574: (Module): Version 0.98d
575:
576: Revision 1.116 2006/03/06 10:29:27 brouard
577: (Module): Variance-covariance wrong links and
578: varian-covariance of ej. is needed (Saito).
579:
580: Revision 1.115 2006/02/27 12:17:45 brouard
581: (Module): One freematrix added in mlikeli! 0.98c
582:
583: Revision 1.114 2006/02/26 12:57:58 brouard
584: (Module): Some improvements in processing parameter
585: filename with strsep.
586:
587: Revision 1.113 2006/02/24 14:20:24 brouard
588: (Module): Memory leaks checks with valgrind and:
589: datafile was not closed, some imatrix were not freed and on matrix
590: allocation too.
591:
592: Revision 1.112 2006/01/30 09:55:26 brouard
593: (Module): Back to gnuplot.exe instead of wgnuplot.exe
594:
595: Revision 1.111 2006/01/25 20:38:18 brouard
596: (Module): Lots of cleaning and bugs added (Gompertz)
597: (Module): Comments can be added in data file. Missing date values
598: can be a simple dot '.'.
599:
600: Revision 1.110 2006/01/25 00:51:50 brouard
601: (Module): Lots of cleaning and bugs added (Gompertz)
602:
603: Revision 1.109 2006/01/24 19:37:15 brouard
604: (Module): Comments (lines starting with a #) are allowed in data.
605:
606: Revision 1.108 2006/01/19 18:05:42 lievre
607: Gnuplot problem appeared...
608: To be fixed
609:
610: Revision 1.107 2006/01/19 16:20:37 brouard
611: Test existence of gnuplot in imach path
612:
613: Revision 1.106 2006/01/19 13:24:36 brouard
614: Some cleaning and links added in html output
615:
616: Revision 1.105 2006/01/05 20:23:19 lievre
617: *** empty log message ***
618:
619: Revision 1.104 2005/09/30 16:11:43 lievre
620: (Module): sump fixed, loop imx fixed, and simplifications.
621: (Module): If the status is missing at the last wave but we know
622: that the person is alive, then we can code his/her status as -2
623: (instead of missing=-1 in earlier versions) and his/her
624: contributions to the likelihood is 1 - Prob of dying from last
625: health status (= 1-p13= p11+p12 in the easiest case of somebody in
626: the healthy state at last known wave). Version is 0.98
627:
628: Revision 1.103 2005/09/30 15:54:49 lievre
629: (Module): sump fixed, loop imx fixed, and simplifications.
630:
631: Revision 1.102 2004/09/15 17:31:30 brouard
632: Add the possibility to read data file including tab characters.
633:
634: Revision 1.101 2004/09/15 10:38:38 brouard
635: Fix on curr_time
636:
637: Revision 1.100 2004/07/12 18:29:06 brouard
638: Add version for Mac OS X. Just define UNIX in Makefile
639:
640: Revision 1.99 2004/06/05 08:57:40 brouard
641: *** empty log message ***
642:
643: Revision 1.98 2004/05/16 15:05:56 brouard
644: New version 0.97 . First attempt to estimate force of mortality
645: directly from the data i.e. without the need of knowing the health
646: state at each age, but using a Gompertz model: log u =a + b*age .
647: This is the basic analysis of mortality and should be done before any
648: other analysis, in order to test if the mortality estimated from the
649: cross-longitudinal survey is different from the mortality estimated
650: from other sources like vital statistic data.
651:
652: The same imach parameter file can be used but the option for mle should be -3.
653:
1.133 brouard 654: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 655: former routines in order to include the new code within the former code.
656:
657: The output is very simple: only an estimate of the intercept and of
658: the slope with 95% confident intervals.
659:
660: Current limitations:
661: A) Even if you enter covariates, i.e. with the
662: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
663: B) There is no computation of Life Expectancy nor Life Table.
664:
665: Revision 1.97 2004/02/20 13:25:42 lievre
666: Version 0.96d. Population forecasting command line is (temporarily)
667: suppressed.
668:
669: Revision 1.96 2003/07/15 15:38:55 brouard
670: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
671: rewritten within the same printf. Workaround: many printfs.
672:
673: Revision 1.95 2003/07/08 07:54:34 brouard
674: * imach.c (Repository):
675: (Repository): Using imachwizard code to output a more meaningful covariance
676: matrix (cov(a12,c31) instead of numbers.
677:
678: Revision 1.94 2003/06/27 13:00:02 brouard
679: Just cleaning
680:
681: Revision 1.93 2003/06/25 16:33:55 brouard
682: (Module): On windows (cygwin) function asctime_r doesn't
683: exist so I changed back to asctime which exists.
684: (Module): Version 0.96b
685:
686: Revision 1.92 2003/06/25 16:30:45 brouard
687: (Module): On windows (cygwin) function asctime_r doesn't
688: exist so I changed back to asctime which exists.
689:
690: Revision 1.91 2003/06/25 15:30:29 brouard
691: * imach.c (Repository): Duplicated warning errors corrected.
692: (Repository): Elapsed time after each iteration is now output. It
693: helps to forecast when convergence will be reached. Elapsed time
694: is stamped in powell. We created a new html file for the graphs
695: concerning matrix of covariance. It has extension -cov.htm.
696:
697: Revision 1.90 2003/06/24 12:34:15 brouard
698: (Module): Some bugs corrected for windows. Also, when
699: mle=-1 a template is output in file "or"mypar.txt with the design
700: of the covariance matrix to be input.
701:
702: Revision 1.89 2003/06/24 12:30:52 brouard
703: (Module): Some bugs corrected for windows. Also, when
704: mle=-1 a template is output in file "or"mypar.txt with the design
705: of the covariance matrix to be input.
706:
707: Revision 1.88 2003/06/23 17:54:56 brouard
708: * 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.
709:
710: Revision 1.87 2003/06/18 12:26:01 brouard
711: Version 0.96
712:
713: Revision 1.86 2003/06/17 20:04:08 brouard
714: (Module): Change position of html and gnuplot routines and added
715: routine fileappend.
716:
717: Revision 1.85 2003/06/17 13:12:43 brouard
718: * imach.c (Repository): Check when date of death was earlier that
719: current date of interview. It may happen when the death was just
720: prior to the death. In this case, dh was negative and likelihood
721: was wrong (infinity). We still send an "Error" but patch by
722: assuming that the date of death was just one stepm after the
723: interview.
724: (Repository): Because some people have very long ID (first column)
725: we changed int to long in num[] and we added a new lvector for
726: memory allocation. But we also truncated to 8 characters (left
727: truncation)
728: (Repository): No more line truncation errors.
729:
730: Revision 1.84 2003/06/13 21:44:43 brouard
731: * imach.c (Repository): Replace "freqsummary" at a correct
732: place. It differs from routine "prevalence" which may be called
733: many times. Probs is memory consuming and must be used with
734: parcimony.
735: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
736:
737: Revision 1.83 2003/06/10 13:39:11 lievre
738: *** empty log message ***
739:
740: Revision 1.82 2003/06/05 15:57:20 brouard
741: Add log in imach.c and fullversion number is now printed.
742:
743: */
744: /*
745: Interpolated Markov Chain
746:
747: Short summary of the programme:
748:
1.227 brouard 749: This program computes Healthy Life Expectancies or State-specific
750: (if states aren't health statuses) Expectancies from
751: cross-longitudinal data. Cross-longitudinal data consist in:
752:
753: -1- a first survey ("cross") where individuals from different ages
754: are interviewed on their health status or degree of disability (in
755: the case of a health survey which is our main interest)
756:
757: -2- at least a second wave of interviews ("longitudinal") which
758: measure each change (if any) in individual health status. Health
759: expectancies are computed from the time spent in each health state
760: according to a model. More health states you consider, more time is
761: necessary to reach the Maximum Likelihood of the parameters involved
762: in the model. The simplest model is the multinomial logistic model
763: where pij is the probability to be observed in state j at the second
764: wave conditional to be observed in state i at the first
765: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
766: etc , where 'age' is age and 'sex' is a covariate. If you want to
767: have a more complex model than "constant and age", you should modify
768: the program where the markup *Covariates have to be included here
769: again* invites you to do it. More covariates you add, slower the
1.126 brouard 770: convergence.
771:
772: The advantage of this computer programme, compared to a simple
773: multinomial logistic model, is clear when the delay between waves is not
774: identical for each individual. Also, if a individual missed an
775: intermediate interview, the information is lost, but taken into
776: account using an interpolation or extrapolation.
777:
778: hPijx is the probability to be observed in state i at age x+h
779: conditional to the observed state i at age x. The delay 'h' can be
780: split into an exact number (nh*stepm) of unobserved intermediate
781: states. This elementary transition (by month, quarter,
782: semester or year) is modelled as a multinomial logistic. The hPx
783: matrix is simply the matrix product of nh*stepm elementary matrices
784: and the contribution of each individual to the likelihood is simply
785: hPijx.
786:
787: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 788: of the life expectancies. It also computes the period (stable) prevalence.
789:
790: Back prevalence and projections:
1.227 brouard 791:
792: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
793: double agemaxpar, double ftolpl, int *ncvyearp, double
794: dateprev1,double dateprev2, int firstpass, int lastpass, int
795: mobilavproj)
796:
797: Computes the back prevalence limit for any combination of
798: covariate values k at any age between ageminpar and agemaxpar and
799: returns it in **bprlim. In the loops,
800:
801: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
802: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
803:
804: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 805: Computes for any combination of covariates k and any age between bage and fage
806: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
807: oldm=oldms;savm=savms;
1.227 brouard 808:
809: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 810: Computes the transition matrix starting at age 'age' over
811: 'nhstepm*hstepm*stepm' months (i.e. until
812: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 813: nhstepm*hstepm matrices.
814:
815: Returns p3mat[i][j][h] after calling
816: p3mat[i][j][h]=matprod2(newm,
817: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
818: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
819: oldm);
1.226 brouard 820:
821: Important routines
822:
823: - func (or funcone), computes logit (pij) distinguishing
824: o fixed variables (single or product dummies or quantitative);
825: o varying variables by:
826: (1) wave (single, product dummies, quantitative),
827: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
828: % fixed dummy (treated) or quantitative (not done because time-consuming);
829: % varying dummy (not done) or quantitative (not done);
830: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
831: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
832: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
833: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
834: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 835:
1.226 brouard 836:
837:
1.133 brouard 838: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
839: Institut national d'études démographiques, Paris.
1.126 brouard 840: This software have been partly granted by Euro-REVES, a concerted action
841: from the European Union.
842: It is copyrighted identically to a GNU software product, ie programme and
843: software can be distributed freely for non commercial use. Latest version
844: can be accessed at http://euroreves.ined.fr/imach .
845:
846: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
847: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
848:
849: **********************************************************************/
850: /*
851: main
852: read parameterfile
853: read datafile
854: concatwav
855: freqsummary
856: if (mle >= 1)
857: mlikeli
858: print results files
859: if mle==1
860: computes hessian
861: read end of parameter file: agemin, agemax, bage, fage, estepm
862: begin-prev-date,...
863: open gnuplot file
864: open html file
1.145 brouard 865: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
866: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
867: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
868: freexexit2 possible for memory heap.
869:
870: h Pij x | pij_nom ficrestpij
871: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
872: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
873: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
874:
875: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
876: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
877: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
878: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
879: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
880:
1.126 brouard 881: forecasting if prevfcast==1 prevforecast call prevalence()
882: health expectancies
883: Variance-covariance of DFLE
884: prevalence()
885: movingaverage()
886: varevsij()
887: if popbased==1 varevsij(,popbased)
888: total life expectancies
889: Variance of period (stable) prevalence
890: end
891: */
892:
1.187 brouard 893: /* #define DEBUG */
894: /* #define DEBUGBRENT */
1.203 brouard 895: /* #define DEBUGLINMIN */
896: /* #define DEBUGHESS */
897: #define DEBUGHESSIJ
1.224 brouard 898: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 899: #define POWELL /* Instead of NLOPT */
1.224 brouard 900: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 901: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
902: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 903:
904: #include <math.h>
905: #include <stdio.h>
906: #include <stdlib.h>
907: #include <string.h>
1.226 brouard 908: #include <ctype.h>
1.159 brouard 909:
910: #ifdef _WIN32
911: #include <io.h>
1.172 brouard 912: #include <windows.h>
913: #include <tchar.h>
1.159 brouard 914: #else
1.126 brouard 915: #include <unistd.h>
1.159 brouard 916: #endif
1.126 brouard 917:
918: #include <limits.h>
919: #include <sys/types.h>
1.171 brouard 920:
921: #if defined(__GNUC__)
922: #include <sys/utsname.h> /* Doesn't work on Windows */
923: #endif
924:
1.126 brouard 925: #include <sys/stat.h>
926: #include <errno.h>
1.159 brouard 927: /* extern int errno; */
1.126 brouard 928:
1.157 brouard 929: /* #ifdef LINUX */
930: /* #include <time.h> */
931: /* #include "timeval.h" */
932: /* #else */
933: /* #include <sys/time.h> */
934: /* #endif */
935:
1.126 brouard 936: #include <time.h>
937:
1.136 brouard 938: #ifdef GSL
939: #include <gsl/gsl_errno.h>
940: #include <gsl/gsl_multimin.h>
941: #endif
942:
1.167 brouard 943:
1.162 brouard 944: #ifdef NLOPT
945: #include <nlopt.h>
946: typedef struct {
947: double (* function)(double [] );
948: } myfunc_data ;
949: #endif
950:
1.126 brouard 951: /* #include <libintl.h> */
952: /* #define _(String) gettext (String) */
953:
1.251 brouard 954: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 955:
956: #define GNUPLOTPROGRAM "gnuplot"
957: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
958: #define FILENAMELENGTH 132
959:
960: #define GLOCK_ERROR_NOPATH -1 /* empty path */
961: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
962:
1.144 brouard 963: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
964: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 965:
966: #define NINTERVMAX 8
1.144 brouard 967: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
968: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
969: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 970: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 971: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
972: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 973: #define MAXN 20000
1.144 brouard 974: #define YEARM 12. /**< Number of months per year */
1.218 brouard 975: /* #define AGESUP 130 */
976: #define AGESUP 150
977: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 978: #define AGEBASE 40
1.194 brouard 979: #define AGEOVERFLOW 1.e20
1.164 brouard 980: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 981: #ifdef _WIN32
982: #define DIRSEPARATOR '\\'
983: #define CHARSEPARATOR "\\"
984: #define ODIRSEPARATOR '/'
985: #else
1.126 brouard 986: #define DIRSEPARATOR '/'
987: #define CHARSEPARATOR "/"
988: #define ODIRSEPARATOR '\\'
989: #endif
990:
1.263 ! brouard 991: /* $Id: imach.c,v 1.262 2017/04/18 16:48:12 brouard Exp $ */
1.126 brouard 992: /* $State: Exp $ */
1.196 brouard 993: #include "version.h"
994: char version[]=__IMACH_VERSION__;
1.224 brouard 995: 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";
1.263 ! brouard 996: char fullversion[]="$Revision: 1.262 $ $Date: 2017/04/18 16:48:12 $";
1.126 brouard 997: char strstart[80];
998: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 999: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1000: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1001: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1002: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1003: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1004: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1005: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1006: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1007: int cptcovprodnoage=0; /**< Number of covariate products without age */
1008: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1009: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1010: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1011: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1012: int nsd=0; /**< Total number of single dummy variables (output) */
1013: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1014: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1015: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1016: int ntveff=0; /**< ntveff number of effective time varying variables */
1017: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1018: int cptcov=0; /* Working variable */
1.218 brouard 1019: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1020: int npar=NPARMAX;
1021: int nlstate=2; /* Number of live states */
1022: int ndeath=1; /* Number of dead states */
1.130 brouard 1023: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1024: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1025: int popbased=0;
1026:
1027: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1028: int maxwav=0; /* Maxim number of waves */
1029: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1030: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1031: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1032: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1033: int mle=1, weightopt=0;
1.126 brouard 1034: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1035: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1036: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1037: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1038: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1039: int selected(int kvar); /* Is covariate kvar selected for printing results */
1040:
1.130 brouard 1041: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1042: double **matprod2(); /* test */
1.126 brouard 1043: double **oldm, **newm, **savm; /* Working pointers to matrices */
1044: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1045: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1046:
1.136 brouard 1047: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1048: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1049: FILE *ficlog, *ficrespow;
1.130 brouard 1050: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1051: double fretone; /* Only one call to likelihood */
1.130 brouard 1052: long ipmx=0; /* Number of contributions */
1.126 brouard 1053: double sw; /* Sum of weights */
1054: char filerespow[FILENAMELENGTH];
1055: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1056: FILE *ficresilk;
1057: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1058: FILE *ficresprobmorprev;
1059: FILE *fichtm, *fichtmcov; /* Html File */
1060: FILE *ficreseij;
1061: char filerese[FILENAMELENGTH];
1062: FILE *ficresstdeij;
1063: char fileresstde[FILENAMELENGTH];
1064: FILE *ficrescveij;
1065: char filerescve[FILENAMELENGTH];
1066: FILE *ficresvij;
1067: char fileresv[FILENAMELENGTH];
1068: FILE *ficresvpl;
1069: char fileresvpl[FILENAMELENGTH];
1070: char title[MAXLINE];
1.234 brouard 1071: char model[MAXLINE]; /**< The model line */
1.217 brouard 1072: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1073: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1074: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1075: char command[FILENAMELENGTH];
1076: int outcmd=0;
1077:
1.217 brouard 1078: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1079: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1080: char filelog[FILENAMELENGTH]; /* Log file */
1081: char filerest[FILENAMELENGTH];
1082: char fileregp[FILENAMELENGTH];
1083: char popfile[FILENAMELENGTH];
1084:
1085: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1086:
1.157 brouard 1087: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1088: /* struct timezone tzp; */
1089: /* extern int gettimeofday(); */
1090: struct tm tml, *gmtime(), *localtime();
1091:
1092: extern time_t time();
1093:
1094: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1095: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1096: struct tm tm;
1097:
1.126 brouard 1098: char strcurr[80], strfor[80];
1099:
1100: char *endptr;
1101: long lval;
1102: double dval;
1103:
1104: #define NR_END 1
1105: #define FREE_ARG char*
1106: #define FTOL 1.0e-10
1107:
1108: #define NRANSI
1.240 brouard 1109: #define ITMAX 200
1110: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1111:
1112: #define TOL 2.0e-4
1113:
1114: #define CGOLD 0.3819660
1115: #define ZEPS 1.0e-10
1116: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1117:
1118: #define GOLD 1.618034
1119: #define GLIMIT 100.0
1120: #define TINY 1.0e-20
1121:
1122: static double maxarg1,maxarg2;
1123: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1124: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1125:
1126: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1127: #define rint(a) floor(a+0.5)
1.166 brouard 1128: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1129: #define mytinydouble 1.0e-16
1.166 brouard 1130: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1131: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1132: /* static double dsqrarg; */
1133: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1134: static double sqrarg;
1135: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1136: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1137: int agegomp= AGEGOMP;
1138:
1139: int imx;
1140: int stepm=1;
1141: /* Stepm, step in month: minimum step interpolation*/
1142:
1143: int estepm;
1144: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1145:
1146: int m,nb;
1147: long *num;
1.197 brouard 1148: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1149: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1150: covariate for which somebody answered excluding
1151: undefined. Usually 2: 0 and 1. */
1152: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1153: covariate for which somebody answered including
1154: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1155: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1156: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1157: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1158: double *ageexmed,*agecens;
1159: double dateintmean=0;
1160:
1161: double *weight;
1162: int **s; /* Status */
1.141 brouard 1163: double *agedc;
1.145 brouard 1164: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1165: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1166: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1167: double **coqvar; /* Fixed quantitative covariate iqv */
1168: double ***cotvar; /* Time varying covariate itv */
1169: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1170: double idx;
1171: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1172: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1173: /*k 1 2 3 4 5 6 7 8 9 */
1174: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1175: /* Tndvar[k] 1 2 3 4 5 */
1176: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1177: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1178: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1179: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1180: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1181: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1182: /* Tprod[i]=k 4 7 */
1183: /* Tage[i]=k 5 8 */
1184: /* */
1185: /* Type */
1186: /* V 1 2 3 4 5 */
1187: /* F F V V V */
1188: /* D Q D D Q */
1189: /* */
1190: int *TvarsD;
1191: int *TvarsDind;
1192: int *TvarsQ;
1193: int *TvarsQind;
1194:
1.235 brouard 1195: #define MAXRESULTLINES 10
1196: int nresult=0;
1.258 brouard 1197: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1198: int TKresult[MAXRESULTLINES];
1.237 brouard 1199: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1200: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1201: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1202: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1203: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1204: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1205:
1.234 brouard 1206: /* 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 1207: 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 */
1208: 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 */
1209: 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 */
1210: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1211: 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 */
1212: 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 1213: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1214: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1215: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1216: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1217: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1218: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1219: 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 */
1220: 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 */
1221:
1.230 brouard 1222: int *Tvarsel; /**< Selected covariates for output */
1223: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1224: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1225: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1226: 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 1227: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1228: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1229: int *Tage;
1.227 brouard 1230: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1231: 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 1232: 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*/
1233: 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 1234: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1235: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1236: int **Tvard;
1237: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1238: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1239: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1240: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1241: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1242: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1243: double *lsurv, *lpop, *tpop;
1244:
1.231 brouard 1245: #define FD 1; /* Fixed dummy covariate */
1246: #define FQ 2; /* Fixed quantitative covariate */
1247: #define FP 3; /* Fixed product covariate */
1248: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1249: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1250: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1251: #define VD 10; /* Varying dummy covariate */
1252: #define VQ 11; /* Varying quantitative covariate */
1253: #define VP 12; /* Varying product covariate */
1254: #define VPDD 13; /* Varying product dummy*dummy covariate */
1255: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1256: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1257: #define APFD 16; /* Age product * fixed dummy covariate */
1258: #define APFQ 17; /* Age product * fixed quantitative covariate */
1259: #define APVD 18; /* Age product * varying dummy covariate */
1260: #define APVQ 19; /* Age product * varying quantitative covariate */
1261:
1262: #define FTYPE 1; /* Fixed covariate */
1263: #define VTYPE 2; /* Varying covariate (loop in wave) */
1264: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1265:
1266: struct kmodel{
1267: int maintype; /* main type */
1268: int subtype; /* subtype */
1269: };
1270: struct kmodel modell[NCOVMAX];
1271:
1.143 brouard 1272: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1273: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1274:
1275: /**************** split *************************/
1276: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1277: {
1278: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1279: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1280: */
1281: char *ss; /* pointer */
1.186 brouard 1282: int l1=0, l2=0; /* length counters */
1.126 brouard 1283:
1284: l1 = strlen(path ); /* length of path */
1285: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1286: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1287: if ( ss == NULL ) { /* no directory, so determine current directory */
1288: strcpy( name, path ); /* we got the fullname name because no directory */
1289: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1290: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1291: /* get current working directory */
1292: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1293: #ifdef WIN32
1294: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1295: #else
1296: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1297: #endif
1.126 brouard 1298: return( GLOCK_ERROR_GETCWD );
1299: }
1300: /* got dirc from getcwd*/
1301: printf(" DIRC = %s \n",dirc);
1.205 brouard 1302: } else { /* strip directory from path */
1.126 brouard 1303: ss++; /* after this, the filename */
1304: l2 = strlen( ss ); /* length of filename */
1305: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1306: strcpy( name, ss ); /* save file name */
1307: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1308: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1309: printf(" DIRC2 = %s \n",dirc);
1310: }
1311: /* We add a separator at the end of dirc if not exists */
1312: l1 = strlen( dirc ); /* length of directory */
1313: if( dirc[l1-1] != DIRSEPARATOR ){
1314: dirc[l1] = DIRSEPARATOR;
1315: dirc[l1+1] = 0;
1316: printf(" DIRC3 = %s \n",dirc);
1317: }
1318: ss = strrchr( name, '.' ); /* find last / */
1319: if (ss >0){
1320: ss++;
1321: strcpy(ext,ss); /* save extension */
1322: l1= strlen( name);
1323: l2= strlen(ss)+1;
1324: strncpy( finame, name, l1-l2);
1325: finame[l1-l2]= 0;
1326: }
1327:
1328: return( 0 ); /* we're done */
1329: }
1330:
1331:
1332: /******************************************/
1333:
1334: void replace_back_to_slash(char *s, char*t)
1335: {
1336: int i;
1337: int lg=0;
1338: i=0;
1339: lg=strlen(t);
1340: for(i=0; i<= lg; i++) {
1341: (s[i] = t[i]);
1342: if (t[i]== '\\') s[i]='/';
1343: }
1344: }
1345:
1.132 brouard 1346: char *trimbb(char *out, char *in)
1.137 brouard 1347: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1348: char *s;
1349: s=out;
1350: while (*in != '\0'){
1.137 brouard 1351: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1352: in++;
1353: }
1354: *out++ = *in++;
1355: }
1356: *out='\0';
1357: return s;
1358: }
1359:
1.187 brouard 1360: /* char *substrchaine(char *out, char *in, char *chain) */
1361: /* { */
1362: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1363: /* char *s, *t; */
1364: /* t=in;s=out; */
1365: /* while ((*in != *chain) && (*in != '\0')){ */
1366: /* *out++ = *in++; */
1367: /* } */
1368:
1369: /* /\* *in matches *chain *\/ */
1370: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1371: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1372: /* } */
1373: /* in--; chain--; */
1374: /* while ( (*in != '\0')){ */
1375: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1376: /* *out++ = *in++; */
1377: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1378: /* } */
1379: /* *out='\0'; */
1380: /* out=s; */
1381: /* return out; */
1382: /* } */
1383: char *substrchaine(char *out, char *in, char *chain)
1384: {
1385: /* Substract chain 'chain' from 'in', return and output 'out' */
1386: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1387:
1388: char *strloc;
1389:
1390: strcpy (out, in);
1391: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1392: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1393: if(strloc != NULL){
1394: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1395: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1396: /* strcpy (strloc, strloc +strlen(chain));*/
1397: }
1398: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1399: return out;
1400: }
1401:
1402:
1.145 brouard 1403: char *cutl(char *blocc, char *alocc, char *in, char occ)
1404: {
1.187 brouard 1405: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1406: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1407: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1408: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1409: */
1.160 brouard 1410: char *s, *t;
1.145 brouard 1411: t=in;s=in;
1412: while ((*in != occ) && (*in != '\0')){
1413: *alocc++ = *in++;
1414: }
1415: if( *in == occ){
1416: *(alocc)='\0';
1417: s=++in;
1418: }
1419:
1420: if (s == t) {/* occ not found */
1421: *(alocc-(in-s))='\0';
1422: in=s;
1423: }
1424: while ( *in != '\0'){
1425: *blocc++ = *in++;
1426: }
1427:
1428: *blocc='\0';
1429: return t;
1430: }
1.137 brouard 1431: char *cutv(char *blocc, char *alocc, char *in, char occ)
1432: {
1.187 brouard 1433: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1434: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1435: gives blocc="abcdef2ghi" and alocc="j".
1436: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1437: */
1438: char *s, *t;
1439: t=in;s=in;
1440: while (*in != '\0'){
1441: while( *in == occ){
1442: *blocc++ = *in++;
1443: s=in;
1444: }
1445: *blocc++ = *in++;
1446: }
1447: if (s == t) /* occ not found */
1448: *(blocc-(in-s))='\0';
1449: else
1450: *(blocc-(in-s)-1)='\0';
1451: in=s;
1452: while ( *in != '\0'){
1453: *alocc++ = *in++;
1454: }
1455:
1456: *alocc='\0';
1457: return s;
1458: }
1459:
1.126 brouard 1460: int nbocc(char *s, char occ)
1461: {
1462: int i,j=0;
1463: int lg=20;
1464: i=0;
1465: lg=strlen(s);
1466: for(i=0; i<= lg; i++) {
1.234 brouard 1467: if (s[i] == occ ) j++;
1.126 brouard 1468: }
1469: return j;
1470: }
1471:
1.137 brouard 1472: /* void cutv(char *u,char *v, char*t, char occ) */
1473: /* { */
1474: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1475: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1476: /* gives u="abcdef2ghi" and v="j" *\/ */
1477: /* int i,lg,j,p=0; */
1478: /* i=0; */
1479: /* lg=strlen(t); */
1480: /* for(j=0; j<=lg-1; j++) { */
1481: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1482: /* } */
1.126 brouard 1483:
1.137 brouard 1484: /* for(j=0; j<p; j++) { */
1485: /* (u[j] = t[j]); */
1486: /* } */
1487: /* u[p]='\0'; */
1.126 brouard 1488:
1.137 brouard 1489: /* for(j=0; j<= lg; j++) { */
1490: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1491: /* } */
1492: /* } */
1.126 brouard 1493:
1.160 brouard 1494: #ifdef _WIN32
1495: char * strsep(char **pp, const char *delim)
1496: {
1497: char *p, *q;
1498:
1499: if ((p = *pp) == NULL)
1500: return 0;
1501: if ((q = strpbrk (p, delim)) != NULL)
1502: {
1503: *pp = q + 1;
1504: *q = '\0';
1505: }
1506: else
1507: *pp = 0;
1508: return p;
1509: }
1510: #endif
1511:
1.126 brouard 1512: /********************** nrerror ********************/
1513:
1514: void nrerror(char error_text[])
1515: {
1516: fprintf(stderr,"ERREUR ...\n");
1517: fprintf(stderr,"%s\n",error_text);
1518: exit(EXIT_FAILURE);
1519: }
1520: /*********************** vector *******************/
1521: double *vector(int nl, int nh)
1522: {
1523: double *v;
1524: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1525: if (!v) nrerror("allocation failure in vector");
1526: return v-nl+NR_END;
1527: }
1528:
1529: /************************ free vector ******************/
1530: void free_vector(double*v, int nl, int nh)
1531: {
1532: free((FREE_ARG)(v+nl-NR_END));
1533: }
1534:
1535: /************************ivector *******************************/
1536: int *ivector(long nl,long nh)
1537: {
1538: int *v;
1539: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1540: if (!v) nrerror("allocation failure in ivector");
1541: return v-nl+NR_END;
1542: }
1543:
1544: /******************free ivector **************************/
1545: void free_ivector(int *v, long nl, long nh)
1546: {
1547: free((FREE_ARG)(v+nl-NR_END));
1548: }
1549:
1550: /************************lvector *******************************/
1551: long *lvector(long nl,long nh)
1552: {
1553: long *v;
1554: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1555: if (!v) nrerror("allocation failure in ivector");
1556: return v-nl+NR_END;
1557: }
1558:
1559: /******************free lvector **************************/
1560: void free_lvector(long *v, long nl, long nh)
1561: {
1562: free((FREE_ARG)(v+nl-NR_END));
1563: }
1564:
1565: /******************* imatrix *******************************/
1566: int **imatrix(long nrl, long nrh, long ncl, long nch)
1567: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1568: {
1569: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1570: int **m;
1571:
1572: /* allocate pointers to rows */
1573: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1574: if (!m) nrerror("allocation failure 1 in matrix()");
1575: m += NR_END;
1576: m -= nrl;
1577:
1578:
1579: /* allocate rows and set pointers to them */
1580: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1581: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1582: m[nrl] += NR_END;
1583: m[nrl] -= ncl;
1584:
1585: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1586:
1587: /* return pointer to array of pointers to rows */
1588: return m;
1589: }
1590:
1591: /****************** free_imatrix *************************/
1592: void free_imatrix(m,nrl,nrh,ncl,nch)
1593: int **m;
1594: long nch,ncl,nrh,nrl;
1595: /* free an int matrix allocated by imatrix() */
1596: {
1597: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1598: free((FREE_ARG) (m+nrl-NR_END));
1599: }
1600:
1601: /******************* matrix *******************************/
1602: double **matrix(long nrl, long nrh, long ncl, long nch)
1603: {
1604: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1605: double **m;
1606:
1607: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1608: if (!m) nrerror("allocation failure 1 in matrix()");
1609: m += NR_END;
1610: m -= nrl;
1611:
1612: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1613: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1614: m[nrl] += NR_END;
1615: m[nrl] -= ncl;
1616:
1617: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1618: return m;
1.145 brouard 1619: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1620: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1621: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1622: */
1623: }
1624:
1625: /*************************free matrix ************************/
1626: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1627: {
1628: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1629: free((FREE_ARG)(m+nrl-NR_END));
1630: }
1631:
1632: /******************* ma3x *******************************/
1633: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1634: {
1635: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1636: double ***m;
1637:
1638: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1639: if (!m) nrerror("allocation failure 1 in matrix()");
1640: m += NR_END;
1641: m -= nrl;
1642:
1643: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1644: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1645: m[nrl] += NR_END;
1646: m[nrl] -= ncl;
1647:
1648: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1649:
1650: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1651: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1652: m[nrl][ncl] += NR_END;
1653: m[nrl][ncl] -= nll;
1654: for (j=ncl+1; j<=nch; j++)
1655: m[nrl][j]=m[nrl][j-1]+nlay;
1656:
1657: for (i=nrl+1; i<=nrh; i++) {
1658: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1659: for (j=ncl+1; j<=nch; j++)
1660: m[i][j]=m[i][j-1]+nlay;
1661: }
1662: return m;
1663: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1664: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1665: */
1666: }
1667:
1668: /*************************free ma3x ************************/
1669: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1670: {
1671: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1672: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1673: free((FREE_ARG)(m+nrl-NR_END));
1674: }
1675:
1676: /*************** function subdirf ***********/
1677: char *subdirf(char fileres[])
1678: {
1679: /* Caution optionfilefiname is hidden */
1680: strcpy(tmpout,optionfilefiname);
1681: strcat(tmpout,"/"); /* Add to the right */
1682: strcat(tmpout,fileres);
1683: return tmpout;
1684: }
1685:
1686: /*************** function subdirf2 ***********/
1687: char *subdirf2(char fileres[], char *preop)
1688: {
1689:
1690: /* Caution optionfilefiname is hidden */
1691: strcpy(tmpout,optionfilefiname);
1692: strcat(tmpout,"/");
1693: strcat(tmpout,preop);
1694: strcat(tmpout,fileres);
1695: return tmpout;
1696: }
1697:
1698: /*************** function subdirf3 ***********/
1699: char *subdirf3(char fileres[], char *preop, char *preop2)
1700: {
1701:
1702: /* Caution optionfilefiname is hidden */
1703: strcpy(tmpout,optionfilefiname);
1704: strcat(tmpout,"/");
1705: strcat(tmpout,preop);
1706: strcat(tmpout,preop2);
1707: strcat(tmpout,fileres);
1708: return tmpout;
1709: }
1.213 brouard 1710:
1711: /*************** function subdirfext ***********/
1712: char *subdirfext(char fileres[], char *preop, char *postop)
1713: {
1714:
1715: strcpy(tmpout,preop);
1716: strcat(tmpout,fileres);
1717: strcat(tmpout,postop);
1718: return tmpout;
1719: }
1.126 brouard 1720:
1.213 brouard 1721: /*************** function subdirfext3 ***********/
1722: char *subdirfext3(char fileres[], char *preop, char *postop)
1723: {
1724:
1725: /* Caution optionfilefiname is hidden */
1726: strcpy(tmpout,optionfilefiname);
1727: strcat(tmpout,"/");
1728: strcat(tmpout,preop);
1729: strcat(tmpout,fileres);
1730: strcat(tmpout,postop);
1731: return tmpout;
1732: }
1733:
1.162 brouard 1734: char *asc_diff_time(long time_sec, char ascdiff[])
1735: {
1736: long sec_left, days, hours, minutes;
1737: days = (time_sec) / (60*60*24);
1738: sec_left = (time_sec) % (60*60*24);
1739: hours = (sec_left) / (60*60) ;
1740: sec_left = (sec_left) %(60*60);
1741: minutes = (sec_left) /60;
1742: sec_left = (sec_left) % (60);
1743: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1744: return ascdiff;
1745: }
1746:
1.126 brouard 1747: /***************** f1dim *************************/
1748: extern int ncom;
1749: extern double *pcom,*xicom;
1750: extern double (*nrfunc)(double []);
1751:
1752: double f1dim(double x)
1753: {
1754: int j;
1755: double f;
1756: double *xt;
1757:
1758: xt=vector(1,ncom);
1759: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1760: f=(*nrfunc)(xt);
1761: free_vector(xt,1,ncom);
1762: return f;
1763: }
1764:
1765: /*****************brent *************************/
1766: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1767: {
1768: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1769: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1770: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1771: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1772: * returned function value.
1773: */
1.126 brouard 1774: int iter;
1775: double a,b,d,etemp;
1.159 brouard 1776: double fu=0,fv,fw,fx;
1.164 brouard 1777: double ftemp=0.;
1.126 brouard 1778: double p,q,r,tol1,tol2,u,v,w,x,xm;
1779: double e=0.0;
1780:
1781: a=(ax < cx ? ax : cx);
1782: b=(ax > cx ? ax : cx);
1783: x=w=v=bx;
1784: fw=fv=fx=(*f)(x);
1785: for (iter=1;iter<=ITMAX;iter++) {
1786: xm=0.5*(a+b);
1787: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1788: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1789: printf(".");fflush(stdout);
1790: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1791: #ifdef DEBUGBRENT
1.126 brouard 1792: 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);
1793: 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);
1794: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1795: #endif
1796: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1797: *xmin=x;
1798: return fx;
1799: }
1800: ftemp=fu;
1801: if (fabs(e) > tol1) {
1802: r=(x-w)*(fx-fv);
1803: q=(x-v)*(fx-fw);
1804: p=(x-v)*q-(x-w)*r;
1805: q=2.0*(q-r);
1806: if (q > 0.0) p = -p;
1807: q=fabs(q);
1808: etemp=e;
1809: e=d;
1810: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1811: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1812: else {
1.224 brouard 1813: d=p/q;
1814: u=x+d;
1815: if (u-a < tol2 || b-u < tol2)
1816: d=SIGN(tol1,xm-x);
1.126 brouard 1817: }
1818: } else {
1819: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1820: }
1821: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1822: fu=(*f)(u);
1823: if (fu <= fx) {
1824: if (u >= x) a=x; else b=x;
1825: SHFT(v,w,x,u)
1.183 brouard 1826: SHFT(fv,fw,fx,fu)
1827: } else {
1828: if (u < x) a=u; else b=u;
1829: if (fu <= fw || w == x) {
1.224 brouard 1830: v=w;
1831: w=u;
1832: fv=fw;
1833: fw=fu;
1.183 brouard 1834: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1835: v=u;
1836: fv=fu;
1.183 brouard 1837: }
1838: }
1.126 brouard 1839: }
1840: nrerror("Too many iterations in brent");
1841: *xmin=x;
1842: return fx;
1843: }
1844:
1845: /****************** mnbrak ***********************/
1846:
1847: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1848: double (*func)(double))
1.183 brouard 1849: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1850: the downhill direction (defined by the function as evaluated at the initial points) and returns
1851: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1852: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1853: */
1.126 brouard 1854: double ulim,u,r,q, dum;
1855: double fu;
1.187 brouard 1856:
1857: double scale=10.;
1858: int iterscale=0;
1859:
1860: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1861: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1862:
1863:
1864: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1865: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1866: /* *bx = *ax - (*ax - *bx)/scale; */
1867: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1868: /* } */
1869:
1.126 brouard 1870: if (*fb > *fa) {
1871: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1872: SHFT(dum,*fb,*fa,dum)
1873: }
1.126 brouard 1874: *cx=(*bx)+GOLD*(*bx-*ax);
1875: *fc=(*func)(*cx);
1.183 brouard 1876: #ifdef DEBUG
1.224 brouard 1877: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1878: 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 1879: #endif
1.224 brouard 1880: 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 1881: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1882: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1883: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1884: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1885: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1886: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1887: fu=(*func)(u);
1.163 brouard 1888: #ifdef DEBUG
1889: /* f(x)=A(x-u)**2+f(u) */
1890: double A, fparabu;
1891: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1892: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1893: 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);
1894: 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 1895: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1896: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1897: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1898: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1899: #endif
1.184 brouard 1900: #ifdef MNBRAKORIGINAL
1.183 brouard 1901: #else
1.191 brouard 1902: /* if (fu > *fc) { */
1903: /* #ifdef DEBUG */
1904: /* printf("mnbrak4 fu > fc \n"); */
1905: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1906: /* #endif */
1907: /* /\* 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 *\\/ *\/ */
1908: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1909: /* dum=u; /\* Shifting c and u *\/ */
1910: /* u = *cx; */
1911: /* *cx = dum; */
1912: /* dum = fu; */
1913: /* fu = *fc; */
1914: /* *fc =dum; */
1915: /* } else { /\* end *\/ */
1916: /* #ifdef DEBUG */
1917: /* printf("mnbrak3 fu < fc \n"); */
1918: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1919: /* #endif */
1920: /* dum=u; /\* Shifting c and u *\/ */
1921: /* u = *cx; */
1922: /* *cx = dum; */
1923: /* dum = fu; */
1924: /* fu = *fc; */
1925: /* *fc =dum; */
1926: /* } */
1.224 brouard 1927: #ifdef DEBUGMNBRAK
1928: double A, fparabu;
1929: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1930: fparabu= *fa - A*(*ax-u)*(*ax-u);
1931: 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);
1932: 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 1933: #endif
1.191 brouard 1934: dum=u; /* Shifting c and u */
1935: u = *cx;
1936: *cx = dum;
1937: dum = fu;
1938: fu = *fc;
1939: *fc =dum;
1.183 brouard 1940: #endif
1.162 brouard 1941: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1942: #ifdef DEBUG
1.224 brouard 1943: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1944: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1945: #endif
1.126 brouard 1946: fu=(*func)(u);
1947: if (fu < *fc) {
1.183 brouard 1948: #ifdef DEBUG
1.224 brouard 1949: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1950: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1951: #endif
1952: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1953: SHFT(*fb,*fc,fu,(*func)(u))
1954: #ifdef DEBUG
1955: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1956: #endif
1957: }
1.162 brouard 1958: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1959: #ifdef DEBUG
1.224 brouard 1960: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1961: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1962: #endif
1.126 brouard 1963: u=ulim;
1964: fu=(*func)(u);
1.183 brouard 1965: } else { /* u could be left to b (if r > q parabola has a maximum) */
1966: #ifdef DEBUG
1.224 brouard 1967: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1968: 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 1969: #endif
1.126 brouard 1970: u=(*cx)+GOLD*(*cx-*bx);
1971: fu=(*func)(u);
1.224 brouard 1972: #ifdef DEBUG
1973: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1974: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1975: #endif
1.183 brouard 1976: } /* end tests */
1.126 brouard 1977: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1978: SHFT(*fa,*fb,*fc,fu)
1979: #ifdef DEBUG
1.224 brouard 1980: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1981: 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 1982: #endif
1983: } /* 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 1984: }
1985:
1986: /*************** linmin ************************/
1.162 brouard 1987: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1988: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1989: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1990: the value of func at the returned location p . This is actually all accomplished by calling the
1991: routines mnbrak and brent .*/
1.126 brouard 1992: int ncom;
1993: double *pcom,*xicom;
1994: double (*nrfunc)(double []);
1995:
1.224 brouard 1996: #ifdef LINMINORIGINAL
1.126 brouard 1997: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1998: #else
1999: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2000: #endif
1.126 brouard 2001: {
2002: double brent(double ax, double bx, double cx,
2003: double (*f)(double), double tol, double *xmin);
2004: double f1dim(double x);
2005: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2006: double *fc, double (*func)(double));
2007: int j;
2008: double xx,xmin,bx,ax;
2009: double fx,fb,fa;
1.187 brouard 2010:
1.203 brouard 2011: #ifdef LINMINORIGINAL
2012: #else
2013: double scale=10., axs, xxs; /* Scale added for infinity */
2014: #endif
2015:
1.126 brouard 2016: ncom=n;
2017: pcom=vector(1,n);
2018: xicom=vector(1,n);
2019: nrfunc=func;
2020: for (j=1;j<=n;j++) {
2021: pcom[j]=p[j];
1.202 brouard 2022: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2023: }
1.187 brouard 2024:
1.203 brouard 2025: #ifdef LINMINORIGINAL
2026: xx=1.;
2027: #else
2028: axs=0.0;
2029: xxs=1.;
2030: do{
2031: xx= xxs;
2032: #endif
1.187 brouard 2033: ax=0.;
2034: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2035: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2036: /* 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)) */
2037: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2038: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2039: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2040: /* 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 2041: #ifdef LINMINORIGINAL
2042: #else
2043: if (fx != fx){
1.224 brouard 2044: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2045: printf("|");
2046: fprintf(ficlog,"|");
1.203 brouard 2047: #ifdef DEBUGLINMIN
1.224 brouard 2048: 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 2049: #endif
2050: }
1.224 brouard 2051: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2052: #endif
2053:
1.191 brouard 2054: #ifdef DEBUGLINMIN
2055: 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 2056: 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 2057: #endif
1.224 brouard 2058: #ifdef LINMINORIGINAL
2059: #else
2060: if(fb == fx){ /* Flat function in the direction */
2061: xmin=xx;
2062: *flat=1;
2063: }else{
2064: *flat=0;
2065: #endif
2066: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2067: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2068: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2069: /* fmin = f(p[j] + xmin * xi[j]) */
2070: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2071: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2072: #ifdef DEBUG
1.224 brouard 2073: 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);
2074: 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);
2075: #endif
2076: #ifdef LINMINORIGINAL
2077: #else
2078: }
1.126 brouard 2079: #endif
1.191 brouard 2080: #ifdef DEBUGLINMIN
2081: printf("linmin end ");
1.202 brouard 2082: fprintf(ficlog,"linmin end ");
1.191 brouard 2083: #endif
1.126 brouard 2084: for (j=1;j<=n;j++) {
1.203 brouard 2085: #ifdef LINMINORIGINAL
2086: xi[j] *= xmin;
2087: #else
2088: #ifdef DEBUGLINMIN
2089: if(xxs <1.0)
2090: printf(" before xi[%d]=%12.8f", j,xi[j]);
2091: #endif
2092: 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) */
2093: #ifdef DEBUGLINMIN
2094: if(xxs <1.0)
2095: 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 );
2096: #endif
2097: #endif
1.187 brouard 2098: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2099: }
1.191 brouard 2100: #ifdef DEBUGLINMIN
1.203 brouard 2101: printf("\n");
1.191 brouard 2102: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2103: 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 2104: for (j=1;j<=n;j++) {
1.202 brouard 2105: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2106: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2107: if(j % ncovmodel == 0){
1.191 brouard 2108: printf("\n");
1.202 brouard 2109: fprintf(ficlog,"\n");
2110: }
1.191 brouard 2111: }
1.203 brouard 2112: #else
1.191 brouard 2113: #endif
1.126 brouard 2114: free_vector(xicom,1,n);
2115: free_vector(pcom,1,n);
2116: }
2117:
2118:
2119: /*************** powell ************************/
1.162 brouard 2120: /*
2121: Minimization of a function func of n variables. Input consists of an initial starting point
2122: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2123: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2124: such that failure to decrease by more than this amount on one iteration signals doneness. On
2125: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2126: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2127: */
1.224 brouard 2128: #ifdef LINMINORIGINAL
2129: #else
2130: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2131: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2132: #endif
1.126 brouard 2133: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2134: double (*func)(double []))
2135: {
1.224 brouard 2136: #ifdef LINMINORIGINAL
2137: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2138: double (*func)(double []));
1.224 brouard 2139: #else
1.241 brouard 2140: void linmin(double p[], double xi[], int n, double *fret,
2141: double (*func)(double []),int *flat);
1.224 brouard 2142: #endif
1.239 brouard 2143: int i,ibig,j,jk,k;
1.126 brouard 2144: double del,t,*pt,*ptt,*xit;
1.181 brouard 2145: double directest;
1.126 brouard 2146: double fp,fptt;
2147: double *xits;
2148: int niterf, itmp;
1.224 brouard 2149: #ifdef LINMINORIGINAL
2150: #else
2151:
2152: flatdir=ivector(1,n);
2153: for (j=1;j<=n;j++) flatdir[j]=0;
2154: #endif
1.126 brouard 2155:
2156: pt=vector(1,n);
2157: ptt=vector(1,n);
2158: xit=vector(1,n);
2159: xits=vector(1,n);
2160: *fret=(*func)(p);
2161: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2162: rcurr_time = time(NULL);
1.126 brouard 2163: for (*iter=1;;++(*iter)) {
1.187 brouard 2164: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2165: ibig=0;
2166: del=0.0;
1.157 brouard 2167: rlast_time=rcurr_time;
2168: /* (void) gettimeofday(&curr_time,&tzp); */
2169: rcurr_time = time(NULL);
2170: curr_time = *localtime(&rcurr_time);
2171: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2172: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2173: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2174: for (i=1;i<=n;i++) {
1.126 brouard 2175: fprintf(ficrespow," %.12lf", p[i]);
2176: }
1.239 brouard 2177: fprintf(ficrespow,"\n");fflush(ficrespow);
2178: printf("\n#model= 1 + age ");
2179: fprintf(ficlog,"\n#model= 1 + age ");
2180: if(nagesqr==1){
1.241 brouard 2181: printf(" + age*age ");
2182: fprintf(ficlog," + age*age ");
1.239 brouard 2183: }
2184: for(j=1;j <=ncovmodel-2;j++){
2185: if(Typevar[j]==0) {
2186: printf(" + V%d ",Tvar[j]);
2187: fprintf(ficlog," + V%d ",Tvar[j]);
2188: }else if(Typevar[j]==1) {
2189: printf(" + V%d*age ",Tvar[j]);
2190: fprintf(ficlog," + V%d*age ",Tvar[j]);
2191: }else if(Typevar[j]==2) {
2192: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2193: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2194: }
2195: }
1.126 brouard 2196: printf("\n");
1.239 brouard 2197: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2198: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2199: fprintf(ficlog,"\n");
1.239 brouard 2200: for(i=1,jk=1; i <=nlstate; i++){
2201: for(k=1; k <=(nlstate+ndeath); k++){
2202: if (k != i) {
2203: printf("%d%d ",i,k);
2204: fprintf(ficlog,"%d%d ",i,k);
2205: for(j=1; j <=ncovmodel; j++){
2206: printf("%12.7f ",p[jk]);
2207: fprintf(ficlog,"%12.7f ",p[jk]);
2208: jk++;
2209: }
2210: printf("\n");
2211: fprintf(ficlog,"\n");
2212: }
2213: }
2214: }
1.241 brouard 2215: if(*iter <=3 && *iter >1){
1.157 brouard 2216: tml = *localtime(&rcurr_time);
2217: strcpy(strcurr,asctime(&tml));
2218: rforecast_time=rcurr_time;
1.126 brouard 2219: itmp = strlen(strcurr);
2220: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2221: strcurr[itmp-1]='\0';
1.162 brouard 2222: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2223: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2224: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2225: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2226: forecast_time = *localtime(&rforecast_time);
2227: strcpy(strfor,asctime(&forecast_time));
2228: itmp = strlen(strfor);
2229: if(strfor[itmp-1]=='\n')
2230: strfor[itmp-1]='\0';
2231: 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);
2232: 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 2233: }
2234: }
1.187 brouard 2235: for (i=1;i<=n;i++) { /* For each direction i */
2236: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2237: fptt=(*fret);
2238: #ifdef DEBUG
1.203 brouard 2239: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2240: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2241: #endif
1.203 brouard 2242: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2243: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2244: #ifdef LINMINORIGINAL
1.188 brouard 2245: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2246: #else
2247: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2248: flatdir[i]=flat; /* Function is vanishing in that direction i */
2249: #endif
2250: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2251: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2252: /* because that direction will be replaced unless the gain del is small */
2253: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2254: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2255: /* with the new direction. */
2256: del=fabs(fptt-(*fret));
2257: ibig=i;
1.126 brouard 2258: }
2259: #ifdef DEBUG
2260: printf("%d %.12e",i,(*fret));
2261: fprintf(ficlog,"%d %.12e",i,(*fret));
2262: for (j=1;j<=n;j++) {
1.224 brouard 2263: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2264: printf(" x(%d)=%.12e",j,xit[j]);
2265: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2266: }
2267: for(j=1;j<=n;j++) {
1.225 brouard 2268: printf(" p(%d)=%.12e",j,p[j]);
2269: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2270: }
2271: printf("\n");
2272: fprintf(ficlog,"\n");
2273: #endif
1.187 brouard 2274: } /* end loop on each direction i */
2275: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2276: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2277: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2278: for(j=1;j<=n;j++) {
1.225 brouard 2279: if(flatdir[j] >0){
2280: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2281: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2282: }
2283: /* printf("\n"); */
2284: /* fprintf(ficlog,"\n"); */
2285: }
1.243 brouard 2286: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2287: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2288: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2289: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2290: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2291: /* decreased of more than 3.84 */
2292: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2293: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2294: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2295:
1.188 brouard 2296: /* Starting the program with initial values given by a former maximization will simply change */
2297: /* the scales of the directions and the directions, because the are reset to canonical directions */
2298: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2299: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2300: #ifdef DEBUG
2301: int k[2],l;
2302: k[0]=1;
2303: k[1]=-1;
2304: printf("Max: %.12e",(*func)(p));
2305: fprintf(ficlog,"Max: %.12e",(*func)(p));
2306: for (j=1;j<=n;j++) {
2307: printf(" %.12e",p[j]);
2308: fprintf(ficlog," %.12e",p[j]);
2309: }
2310: printf("\n");
2311: fprintf(ficlog,"\n");
2312: for(l=0;l<=1;l++) {
2313: for (j=1;j<=n;j++) {
2314: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2315: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2316: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2317: }
2318: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2319: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2320: }
2321: #endif
2322:
1.224 brouard 2323: #ifdef LINMINORIGINAL
2324: #else
2325: free_ivector(flatdir,1,n);
2326: #endif
1.126 brouard 2327: free_vector(xit,1,n);
2328: free_vector(xits,1,n);
2329: free_vector(ptt,1,n);
2330: free_vector(pt,1,n);
2331: return;
1.192 brouard 2332: } /* enough precision */
1.240 brouard 2333: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2334: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2335: ptt[j]=2.0*p[j]-pt[j];
2336: xit[j]=p[j]-pt[j];
2337: pt[j]=p[j];
2338: }
1.181 brouard 2339: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2340: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2341: if (*iter <=4) {
1.225 brouard 2342: #else
2343: #endif
1.224 brouard 2344: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2345: #else
1.161 brouard 2346: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2347: #endif
1.162 brouard 2348: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2349: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2350: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2351: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2352: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2353: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2354: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2355: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2356: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2357: /* Even if f3 <f1, directest can be negative and t >0 */
2358: /* mu² and del² are equal when f3=f1 */
2359: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2360: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2361: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2362: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2363: #ifdef NRCORIGINAL
2364: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2365: #else
2366: 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 2367: t= t- del*SQR(fp-fptt);
1.183 brouard 2368: #endif
1.202 brouard 2369: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2370: #ifdef DEBUG
1.181 brouard 2371: 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);
2372: 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 2373: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2374: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2375: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2376: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2377: 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);
2378: 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);
2379: #endif
1.183 brouard 2380: #ifdef POWELLORIGINAL
2381: if (t < 0.0) { /* Then we use it for new direction */
2382: #else
1.182 brouard 2383: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2384: 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 2385: 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 2386: 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 2387: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2388: }
1.181 brouard 2389: if (directest < 0.0) { /* Then we use it for new direction */
2390: #endif
1.191 brouard 2391: #ifdef DEBUGLINMIN
1.234 brouard 2392: printf("Before linmin in direction P%d-P0\n",n);
2393: for (j=1;j<=n;j++) {
2394: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2395: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2396: if(j % ncovmodel == 0){
2397: printf("\n");
2398: fprintf(ficlog,"\n");
2399: }
2400: }
1.224 brouard 2401: #endif
2402: #ifdef LINMINORIGINAL
1.234 brouard 2403: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2404: #else
1.234 brouard 2405: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2406: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2407: #endif
1.234 brouard 2408:
1.191 brouard 2409: #ifdef DEBUGLINMIN
1.234 brouard 2410: for (j=1;j<=n;j++) {
2411: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2412: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2413: if(j % ncovmodel == 0){
2414: printf("\n");
2415: fprintf(ficlog,"\n");
2416: }
2417: }
1.224 brouard 2418: #endif
1.234 brouard 2419: for (j=1;j<=n;j++) {
2420: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2421: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2422: }
1.224 brouard 2423: #ifdef LINMINORIGINAL
2424: #else
1.234 brouard 2425: for (j=1, flatd=0;j<=n;j++) {
2426: if(flatdir[j]>0)
2427: flatd++;
2428: }
2429: if(flatd >0){
1.255 brouard 2430: printf("%d flat directions: ",flatd);
2431: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2432: for (j=1;j<=n;j++) {
2433: if(flatdir[j]>0){
2434: printf("%d ",j);
2435: fprintf(ficlog,"%d ",j);
2436: }
2437: }
2438: printf("\n");
2439: fprintf(ficlog,"\n");
2440: }
1.191 brouard 2441: #endif
1.234 brouard 2442: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2443: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2444:
1.126 brouard 2445: #ifdef DEBUG
1.234 brouard 2446: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2447: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2448: for(j=1;j<=n;j++){
2449: printf(" %lf",xit[j]);
2450: fprintf(ficlog," %lf",xit[j]);
2451: }
2452: printf("\n");
2453: fprintf(ficlog,"\n");
1.126 brouard 2454: #endif
1.192 brouard 2455: } /* end of t or directest negative */
1.224 brouard 2456: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2457: #else
1.234 brouard 2458: } /* end if (fptt < fp) */
1.192 brouard 2459: #endif
1.225 brouard 2460: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2461: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2462: #else
1.224 brouard 2463: #endif
1.234 brouard 2464: } /* loop iteration */
1.126 brouard 2465: }
1.234 brouard 2466:
1.126 brouard 2467: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2468:
1.235 brouard 2469: 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 2470: {
1.235 brouard 2471: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2472: (and selected quantitative values in nres)
2473: by left multiplying the unit
1.234 brouard 2474: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2475: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2476: /* Wx is row vector: population in state 1, population in state 2, population dead */
2477: /* or prevalence in state 1, prevalence in state 2, 0 */
2478: /* newm is the matrix after multiplications, its rows are identical at a factor */
2479: /* Initial matrix pimij */
2480: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2481: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2482: /* 0, 0 , 1} */
2483: /*
2484: * and after some iteration: */
2485: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2486: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2487: /* 0, 0 , 1} */
2488: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2489: /* {0.51571254859325999, 0.4842874514067399, */
2490: /* 0.51326036147820708, 0.48673963852179264} */
2491: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2492:
1.126 brouard 2493: int i, ii,j,k;
1.209 brouard 2494: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2495: /* double **matprod2(); */ /* test */
1.218 brouard 2496: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2497: double **newm;
1.209 brouard 2498: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2499: int ncvloop=0;
1.169 brouard 2500:
1.209 brouard 2501: min=vector(1,nlstate);
2502: max=vector(1,nlstate);
2503: meandiff=vector(1,nlstate);
2504:
1.218 brouard 2505: /* Starting with matrix unity */
1.126 brouard 2506: for (ii=1;ii<=nlstate+ndeath;ii++)
2507: for (j=1;j<=nlstate+ndeath;j++){
2508: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2509: }
1.169 brouard 2510:
2511: cov[1]=1.;
2512:
2513: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2514: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2515: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2516: ncvloop++;
1.126 brouard 2517: newm=savm;
2518: /* Covariates have to be included here again */
1.138 brouard 2519: cov[2]=agefin;
1.187 brouard 2520: if(nagesqr==1)
2521: cov[3]= agefin*agefin;;
1.234 brouard 2522: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2523: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2524: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2525: /* 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 2526: }
2527: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2528: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2529: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2530: /* 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 2531: }
1.237 brouard 2532: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2533: if(Dummy[Tvar[Tage[k]]]){
2534: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2535: } else{
1.235 brouard 2536: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2537: }
1.235 brouard 2538: /* 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 2539: }
1.237 brouard 2540: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2541: /* 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 2542: if(Dummy[Tvard[k][1]==0]){
2543: if(Dummy[Tvard[k][2]==0]){
2544: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2545: }else{
2546: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2547: }
2548: }else{
2549: if(Dummy[Tvard[k][2]==0]){
2550: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2551: }else{
2552: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2553: }
2554: }
1.234 brouard 2555: }
1.138 brouard 2556: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2557: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2558: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2559: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2560: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2561: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2562: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2563:
1.126 brouard 2564: savm=oldm;
2565: oldm=newm;
1.209 brouard 2566:
2567: for(j=1; j<=nlstate; j++){
2568: max[j]=0.;
2569: min[j]=1.;
2570: }
2571: for(i=1;i<=nlstate;i++){
2572: sumnew=0;
2573: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2574: for(j=1; j<=nlstate; j++){
2575: prlim[i][j]= newm[i][j]/(1-sumnew);
2576: max[j]=FMAX(max[j],prlim[i][j]);
2577: min[j]=FMIN(min[j],prlim[i][j]);
2578: }
2579: }
2580:
1.126 brouard 2581: maxmax=0.;
1.209 brouard 2582: for(j=1; j<=nlstate; j++){
2583: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2584: maxmax=FMAX(maxmax,meandiff[j]);
2585: /* 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 2586: } /* j loop */
1.203 brouard 2587: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2588: /* 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 2589: if(maxmax < ftolpl){
1.209 brouard 2590: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2591: free_vector(min,1,nlstate);
2592: free_vector(max,1,nlstate);
2593: free_vector(meandiff,1,nlstate);
1.126 brouard 2594: return prlim;
2595: }
1.169 brouard 2596: } /* age loop */
1.208 brouard 2597: /* After some age loop it doesn't converge */
1.209 brouard 2598: 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 2599: 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 2600: /* 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); */
2601: free_vector(min,1,nlstate);
2602: free_vector(max,1,nlstate);
2603: free_vector(meandiff,1,nlstate);
1.208 brouard 2604:
1.169 brouard 2605: return prlim; /* should not reach here */
1.126 brouard 2606: }
2607:
1.217 brouard 2608:
2609: /**** Back Prevalence limit (stable or period prevalence) ****************/
2610:
1.218 brouard 2611: /* 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) */
2612: /* 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 2613: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2614: {
1.218 brouard 2615: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2616: matrix by transitions matrix until convergence is reached with precision ftolpl */
2617: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2618: /* Wx is row vector: population in state 1, population in state 2, population dead */
2619: /* or prevalence in state 1, prevalence in state 2, 0 */
2620: /* newm is the matrix after multiplications, its rows are identical at a factor */
2621: /* Initial matrix pimij */
2622: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2623: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2624: /* 0, 0 , 1} */
2625: /*
2626: * and after some iteration: */
2627: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2628: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2629: /* 0, 0 , 1} */
2630: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2631: /* {0.51571254859325999, 0.4842874514067399, */
2632: /* 0.51326036147820708, 0.48673963852179264} */
2633: /* If we start from prlim again, prlim tends to a constant matrix */
2634:
2635: int i, ii,j,k;
1.247 brouard 2636: int first=0;
1.217 brouard 2637: double *min, *max, *meandiff, maxmax,sumnew=0.;
2638: /* double **matprod2(); */ /* test */
2639: double **out, cov[NCOVMAX+1], **bmij();
2640: double **newm;
1.218 brouard 2641: double **dnewm, **doldm, **dsavm; /* for use */
2642: double **oldm, **savm; /* for use */
2643:
1.217 brouard 2644: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2645: int ncvloop=0;
2646:
2647: min=vector(1,nlstate);
2648: max=vector(1,nlstate);
2649: meandiff=vector(1,nlstate);
2650:
1.218 brouard 2651: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2652: oldm=oldms; savm=savms;
2653:
2654: /* Starting with matrix unity */
2655: for (ii=1;ii<=nlstate+ndeath;ii++)
2656: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2657: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2658: }
2659:
2660: cov[1]=1.;
2661:
2662: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2663: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2664: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2665: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2666: ncvloop++;
1.218 brouard 2667: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2668: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2669: /* Covariates have to be included here again */
2670: cov[2]=agefin;
2671: if(nagesqr==1)
2672: cov[3]= agefin*agefin;;
1.242 brouard 2673: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2674: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2675: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2676: /* 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)); */
2677: }
2678: /* for (k=1; k<=cptcovn;k++) { */
2679: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2680: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2681: /* /\* 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])]); *\/ */
2682: /* } */
2683: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2684: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2685: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2686: /* 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]); */
2687: }
2688: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2689: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2690: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2691: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2692: for (k=1; k<=cptcovage;k++){ /* For product with age */
2693: if(Dummy[Tvar[Tage[k]]]){
2694: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2695: } else{
2696: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2697: }
2698: /* 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]); */
2699: }
2700: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2701: /* 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]); */
2702: if(Dummy[Tvard[k][1]==0]){
2703: if(Dummy[Tvard[k][2]==0]){
2704: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2705: }else{
2706: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2707: }
2708: }else{
2709: if(Dummy[Tvard[k][2]==0]){
2710: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2711: }else{
2712: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2713: }
2714: }
1.217 brouard 2715: }
2716:
2717: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2718: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2719: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2720: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2721: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2722: /* ij should be linked to the correct index of cov */
2723: /* age and covariate values ij are in 'cov', but we need to pass
2724: * ij for the observed prevalence at age and status and covariate
2725: * number: prevacurrent[(int)agefin][ii][ij]
2726: */
2727: /* 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 *\/ */
2728: /* 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 *\/ */
2729: 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 2730: savm=oldm;
2731: oldm=newm;
2732: for(j=1; j<=nlstate; j++){
2733: max[j]=0.;
2734: min[j]=1.;
2735: }
2736: for(j=1; j<=nlstate; j++){
2737: for(i=1;i<=nlstate;i++){
1.234 brouard 2738: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2739: bprlim[i][j]= newm[i][j];
2740: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2741: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2742: }
2743: }
1.218 brouard 2744:
1.217 brouard 2745: maxmax=0.;
2746: for(i=1; i<=nlstate; i++){
2747: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2748: maxmax=FMAX(maxmax,meandiff[i]);
2749: /* 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); */
2750: } /* j loop */
2751: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2752: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2753: if(maxmax < ftolpl){
1.220 brouard 2754: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2755: free_vector(min,1,nlstate);
2756: free_vector(max,1,nlstate);
2757: free_vector(meandiff,1,nlstate);
2758: return bprlim;
2759: }
2760: } /* age loop */
2761: /* After some age loop it doesn't converge */
1.247 brouard 2762: if(first){
2763: first=1;
2764: 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\
2765: 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);
2766: }
2767: 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 2768: 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);
2769: /* 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); */
2770: free_vector(min,1,nlstate);
2771: free_vector(max,1,nlstate);
2772: free_vector(meandiff,1,nlstate);
2773:
2774: return bprlim; /* should not reach here */
2775: }
2776:
1.126 brouard 2777: /*************** transition probabilities ***************/
2778:
2779: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2780: {
1.138 brouard 2781: /* According to parameters values stored in x and the covariate's values stored in cov,
2782: computes the probability to be observed in state j being in state i by appying the
2783: model to the ncovmodel covariates (including constant and age).
2784: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2785: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2786: ncth covariate in the global vector x is given by the formula:
2787: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2788: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2789: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2790: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2791: Outputs ps[i][j] the probability to be observed in j being in j according to
2792: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2793: */
2794: double s1, lnpijopii;
1.126 brouard 2795: /*double t34;*/
1.164 brouard 2796: int i,j, nc, ii, jj;
1.126 brouard 2797:
1.223 brouard 2798: for(i=1; i<= nlstate; i++){
2799: for(j=1; j<i;j++){
2800: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2801: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2802: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2803: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2804: }
2805: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2806: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2807: }
2808: for(j=i+1; j<=nlstate+ndeath;j++){
2809: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2810: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2811: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2812: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2813: }
2814: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2815: }
2816: }
1.218 brouard 2817:
1.223 brouard 2818: for(i=1; i<= nlstate; i++){
2819: s1=0;
2820: for(j=1; j<i; j++){
2821: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2822: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2823: }
2824: for(j=i+1; j<=nlstate+ndeath; j++){
2825: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2826: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2827: }
2828: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2829: ps[i][i]=1./(s1+1.);
2830: /* Computing other pijs */
2831: for(j=1; j<i; j++)
2832: ps[i][j]= exp(ps[i][j])*ps[i][i];
2833: for(j=i+1; j<=nlstate+ndeath; j++)
2834: ps[i][j]= exp(ps[i][j])*ps[i][i];
2835: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2836: } /* end i */
1.218 brouard 2837:
1.223 brouard 2838: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2839: for(jj=1; jj<= nlstate+ndeath; jj++){
2840: ps[ii][jj]=0;
2841: ps[ii][ii]=1;
2842: }
2843: }
1.218 brouard 2844:
2845:
1.223 brouard 2846: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2847: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2848: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2849: /* } */
2850: /* printf("\n "); */
2851: /* } */
2852: /* printf("\n ");printf("%lf ",cov[2]);*/
2853: /*
2854: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2855: goto end;*/
1.223 brouard 2856: return ps;
1.126 brouard 2857: }
2858:
1.218 brouard 2859: /*************** backward transition probabilities ***************/
2860:
2861: /* 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 ) */
2862: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2863: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2864: {
1.222 brouard 2865: /* Computes the backward probability at age agefin and covariate ij
2866: * and returns in **ps as well as **bmij.
2867: */
1.218 brouard 2868: int i, ii, j,k;
1.222 brouard 2869:
2870: double **out, **pmij();
2871: double sumnew=0.;
1.218 brouard 2872: double agefin;
1.222 brouard 2873:
2874: double **dnewm, **dsavm, **doldm;
2875: double **bbmij;
2876:
1.218 brouard 2877: doldm=ddoldms; /* global pointers */
1.222 brouard 2878: dnewm=ddnewms;
2879: dsavm=ddsavms;
2880:
2881: agefin=cov[2];
2882: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2883: the observed prevalence (with this covariate ij) */
2884: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2885: /* We do have the matrix Px in savm and we need pij */
2886: for (j=1;j<=nlstate+ndeath;j++){
2887: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2888: for (ii=1;ii<=nlstate;ii++){
2889: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2890: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2891: for (ii=1;ii<=nlstate+ndeath;ii++){
2892: if(sumnew >= 1.e-10){
2893: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2894: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2895: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2896: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2897: /* }else */
2898: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2899: }else{
1.242 brouard 2900: ;
2901: /* 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 2902: }
2903: } /*End ii */
2904: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2905: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2906: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2907: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2908: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2909: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2910: /* left Product of this matrix by diag matrix of prevalences (savm) */
2911: for (j=1;j<=nlstate+ndeath;j++){
2912: for (ii=1;ii<=nlstate+ndeath;ii++){
2913: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2914: }
2915: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2916: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2917: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2918: /* end bmij */
2919: return ps;
1.218 brouard 2920: }
1.217 brouard 2921: /*************** transition probabilities ***************/
2922:
1.218 brouard 2923: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2924: {
2925: /* According to parameters values stored in x and the covariate's values stored in cov,
2926: computes the probability to be observed in state j being in state i by appying the
2927: model to the ncovmodel covariates (including constant and age).
2928: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2929: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2930: ncth covariate in the global vector x is given by the formula:
2931: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2932: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2933: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2934: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2935: Outputs ps[i][j] the probability to be observed in j being in j according to
2936: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2937: */
2938: double s1, lnpijopii;
2939: /*double t34;*/
2940: int i,j, nc, ii, jj;
2941:
1.234 brouard 2942: for(i=1; i<= nlstate; i++){
2943: for(j=1; j<i;j++){
2944: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2945: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2946: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2947: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2948: }
2949: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2950: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2951: }
2952: for(j=i+1; j<=nlstate+ndeath;j++){
2953: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2954: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2955: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2956: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2957: }
2958: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2959: }
2960: }
2961:
2962: for(i=1; i<= nlstate; i++){
2963: s1=0;
2964: for(j=1; j<i; j++){
2965: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2966: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2967: }
2968: for(j=i+1; j<=nlstate+ndeath; j++){
2969: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2970: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2971: }
2972: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2973: ps[i][i]=1./(s1+1.);
2974: /* Computing other pijs */
2975: for(j=1; j<i; j++)
2976: ps[i][j]= exp(ps[i][j])*ps[i][i];
2977: for(j=i+1; j<=nlstate+ndeath; j++)
2978: ps[i][j]= exp(ps[i][j])*ps[i][i];
2979: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2980: } /* end i */
2981:
2982: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2983: for(jj=1; jj<= nlstate+ndeath; jj++){
2984: ps[ii][jj]=0;
2985: ps[ii][ii]=1;
2986: }
2987: }
2988: /* Added for backcast */ /* Transposed matrix too */
2989: for(jj=1; jj<= nlstate+ndeath; jj++){
2990: s1=0.;
2991: for(ii=1; ii<= nlstate+ndeath; ii++){
2992: s1+=ps[ii][jj];
2993: }
2994: for(ii=1; ii<= nlstate; ii++){
2995: ps[ii][jj]=ps[ii][jj]/s1;
2996: }
2997: }
2998: /* Transposition */
2999: for(jj=1; jj<= nlstate+ndeath; jj++){
3000: for(ii=jj; ii<= nlstate+ndeath; ii++){
3001: s1=ps[ii][jj];
3002: ps[ii][jj]=ps[jj][ii];
3003: ps[jj][ii]=s1;
3004: }
3005: }
3006: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3007: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3008: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3009: /* } */
3010: /* printf("\n "); */
3011: /* } */
3012: /* printf("\n ");printf("%lf ",cov[2]);*/
3013: /*
3014: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3015: goto end;*/
3016: return ps;
1.217 brouard 3017: }
3018:
3019:
1.126 brouard 3020: /**************** Product of 2 matrices ******************/
3021:
1.145 brouard 3022: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3023: {
3024: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3025: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3026: /* in, b, out are matrice of pointers which should have been initialized
3027: before: only the contents of out is modified. The function returns
3028: a pointer to pointers identical to out */
1.145 brouard 3029: int i, j, k;
1.126 brouard 3030: for(i=nrl; i<= nrh; i++)
1.145 brouard 3031: for(k=ncolol; k<=ncoloh; k++){
3032: out[i][k]=0.;
3033: for(j=ncl; j<=nch; j++)
3034: out[i][k] +=in[i][j]*b[j][k];
3035: }
1.126 brouard 3036: return out;
3037: }
3038:
3039:
3040: /************* Higher Matrix Product ***************/
3041:
1.235 brouard 3042: 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 3043: {
1.218 brouard 3044: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3045: 'nhstepm*hstepm*stepm' months (i.e. until
3046: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3047: nhstepm*hstepm matrices.
3048: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3049: (typically every 2 years instead of every month which is too big
3050: for the memory).
3051: Model is determined by parameters x and covariates have to be
3052: included manually here.
3053:
3054: */
3055:
3056: int i, j, d, h, k;
1.131 brouard 3057: double **out, cov[NCOVMAX+1];
1.126 brouard 3058: double **newm;
1.187 brouard 3059: double agexact;
1.214 brouard 3060: double agebegin, ageend;
1.126 brouard 3061:
3062: /* Hstepm could be zero and should return the unit matrix */
3063: for (i=1;i<=nlstate+ndeath;i++)
3064: for (j=1;j<=nlstate+ndeath;j++){
3065: oldm[i][j]=(i==j ? 1.0 : 0.0);
3066: po[i][j][0]=(i==j ? 1.0 : 0.0);
3067: }
3068: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3069: for(h=1; h <=nhstepm; h++){
3070: for(d=1; d <=hstepm; d++){
3071: newm=savm;
3072: /* Covariates have to be included here again */
3073: cov[1]=1.;
1.214 brouard 3074: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3075: cov[2]=agexact;
3076: if(nagesqr==1)
1.227 brouard 3077: cov[3]= agexact*agexact;
1.235 brouard 3078: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3079: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3080: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3081: /* 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)); */
3082: }
3083: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3084: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3085: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3086: /* 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]); */
3087: }
3088: for (k=1; k<=cptcovage;k++){
3089: if(Dummy[Tvar[Tage[k]]]){
3090: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3091: } else{
3092: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3093: }
3094: /* 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]); */
3095: }
3096: for (k=1; k<=cptcovprod;k++){ /* */
3097: /* 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]); */
3098: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3099: }
3100: /* for (k=1; k<=cptcovn;k++) */
3101: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3102: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3103: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3104: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3105: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3106:
3107:
1.126 brouard 3108: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3109: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3110: /* right multiplication of oldm by the current matrix */
1.126 brouard 3111: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3112: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3113: /* if((int)age == 70){ */
3114: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3115: /* for(i=1; i<=nlstate+ndeath; i++) { */
3116: /* printf("%d pmmij ",i); */
3117: /* for(j=1;j<=nlstate+ndeath;j++) { */
3118: /* printf("%f ",pmmij[i][j]); */
3119: /* } */
3120: /* printf(" oldm "); */
3121: /* for(j=1;j<=nlstate+ndeath;j++) { */
3122: /* printf("%f ",oldm[i][j]); */
3123: /* } */
3124: /* printf("\n"); */
3125: /* } */
3126: /* } */
1.126 brouard 3127: savm=oldm;
3128: oldm=newm;
3129: }
3130: for(i=1; i<=nlstate+ndeath; i++)
3131: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3132: po[i][j][h]=newm[i][j];
3133: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3134: }
1.128 brouard 3135: /*printf("h=%d ",h);*/
1.126 brouard 3136: } /* end h */
1.218 brouard 3137: /* printf("\n H=%d \n",h); */
1.126 brouard 3138: return po;
3139: }
3140:
1.217 brouard 3141: /************* Higher Back Matrix Product ***************/
1.218 brouard 3142: /* 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 3143: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3144: {
1.218 brouard 3145: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3146: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3147: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3148: nhstepm*hstepm matrices.
3149: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3150: (typically every 2 years instead of every month which is too big
1.217 brouard 3151: for the memory).
1.218 brouard 3152: Model is determined by parameters x and covariates have to be
3153: included manually here.
1.217 brouard 3154:
1.222 brouard 3155: */
1.217 brouard 3156:
3157: int i, j, d, h, k;
3158: double **out, cov[NCOVMAX+1];
3159: double **newm;
3160: double agexact;
3161: double agebegin, ageend;
1.222 brouard 3162: double **oldm, **savm;
1.217 brouard 3163:
1.222 brouard 3164: oldm=oldms;savm=savms;
1.217 brouard 3165: /* Hstepm could be zero and should return the unit matrix */
3166: for (i=1;i<=nlstate+ndeath;i++)
3167: for (j=1;j<=nlstate+ndeath;j++){
3168: oldm[i][j]=(i==j ? 1.0 : 0.0);
3169: po[i][j][0]=(i==j ? 1.0 : 0.0);
3170: }
3171: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3172: for(h=1; h <=nhstepm; h++){
3173: for(d=1; d <=hstepm; d++){
3174: newm=savm;
3175: /* Covariates have to be included here again */
3176: cov[1]=1.;
3177: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3178: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3179: cov[2]=agexact;
3180: if(nagesqr==1)
1.222 brouard 3181: cov[3]= agexact*agexact;
1.218 brouard 3182: for (k=1; k<=cptcovn;k++)
1.222 brouard 3183: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3184: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3185: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3186: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3187: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3188: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3189: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3190: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3191: /* 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 3192:
3193:
1.217 brouard 3194: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3195: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3196: /* Careful transposed matrix */
1.222 brouard 3197: /* age is in cov[2] */
1.218 brouard 3198: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3199: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3200: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3201: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3202: /* if((int)age == 70){ */
3203: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3204: /* for(i=1; i<=nlstate+ndeath; i++) { */
3205: /* printf("%d pmmij ",i); */
3206: /* for(j=1;j<=nlstate+ndeath;j++) { */
3207: /* printf("%f ",pmmij[i][j]); */
3208: /* } */
3209: /* printf(" oldm "); */
3210: /* for(j=1;j<=nlstate+ndeath;j++) { */
3211: /* printf("%f ",oldm[i][j]); */
3212: /* } */
3213: /* printf("\n"); */
3214: /* } */
3215: /* } */
3216: savm=oldm;
3217: oldm=newm;
3218: }
3219: for(i=1; i<=nlstate+ndeath; i++)
3220: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3221: po[i][j][h]=newm[i][j];
3222: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3223: }
3224: /*printf("h=%d ",h);*/
3225: } /* end h */
1.222 brouard 3226: /* printf("\n H=%d \n",h); */
1.217 brouard 3227: return po;
3228: }
3229:
3230:
1.162 brouard 3231: #ifdef NLOPT
3232: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3233: double fret;
3234: double *xt;
3235: int j;
3236: myfunc_data *d2 = (myfunc_data *) pd;
3237: /* xt = (p1-1); */
3238: xt=vector(1,n);
3239: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3240:
3241: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3242: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3243: printf("Function = %.12lf ",fret);
3244: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3245: printf("\n");
3246: free_vector(xt,1,n);
3247: return fret;
3248: }
3249: #endif
1.126 brouard 3250:
3251: /*************** log-likelihood *************/
3252: double func( double *x)
3253: {
1.226 brouard 3254: int i, ii, j, k, mi, d, kk;
3255: int ioffset=0;
3256: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3257: double **out;
3258: double lli; /* Individual log likelihood */
3259: int s1, s2;
1.228 brouard 3260: 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 3261: double bbh, survp;
3262: long ipmx;
3263: double agexact;
3264: /*extern weight */
3265: /* We are differentiating ll according to initial status */
3266: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3267: /*for(i=1;i<imx;i++)
3268: printf(" %d\n",s[4][i]);
3269: */
1.162 brouard 3270:
1.226 brouard 3271: ++countcallfunc;
1.162 brouard 3272:
1.226 brouard 3273: cov[1]=1.;
1.126 brouard 3274:
1.226 brouard 3275: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3276: ioffset=0;
1.226 brouard 3277: if(mle==1){
3278: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3279: /* Computes the values of the ncovmodel covariates of the model
3280: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3281: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3282: to be observed in j being in i according to the model.
3283: */
1.243 brouard 3284: ioffset=2+nagesqr ;
1.233 brouard 3285: /* Fixed */
1.234 brouard 3286: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3287: 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)*/
3288: }
1.226 brouard 3289: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3290: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3291: has been calculated etc */
3292: /* For an individual i, wav[i] gives the number of effective waves */
3293: /* We compute the contribution to Likelihood of each effective transition
3294: mw[mi][i] is real wave of the mi th effectve wave */
3295: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3296: s2=s[mw[mi+1][i]][i];
3297: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3298: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3299: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3300: */
3301: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3302: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3303: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3304: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3305: }
3306: for (ii=1;ii<=nlstate+ndeath;ii++)
3307: for (j=1;j<=nlstate+ndeath;j++){
3308: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3309: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3310: }
3311: for(d=0; d<dh[mi][i]; d++){
3312: newm=savm;
3313: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3314: cov[2]=agexact;
3315: if(nagesqr==1)
3316: cov[3]= agexact*agexact; /* Should be changed here */
3317: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3318: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3319: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3320: else
3321: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3322: }
3323: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3324: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3325: savm=oldm;
3326: oldm=newm;
3327: } /* end mult */
3328:
3329: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3330: /* But now since version 0.9 we anticipate for bias at large stepm.
3331: * If stepm is larger than one month (smallest stepm) and if the exact delay
3332: * (in months) between two waves is not a multiple of stepm, we rounded to
3333: * the nearest (and in case of equal distance, to the lowest) interval but now
3334: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3335: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3336: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3337: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3338: * -stepm/2 to stepm/2 .
3339: * For stepm=1 the results are the same as for previous versions of Imach.
3340: * For stepm > 1 the results are less biased than in previous versions.
3341: */
1.234 brouard 3342: s1=s[mw[mi][i]][i];
3343: s2=s[mw[mi+1][i]][i];
3344: bbh=(double)bh[mi][i]/(double)stepm;
3345: /* bias bh is positive if real duration
3346: * is higher than the multiple of stepm and negative otherwise.
3347: */
3348: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3349: if( s2 > nlstate){
3350: /* i.e. if s2 is a death state and if the date of death is known
3351: then the contribution to the likelihood is the probability to
3352: die between last step unit time and current step unit time,
3353: which is also equal to probability to die before dh
3354: minus probability to die before dh-stepm .
3355: In version up to 0.92 likelihood was computed
3356: as if date of death was unknown. Death was treated as any other
3357: health state: the date of the interview describes the actual state
3358: and not the date of a change in health state. The former idea was
3359: to consider that at each interview the state was recorded
3360: (healthy, disable or death) and IMaCh was corrected; but when we
3361: introduced the exact date of death then we should have modified
3362: the contribution of an exact death to the likelihood. This new
3363: contribution is smaller and very dependent of the step unit
3364: stepm. It is no more the probability to die between last interview
3365: and month of death but the probability to survive from last
3366: interview up to one month before death multiplied by the
3367: probability to die within a month. Thanks to Chris
3368: Jackson for correcting this bug. Former versions increased
3369: mortality artificially. The bad side is that we add another loop
3370: which slows down the processing. The difference can be up to 10%
3371: lower mortality.
3372: */
3373: /* If, at the beginning of the maximization mostly, the
3374: cumulative probability or probability to be dead is
3375: constant (ie = 1) over time d, the difference is equal to
3376: 0. out[s1][3] = savm[s1][3]: probability, being at state
3377: s1 at precedent wave, to be dead a month before current
3378: wave is equal to probability, being at state s1 at
3379: precedent wave, to be dead at mont of the current
3380: wave. Then the observed probability (that this person died)
3381: is null according to current estimated parameter. In fact,
3382: it should be very low but not zero otherwise the log go to
3383: infinity.
3384: */
1.183 brouard 3385: /* #ifdef INFINITYORIGINAL */
3386: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3387: /* #else */
3388: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3389: /* lli=log(mytinydouble); */
3390: /* else */
3391: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3392: /* #endif */
1.226 brouard 3393: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3394:
1.226 brouard 3395: } else if ( s2==-1 ) { /* alive */
3396: for (j=1,survp=0. ; j<=nlstate; j++)
3397: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3398: /*survp += out[s1][j]; */
3399: lli= log(survp);
3400: }
3401: else if (s2==-4) {
3402: for (j=3,survp=0. ; j<=nlstate; j++)
3403: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3404: lli= log(survp);
3405: }
3406: else if (s2==-5) {
3407: for (j=1,survp=0. ; j<=2; j++)
3408: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3409: lli= log(survp);
3410: }
3411: else{
3412: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3413: /* 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 */
3414: }
3415: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3416: /*if(lli ==000.0)*/
3417: /*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); */
3418: ipmx +=1;
3419: sw += weight[i];
3420: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3421: /* if (lli < log(mytinydouble)){ */
3422: /* 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); */
3423: /* 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]); */
3424: /* } */
3425: } /* end of wave */
3426: } /* end of individual */
3427: } else if(mle==2){
3428: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3429: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3430: for(mi=1; mi<= wav[i]-1; mi++){
3431: for (ii=1;ii<=nlstate+ndeath;ii++)
3432: for (j=1;j<=nlstate+ndeath;j++){
3433: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3434: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3435: }
3436: for(d=0; d<=dh[mi][i]; d++){
3437: newm=savm;
3438: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3439: cov[2]=agexact;
3440: if(nagesqr==1)
3441: cov[3]= agexact*agexact;
3442: for (kk=1; kk<=cptcovage;kk++) {
3443: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3444: }
3445: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3446: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3447: savm=oldm;
3448: oldm=newm;
3449: } /* end mult */
3450:
3451: s1=s[mw[mi][i]][i];
3452: s2=s[mw[mi+1][i]][i];
3453: bbh=(double)bh[mi][i]/(double)stepm;
3454: 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 */
3455: ipmx +=1;
3456: sw += weight[i];
3457: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3458: } /* end of wave */
3459: } /* end of individual */
3460: } else if(mle==3){ /* exponential inter-extrapolation */
3461: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3462: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3463: for(mi=1; mi<= wav[i]-1; mi++){
3464: for (ii=1;ii<=nlstate+ndeath;ii++)
3465: for (j=1;j<=nlstate+ndeath;j++){
3466: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3467: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3468: }
3469: for(d=0; d<dh[mi][i]; d++){
3470: newm=savm;
3471: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3472: cov[2]=agexact;
3473: if(nagesqr==1)
3474: cov[3]= agexact*agexact;
3475: for (kk=1; kk<=cptcovage;kk++) {
3476: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3477: }
3478: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3479: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3480: savm=oldm;
3481: oldm=newm;
3482: } /* end mult */
3483:
3484: s1=s[mw[mi][i]][i];
3485: s2=s[mw[mi+1][i]][i];
3486: bbh=(double)bh[mi][i]/(double)stepm;
3487: 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 */
3488: ipmx +=1;
3489: sw += weight[i];
3490: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3491: } /* end of wave */
3492: } /* end of individual */
3493: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3494: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3495: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3496: for(mi=1; mi<= wav[i]-1; mi++){
3497: for (ii=1;ii<=nlstate+ndeath;ii++)
3498: for (j=1;j<=nlstate+ndeath;j++){
3499: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3500: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3501: }
3502: for(d=0; d<dh[mi][i]; d++){
3503: newm=savm;
3504: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3505: cov[2]=agexact;
3506: if(nagesqr==1)
3507: cov[3]= agexact*agexact;
3508: for (kk=1; kk<=cptcovage;kk++) {
3509: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3510: }
1.126 brouard 3511:
1.226 brouard 3512: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3513: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3514: savm=oldm;
3515: oldm=newm;
3516: } /* end mult */
3517:
3518: s1=s[mw[mi][i]][i];
3519: s2=s[mw[mi+1][i]][i];
3520: if( s2 > nlstate){
3521: lli=log(out[s1][s2] - savm[s1][s2]);
3522: } else if ( s2==-1 ) { /* alive */
3523: for (j=1,survp=0. ; j<=nlstate; j++)
3524: survp += out[s1][j];
3525: lli= log(survp);
3526: }else{
3527: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3528: }
3529: ipmx +=1;
3530: sw += weight[i];
3531: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3532: /* 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 3533: } /* end of wave */
3534: } /* end of individual */
3535: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3536: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3537: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3538: for(mi=1; mi<= wav[i]-1; mi++){
3539: for (ii=1;ii<=nlstate+ndeath;ii++)
3540: for (j=1;j<=nlstate+ndeath;j++){
3541: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3542: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3543: }
3544: for(d=0; d<dh[mi][i]; d++){
3545: newm=savm;
3546: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3547: cov[2]=agexact;
3548: if(nagesqr==1)
3549: cov[3]= agexact*agexact;
3550: for (kk=1; kk<=cptcovage;kk++) {
3551: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3552: }
1.126 brouard 3553:
1.226 brouard 3554: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3555: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3556: savm=oldm;
3557: oldm=newm;
3558: } /* end mult */
3559:
3560: s1=s[mw[mi][i]][i];
3561: s2=s[mw[mi+1][i]][i];
3562: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3563: ipmx +=1;
3564: sw += weight[i];
3565: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3566: /*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]);*/
3567: } /* end of wave */
3568: } /* end of individual */
3569: } /* End of if */
3570: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3571: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3572: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3573: return -l;
1.126 brouard 3574: }
3575:
3576: /*************** log-likelihood *************/
3577: double funcone( double *x)
3578: {
1.228 brouard 3579: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3580: int i, ii, j, k, mi, d, kk;
1.228 brouard 3581: int ioffset=0;
1.131 brouard 3582: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3583: double **out;
3584: double lli; /* Individual log likelihood */
3585: double llt;
3586: int s1, s2;
1.228 brouard 3587: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3588:
1.126 brouard 3589: double bbh, survp;
1.187 brouard 3590: double agexact;
1.214 brouard 3591: double agebegin, ageend;
1.126 brouard 3592: /*extern weight */
3593: /* We are differentiating ll according to initial status */
3594: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3595: /*for(i=1;i<imx;i++)
3596: printf(" %d\n",s[4][i]);
3597: */
3598: cov[1]=1.;
3599:
3600: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3601: ioffset=0;
3602: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3603: /* ioffset=2+nagesqr+cptcovage; */
3604: ioffset=2+nagesqr;
1.232 brouard 3605: /* Fixed */
1.224 brouard 3606: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3607: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3608: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3609: 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)*/
3610: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3611: /* cov[2+6]=covar[Tvar[6]][i]; */
3612: /* cov[2+6]=covar[2][i]; V2 */
3613: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3614: /* cov[2+7]=covar[Tvar[7]][i]; */
3615: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3616: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3617: /* cov[2+9]=covar[Tvar[9]][i]; */
3618: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3619: }
1.232 brouard 3620: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3621: /* 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?)*\/ */
3622: /* } */
1.231 brouard 3623: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3624: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3625: /* } */
1.225 brouard 3626:
1.233 brouard 3627:
3628: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3629: /* Wave varying (but not age varying) */
3630: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3631: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3632: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3633: }
1.232 brouard 3634: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3635: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3636: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3637: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3638: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3639: /* 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 3640: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3641: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3642: /* /\* 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]); *\/ */
3643: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3644: /* } */
1.126 brouard 3645: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3646: for (j=1;j<=nlstate+ndeath;j++){
3647: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3648: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3649: }
1.214 brouard 3650:
3651: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3652: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3653: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3654: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3655: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3656: and mw[mi+1][i]. dh depends on stepm.*/
3657: newm=savm;
1.247 brouard 3658: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3659: cov[2]=agexact;
3660: if(nagesqr==1)
3661: cov[3]= agexact*agexact;
3662: for (kk=1; kk<=cptcovage;kk++) {
3663: if(!FixedV[Tvar[Tage[kk]]])
3664: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3665: else
3666: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3667: }
3668: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3669: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3670: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3671: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3672: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3673: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3674: savm=oldm;
3675: oldm=newm;
1.126 brouard 3676: } /* end mult */
3677:
3678: s1=s[mw[mi][i]][i];
3679: s2=s[mw[mi+1][i]][i];
1.217 brouard 3680: /* if(s2==-1){ */
3681: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3682: /* /\* exit(1); *\/ */
3683: /* } */
1.126 brouard 3684: bbh=(double)bh[mi][i]/(double)stepm;
3685: /* bias is positive if real duration
3686: * is higher than the multiple of stepm and negative otherwise.
3687: */
3688: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3689: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3690: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3691: for (j=1,survp=0. ; j<=nlstate; j++)
3692: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3693: lli= log(survp);
1.126 brouard 3694: }else if (mle==1){
1.242 brouard 3695: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3696: } else if(mle==2){
1.242 brouard 3697: 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 3698: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3699: 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 3700: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3701: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3702: } else{ /* mle=0 back to 1 */
1.242 brouard 3703: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3704: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3705: } /* End of if */
3706: ipmx +=1;
3707: sw += weight[i];
3708: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3709: /*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 3710: if(globpr){
1.246 brouard 3711: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3712: %11.6f %11.6f %11.6f ", \
1.242 brouard 3713: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3714: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3715: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3716: llt +=ll[k]*gipmx/gsw;
3717: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3718: }
3719: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3720: }
1.232 brouard 3721: } /* end of wave */
3722: } /* end of individual */
3723: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3724: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3725: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3726: if(globpr==0){ /* First time we count the contributions and weights */
3727: gipmx=ipmx;
3728: gsw=sw;
3729: }
3730: return -l;
1.126 brouard 3731: }
3732:
3733:
3734: /*************** function likelione ***********/
3735: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3736: {
3737: /* This routine should help understanding what is done with
3738: the selection of individuals/waves and
3739: to check the exact contribution to the likelihood.
3740: Plotting could be done.
3741: */
3742: int k;
3743:
3744: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3745: strcpy(fileresilk,"ILK_");
1.202 brouard 3746: strcat(fileresilk,fileresu);
1.126 brouard 3747: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3748: printf("Problem with resultfile: %s\n", fileresilk);
3749: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3750: }
1.214 brouard 3751: 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");
3752: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3753: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3754: for(k=1; k<=nlstate; k++)
3755: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3756: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3757: }
3758:
3759: *fretone=(*funcone)(p);
3760: if(*globpri !=0){
3761: fclose(ficresilk);
1.205 brouard 3762: if (mle ==0)
3763: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3764: else if(mle >=1)
3765: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3766: 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 3767:
1.208 brouard 3768:
3769: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3770: 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 3771: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3772: }
1.207 brouard 3773: 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 3774: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3775: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3776: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3777: fflush(fichtm);
1.205 brouard 3778: }
1.126 brouard 3779: return;
3780: }
3781:
3782:
3783: /*********** Maximum Likelihood Estimation ***************/
3784:
3785: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3786: {
1.165 brouard 3787: int i,j, iter=0;
1.126 brouard 3788: double **xi;
3789: double fret;
3790: double fretone; /* Only one call to likelihood */
3791: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3792:
3793: #ifdef NLOPT
3794: int creturn;
3795: nlopt_opt opt;
3796: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3797: double *lb;
3798: double minf; /* the minimum objective value, upon return */
3799: double * p1; /* Shifted parameters from 0 instead of 1 */
3800: myfunc_data dinst, *d = &dinst;
3801: #endif
3802:
3803:
1.126 brouard 3804: xi=matrix(1,npar,1,npar);
3805: for (i=1;i<=npar;i++)
3806: for (j=1;j<=npar;j++)
3807: xi[i][j]=(i==j ? 1.0 : 0.0);
3808: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3809: strcpy(filerespow,"POW_");
1.126 brouard 3810: strcat(filerespow,fileres);
3811: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3812: printf("Problem with resultfile: %s\n", filerespow);
3813: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3814: }
3815: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3816: for (i=1;i<=nlstate;i++)
3817: for(j=1;j<=nlstate+ndeath;j++)
3818: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3819: fprintf(ficrespow,"\n");
1.162 brouard 3820: #ifdef POWELL
1.126 brouard 3821: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3822: #endif
1.126 brouard 3823:
1.162 brouard 3824: #ifdef NLOPT
3825: #ifdef NEWUOA
3826: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3827: #else
3828: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3829: #endif
3830: lb=vector(0,npar-1);
3831: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3832: nlopt_set_lower_bounds(opt, lb);
3833: nlopt_set_initial_step1(opt, 0.1);
3834:
3835: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3836: d->function = func;
3837: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3838: nlopt_set_min_objective(opt, myfunc, d);
3839: nlopt_set_xtol_rel(opt, ftol);
3840: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3841: printf("nlopt failed! %d\n",creturn);
3842: }
3843: else {
3844: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3845: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3846: iter=1; /* not equal */
3847: }
3848: nlopt_destroy(opt);
3849: #endif
1.126 brouard 3850: free_matrix(xi,1,npar,1,npar);
3851: fclose(ficrespow);
1.203 brouard 3852: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3853: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3854: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3855:
3856: }
3857:
3858: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3859: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3860: {
3861: double **a,**y,*x,pd;
1.203 brouard 3862: /* double **hess; */
1.164 brouard 3863: int i, j;
1.126 brouard 3864: int *indx;
3865:
3866: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3867: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3868: void lubksb(double **a, int npar, int *indx, double b[]) ;
3869: void ludcmp(double **a, int npar, int *indx, double *d) ;
3870: double gompertz(double p[]);
1.203 brouard 3871: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3872:
3873: printf("\nCalculation of the hessian matrix. Wait...\n");
3874: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3875: for (i=1;i<=npar;i++){
1.203 brouard 3876: printf("%d-",i);fflush(stdout);
3877: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3878:
3879: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3880:
3881: /* printf(" %f ",p[i]);
3882: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3883: }
3884:
3885: for (i=1;i<=npar;i++) {
3886: for (j=1;j<=npar;j++) {
3887: if (j>i) {
1.203 brouard 3888: printf(".%d-%d",i,j);fflush(stdout);
3889: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3890: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3891:
3892: hess[j][i]=hess[i][j];
3893: /*printf(" %lf ",hess[i][j]);*/
3894: }
3895: }
3896: }
3897: printf("\n");
3898: fprintf(ficlog,"\n");
3899:
3900: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3901: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3902:
3903: a=matrix(1,npar,1,npar);
3904: y=matrix(1,npar,1,npar);
3905: x=vector(1,npar);
3906: indx=ivector(1,npar);
3907: for (i=1;i<=npar;i++)
3908: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3909: ludcmp(a,npar,indx,&pd);
3910:
3911: for (j=1;j<=npar;j++) {
3912: for (i=1;i<=npar;i++) x[i]=0;
3913: x[j]=1;
3914: lubksb(a,npar,indx,x);
3915: for (i=1;i<=npar;i++){
3916: matcov[i][j]=x[i];
3917: }
3918: }
3919:
3920: printf("\n#Hessian matrix#\n");
3921: fprintf(ficlog,"\n#Hessian matrix#\n");
3922: for (i=1;i<=npar;i++) {
3923: for (j=1;j<=npar;j++) {
1.203 brouard 3924: printf("%.6e ",hess[i][j]);
3925: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3926: }
3927: printf("\n");
3928: fprintf(ficlog,"\n");
3929: }
3930:
1.203 brouard 3931: /* printf("\n#Covariance matrix#\n"); */
3932: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3933: /* for (i=1;i<=npar;i++) { */
3934: /* for (j=1;j<=npar;j++) { */
3935: /* printf("%.6e ",matcov[i][j]); */
3936: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3937: /* } */
3938: /* printf("\n"); */
3939: /* fprintf(ficlog,"\n"); */
3940: /* } */
3941:
1.126 brouard 3942: /* Recompute Inverse */
1.203 brouard 3943: /* for (i=1;i<=npar;i++) */
3944: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3945: /* ludcmp(a,npar,indx,&pd); */
3946:
3947: /* printf("\n#Hessian matrix recomputed#\n"); */
3948:
3949: /* for (j=1;j<=npar;j++) { */
3950: /* for (i=1;i<=npar;i++) x[i]=0; */
3951: /* x[j]=1; */
3952: /* lubksb(a,npar,indx,x); */
3953: /* for (i=1;i<=npar;i++){ */
3954: /* y[i][j]=x[i]; */
3955: /* printf("%.3e ",y[i][j]); */
3956: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3957: /* } */
3958: /* printf("\n"); */
3959: /* fprintf(ficlog,"\n"); */
3960: /* } */
3961:
3962: /* Verifying the inverse matrix */
3963: #ifdef DEBUGHESS
3964: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3965:
1.203 brouard 3966: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3967: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3968:
3969: for (j=1;j<=npar;j++) {
3970: for (i=1;i<=npar;i++){
1.203 brouard 3971: printf("%.2f ",y[i][j]);
3972: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3973: }
3974: printf("\n");
3975: fprintf(ficlog,"\n");
3976: }
1.203 brouard 3977: #endif
1.126 brouard 3978:
3979: free_matrix(a,1,npar,1,npar);
3980: free_matrix(y,1,npar,1,npar);
3981: free_vector(x,1,npar);
3982: free_ivector(indx,1,npar);
1.203 brouard 3983: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3984:
3985:
3986: }
3987:
3988: /*************** hessian matrix ****************/
3989: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3990: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3991: int i;
3992: int l=1, lmax=20;
1.203 brouard 3993: double k1,k2, res, fx;
1.132 brouard 3994: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3995: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3996: int k=0,kmax=10;
3997: double l1;
3998:
3999: fx=func(x);
4000: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4001: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4002: l1=pow(10,l);
4003: delts=delt;
4004: for(k=1 ; k <kmax; k=k+1){
4005: delt = delta*(l1*k);
4006: p2[theta]=x[theta] +delt;
1.145 brouard 4007: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4008: p2[theta]=x[theta]-delt;
4009: k2=func(p2)-fx;
4010: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4011: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4012:
1.203 brouard 4013: #ifdef DEBUGHESSII
1.126 brouard 4014: 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);
4015: 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);
4016: #endif
4017: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4018: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4019: k=kmax;
4020: }
4021: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4022: k=kmax; l=lmax*10;
1.126 brouard 4023: }
4024: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4025: delts=delt;
4026: }
1.203 brouard 4027: } /* End loop k */
1.126 brouard 4028: }
4029: delti[theta]=delts;
4030: return res;
4031:
4032: }
4033:
1.203 brouard 4034: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4035: {
4036: int i;
1.164 brouard 4037: int l=1, lmax=20;
1.126 brouard 4038: double k1,k2,k3,k4,res,fx;
1.132 brouard 4039: double p2[MAXPARM+1];
1.203 brouard 4040: int k, kmax=1;
4041: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4042:
4043: int firstime=0;
1.203 brouard 4044:
1.126 brouard 4045: fx=func(x);
1.203 brouard 4046: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4047: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4048: p2[thetai]=x[thetai]+delti[thetai]*k;
4049: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4050: k1=func(p2)-fx;
4051:
1.203 brouard 4052: p2[thetai]=x[thetai]+delti[thetai]*k;
4053: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4054: k2=func(p2)-fx;
4055:
1.203 brouard 4056: p2[thetai]=x[thetai]-delti[thetai]*k;
4057: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4058: k3=func(p2)-fx;
4059:
1.203 brouard 4060: p2[thetai]=x[thetai]-delti[thetai]*k;
4061: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4062: k4=func(p2)-fx;
1.203 brouard 4063: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4064: if(k1*k2*k3*k4 <0.){
1.208 brouard 4065: firstime=1;
1.203 brouard 4066: kmax=kmax+10;
1.208 brouard 4067: }
4068: if(kmax >=10 || firstime ==1){
1.246 brouard 4069: 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);
4070: 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 4071: 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);
4072: 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);
4073: }
4074: #ifdef DEBUGHESSIJ
4075: v1=hess[thetai][thetai];
4076: v2=hess[thetaj][thetaj];
4077: cv12=res;
4078: /* Computing eigen value of Hessian matrix */
4079: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4080: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4081: if ((lc2 <0) || (lc1 <0) ){
4082: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4083: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4084: 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);
4085: 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);
4086: }
1.126 brouard 4087: #endif
4088: }
4089: return res;
4090: }
4091:
1.203 brouard 4092: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4093: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4094: /* { */
4095: /* int i; */
4096: /* int l=1, lmax=20; */
4097: /* double k1,k2,k3,k4,res,fx; */
4098: /* double p2[MAXPARM+1]; */
4099: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4100: /* int k=0,kmax=10; */
4101: /* double l1; */
4102:
4103: /* fx=func(x); */
4104: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4105: /* l1=pow(10,l); */
4106: /* delts=delt; */
4107: /* for(k=1 ; k <kmax; k=k+1){ */
4108: /* delt = delti*(l1*k); */
4109: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4110: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4111: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4112: /* k1=func(p2)-fx; */
4113:
4114: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4115: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4116: /* k2=func(p2)-fx; */
4117:
4118: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4119: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4120: /* k3=func(p2)-fx; */
4121:
4122: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4123: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4124: /* k4=func(p2)-fx; */
4125: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4126: /* #ifdef DEBUGHESSIJ */
4127: /* 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); */
4128: /* 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); */
4129: /* #endif */
4130: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4131: /* k=kmax; */
4132: /* } */
4133: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4134: /* k=kmax; l=lmax*10; */
4135: /* } */
4136: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4137: /* delts=delt; */
4138: /* } */
4139: /* } /\* End loop k *\/ */
4140: /* } */
4141: /* delti[theta]=delts; */
4142: /* return res; */
4143: /* } */
4144:
4145:
1.126 brouard 4146: /************** Inverse of matrix **************/
4147: void ludcmp(double **a, int n, int *indx, double *d)
4148: {
4149: int i,imax,j,k;
4150: double big,dum,sum,temp;
4151: double *vv;
4152:
4153: vv=vector(1,n);
4154: *d=1.0;
4155: for (i=1;i<=n;i++) {
4156: big=0.0;
4157: for (j=1;j<=n;j++)
4158: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4159: if (big == 0.0){
4160: printf(" Singular Hessian matrix at row %d:\n",i);
4161: for (j=1;j<=n;j++) {
4162: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4163: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4164: }
4165: fflush(ficlog);
4166: fclose(ficlog);
4167: nrerror("Singular matrix in routine ludcmp");
4168: }
1.126 brouard 4169: vv[i]=1.0/big;
4170: }
4171: for (j=1;j<=n;j++) {
4172: for (i=1;i<j;i++) {
4173: sum=a[i][j];
4174: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4175: a[i][j]=sum;
4176: }
4177: big=0.0;
4178: for (i=j;i<=n;i++) {
4179: sum=a[i][j];
4180: for (k=1;k<j;k++)
4181: sum -= a[i][k]*a[k][j];
4182: a[i][j]=sum;
4183: if ( (dum=vv[i]*fabs(sum)) >= big) {
4184: big=dum;
4185: imax=i;
4186: }
4187: }
4188: if (j != imax) {
4189: for (k=1;k<=n;k++) {
4190: dum=a[imax][k];
4191: a[imax][k]=a[j][k];
4192: a[j][k]=dum;
4193: }
4194: *d = -(*d);
4195: vv[imax]=vv[j];
4196: }
4197: indx[j]=imax;
4198: if (a[j][j] == 0.0) a[j][j]=TINY;
4199: if (j != n) {
4200: dum=1.0/(a[j][j]);
4201: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4202: }
4203: }
4204: free_vector(vv,1,n); /* Doesn't work */
4205: ;
4206: }
4207:
4208: void lubksb(double **a, int n, int *indx, double b[])
4209: {
4210: int i,ii=0,ip,j;
4211: double sum;
4212:
4213: for (i=1;i<=n;i++) {
4214: ip=indx[i];
4215: sum=b[ip];
4216: b[ip]=b[i];
4217: if (ii)
4218: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4219: else if (sum) ii=i;
4220: b[i]=sum;
4221: }
4222: for (i=n;i>=1;i--) {
4223: sum=b[i];
4224: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4225: b[i]=sum/a[i][i];
4226: }
4227: }
4228:
4229: void pstamp(FILE *fichier)
4230: {
1.196 brouard 4231: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4232: }
4233:
1.253 brouard 4234: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4235:
4236: /* y=a+bx regression */
4237: double sumx = 0.0; /* sum of x */
4238: double sumx2 = 0.0; /* sum of x**2 */
4239: double sumxy = 0.0; /* sum of x * y */
4240: double sumy = 0.0; /* sum of y */
4241: double sumy2 = 0.0; /* sum of y**2 */
4242: double sume2; /* sum of square or residuals */
4243: double yhat;
4244:
4245: double denom=0;
4246: int i;
4247: int ne=*no;
4248:
4249: for ( i=ifi, ne=0;i<=ila;i++) {
4250: if(!isfinite(x[i]) || !isfinite(y[i])){
4251: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4252: continue;
4253: }
4254: ne=ne+1;
4255: sumx += x[i];
4256: sumx2 += x[i]*x[i];
4257: sumxy += x[i] * y[i];
4258: sumy += y[i];
4259: sumy2 += y[i]*y[i];
4260: denom = (ne * sumx2 - sumx*sumx);
4261: /* 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); */
4262: }
4263:
4264: denom = (ne * sumx2 - sumx*sumx);
4265: if (denom == 0) {
4266: // vertical, slope m is infinity
4267: *b = INFINITY;
4268: *a = 0;
4269: if (r) *r = 0;
4270: return 1;
4271: }
4272:
4273: *b = (ne * sumxy - sumx * sumy) / denom;
4274: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4275: if (r!=NULL) {
4276: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4277: sqrt((sumx2 - sumx*sumx/ne) *
4278: (sumy2 - sumy*sumy/ne));
4279: }
4280: *no=ne;
4281: for ( i=ifi, ne=0;i<=ila;i++) {
4282: if(!isfinite(x[i]) || !isfinite(y[i])){
4283: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4284: continue;
4285: }
4286: ne=ne+1;
4287: yhat = y[i] - *a -*b* x[i];
4288: sume2 += yhat * yhat ;
4289:
4290: denom = (ne * sumx2 - sumx*sumx);
4291: /* 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); */
4292: }
4293: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4294: *sa= *sb * sqrt(sumx2/ne);
4295:
4296: return 0;
4297: }
4298:
1.126 brouard 4299: /************ Frequencies ********************/
1.251 brouard 4300: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4301: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4302: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4303: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4304:
1.253 brouard 4305: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4306: int iind=0, iage=0;
4307: int mi; /* Effective wave */
4308: int first;
4309: double ***freq; /* Frequencies */
1.253 brouard 4310: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4311: int no;
1.226 brouard 4312: double *meanq;
4313: double **meanqt;
4314: double *pp, **prop, *posprop, *pospropt;
4315: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4316: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4317: double agebegin, ageend;
4318:
4319: pp=vector(1,nlstate);
1.251 brouard 4320: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4321: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4322: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4323: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4324: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4325: meanqt=matrix(1,lastpass,1,nqtveff);
4326: strcpy(fileresp,"P_");
4327: strcat(fileresp,fileresu);
4328: /*strcat(fileresphtm,fileresu);*/
4329: if((ficresp=fopen(fileresp,"w"))==NULL) {
4330: printf("Problem with prevalence resultfile: %s\n", fileresp);
4331: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4332: exit(0);
4333: }
1.240 brouard 4334:
1.226 brouard 4335: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4336: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4337: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4338: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4339: fflush(ficlog);
4340: exit(70);
4341: }
4342: else{
4343: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4344: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4345: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4346: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4347: }
1.237 brouard 4348: 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 4349:
1.226 brouard 4350: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4351: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4352: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4353: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4354: fflush(ficlog);
4355: exit(70);
1.240 brouard 4356: } else{
1.226 brouard 4357: 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 4358: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4359: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4360: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4361: }
1.240 brouard 4362: 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);
4363:
1.253 brouard 4364: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4365: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4366: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4367: j1=0;
1.126 brouard 4368:
1.227 brouard 4369: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4370: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4371: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4372:
4373:
1.226 brouard 4374: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4375: reference=low_education V1=0,V2=0
4376: med_educ V1=1 V2=0,
4377: high_educ V1=0 V2=1
4378: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4379: */
1.249 brouard 4380: dateintsum=0;
4381: k2cpt=0;
4382:
1.253 brouard 4383: if(cptcoveff == 0 )
4384: nl=1; /* Constant model only */
4385: else
4386: nl=2;
4387: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4388: if(nj==1)
4389: j=0; /* First pass for the constant */
4390: else
4391: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4392: first=1;
4393: 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 */
4394: posproptt=0.;
4395: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4396: scanf("%d", i);*/
4397: for (i=-5; i<=nlstate+ndeath; i++)
4398: for (jk=-5; jk<=nlstate+ndeath; jk++)
4399: for(m=iagemin; m <= iagemax+3; m++)
4400: freq[i][jk][m]=0;
4401:
4402: for (i=1; i<=nlstate; i++) {
1.240 brouard 4403: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4404: prop[i][m]=0;
4405: posprop[i]=0;
4406: pospropt[i]=0;
4407: }
4408: /* for (z1=1; z1<= nqfveff; z1++) { */
4409: /* meanq[z1]+=0.; */
4410: /* for(m=1;m<=lastpass;m++){ */
4411: /* meanqt[m][z1]=0.; */
4412: /* } */
4413: /* } */
4414:
4415: /* dateintsum=0; */
4416: /* k2cpt=0; */
4417:
4418: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4419: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4420: bool=1;
4421: if(j !=0){
4422: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4423: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4424: /* for (z1=1; z1<= nqfveff; z1++) { */
4425: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4426: /* } */
4427: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4428: /* if(Tvaraff[z1] ==-20){ */
4429: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4430: /* }else if(Tvaraff[z1] ==-10){ */
4431: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4432: /* }else */
4433: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4434: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4435: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4436: /* 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",
4437: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4438: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4439: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4440: } /* Onlyf fixed */
4441: } /* end z1 */
4442: } /* cptcovn > 0 */
4443: } /* end any */
4444: }/* end j==0 */
4445: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4446: /* for(m=firstpass; m<=lastpass; m++){ */
4447: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4448: m=mw[mi][iind];
4449: if(j!=0){
4450: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4451: for (z1=1; z1<=cptcoveff; z1++) {
4452: if( Fixed[Tmodelind[z1]]==1){
4453: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4454: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4455: value is -1, we don't select. It differs from the
4456: constant and age model which counts them. */
4457: bool=0; /* not selected */
4458: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4459: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4460: bool=0;
4461: }
4462: }
4463: }
4464: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4465: } /* end j==0 */
4466: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4467: if(bool==1){
4468: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4469: and mw[mi+1][iind]. dh depends on stepm. */
4470: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4471: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4472: if(m >=firstpass && m <=lastpass){
4473: k2=anint[m][iind]+(mint[m][iind]/12.);
4474: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4475: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4476: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4477: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4478: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4479: if (m<lastpass) {
4480: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4481: /* 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]); */
4482: if(s[m][iind]==-1)
4483: 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.));
4484: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4485: /* if((int)agev[m][iind] == 55) */
4486: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4487: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4488: 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 4489: }
1.251 brouard 4490: } /* end if between passes */
4491: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4492: dateintsum=dateintsum+k2; /* on all covariates ?*/
4493: k2cpt++;
4494: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4495: }
1.251 brouard 4496: }else{
4497: bool=1;
4498: }/* end bool 2 */
4499: } /* end m */
4500: } /* end bool */
4501: } /* end iind = 1 to imx */
4502: /* prop[s][age] is feeded for any initial and valid live state as well as
4503: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4504:
4505:
4506: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4507: pstamp(ficresp);
4508: if (cptcoveff>0 && j!=0){
4509: printf( "\n#********** Variable ");
4510: fprintf(ficresp, "\n#********** Variable ");
4511: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4512: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4513: fprintf(ficlog, "\n#********** Variable ");
4514: for (z1=1; z1<=cptcoveff; z1++){
4515: if(!FixedV[Tvaraff[z1]]){
4516: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4517: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4518: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4519: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4520: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4521: }else{
1.251 brouard 4522: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4523: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4524: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4525: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4526: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4527: }
4528: }
4529: printf( "**********\n#");
4530: fprintf(ficresp, "**********\n#");
4531: fprintf(ficresphtm, "**********</h3>\n");
4532: fprintf(ficresphtmfr, "**********</h3>\n");
4533: fprintf(ficlog, "**********\n");
4534: }
4535: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4536: for(i=1; i<=nlstate;i++) {
4537: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4538: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4539: }
4540: fprintf(ficresp, "\n");
4541: fprintf(ficresphtm, "\n");
4542:
4543: /* Header of frequency table by age */
4544: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4545: fprintf(ficresphtmfr,"<th>Age</th> ");
4546: for(jk=-1; jk <=nlstate+ndeath; jk++){
4547: for(m=-1; m <=nlstate+ndeath; m++){
4548: if(jk!=0 && m!=0)
4549: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4550: }
1.226 brouard 4551: }
1.251 brouard 4552: fprintf(ficresphtmfr, "\n");
4553:
4554: /* For each age */
4555: for(iage=iagemin; iage <= iagemax+3; iage++){
4556: fprintf(ficresphtm,"<tr>");
4557: if(iage==iagemax+1){
4558: fprintf(ficlog,"1");
4559: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4560: }else if(iage==iagemax+2){
4561: fprintf(ficlog,"0");
4562: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4563: }else if(iage==iagemax+3){
4564: fprintf(ficlog,"Total");
4565: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4566: }else{
1.240 brouard 4567: if(first==1){
1.251 brouard 4568: first=0;
4569: printf("See log file for details...\n");
4570: }
4571: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4572: fprintf(ficlog,"Age %d", iage);
4573: }
4574: for(jk=1; jk <=nlstate ; jk++){
4575: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4576: pp[jk] += freq[jk][m][iage];
4577: }
4578: for(jk=1; jk <=nlstate ; jk++){
4579: for(m=-1, pos=0; m <=0 ; m++)
4580: pos += freq[jk][m][iage];
4581: if(pp[jk]>=1.e-10){
4582: if(first==1){
4583: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4584: }
4585: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4586: }else{
4587: if(first==1)
4588: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4589: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4590: }
4591: }
4592:
1.251 brouard 4593: for(jk=1; jk <=nlstate ; jk++){
4594: /* posprop[jk]=0; */
4595: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4596: pp[jk] += freq[jk][m][iage];
4597: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4598:
4599: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4600: pos += pp[jk]; /* pos is the total number of transitions until this age */
4601: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4602: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4603: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4604: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4605: }
1.251 brouard 4606: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4607: if(pos>=1.e-5){
1.251 brouard 4608: if(first==1)
4609: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4610: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4611: }else{
4612: if(first==1)
4613: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4614: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4615: }
4616: if( iage <= iagemax){
4617: if(pos>=1.e-5){
4618: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4619: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4620: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4621: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4622: }
4623: else{
4624: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4625: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4626: }
1.240 brouard 4627: }
1.251 brouard 4628: pospropt[jk] +=posprop[jk];
4629: } /* end loop jk */
4630: /* pospropt=0.; */
4631: for(jk=-1; jk <=nlstate+ndeath; jk++){
4632: for(m=-1; m <=nlstate+ndeath; m++){
4633: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4634: if(first==1){
4635: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4636: }
1.253 brouard 4637: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4638: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4639: }
4640: if(jk!=0 && m!=0)
4641: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4642: }
1.251 brouard 4643: } /* end loop jk */
4644: posproptt=0.;
4645: for(jk=1; jk <=nlstate; jk++){
4646: posproptt += pospropt[jk];
4647: }
4648: fprintf(ficresphtmfr,"</tr>\n ");
4649: if(iage <= iagemax){
4650: fprintf(ficresp,"\n");
4651: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4652: }
1.251 brouard 4653: if(first==1)
4654: printf("Others in log...\n");
4655: fprintf(ficlog,"\n");
4656: } /* end loop age iage */
4657: fprintf(ficresphtm,"<tr><th>Tot</th>");
4658: for(jk=1; jk <=nlstate ; jk++){
4659: if(posproptt < 1.e-5){
4660: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4661: }else{
4662: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4663: }
1.226 brouard 4664: }
1.251 brouard 4665: fprintf(ficresphtm,"</tr>\n");
4666: fprintf(ficresphtm,"</table>\n");
4667: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4668: if(posproptt < 1.e-5){
1.251 brouard 4669: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4670: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4671: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4672: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4673: invalidvarcomb[j1]=1;
1.226 brouard 4674: }else{
1.251 brouard 4675: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4676: invalidvarcomb[j1]=0;
1.226 brouard 4677: }
1.251 brouard 4678: fprintf(ficresphtmfr,"</table>\n");
4679: fprintf(ficlog,"\n");
4680: if(j!=0){
4681: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4682: for(i=1,jk=1; i <=nlstate; i++){
4683: for(k=1; k <=(nlstate+ndeath); k++){
4684: if (k != i) {
4685: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4686: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4687: if(j1==1){ /* All dummy covariates to zero */
4688: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4689: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4690: printf("%d%d ",i,k);
4691: fprintf(ficlog,"%d%d ",i,k);
4692: 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]));
4693: 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]));
4694: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4695: }
1.253 brouard 4696: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4697: for(iage=iagemin; iage <= iagemax+3; iage++){
4698: x[iage]= (double)iage;
4699: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4700: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4701: }
4702: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4703: pstart[jk]=b;
4704: pstart[jk-1]=a;
1.252 brouard 4705: }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 */
4706: 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]);
4707: 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 4708: 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 4709: printf("%d%d ",i,k);
4710: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4711: 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]));
4712: }else{ /* Other cases, like quantitative fixed or varying covariates */
4713: ;
4714: }
4715: /* printf("%12.7f )", param[i][jj][k]); */
4716: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4717: jk++;
4718: } /* end jj */
4719: } /* end k!= i */
4720: } /* end k */
4721: } /* end i, jk */
4722: } /* end j !=0 */
4723: } /* end selected combination of covariate j1 */
4724: if(j==0){ /* We can estimate starting values from the occurences in each case */
4725: printf("#Freqsummary: Starting values for the constants:\n");
4726: fprintf(ficlog,"\n");
4727: for(i=1,jk=1; i <=nlstate; i++){
4728: for(k=1; k <=(nlstate+ndeath); k++){
4729: if (k != i) {
4730: printf("%d%d ",i,k);
4731: fprintf(ficlog,"%d%d ",i,k);
4732: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4733: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4734: if(jj==1){ /* Age has to be done */
4735: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4736: 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]));
4737: 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]));
4738: }
4739: /* printf("%12.7f )", param[i][jj][k]); */
4740: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4741: jk++;
1.250 brouard 4742: }
1.251 brouard 4743: printf("\n");
4744: fprintf(ficlog,"\n");
1.250 brouard 4745: }
4746: }
4747: }
1.251 brouard 4748: printf("#Freqsummary\n");
4749: fprintf(ficlog,"\n");
4750: for(jk=-1; jk <=nlstate+ndeath; jk++){
4751: for(m=-1; m <=nlstate+ndeath; m++){
4752: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4753: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4754: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4755: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4756: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4757: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4758: /* } */
4759: }
4760: } /* end loop jk */
4761:
4762: printf("\n");
4763: fprintf(ficlog,"\n");
4764: } /* end j=0 */
1.249 brouard 4765: } /* end j */
1.252 brouard 4766:
1.253 brouard 4767: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4768: for(i=1, jk=1; i <=nlstate; i++){
4769: for(j=1; j <=nlstate+ndeath; j++){
4770: if(j!=i){
4771: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4772: printf("%1d%1d",i,j);
4773: fprintf(ficparo,"%1d%1d",i,j);
4774: for(k=1; k<=ncovmodel;k++){
4775: /* printf(" %lf",param[i][j][k]); */
4776: /* fprintf(ficparo," %lf",param[i][j][k]); */
4777: p[jk]=pstart[jk];
4778: printf(" %f ",pstart[jk]);
4779: fprintf(ficparo," %f ",pstart[jk]);
4780: jk++;
4781: }
4782: printf("\n");
4783: fprintf(ficparo,"\n");
4784: }
4785: }
4786: }
4787: } /* end mle=-2 */
1.226 brouard 4788: dateintmean=dateintsum/k2cpt;
1.240 brouard 4789:
1.226 brouard 4790: fclose(ficresp);
4791: fclose(ficresphtm);
4792: fclose(ficresphtmfr);
4793: free_vector(meanq,1,nqfveff);
4794: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4795: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4796: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4797: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4798: free_vector(pospropt,1,nlstate);
4799: free_vector(posprop,1,nlstate);
1.251 brouard 4800: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4801: free_vector(pp,1,nlstate);
4802: /* End of freqsummary */
4803: }
1.126 brouard 4804:
4805: /************ Prevalence ********************/
1.227 brouard 4806: 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)
4807: {
4808: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4809: in each health status at the date of interview (if between dateprev1 and dateprev2).
4810: We still use firstpass and lastpass as another selection.
4811: */
1.126 brouard 4812:
1.227 brouard 4813: int i, m, jk, j1, bool, z1,j, iv;
4814: int mi; /* Effective wave */
4815: int iage;
4816: double agebegin, ageend;
4817:
4818: double **prop;
4819: double posprop;
4820: double y2; /* in fractional years */
4821: int iagemin, iagemax;
4822: int first; /** to stop verbosity which is redirected to log file */
4823:
4824: iagemin= (int) agemin;
4825: iagemax= (int) agemax;
4826: /*pp=vector(1,nlstate);*/
1.251 brouard 4827: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4828: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4829: j1=0;
1.222 brouard 4830:
1.227 brouard 4831: /*j=cptcoveff;*/
4832: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4833:
1.227 brouard 4834: first=1;
4835: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4836: for (i=1; i<=nlstate; i++)
1.251 brouard 4837: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4838: prop[i][iage]=0.0;
4839: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4840: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4841: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4842:
4843: for (i=1; i<=imx; i++) { /* Each individual */
4844: bool=1;
4845: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4846: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4847: m=mw[mi][i];
4848: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4849: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4850: for (z1=1; z1<=cptcoveff; z1++){
4851: if( Fixed[Tmodelind[z1]]==1){
4852: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4853: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4854: bool=0;
4855: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4856: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4857: bool=0;
4858: }
4859: }
4860: if(bool==1){ /* Otherwise we skip that wave/person */
4861: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4862: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4863: if(m >=firstpass && m <=lastpass){
4864: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4865: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4866: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4867: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4868: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4869: 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);
4870: exit(1);
4871: }
4872: if (s[m][i]>0 && s[m][i]<=nlstate) {
4873: /*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]]);*/
4874: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4875: prop[s[m][i]][iagemax+3] += weight[i];
4876: } /* end valid statuses */
4877: } /* end selection of dates */
4878: } /* end selection of waves */
4879: } /* end bool */
4880: } /* end wave */
4881: } /* end individual */
4882: for(i=iagemin; i <= iagemax+3; i++){
4883: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4884: posprop += prop[jk][i];
4885: }
4886:
4887: for(jk=1; jk <=nlstate ; jk++){
4888: if( i <= iagemax){
4889: if(posprop>=1.e-5){
4890: probs[i][jk][j1]= prop[jk][i]/posprop;
4891: } else{
4892: if(first==1){
4893: first=0;
4894: 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]);
4895: }
4896: }
4897: }
4898: }/* end jk */
4899: }/* end i */
1.222 brouard 4900: /*} *//* end i1 */
1.227 brouard 4901: } /* end j1 */
1.222 brouard 4902:
1.227 brouard 4903: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4904: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4905: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4906: } /* End of prevalence */
1.126 brouard 4907:
4908: /************* Waves Concatenation ***************/
4909:
4910: 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)
4911: {
4912: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4913: Death is a valid wave (if date is known).
4914: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4915: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4916: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4917: */
1.126 brouard 4918:
1.224 brouard 4919: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4920: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4921: double sum=0., jmean=0.;*/
1.224 brouard 4922: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4923: int j, k=0,jk, ju, jl;
4924: double sum=0.;
4925: first=0;
1.214 brouard 4926: firstwo=0;
1.217 brouard 4927: firsthree=0;
1.218 brouard 4928: firstfour=0;
1.164 brouard 4929: jmin=100000;
1.126 brouard 4930: jmax=-1;
4931: jmean=0.;
1.224 brouard 4932:
4933: /* Treating live states */
1.214 brouard 4934: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4935: mi=0; /* First valid wave */
1.227 brouard 4936: mli=0; /* Last valid wave */
1.126 brouard 4937: m=firstpass;
1.214 brouard 4938: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4939: 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 */
4940: mli=m-1;/* mw[++mi][i]=m-1; */
4941: }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 */
4942: mw[++mi][i]=m;
4943: mli=m;
1.224 brouard 4944: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4945: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4946: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4947: }
1.227 brouard 4948: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4949: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4950: break;
1.224 brouard 4951: #else
1.227 brouard 4952: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4953: if(firsthree == 0){
1.262 brouard 4954: 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 1-p%d%d .\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, s[m][i], nlstate+ndeath);
1.227 brouard 4955: firsthree=1;
4956: }
1.262 brouard 4957: 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 1-p%d%d .\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, s[m][i], nlstate+ndeath);
1.227 brouard 4958: mw[++mi][i]=m;
4959: mli=m;
4960: }
4961: if(s[m][i]==-2){ /* Vital status is really unknown */
4962: nbwarn++;
4963: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4964: 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);
4965: 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);
4966: }
4967: break;
4968: }
4969: break;
1.224 brouard 4970: #endif
1.227 brouard 4971: }/* End m >= lastpass */
1.126 brouard 4972: }/* end while */
1.224 brouard 4973:
1.227 brouard 4974: /* 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 4975: /* After last pass */
1.224 brouard 4976: /* Treating death states */
1.214 brouard 4977: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4978: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4979: /* } */
1.126 brouard 4980: mi++; /* Death is another wave */
4981: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4982: /* Only death is a correct wave */
1.126 brouard 4983: mw[mi][i]=m;
1.257 brouard 4984: } /* else not in a death state */
1.224 brouard 4985: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 4986: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 4987: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4988: 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 */
4989: nbwarn++;
4990: if(firstfiv==0){
4991: 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 );
4992: firstfiv=1;
4993: }else{
4994: 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 );
4995: }
4996: }else{ /* Death occured afer last wave potential bias */
4997: nberr++;
4998: if(firstwo==0){
1.257 brouard 4999: 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 5000: firstwo=1;
5001: }
1.257 brouard 5002: 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 5003: }
1.257 brouard 5004: }else{ /* if date of interview is unknown */
1.227 brouard 5005: /* death is known but not confirmed by death status at any wave */
5006: if(firstfour==0){
5007: 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 );
5008: firstfour=1;
5009: }
5010: 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 5011: }
1.224 brouard 5012: } /* end if date of death is known */
5013: #endif
5014: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5015: /* wav[i]=mw[mi][i]; */
1.126 brouard 5016: if(mi==0){
5017: nbwarn++;
5018: if(first==0){
1.227 brouard 5019: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5020: first=1;
1.126 brouard 5021: }
5022: if(first==1){
1.227 brouard 5023: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5024: }
5025: } /* end mi==0 */
5026: } /* End individuals */
1.214 brouard 5027: /* wav and mw are no more changed */
1.223 brouard 5028:
1.214 brouard 5029:
1.126 brouard 5030: for(i=1; i<=imx; i++){
5031: for(mi=1; mi<wav[i];mi++){
5032: if (stepm <=0)
1.227 brouard 5033: dh[mi][i]=1;
1.126 brouard 5034: else{
1.260 brouard 5035: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5036: if (agedc[i] < 2*AGESUP) {
5037: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5038: if(j==0) j=1; /* Survives at least one month after exam */
5039: else if(j<0){
5040: nberr++;
5041: 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]);
5042: j=1; /* Temporary Dangerous patch */
5043: 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);
5044: 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]);
5045: 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);
5046: }
5047: k=k+1;
5048: if (j >= jmax){
5049: jmax=j;
5050: ijmax=i;
5051: }
5052: if (j <= jmin){
5053: jmin=j;
5054: ijmin=i;
5055: }
5056: sum=sum+j;
5057: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5058: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5059: }
5060: }
5061: else{
5062: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5063: /* 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 5064:
1.227 brouard 5065: k=k+1;
5066: if (j >= jmax) {
5067: jmax=j;
5068: ijmax=i;
5069: }
5070: else if (j <= jmin){
5071: jmin=j;
5072: ijmin=i;
5073: }
5074: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5075: /*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]);*/
5076: if(j<0){
5077: nberr++;
5078: 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]);
5079: 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]);
5080: }
5081: sum=sum+j;
5082: }
5083: jk= j/stepm;
5084: jl= j -jk*stepm;
5085: ju= j -(jk+1)*stepm;
5086: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5087: if(jl==0){
5088: dh[mi][i]=jk;
5089: bh[mi][i]=0;
5090: }else{ /* We want a negative bias in order to only have interpolation ie
5091: * to avoid the price of an extra matrix product in likelihood */
5092: dh[mi][i]=jk+1;
5093: bh[mi][i]=ju;
5094: }
5095: }else{
5096: if(jl <= -ju){
5097: dh[mi][i]=jk;
5098: bh[mi][i]=jl; /* bias is positive if real duration
5099: * is higher than the multiple of stepm and negative otherwise.
5100: */
5101: }
5102: else{
5103: dh[mi][i]=jk+1;
5104: bh[mi][i]=ju;
5105: }
5106: if(dh[mi][i]==0){
5107: dh[mi][i]=1; /* At least one step */
5108: bh[mi][i]=ju; /* At least one step */
5109: /* 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);*/
5110: }
5111: } /* end if mle */
1.126 brouard 5112: }
5113: } /* end wave */
5114: }
5115: jmean=sum/k;
5116: 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 5117: 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 5118: }
1.126 brouard 5119:
5120: /*********** Tricode ****************************/
1.220 brouard 5121: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5122: {
5123: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5124: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5125: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5126: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5127: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5128: */
1.130 brouard 5129:
1.242 brouard 5130: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5131: int modmaxcovj=0; /* Modality max of covariates j */
5132: int cptcode=0; /* Modality max of covariates j */
5133: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5134:
5135:
1.242 brouard 5136: /* cptcoveff=0; */
5137: /* *cptcov=0; */
1.126 brouard 5138:
1.242 brouard 5139: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5140:
1.242 brouard 5141: /* Loop on covariates without age and products and no quantitative variable */
5142: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5143: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5144: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5145: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5146: switch(Fixed[k]) {
5147: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5148: 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*/
5149: ij=(int)(covar[Tvar[k]][i]);
5150: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5151: * If product of Vn*Vm, still boolean *:
5152: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5153: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5154: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5155: modality of the nth covariate of individual i. */
5156: if (ij > modmaxcovj)
5157: modmaxcovj=ij;
5158: else if (ij < modmincovj)
5159: modmincovj=ij;
5160: if ((ij < -1) && (ij > NCOVMAX)){
5161: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5162: exit(1);
5163: }else
5164: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5165: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5166: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5167: /* getting the maximum value of the modality of the covariate
5168: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5169: female ies 1, then modmaxcovj=1.
5170: */
5171: } /* end for loop on individuals i */
5172: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5173: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5174: cptcode=modmaxcovj;
5175: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5176: /*for (i=0; i<=cptcode; i++) {*/
5177: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5178: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5179: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5180: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5181: if( j != -1){
5182: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5183: covariate for which somebody answered excluding
5184: undefined. Usually 2: 0 and 1. */
5185: }
5186: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5187: covariate for which somebody answered including
5188: undefined. Usually 3: -1, 0 and 1. */
5189: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5190: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5191: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5192:
1.242 brouard 5193: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5194: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5195: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5196: /* modmincovj=3; modmaxcovj = 7; */
5197: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5198: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5199: /* defining two dummy variables: variables V1_1 and V1_2.*/
5200: /* nbcode[Tvar[j]][ij]=k; */
5201: /* nbcode[Tvar[j]][1]=0; */
5202: /* nbcode[Tvar[j]][2]=1; */
5203: /* nbcode[Tvar[j]][3]=2; */
5204: /* To be continued (not working yet). */
5205: ij=0; /* ij is similar to i but can jump over null modalities */
5206: 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*/
5207: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5208: break;
5209: }
5210: ij++;
5211: 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*/
5212: cptcode = ij; /* New max modality for covar j */
5213: } /* end of loop on modality i=-1 to 1 or more */
5214: break;
5215: case 1: /* Testing on varying covariate, could be simple and
5216: * should look at waves or product of fixed *
5217: * varying. No time to test -1, assuming 0 and 1 only */
5218: ij=0;
5219: for(i=0; i<=1;i++){
5220: nbcode[Tvar[k]][++ij]=i;
5221: }
5222: break;
5223: default:
5224: break;
5225: } /* end switch */
5226: } /* end dummy test */
5227:
5228: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5229: /* /\*recode from 0 *\/ */
5230: /* k is a modality. If we have model=V1+V1*sex */
5231: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5232: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5233: /* } */
5234: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5235: /* if (ij > ncodemax[j]) { */
5236: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5237: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5238: /* break; */
5239: /* } */
5240: /* } /\* end of loop on modality k *\/ */
5241: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5242:
5243: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5244: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5245: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5246: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5247: 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 */
5248: 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 */
5249: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5250: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5251:
5252: ij=0;
5253: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5254: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5255: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5256: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5257: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5258: /* If product not in single variable we don't print results */
5259: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5260: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5261: 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*/
5262: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5263: 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 */
5264: if(Fixed[k]!=0)
5265: anyvaryingduminmodel=1;
5266: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5267: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5268: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5269: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5270: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5271: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5272: }
5273: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5274: /* ij--; */
5275: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5276: *cptcov=ij; /*Number of total real effective covariates: effective
5277: * because they can be excluded from the model and real
5278: * if in the model but excluded because missing values, but how to get k from ij?*/
5279: for(j=ij+1; j<= cptcovt; j++){
5280: Tvaraff[j]=0;
5281: Tmodelind[j]=0;
5282: }
5283: for(j=ntveff+1; j<= cptcovt; j++){
5284: TmodelInvind[j]=0;
5285: }
5286: /* To be sorted */
5287: ;
5288: }
1.126 brouard 5289:
1.145 brouard 5290:
1.126 brouard 5291: /*********** Health Expectancies ****************/
5292:
1.235 brouard 5293: 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 5294:
5295: {
5296: /* Health expectancies, no variances */
1.164 brouard 5297: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5298: int nhstepma, nstepma; /* Decreasing with age */
5299: double age, agelim, hf;
5300: double ***p3mat;
5301: double eip;
5302:
1.238 brouard 5303: /* pstamp(ficreseij); */
1.126 brouard 5304: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5305: fprintf(ficreseij,"# Age");
5306: for(i=1; i<=nlstate;i++){
5307: for(j=1; j<=nlstate;j++){
5308: fprintf(ficreseij," e%1d%1d ",i,j);
5309: }
5310: fprintf(ficreseij," e%1d. ",i);
5311: }
5312: fprintf(ficreseij,"\n");
5313:
5314:
5315: if(estepm < stepm){
5316: printf ("Problem %d lower than %d\n",estepm, stepm);
5317: }
5318: else hstepm=estepm;
5319: /* We compute the life expectancy from trapezoids spaced every estepm months
5320: * This is mainly to measure the difference between two models: for example
5321: * if stepm=24 months pijx are given only every 2 years and by summing them
5322: * we are calculating an estimate of the Life Expectancy assuming a linear
5323: * progression in between and thus overestimating or underestimating according
5324: * to the curvature of the survival function. If, for the same date, we
5325: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5326: * to compare the new estimate of Life expectancy with the same linear
5327: * hypothesis. A more precise result, taking into account a more precise
5328: * curvature will be obtained if estepm is as small as stepm. */
5329:
5330: /* For example we decided to compute the life expectancy with the smallest unit */
5331: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5332: nhstepm is the number of hstepm from age to agelim
5333: nstepm is the number of stepm from age to agelin.
5334: Look at hpijx to understand the reason of that which relies in memory size
5335: and note for a fixed period like estepm months */
5336: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5337: survival function given by stepm (the optimization length). Unfortunately it
5338: means that if the survival funtion is printed only each two years of age and if
5339: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5340: results. So we changed our mind and took the option of the best precision.
5341: */
5342: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5343:
5344: agelim=AGESUP;
5345: /* If stepm=6 months */
5346: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5347: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5348:
5349: /* nhstepm age range expressed in number of stepm */
5350: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5351: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5352: /* if (stepm >= YEARM) hstepm=1;*/
5353: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5354: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5355:
5356: for (age=bage; age<=fage; age ++){
5357: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5358: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5359: /* if (stepm >= YEARM) hstepm=1;*/
5360: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5361:
5362: /* If stepm=6 months */
5363: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5364: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5365:
1.235 brouard 5366: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5367:
5368: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5369:
5370: printf("%d|",(int)age);fflush(stdout);
5371: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5372:
5373: /* Computing expectancies */
5374: for(i=1; i<=nlstate;i++)
5375: for(j=1; j<=nlstate;j++)
5376: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5377: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5378:
5379: /* 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]);*/
5380:
5381: }
5382:
5383: fprintf(ficreseij,"%3.0f",age );
5384: for(i=1; i<=nlstate;i++){
5385: eip=0;
5386: for(j=1; j<=nlstate;j++){
5387: eip +=eij[i][j][(int)age];
5388: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5389: }
5390: fprintf(ficreseij,"%9.4f", eip );
5391: }
5392: fprintf(ficreseij,"\n");
5393:
5394: }
5395: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5396: printf("\n");
5397: fprintf(ficlog,"\n");
5398:
5399: }
5400:
1.235 brouard 5401: 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 5402:
5403: {
5404: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5405: to initial status i, ei. .
1.126 brouard 5406: */
5407: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5408: int nhstepma, nstepma; /* Decreasing with age */
5409: double age, agelim, hf;
5410: double ***p3matp, ***p3matm, ***varhe;
5411: double **dnewm,**doldm;
5412: double *xp, *xm;
5413: double **gp, **gm;
5414: double ***gradg, ***trgradg;
5415: int theta;
5416:
5417: double eip, vip;
5418:
5419: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5420: xp=vector(1,npar);
5421: xm=vector(1,npar);
5422: dnewm=matrix(1,nlstate*nlstate,1,npar);
5423: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5424:
5425: pstamp(ficresstdeij);
5426: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5427: fprintf(ficresstdeij,"# Age");
5428: for(i=1; i<=nlstate;i++){
5429: for(j=1; j<=nlstate;j++)
5430: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5431: fprintf(ficresstdeij," e%1d. ",i);
5432: }
5433: fprintf(ficresstdeij,"\n");
5434:
5435: pstamp(ficrescveij);
5436: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5437: fprintf(ficrescveij,"# Age");
5438: for(i=1; i<=nlstate;i++)
5439: for(j=1; j<=nlstate;j++){
5440: cptj= (j-1)*nlstate+i;
5441: for(i2=1; i2<=nlstate;i2++)
5442: for(j2=1; j2<=nlstate;j2++){
5443: cptj2= (j2-1)*nlstate+i2;
5444: if(cptj2 <= cptj)
5445: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5446: }
5447: }
5448: fprintf(ficrescveij,"\n");
5449:
5450: if(estepm < stepm){
5451: printf ("Problem %d lower than %d\n",estepm, stepm);
5452: }
5453: else hstepm=estepm;
5454: /* We compute the life expectancy from trapezoids spaced every estepm months
5455: * This is mainly to measure the difference between two models: for example
5456: * if stepm=24 months pijx are given only every 2 years and by summing them
5457: * we are calculating an estimate of the Life Expectancy assuming a linear
5458: * progression in between and thus overestimating or underestimating according
5459: * to the curvature of the survival function. If, for the same date, we
5460: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5461: * to compare the new estimate of Life expectancy with the same linear
5462: * hypothesis. A more precise result, taking into account a more precise
5463: * curvature will be obtained if estepm is as small as stepm. */
5464:
5465: /* For example we decided to compute the life expectancy with the smallest unit */
5466: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5467: nhstepm is the number of hstepm from age to agelim
5468: nstepm is the number of stepm from age to agelin.
5469: Look at hpijx to understand the reason of that which relies in memory size
5470: and note for a fixed period like estepm months */
5471: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5472: survival function given by stepm (the optimization length). Unfortunately it
5473: means that if the survival funtion is printed only each two years of age and if
5474: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5475: results. So we changed our mind and took the option of the best precision.
5476: */
5477: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5478:
5479: /* If stepm=6 months */
5480: /* nhstepm age range expressed in number of stepm */
5481: agelim=AGESUP;
5482: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5483: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5484: /* if (stepm >= YEARM) hstepm=1;*/
5485: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5486:
5487: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5488: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5489: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5490: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5491: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5492: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5493:
5494: for (age=bage; age<=fage; age ++){
5495: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5496: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5497: /* if (stepm >= YEARM) hstepm=1;*/
5498: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5499:
1.126 brouard 5500: /* If stepm=6 months */
5501: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5502: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5503:
5504: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5505:
1.126 brouard 5506: /* Computing Variances of health expectancies */
5507: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5508: decrease memory allocation */
5509: for(theta=1; theta <=npar; theta++){
5510: for(i=1; i<=npar; i++){
1.222 brouard 5511: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5512: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5513: }
1.235 brouard 5514: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5515: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5516:
1.126 brouard 5517: for(j=1; j<= nlstate; j++){
1.222 brouard 5518: for(i=1; i<=nlstate; i++){
5519: for(h=0; h<=nhstepm-1; h++){
5520: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5521: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5522: }
5523: }
1.126 brouard 5524: }
1.218 brouard 5525:
1.126 brouard 5526: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5527: for(h=0; h<=nhstepm-1; h++){
5528: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5529: }
1.126 brouard 5530: }/* End theta */
5531:
5532:
5533: for(h=0; h<=nhstepm-1; h++)
5534: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5535: for(theta=1; theta <=npar; theta++)
5536: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5537:
1.218 brouard 5538:
1.222 brouard 5539: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5540: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5541: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5542:
1.222 brouard 5543: printf("%d|",(int)age);fflush(stdout);
5544: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5545: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5546: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5547: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5548: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5549: for(ij=1;ij<=nlstate*nlstate;ij++)
5550: for(ji=1;ji<=nlstate*nlstate;ji++)
5551: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5552: }
5553: }
1.218 brouard 5554:
1.126 brouard 5555: /* Computing expectancies */
1.235 brouard 5556: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5557: for(i=1; i<=nlstate;i++)
5558: for(j=1; j<=nlstate;j++)
1.222 brouard 5559: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5560: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5561:
1.222 brouard 5562: /* 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 5563:
1.222 brouard 5564: }
1.218 brouard 5565:
1.126 brouard 5566: fprintf(ficresstdeij,"%3.0f",age );
5567: for(i=1; i<=nlstate;i++){
5568: eip=0.;
5569: vip=0.;
5570: for(j=1; j<=nlstate;j++){
1.222 brouard 5571: eip += eij[i][j][(int)age];
5572: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5573: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5574: 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 5575: }
5576: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5577: }
5578: fprintf(ficresstdeij,"\n");
1.218 brouard 5579:
1.126 brouard 5580: fprintf(ficrescveij,"%3.0f",age );
5581: for(i=1; i<=nlstate;i++)
5582: for(j=1; j<=nlstate;j++){
1.222 brouard 5583: cptj= (j-1)*nlstate+i;
5584: for(i2=1; i2<=nlstate;i2++)
5585: for(j2=1; j2<=nlstate;j2++){
5586: cptj2= (j2-1)*nlstate+i2;
5587: if(cptj2 <= cptj)
5588: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5589: }
1.126 brouard 5590: }
5591: fprintf(ficrescveij,"\n");
1.218 brouard 5592:
1.126 brouard 5593: }
5594: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5595: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5596: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5597: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5598: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5599: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5600: printf("\n");
5601: fprintf(ficlog,"\n");
1.218 brouard 5602:
1.126 brouard 5603: free_vector(xm,1,npar);
5604: free_vector(xp,1,npar);
5605: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5606: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5607: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5608: }
1.218 brouard 5609:
1.126 brouard 5610: /************ Variance ******************/
1.235 brouard 5611: 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 5612: {
5613: /* Variance of health expectancies */
5614: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5615: /* double **newm;*/
5616: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5617:
5618: /* int movingaverage(); */
5619: double **dnewm,**doldm;
5620: double **dnewmp,**doldmp;
5621: int i, j, nhstepm, hstepm, h, nstepm ;
5622: int k;
5623: double *xp;
5624: double **gp, **gm; /* for var eij */
5625: double ***gradg, ***trgradg; /*for var eij */
5626: double **gradgp, **trgradgp; /* for var p point j */
5627: double *gpp, *gmp; /* for var p point j */
5628: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5629: double ***p3mat;
5630: double age,agelim, hf;
5631: /* double ***mobaverage; */
5632: int theta;
5633: char digit[4];
5634: char digitp[25];
5635:
5636: char fileresprobmorprev[FILENAMELENGTH];
5637:
5638: if(popbased==1){
5639: if(mobilav!=0)
5640: strcpy(digitp,"-POPULBASED-MOBILAV_");
5641: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5642: }
5643: else
5644: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5645:
1.218 brouard 5646: /* if (mobilav!=0) { */
5647: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5648: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5649: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5650: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5651: /* } */
5652: /* } */
5653:
5654: strcpy(fileresprobmorprev,"PRMORPREV-");
5655: sprintf(digit,"%-d",ij);
5656: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5657: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5658: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5659: strcat(fileresprobmorprev,fileresu);
5660: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5661: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5662: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5663: }
5664: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5665: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5666: pstamp(ficresprobmorprev);
5667: 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 5668: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5669: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5670: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5671: }
5672: for(j=1;j<=cptcoveff;j++)
5673: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5674: fprintf(ficresprobmorprev,"\n");
5675:
1.218 brouard 5676: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5677: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5678: fprintf(ficresprobmorprev," p.%-d SE",j);
5679: for(i=1; i<=nlstate;i++)
5680: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5681: }
5682: fprintf(ficresprobmorprev,"\n");
5683:
5684: fprintf(ficgp,"\n# Routine varevsij");
5685: fprintf(ficgp,"\nunset title \n");
5686: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5687: 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");
5688: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5689: /* } */
5690: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5691: pstamp(ficresvij);
5692: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5693: if(popbased==1)
5694: 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);
5695: else
5696: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5697: fprintf(ficresvij,"# Age");
5698: for(i=1; i<=nlstate;i++)
5699: for(j=1; j<=nlstate;j++)
5700: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5701: fprintf(ficresvij,"\n");
5702:
5703: xp=vector(1,npar);
5704: dnewm=matrix(1,nlstate,1,npar);
5705: doldm=matrix(1,nlstate,1,nlstate);
5706: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5707: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5708:
5709: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5710: gpp=vector(nlstate+1,nlstate+ndeath);
5711: gmp=vector(nlstate+1,nlstate+ndeath);
5712: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5713:
1.218 brouard 5714: if(estepm < stepm){
5715: printf ("Problem %d lower than %d\n",estepm, stepm);
5716: }
5717: else hstepm=estepm;
5718: /* For example we decided to compute the life expectancy with the smallest unit */
5719: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5720: nhstepm is the number of hstepm from age to agelim
5721: nstepm is the number of stepm from age to agelim.
5722: Look at function hpijx to understand why because of memory size limitations,
5723: we decided (b) to get a life expectancy respecting the most precise curvature of the
5724: survival function given by stepm (the optimization length). Unfortunately it
5725: means that if the survival funtion is printed every two years of age and if
5726: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5727: results. So we changed our mind and took the option of the best precision.
5728: */
5729: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5730: agelim = AGESUP;
5731: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5732: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5733: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5734: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5735: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5736: gp=matrix(0,nhstepm,1,nlstate);
5737: gm=matrix(0,nhstepm,1,nlstate);
5738:
5739:
5740: for(theta=1; theta <=npar; theta++){
5741: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5742: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5743: }
5744:
1.242 brouard 5745: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5746:
5747: if (popbased==1) {
5748: if(mobilav ==0){
5749: for(i=1; i<=nlstate;i++)
5750: prlim[i][i]=probs[(int)age][i][ij];
5751: }else{ /* mobilav */
5752: for(i=1; i<=nlstate;i++)
5753: prlim[i][i]=mobaverage[(int)age][i][ij];
5754: }
5755: }
5756:
1.235 brouard 5757: 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 5758: for(j=1; j<= nlstate; j++){
5759: for(h=0; h<=nhstepm; h++){
5760: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5761: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5762: }
5763: }
5764: /* Next for computing probability of death (h=1 means
5765: computed over hstepm matrices product = hstepm*stepm months)
5766: as a weighted average of prlim.
5767: */
5768: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5769: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5770: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5771: }
5772: /* end probability of death */
5773:
5774: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5775: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5776:
1.242 brouard 5777: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5778:
5779: if (popbased==1) {
5780: if(mobilav ==0){
5781: for(i=1; i<=nlstate;i++)
5782: prlim[i][i]=probs[(int)age][i][ij];
5783: }else{ /* mobilav */
5784: for(i=1; i<=nlstate;i++)
5785: prlim[i][i]=mobaverage[(int)age][i][ij];
5786: }
5787: }
5788:
1.235 brouard 5789: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5790:
5791: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5792: for(h=0; h<=nhstepm; h++){
5793: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5794: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5795: }
5796: }
5797: /* This for computing probability of death (h=1 means
5798: computed over hstepm matrices product = hstepm*stepm months)
5799: as a weighted average of prlim.
5800: */
5801: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5802: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5803: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5804: }
5805: /* end probability of death */
5806:
5807: for(j=1; j<= nlstate; j++) /* vareij */
5808: for(h=0; h<=nhstepm; h++){
5809: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5810: }
5811:
5812: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5813: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5814: }
5815:
5816: } /* End theta */
5817:
5818: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5819:
5820: for(h=0; h<=nhstepm; h++) /* veij */
5821: for(j=1; j<=nlstate;j++)
5822: for(theta=1; theta <=npar; theta++)
5823: trgradg[h][j][theta]=gradg[h][theta][j];
5824:
5825: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5826: for(theta=1; theta <=npar; theta++)
5827: trgradgp[j][theta]=gradgp[theta][j];
5828:
5829:
5830: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5831: for(i=1;i<=nlstate;i++)
5832: for(j=1;j<=nlstate;j++)
5833: vareij[i][j][(int)age] =0.;
5834:
5835: for(h=0;h<=nhstepm;h++){
5836: for(k=0;k<=nhstepm;k++){
5837: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5838: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5839: for(i=1;i<=nlstate;i++)
5840: for(j=1;j<=nlstate;j++)
5841: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5842: }
5843: }
5844:
5845: /* pptj */
5846: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5847: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5848: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5849: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5850: varppt[j][i]=doldmp[j][i];
5851: /* end ppptj */
5852: /* x centered again */
5853:
1.242 brouard 5854: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5855:
5856: if (popbased==1) {
5857: if(mobilav ==0){
5858: for(i=1; i<=nlstate;i++)
5859: prlim[i][i]=probs[(int)age][i][ij];
5860: }else{ /* mobilav */
5861: for(i=1; i<=nlstate;i++)
5862: prlim[i][i]=mobaverage[(int)age][i][ij];
5863: }
5864: }
5865:
5866: /* This for computing probability of death (h=1 means
5867: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5868: as a weighted average of prlim.
5869: */
1.235 brouard 5870: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5871: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5872: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5873: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5874: }
5875: /* end probability of death */
5876:
5877: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5878: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5879: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5880: for(i=1; i<=nlstate;i++){
5881: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5882: }
5883: }
5884: fprintf(ficresprobmorprev,"\n");
5885:
5886: fprintf(ficresvij,"%.0f ",age );
5887: for(i=1; i<=nlstate;i++)
5888: for(j=1; j<=nlstate;j++){
5889: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5890: }
5891: fprintf(ficresvij,"\n");
5892: free_matrix(gp,0,nhstepm,1,nlstate);
5893: free_matrix(gm,0,nhstepm,1,nlstate);
5894: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5895: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5896: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5897: } /* End age */
5898: free_vector(gpp,nlstate+1,nlstate+ndeath);
5899: free_vector(gmp,nlstate+1,nlstate+ndeath);
5900: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5901: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5902: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5903: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5904: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5905: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5906: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5907: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5908: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5909: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5910: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5911: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5912: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5913: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5914: 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);
5915: /* 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 5916: */
1.218 brouard 5917: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5918: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5919:
1.218 brouard 5920: free_vector(xp,1,npar);
5921: free_matrix(doldm,1,nlstate,1,nlstate);
5922: free_matrix(dnewm,1,nlstate,1,npar);
5923: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5924: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5925: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5926: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5927: fclose(ficresprobmorprev);
5928: fflush(ficgp);
5929: fflush(fichtm);
5930: } /* end varevsij */
1.126 brouard 5931:
5932: /************ Variance of prevlim ******************/
1.235 brouard 5933: 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 5934: {
1.205 brouard 5935: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5936: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5937:
1.126 brouard 5938: double **dnewm,**doldm;
5939: int i, j, nhstepm, hstepm;
5940: double *xp;
5941: double *gp, *gm;
5942: double **gradg, **trgradg;
1.208 brouard 5943: double **mgm, **mgp;
1.126 brouard 5944: double age,agelim;
5945: int theta;
5946:
5947: pstamp(ficresvpl);
5948: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5949: fprintf(ficresvpl,"# Age ");
5950: if(nresult >=1)
5951: fprintf(ficresvpl," Result# ");
1.126 brouard 5952: for(i=1; i<=nlstate;i++)
5953: fprintf(ficresvpl," %1d-%1d",i,i);
5954: fprintf(ficresvpl,"\n");
5955:
5956: xp=vector(1,npar);
5957: dnewm=matrix(1,nlstate,1,npar);
5958: doldm=matrix(1,nlstate,1,nlstate);
5959:
5960: hstepm=1*YEARM; /* Every year of age */
5961: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5962: agelim = AGESUP;
5963: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5964: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5965: if (stepm >= YEARM) hstepm=1;
5966: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5967: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5968: mgp=matrix(1,npar,1,nlstate);
5969: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5970: gp=vector(1,nlstate);
5971: gm=vector(1,nlstate);
5972:
5973: for(theta=1; theta <=npar; theta++){
5974: for(i=1; i<=npar; i++){ /* Computes gradient */
5975: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5976: }
1.209 brouard 5977: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5978: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5979: else
1.235 brouard 5980: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5981: for(i=1;i<=nlstate;i++){
1.126 brouard 5982: gp[i] = prlim[i][i];
1.208 brouard 5983: mgp[theta][i] = prlim[i][i];
5984: }
1.126 brouard 5985: for(i=1; i<=npar; i++) /* Computes gradient */
5986: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5987: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5988: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5989: else
1.235 brouard 5990: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5991: for(i=1;i<=nlstate;i++){
1.126 brouard 5992: gm[i] = prlim[i][i];
1.208 brouard 5993: mgm[theta][i] = prlim[i][i];
5994: }
1.126 brouard 5995: for(i=1;i<=nlstate;i++)
5996: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5997: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5998: } /* End theta */
5999:
6000: trgradg =matrix(1,nlstate,1,npar);
6001:
6002: for(j=1; j<=nlstate;j++)
6003: for(theta=1; theta <=npar; theta++)
6004: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6005: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6006: /* printf("\nmgm mgp %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 %lf",theta,mgm[theta][j],mgp[theta][j]); */
6011: /* printf("\n "); */
6012: /* } */
6013: /* } */
6014: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6015: /* printf("\n gradg %d ",(int)age); */
6016: /* for(j=1; j<=nlstate;j++){ */
6017: /* printf("%d ",j); */
6018: /* for(theta=1; theta <=npar; theta++) */
6019: /* printf("%d %lf ",theta,gradg[theta][j]); */
6020: /* printf("\n "); */
6021: /* } */
6022: /* } */
1.126 brouard 6023:
6024: for(i=1;i<=nlstate;i++)
6025: varpl[i][(int)age] =0.;
1.209 brouard 6026: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6027: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6028: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6029: }else{
1.126 brouard 6030: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6031: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6032: }
1.126 brouard 6033: for(i=1;i<=nlstate;i++)
6034: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6035:
6036: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6037: if(nresult >=1)
6038: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6039: for(i=1; i<=nlstate;i++)
6040: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6041: fprintf(ficresvpl,"\n");
6042: free_vector(gp,1,nlstate);
6043: free_vector(gm,1,nlstate);
1.208 brouard 6044: free_matrix(mgm,1,npar,1,nlstate);
6045: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6046: free_matrix(gradg,1,npar,1,nlstate);
6047: free_matrix(trgradg,1,nlstate,1,npar);
6048: } /* End age */
6049:
6050: free_vector(xp,1,npar);
6051: free_matrix(doldm,1,nlstate,1,npar);
6052: free_matrix(dnewm,1,nlstate,1,nlstate);
6053:
6054: }
6055:
6056: /************ Variance of one-step probabilities ******************/
6057: 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 6058: {
6059: int i, j=0, k1, l1, tj;
6060: int k2, l2, j1, z1;
6061: int k=0, l;
6062: int first=1, first1, first2;
6063: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6064: double **dnewm,**doldm;
6065: double *xp;
6066: double *gp, *gm;
6067: double **gradg, **trgradg;
6068: double **mu;
6069: double age, cov[NCOVMAX+1];
6070: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6071: int theta;
6072: char fileresprob[FILENAMELENGTH];
6073: char fileresprobcov[FILENAMELENGTH];
6074: char fileresprobcor[FILENAMELENGTH];
6075: double ***varpij;
6076:
6077: strcpy(fileresprob,"PROB_");
6078: strcat(fileresprob,fileres);
6079: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6080: printf("Problem with resultfile: %s\n", fileresprob);
6081: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6082: }
6083: strcpy(fileresprobcov,"PROBCOV_");
6084: strcat(fileresprobcov,fileresu);
6085: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6086: printf("Problem with resultfile: %s\n", fileresprobcov);
6087: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6088: }
6089: strcpy(fileresprobcor,"PROBCOR_");
6090: strcat(fileresprobcor,fileresu);
6091: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6092: printf("Problem with resultfile: %s\n", fileresprobcor);
6093: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6094: }
6095: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6096: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6097: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6098: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6099: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6100: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6101: pstamp(ficresprob);
6102: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6103: fprintf(ficresprob,"# Age");
6104: pstamp(ficresprobcov);
6105: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6106: fprintf(ficresprobcov,"# Age");
6107: pstamp(ficresprobcor);
6108: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6109: fprintf(ficresprobcor,"# Age");
1.126 brouard 6110:
6111:
1.222 brouard 6112: for(i=1; i<=nlstate;i++)
6113: for(j=1; j<=(nlstate+ndeath);j++){
6114: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6115: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6116: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6117: }
6118: /* fprintf(ficresprob,"\n");
6119: fprintf(ficresprobcov,"\n");
6120: fprintf(ficresprobcor,"\n");
6121: */
6122: xp=vector(1,npar);
6123: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6124: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6125: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6126: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6127: first=1;
6128: fprintf(ficgp,"\n# Routine varprob");
6129: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6130: fprintf(fichtm,"\n");
6131:
6132: 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);
6133: 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);
6134: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6135: and drawn. It helps understanding how is the covariance between two incidences.\
6136: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6137: 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 6138: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6139: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6140: standard deviations wide on each axis. <br>\
6141: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6142: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6143: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6144:
1.222 brouard 6145: cov[1]=1;
6146: /* tj=cptcoveff; */
1.225 brouard 6147: tj = (int) pow(2,cptcoveff);
1.222 brouard 6148: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6149: j1=0;
1.224 brouard 6150: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6151: if (cptcovn>0) {
6152: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6153: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6154: fprintf(ficresprob, "**********\n#\n");
6155: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6156: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6157: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6158:
1.222 brouard 6159: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6160: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6161: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6162:
6163:
1.222 brouard 6164: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6165: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6166: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6167:
1.222 brouard 6168: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6169: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6170: fprintf(ficresprobcor, "**********\n#");
6171: if(invalidvarcomb[j1]){
6172: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6173: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6174: continue;
6175: }
6176: }
6177: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6178: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6179: gp=vector(1,(nlstate)*(nlstate+ndeath));
6180: gm=vector(1,(nlstate)*(nlstate+ndeath));
6181: for (age=bage; age<=fage; age ++){
6182: cov[2]=age;
6183: if(nagesqr==1)
6184: cov[3]= age*age;
6185: for (k=1; k<=cptcovn;k++) {
6186: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6187: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6188: * 1 1 1 1 1
6189: * 2 2 1 1 1
6190: * 3 1 2 1 1
6191: */
6192: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6193: }
6194: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6195: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6196: for (k=1; k<=cptcovprod;k++)
6197: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6198:
6199:
1.222 brouard 6200: for(theta=1; theta <=npar; theta++){
6201: for(i=1; i<=npar; i++)
6202: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6203:
1.222 brouard 6204: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6205:
1.222 brouard 6206: k=0;
6207: for(i=1; i<= (nlstate); i++){
6208: for(j=1; j<=(nlstate+ndeath);j++){
6209: k=k+1;
6210: gp[k]=pmmij[i][j];
6211: }
6212: }
1.220 brouard 6213:
1.222 brouard 6214: for(i=1; i<=npar; i++)
6215: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6216:
1.222 brouard 6217: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6218: k=0;
6219: for(i=1; i<=(nlstate); i++){
6220: for(j=1; j<=(nlstate+ndeath);j++){
6221: k=k+1;
6222: gm[k]=pmmij[i][j];
6223: }
6224: }
1.220 brouard 6225:
1.222 brouard 6226: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6227: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6228: }
1.126 brouard 6229:
1.222 brouard 6230: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6231: for(theta=1; theta <=npar; theta++)
6232: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6233:
1.222 brouard 6234: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6235: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6236:
1.222 brouard 6237: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6238:
1.222 brouard 6239: k=0;
6240: for(i=1; i<=(nlstate); i++){
6241: for(j=1; j<=(nlstate+ndeath);j++){
6242: k=k+1;
6243: mu[k][(int) age]=pmmij[i][j];
6244: }
6245: }
6246: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6247: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6248: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6249:
1.222 brouard 6250: /*printf("\n%d ",(int)age);
6251: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6252: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6253: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6254: }*/
1.220 brouard 6255:
1.222 brouard 6256: fprintf(ficresprob,"\n%d ",(int)age);
6257: fprintf(ficresprobcov,"\n%d ",(int)age);
6258: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6259:
1.222 brouard 6260: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6261: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6262: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6263: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6264: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6265: }
6266: i=0;
6267: for (k=1; k<=(nlstate);k++){
6268: for (l=1; l<=(nlstate+ndeath);l++){
6269: i++;
6270: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6271: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6272: for (j=1; j<=i;j++){
6273: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6274: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6275: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6276: }
6277: }
6278: }/* end of loop for state */
6279: } /* end of loop for age */
6280: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6281: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6282: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6283: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6284:
6285: /* Confidence intervalle of pij */
6286: /*
6287: fprintf(ficgp,"\nunset parametric;unset label");
6288: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6289: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6290: 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);
6291: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6292: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6293: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6294: */
6295:
6296: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6297: first1=1;first2=2;
6298: for (k2=1; k2<=(nlstate);k2++){
6299: for (l2=1; l2<=(nlstate+ndeath);l2++){
6300: if(l2==k2) continue;
6301: j=(k2-1)*(nlstate+ndeath)+l2;
6302: for (k1=1; k1<=(nlstate);k1++){
6303: for (l1=1; l1<=(nlstate+ndeath);l1++){
6304: if(l1==k1) continue;
6305: i=(k1-1)*(nlstate+ndeath)+l1;
6306: if(i<=j) continue;
6307: for (age=bage; age<=fage; age ++){
6308: if ((int)age %5==0){
6309: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6310: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6311: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6312: mu1=mu[i][(int) age]/stepm*YEARM ;
6313: mu2=mu[j][(int) age]/stepm*YEARM;
6314: c12=cv12/sqrt(v1*v2);
6315: /* Computing eigen value of matrix of covariance */
6316: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6317: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6318: if ((lc2 <0) || (lc1 <0) ){
6319: if(first2==1){
6320: first1=0;
6321: 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);
6322: }
6323: 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);
6324: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6325: /* lc2=fabs(lc2); */
6326: }
1.220 brouard 6327:
1.222 brouard 6328: /* Eigen vectors */
6329: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6330: /*v21=sqrt(1.-v11*v11); *//* error */
6331: v21=(lc1-v1)/cv12*v11;
6332: v12=-v21;
6333: v22=v11;
6334: tnalp=v21/v11;
6335: if(first1==1){
6336: first1=0;
6337: 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);
6338: }
6339: 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);
6340: /*printf(fignu*/
6341: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6342: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6343: if(first==1){
6344: first=0;
6345: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6346: fprintf(ficgp,"\nset parametric;unset label");
6347: 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);
6348: fprintf(ficgp,"\nset ter svg size 640, 480");
6349: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6350: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6351: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6352: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6353: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6354: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6355: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6356: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6357: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6358: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6359: 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", \
6360: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6361: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6362: }else{
6363: first=0;
6364: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6365: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6366: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6367: 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", \
6368: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6369: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6370: }/* if first */
6371: } /* age mod 5 */
6372: } /* end loop age */
6373: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6374: first=1;
6375: } /*l12 */
6376: } /* k12 */
6377: } /*l1 */
6378: }/* k1 */
6379: } /* loop on combination of covariates j1 */
6380: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6381: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6382: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6383: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6384: free_vector(xp,1,npar);
6385: fclose(ficresprob);
6386: fclose(ficresprobcov);
6387: fclose(ficresprobcor);
6388: fflush(ficgp);
6389: fflush(fichtmcov);
6390: }
1.126 brouard 6391:
6392:
6393: /******************* Printing html file ***********/
1.201 brouard 6394: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6395: int lastpass, int stepm, int weightopt, char model[],\
6396: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6397: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6398: double jprev1, double mprev1,double anprev1, double dateprev1, \
6399: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6400: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6401:
6402: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6403: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6404: </ul>");
1.237 brouard 6405: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6406: </ul>", model);
1.214 brouard 6407: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6408: 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",
6409: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6410: 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 6411: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6412: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6413: fprintf(fichtm,"\
6414: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6415: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6416: fprintf(fichtm,"\
1.217 brouard 6417: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6418: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6419: fprintf(fichtm,"\
1.126 brouard 6420: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6421: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6422: fprintf(fichtm,"\
1.217 brouard 6423: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6424: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6425: fprintf(fichtm,"\
1.211 brouard 6426: - (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 6427: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6428: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6429: if(prevfcast==1){
6430: fprintf(fichtm,"\
6431: - Prevalence projections by age and states: \
1.201 brouard 6432: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6433: }
1.126 brouard 6434:
1.222 brouard 6435: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6436:
1.225 brouard 6437: m=pow(2,cptcoveff);
1.222 brouard 6438: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6439:
1.222 brouard 6440: jj1=0;
1.237 brouard 6441:
6442: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6443: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6444: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6445: continue;
1.220 brouard 6446:
1.222 brouard 6447: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6448: jj1++;
6449: if (cptcovn > 0) {
6450: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6451: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6452: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6453: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6454: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6455: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6456: }
1.237 brouard 6457: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6458: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6459: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6460: }
6461:
1.230 brouard 6462: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6463: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6464: if(invalidvarcomb[k1]){
6465: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6466: printf("\nCombination (%d) ignored because no cases \n",k1);
6467: continue;
6468: }
6469: }
6470: /* aij, bij */
1.259 brouard 6471: 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 6472: <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 6473: /* Pij */
1.241 brouard 6474: 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> \
6475: <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 6476: /* Quasi-incidences */
6477: 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 6478: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6479: 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 6480: 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> \
6481: <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 6482: /* Survival functions (period) in state j */
6483: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6484: 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> \
6485: <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 6486: }
6487: /* State specific survival functions (period) */
6488: for(cpt=1; cpt<=nlstate;cpt++){
6489: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6490: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6491: <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 6492: }
6493: /* Period (stable) prevalence in each health state */
6494: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6495: 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 6496: <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 6497: }
6498: if(backcast==1){
6499: /* Period (stable) back prevalence in each health state */
6500: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6501: 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 6502: <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 6503: }
1.217 brouard 6504: }
1.222 brouard 6505: if(prevfcast==1){
6506: /* Projection of prevalence up to period (stable) prevalence in each health state */
6507: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6508: 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> \
6509: <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 6510: }
6511: }
1.220 brouard 6512:
1.222 brouard 6513: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6514: 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> \
6515: <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 6516: }
6517: /* } /\* end i1 *\/ */
6518: }/* End k1 */
6519: fprintf(fichtm,"</ul>");
1.126 brouard 6520:
1.222 brouard 6521: fprintf(fichtm,"\
1.126 brouard 6522: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6523: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6524: - 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 6525: But because parameters are usually highly correlated (a higher incidence of disability \
6526: and a higher incidence of recovery can give very close observed transition) it might \
6527: be very useful to look not only at linear confidence intervals estimated from the \
6528: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6529: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6530: covariance matrix of the one-step probabilities. \
6531: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6532:
1.222 brouard 6533: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6534: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6535: fprintf(fichtm,"\
1.126 brouard 6536: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6537: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6538:
1.222 brouard 6539: fprintf(fichtm,"\
1.126 brouard 6540: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6541: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6542: fprintf(fichtm,"\
1.126 brouard 6543: - 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): \
6544: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6545: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6546: fprintf(fichtm,"\
1.126 brouard 6547: - (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): \
6548: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6549: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6550: fprintf(fichtm,"\
1.128 brouard 6551: - 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 6552: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6553: fprintf(fichtm,"\
1.128 brouard 6554: - 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 6555: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6556: fprintf(fichtm,"\
1.126 brouard 6557: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6558: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6559:
6560: /* if(popforecast==1) fprintf(fichtm,"\n */
6561: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6562: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6563: /* <br>",fileres,fileres,fileres,fileres); */
6564: /* else */
6565: /* 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 6566: fflush(fichtm);
6567: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6568:
1.225 brouard 6569: m=pow(2,cptcoveff);
1.222 brouard 6570: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6571:
1.222 brouard 6572: jj1=0;
1.237 brouard 6573:
1.241 brouard 6574: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6575: for(k1=1; k1<=m;k1++){
1.253 brouard 6576: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6577: continue;
1.222 brouard 6578: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6579: jj1++;
1.126 brouard 6580: if (cptcovn > 0) {
6581: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6582: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6583: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6584: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6585: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6586: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6587: }
6588:
1.126 brouard 6589: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6590:
1.222 brouard 6591: if(invalidvarcomb[k1]){
6592: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6593: continue;
6594: }
1.126 brouard 6595: }
6596: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6597: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6598: 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 6599: <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 6600: }
6601: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6602: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6603: true period expectancies (those weighted with period prevalences are also\
6604: drawn in addition to the population based expectancies computed using\
1.241 brouard 6605: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6606: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6607: /* } /\* end i1 *\/ */
6608: }/* End k1 */
1.241 brouard 6609: }/* End nres */
1.222 brouard 6610: fprintf(fichtm,"</ul>");
6611: fflush(fichtm);
1.126 brouard 6612: }
6613:
6614: /******************* Gnuplot file **************/
1.223 brouard 6615: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6616:
6617: char dirfileres[132],optfileres[132];
1.223 brouard 6618: char gplotcondition[132];
1.237 brouard 6619: 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 6620: int lv=0, vlv=0, kl=0;
1.130 brouard 6621: int ng=0;
1.201 brouard 6622: int vpopbased;
1.223 brouard 6623: int ioffset; /* variable offset for columns */
1.235 brouard 6624: int nres=0; /* Index of resultline */
1.219 brouard 6625:
1.126 brouard 6626: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6627: /* printf("Problem with file %s",optionfilegnuplot); */
6628: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6629: /* } */
6630:
6631: /*#ifdef windows */
6632: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6633: /*#endif */
1.225 brouard 6634: m=pow(2,cptcoveff);
1.126 brouard 6635:
1.202 brouard 6636: /* Contribution to likelihood */
6637: /* Plot the probability implied in the likelihood */
1.223 brouard 6638: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6639: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6640: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6641: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6642: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6643: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6644: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6645: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6646: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6647: 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));
6648: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6649: 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));
6650: for (i=1; i<= nlstate ; i ++) {
6651: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6652: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6653: 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);
6654: for (j=2; j<= nlstate+ndeath ; j ++) {
6655: 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);
6656: }
6657: fprintf(ficgp,";\nset out; unset ylabel;\n");
6658: }
6659: /* 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 */
6660: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6661: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6662: fprintf(ficgp,"\nset out;unset log\n");
6663: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6664:
1.126 brouard 6665: strcpy(dirfileres,optionfilefiname);
6666: strcpy(optfileres,"vpl");
1.223 brouard 6667: /* 1eme*/
1.238 brouard 6668: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6669: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6670: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6671: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6672: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6673: continue;
6674: /* We are interested in selected combination by the resultline */
1.246 brouard 6675: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6676: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6677: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6678: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6679: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6680: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6681: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6682: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6683: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6684: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6685: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6686: }
6687: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6688: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6689: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6690: }
1.246 brouard 6691: /* printf("\n#\n"); */
1.238 brouard 6692: fprintf(ficgp,"\n#\n");
6693: if(invalidvarcomb[k1]){
1.260 brouard 6694: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 6695: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6696: continue;
6697: }
1.235 brouard 6698:
1.241 brouard 6699: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6700: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.260 brouard 6701: 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);
6702: /* 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); */
6703: /* k1-1 error should be nres-1*/
1.238 brouard 6704: for (i=1; i<= nlstate ; i ++) {
6705: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6706: else fprintf(ficgp," %%*lf (%%*lf)");
6707: }
1.260 brouard 6708: 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 6709: for (i=1; i<= nlstate ; i ++) {
6710: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6711: else fprintf(ficgp," %%*lf (%%*lf)");
6712: }
1.260 brouard 6713: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6714: for (i=1; i<= nlstate ; i ++) {
6715: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6716: else fprintf(ficgp," %%*lf (%%*lf)");
6717: }
6718: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6719: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6720: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6721: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6722: if(cptcoveff ==0){
1.245 brouard 6723: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6724: }else{
6725: kl=0;
6726: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6727: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6728: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6729: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6730: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6731: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6732: kl++;
1.238 brouard 6733: /* 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 *\/ */
6734: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6735: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6736: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6737: if(k==cptcoveff){
1.245 brouard 6738: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6739: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6740: }else{
6741: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6742: kl++;
6743: }
6744: } /* end covariate */
6745: } /* end if no covariate */
6746: } /* end if backcast */
6747: fprintf(ficgp,"\nset out \n");
6748: } /* nres */
1.201 brouard 6749: } /* k1 */
6750: } /* cpt */
1.235 brouard 6751:
6752:
1.126 brouard 6753: /*2 eme*/
1.238 brouard 6754: for (k1=1; k1<= m ; k1 ++){
6755: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6756: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6757: continue;
6758: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6759: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6760: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6761: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6762: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6763: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6764: vlv= nbcode[Tvaraff[k]][lv];
6765: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6766: }
1.237 brouard 6767: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6768: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6769: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6770: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6771: }
1.211 brouard 6772: fprintf(ficgp,"\n#\n");
1.223 brouard 6773: if(invalidvarcomb[k1]){
6774: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6775: continue;
6776: }
1.219 brouard 6777:
1.241 brouard 6778: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6779: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6780: if(vpopbased==0)
6781: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6782: else
6783: fprintf(ficgp,"\nreplot ");
6784: for (i=1; i<= nlstate+1 ; i ++) {
6785: k=2*i;
1.261 brouard 6786: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased);
1.238 brouard 6787: for (j=1; j<= nlstate+1 ; j ++) {
6788: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6789: else fprintf(ficgp," %%*lf (%%*lf)");
6790: }
6791: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6792: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 6793: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 6794: for (j=1; j<= nlstate+1 ; j ++) {
6795: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6796: else fprintf(ficgp," %%*lf (%%*lf)");
6797: }
6798: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 6799: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 6800: for (j=1; j<= nlstate+1 ; j ++) {
6801: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6802: else fprintf(ficgp," %%*lf (%%*lf)");
6803: }
6804: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6805: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6806: } /* state */
6807: } /* vpopbased */
1.244 brouard 6808: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6809: } /* end nres */
6810: } /* k1 end 2 eme*/
6811:
6812:
6813: /*3eme*/
6814: for (k1=1; k1<= m ; k1 ++){
6815: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6816: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6817: continue;
6818:
6819: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 6820: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.238 brouard 6821: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6822: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6823: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6824: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6825: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6826: vlv= nbcode[Tvaraff[k]][lv];
6827: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6828: }
6829: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6830: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6831: }
6832: fprintf(ficgp,"\n#\n");
6833: if(invalidvarcomb[k1]){
6834: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6835: continue;
6836: }
6837:
6838: /* k=2+nlstate*(2*cpt-2); */
6839: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6840: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6841: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 6842: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),nres-1,nres-1,k,cpt);
1.238 brouard 6843: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6844: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6845: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6846: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6847: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6848: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6849:
1.238 brouard 6850: */
6851: for (i=1; i< nlstate ; i ++) {
1.261 brouard 6852: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+i,cpt,i+1);
1.238 brouard 6853: /* 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 6854:
1.238 brouard 6855: }
1.261 brouard 6856: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+nlstate,cpt);
1.238 brouard 6857: }
6858: } /* end nres */
6859: } /* end kl 3eme */
1.126 brouard 6860:
1.223 brouard 6861: /* 4eme */
1.201 brouard 6862: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6863: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6864: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6865: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6866: continue;
1.238 brouard 6867: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6868: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6869: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6870: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6871: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6872: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6873: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6874: vlv= nbcode[Tvaraff[k]][lv];
6875: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6876: }
6877: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6878: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6879: }
6880: fprintf(ficgp,"\n#\n");
6881: if(invalidvarcomb[k1]){
6882: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6883: continue;
1.223 brouard 6884: }
1.238 brouard 6885:
1.241 brouard 6886: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6887: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6888: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6889: k=3;
6890: for (i=1; i<= nlstate ; i ++){
6891: if(i==1){
6892: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6893: }else{
6894: fprintf(ficgp,", '' ");
6895: }
6896: l=(nlstate+ndeath)*(i-1)+1;
6897: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6898: for (j=2; j<= nlstate+ndeath ; j ++)
6899: fprintf(ficgp,"+$%d",k+l+j-1);
6900: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6901: } /* nlstate */
6902: fprintf(ficgp,"\nset out\n");
6903: } /* end cpt state*/
6904: } /* end nres */
6905: } /* end covariate k1 */
6906:
1.220 brouard 6907: /* 5eme */
1.201 brouard 6908: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6909: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6910: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6911: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6912: continue;
1.238 brouard 6913: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6914: 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);
6915: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6916: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6917: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6918: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6919: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6920: vlv= nbcode[Tvaraff[k]][lv];
6921: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6922: }
6923: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6924: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6925: }
6926: fprintf(ficgp,"\n#\n");
6927: if(invalidvarcomb[k1]){
6928: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6929: continue;
6930: }
1.227 brouard 6931:
1.241 brouard 6932: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6933: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6934: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6935: k=3;
6936: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6937: if(j==1)
6938: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6939: else
6940: fprintf(ficgp,", '' ");
6941: l=(nlstate+ndeath)*(cpt-1) +j;
6942: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6943: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6944: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6945: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6946: } /* nlstate */
6947: fprintf(ficgp,", '' ");
6948: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6949: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6950: l=(nlstate+ndeath)*(cpt-1) +j;
6951: if(j < nlstate)
6952: fprintf(ficgp,"$%d +",k+l);
6953: else
6954: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6955: }
6956: fprintf(ficgp,"\nset out\n");
6957: } /* end cpt state*/
6958: } /* end covariate */
6959: } /* end nres */
1.227 brouard 6960:
1.220 brouard 6961: /* 6eme */
1.202 brouard 6962: /* CV preval stable (period) for each covariate */
1.237 brouard 6963: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6964: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6965: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6966: continue;
1.255 brouard 6967: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6968:
1.211 brouard 6969: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6970: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6971: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6972: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6973: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6974: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6975: vlv= nbcode[Tvaraff[k]][lv];
6976: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6977: }
1.237 brouard 6978: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6979: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6980: }
1.211 brouard 6981: fprintf(ficgp,"\n#\n");
1.223 brouard 6982: if(invalidvarcomb[k1]){
1.227 brouard 6983: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6984: continue;
1.223 brouard 6985: }
1.227 brouard 6986:
1.241 brouard 6987: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6988: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6989: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6990: k=3; /* Offset */
1.255 brouard 6991: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6992: if(i==1)
6993: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6994: else
6995: fprintf(ficgp,", '' ");
1.255 brouard 6996: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6997: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6998: for (j=2; j<= nlstate ; j ++)
6999: fprintf(ficgp,"+$%d",k+l+j-1);
7000: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7001: } /* nlstate */
1.201 brouard 7002: fprintf(ficgp,"\nset out\n");
1.153 brouard 7003: } /* end cpt state*/
7004: } /* end covariate */
1.227 brouard 7005:
7006:
1.220 brouard 7007: /* 7eme */
1.218 brouard 7008: if(backcast == 1){
1.217 brouard 7009: /* CV back preval stable (period) for each covariate */
1.237 brouard 7010: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7011: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7012: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7013: continue;
1.255 brouard 7014: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
7015: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7016: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7017: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7018: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7019: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7020: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7021: vlv= nbcode[Tvaraff[k]][lv];
7022: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7023: }
1.237 brouard 7024: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7025: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7026: }
1.227 brouard 7027: fprintf(ficgp,"\n#\n");
7028: if(invalidvarcomb[k1]){
7029: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7030: continue;
7031: }
7032:
1.241 brouard 7033: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7034: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7035: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7036: k=3; /* Offset */
1.255 brouard 7037: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7038: if(i==1)
7039: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7040: else
7041: fprintf(ficgp,", '' ");
7042: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7043: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7044: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7045: /* 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 7046: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7047: /* for (j=2; j<= nlstate ; j ++) */
7048: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7049: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7050: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7051: } /* nlstate */
7052: fprintf(ficgp,"\nset out\n");
1.218 brouard 7053: } /* end cpt state*/
7054: } /* end covariate */
7055: } /* End if backcast */
7056:
1.223 brouard 7057: /* 8eme */
1.218 brouard 7058: if(prevfcast==1){
7059: /* Projection from cross-sectional to stable (period) for each covariate */
7060:
1.237 brouard 7061: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7062: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7063: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7064: continue;
1.211 brouard 7065: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7066: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7067: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7068: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7069: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7070: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7071: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7072: vlv= nbcode[Tvaraff[k]][lv];
7073: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7074: }
1.237 brouard 7075: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7076: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7077: }
1.227 brouard 7078: fprintf(ficgp,"\n#\n");
7079: if(invalidvarcomb[k1]){
7080: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7081: continue;
7082: }
7083:
7084: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7085: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7086: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7087: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7088: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7089: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7090: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7091: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7092: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7093: if(i==1){
7094: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7095: }else{
7096: fprintf(ficgp,",\\\n '' ");
7097: }
7098: if(cptcoveff ==0){ /* No covariate */
7099: ioffset=2; /* Age is in 2 */
7100: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7101: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7102: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7103: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7104: fprintf(ficgp," u %d:(", ioffset);
7105: if(i==nlstate+1)
7106: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7107: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7108: else
7109: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7110: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7111: }else{ /* more than 2 covariates */
7112: if(cptcoveff ==1){
7113: ioffset=4; /* Age is in 4 */
7114: }else{
7115: ioffset=6; /* Age is in 6 */
7116: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7117: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7118: }
7119: fprintf(ficgp," u %d:(",ioffset);
7120: kl=0;
7121: strcpy(gplotcondition,"(");
7122: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7123: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7124: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7125: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7126: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7127: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7128: kl++;
7129: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7130: kl++;
7131: if(k <cptcoveff && cptcoveff>1)
7132: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7133: }
7134: strcpy(gplotcondition+strlen(gplotcondition),")");
7135: /* 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 *\/ */
7136: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7137: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7138: /* '' 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*/
7139: if(i==nlstate+1){
7140: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7141: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7142: }else{
7143: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7144: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7145: }
7146: } /* end if covariate */
7147: } /* nlstate */
7148: fprintf(ficgp,"\nset out\n");
1.223 brouard 7149: } /* end cpt state*/
7150: } /* end covariate */
7151: } /* End if prevfcast */
1.227 brouard 7152:
7153:
1.238 brouard 7154: /* 9eme writing MLE parameters */
7155: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7156: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7157: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7158: for(k=1; k <=(nlstate+ndeath); k++){
7159: if (k != i) {
1.227 brouard 7160: fprintf(ficgp,"# current state %d\n",k);
7161: for(j=1; j <=ncovmodel; j++){
7162: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7163: jk++;
7164: }
7165: fprintf(ficgp,"\n");
1.126 brouard 7166: }
7167: }
1.223 brouard 7168: }
1.187 brouard 7169: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7170:
1.145 brouard 7171: /*goto avoid;*/
1.238 brouard 7172: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7173: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7174: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7175: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7176: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7177: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7178: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7179: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7180: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7181: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7182: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7183: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7184: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7185: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7186: fprintf(ficgp,"#\n");
1.223 brouard 7187: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7188: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7189: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7190: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7191: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7192: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7193: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7194: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7195: continue;
7196: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7197: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7198: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7199: }
7200: fprintf(ficgp,"\n#\n");
1.241 brouard 7201: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7202: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7203: if (ng==1){
7204: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7205: fprintf(ficgp,"\nunset log y");
7206: }else if (ng==2){
7207: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7208: fprintf(ficgp,"\nset log y");
7209: }else if (ng==3){
7210: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7211: fprintf(ficgp,"\nset log y");
7212: }else
7213: fprintf(ficgp,"\nunset title ");
7214: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7215: i=1;
7216: for(k2=1; k2<=nlstate; k2++) {
7217: k3=i;
7218: for(k=1; k<=(nlstate+ndeath); k++) {
7219: if (k != k2){
7220: switch( ng) {
7221: case 1:
7222: if(nagesqr==0)
7223: fprintf(ficgp," p%d+p%d*x",i,i+1);
7224: else /* nagesqr =1 */
7225: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7226: break;
7227: case 2: /* ng=2 */
7228: if(nagesqr==0)
7229: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7230: else /* nagesqr =1 */
7231: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7232: break;
7233: case 3:
7234: if(nagesqr==0)
7235: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7236: else /* nagesqr =1 */
7237: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7238: break;
7239: }
7240: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7241: ijp=1; /* product no age */
7242: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7243: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7244: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7245: if(j==Tage[ij]) { /* Product by age */
7246: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7247: if(DummyV[j]==0){
1.237 brouard 7248: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7249: }else{ /* quantitative */
7250: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7251: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7252: }
7253: ij++;
7254: }
7255: }else if(j==Tprod[ijp]) { /* */
7256: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7257: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7258: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7259: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7260: /* 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)]); */
7261: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7262: }else{ /* Vn is dummy and Vm is quanti */
7263: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7264: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7265: }
7266: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7267: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7268: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7269: }else{ /* Both quanti */
7270: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7271: }
7272: }
1.238 brouard 7273: ijp++;
1.237 brouard 7274: }
7275: } else{ /* simple covariate */
7276: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7277: if(Dummy[j]==0){
7278: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7279: }else{ /* quantitative */
7280: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7281: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7282: }
1.237 brouard 7283: } /* end simple */
7284: } /* end j */
1.223 brouard 7285: }else{
7286: i=i-ncovmodel;
7287: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7288: fprintf(ficgp," (1.");
7289: }
1.227 brouard 7290:
1.223 brouard 7291: if(ng != 1){
7292: fprintf(ficgp,")/(1");
1.227 brouard 7293:
1.223 brouard 7294: for(k1=1; k1 <=nlstate; k1++){
7295: if(nagesqr==0)
7296: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7297: else /* nagesqr =1 */
7298: 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 7299:
1.223 brouard 7300: ij=1;
7301: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7302: if((j-2)==Tage[ij]) { /* Bug valgrind */
7303: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7304: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7305: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7306: ij++;
7307: }
7308: }
7309: else
1.225 brouard 7310: 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 7311: }
7312: fprintf(ficgp,")");
7313: }
7314: fprintf(ficgp,")");
7315: if(ng ==2)
7316: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7317: else /* ng= 3 */
7318: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7319: }else{ /* end ng <> 1 */
7320: if( k !=k2) /* logit p11 is hard to draw */
7321: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7322: }
7323: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7324: fprintf(ficgp,",");
7325: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7326: fprintf(ficgp,",");
7327: i=i+ncovmodel;
7328: } /* end k */
7329: } /* end k2 */
7330: fprintf(ficgp,"\n set out\n");
7331: } /* end jk */
7332: } /* end ng */
7333: /* avoid: */
7334: fflush(ficgp);
1.126 brouard 7335: } /* end gnuplot */
7336:
7337:
7338: /*************** Moving average **************/
1.219 brouard 7339: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7340: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7341:
1.222 brouard 7342: int i, cpt, cptcod;
7343: int modcovmax =1;
7344: int mobilavrange, mob;
7345: int iage=0;
7346:
7347: double sum=0.;
7348: double age;
7349: double *sumnewp, *sumnewm;
7350: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7351:
7352:
1.225 brouard 7353: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7354: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7355:
7356: sumnewp = vector(1,ncovcombmax);
7357: sumnewm = vector(1,ncovcombmax);
7358: agemingood = vector(1,ncovcombmax);
7359: agemaxgood = vector(1,ncovcombmax);
7360:
7361: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7362: sumnewm[cptcod]=0.;
7363: sumnewp[cptcod]=0.;
7364: agemingood[cptcod]=0;
7365: agemaxgood[cptcod]=0;
7366: }
7367: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7368:
7369: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7370: if(mobilav==1) mobilavrange=5; /* default */
7371: else mobilavrange=mobilav;
7372: for (age=bage; age<=fage; 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: /* We keep the original values on the extreme ages bage, fage and for
7377: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7378: we use a 5 terms etc. until the borders are no more concerned.
7379: */
7380: for (mob=3;mob <=mobilavrange;mob=mob+2){
7381: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7382: for (i=1; i<=nlstate;i++){
7383: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7384: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7385: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7386: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7387: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7388: }
7389: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7390: }
7391: }
7392: }/* end age */
7393: }/* end mob */
7394: }else
7395: return -1;
7396: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7397: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7398: if(invalidvarcomb[cptcod]){
7399: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7400: continue;
7401: }
1.219 brouard 7402:
1.222 brouard 7403: agemingood[cptcod]=fage-(mob-1)/2;
7404: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7405: sumnewm[cptcod]=0.;
7406: for (i=1; i<=nlstate;i++){
7407: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7408: }
7409: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7410: agemingood[cptcod]=age;
7411: }else{ /* bad */
7412: for (i=1; i<=nlstate;i++){
7413: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7414: } /* i */
7415: } /* end bad */
7416: }/* age */
7417: sum=0.;
7418: for (i=1; i<=nlstate;i++){
7419: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7420: }
7421: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7422: 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);
7423: /* for (i=1; i<=nlstate;i++){ */
7424: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7425: /* } /\* i *\/ */
7426: } /* end bad */
7427: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7428: /* From youngest, finding the oldest wrong */
7429: agemaxgood[cptcod]=bage+(mob-1)/2;
7430: for (age=bage+(mob-1)/2; age<=fage; age++){
7431: sumnewm[cptcod]=0.;
7432: for (i=1; i<=nlstate;i++){
7433: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7434: }
7435: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7436: agemaxgood[cptcod]=age;
7437: }else{ /* bad */
7438: for (i=1; i<=nlstate;i++){
7439: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7440: } /* i */
7441: } /* end bad */
7442: }/* age */
7443: sum=0.;
7444: for (i=1; i<=nlstate;i++){
7445: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7446: }
7447: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7448: 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);
7449: /* for (i=1; i<=nlstate;i++){ */
7450: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7451: /* } /\* i *\/ */
7452: } /* end bad */
7453:
7454: for (age=bage; age<=fage; age++){
1.235 brouard 7455: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7456: sumnewp[cptcod]=0.;
7457: sumnewm[cptcod]=0.;
7458: for (i=1; i<=nlstate;i++){
7459: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7460: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7461: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7462: }
7463: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7464: }
7465: /* printf("\n"); */
7466: /* } */
7467: /* brutal averaging */
7468: for (i=1; i<=nlstate;i++){
7469: for (age=1; age<=bage; age++){
7470: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7471: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7472: }
7473: for (age=fage; age<=AGESUP; age++){
7474: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7475: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7476: }
7477: } /* end i status */
7478: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7479: for (age=1; age<=AGESUP; age++){
7480: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7481: mobaverage[(int)age][i][cptcod]=0.;
7482: }
7483: }
7484: }/* end cptcod */
7485: free_vector(sumnewm,1, ncovcombmax);
7486: free_vector(sumnewp,1, ncovcombmax);
7487: free_vector(agemaxgood,1, ncovcombmax);
7488: free_vector(agemingood,1, ncovcombmax);
7489: return 0;
7490: }/* End movingaverage */
1.218 brouard 7491:
1.126 brouard 7492:
7493: /************** Forecasting ******************/
1.235 brouard 7494: 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 7495: /* proj1, year, month, day of starting projection
7496: agemin, agemax range of age
7497: dateprev1 dateprev2 range of dates during which prevalence is computed
7498: anproj2 year of en of projection (same day and month as proj1).
7499: */
1.235 brouard 7500: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7501: double agec; /* generic age */
7502: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7503: double *popeffectif,*popcount;
7504: double ***p3mat;
1.218 brouard 7505: /* double ***mobaverage; */
1.126 brouard 7506: char fileresf[FILENAMELENGTH];
7507:
7508: agelim=AGESUP;
1.211 brouard 7509: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7510: in each health status at the date of interview (if between dateprev1 and dateprev2).
7511: We still use firstpass and lastpass as another selection.
7512: */
1.214 brouard 7513: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7514: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7515:
1.201 brouard 7516: strcpy(fileresf,"F_");
7517: strcat(fileresf,fileresu);
1.126 brouard 7518: if((ficresf=fopen(fileresf,"w"))==NULL) {
7519: printf("Problem with forecast resultfile: %s\n", fileresf);
7520: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7521: }
1.235 brouard 7522: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7523: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7524:
1.225 brouard 7525: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7526:
7527:
7528: stepsize=(int) (stepm+YEARM-1)/YEARM;
7529: if (stepm<=12) stepsize=1;
7530: if(estepm < stepm){
7531: printf ("Problem %d lower than %d\n",estepm, stepm);
7532: }
7533: else hstepm=estepm;
7534:
7535: hstepm=hstepm/stepm;
7536: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7537: fractional in yp1 */
7538: anprojmean=yp;
7539: yp2=modf((yp1*12),&yp);
7540: mprojmean=yp;
7541: yp1=modf((yp2*30.5),&yp);
7542: jprojmean=yp;
7543: if(jprojmean==0) jprojmean=1;
7544: if(mprojmean==0) jprojmean=1;
7545:
1.227 brouard 7546: i1=pow(2,cptcoveff);
1.126 brouard 7547: if (cptcovn < 1){i1=1;}
7548:
7549: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7550:
7551: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7552:
1.126 brouard 7553: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7554: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7555: for(k=1; k<=i1;k++){
1.253 brouard 7556: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7557: continue;
1.227 brouard 7558: if(invalidvarcomb[k]){
7559: printf("\nCombination (%d) projection ignored because no cases \n",k);
7560: continue;
7561: }
7562: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7563: for(j=1;j<=cptcoveff;j++) {
7564: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7565: }
1.235 brouard 7566: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7567: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7568: }
1.227 brouard 7569: fprintf(ficresf," yearproj age");
7570: for(j=1; j<=nlstate+ndeath;j++){
7571: for(i=1; i<=nlstate;i++)
7572: fprintf(ficresf," p%d%d",i,j);
7573: fprintf(ficresf," wp.%d",j);
7574: }
7575: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7576: fprintf(ficresf,"\n");
7577: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7578: for (agec=fage; agec>=(ageminpar-1); agec--){
7579: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7580: nhstepm = nhstepm/hstepm;
7581: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7582: oldm=oldms;savm=savms;
1.235 brouard 7583: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7584:
7585: for (h=0; h<=nhstepm; h++){
7586: if (h*hstepm/YEARM*stepm ==yearp) {
7587: fprintf(ficresf,"\n");
7588: for(j=1;j<=cptcoveff;j++)
7589: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7590: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7591: }
7592: for(j=1; j<=nlstate+ndeath;j++) {
7593: ppij=0.;
7594: for(i=1; i<=nlstate;i++) {
7595: if (mobilav==1)
7596: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7597: else {
7598: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7599: }
7600: if (h*hstepm/YEARM*stepm== yearp) {
7601: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7602: }
7603: } /* end i */
7604: if (h*hstepm/YEARM*stepm==yearp) {
7605: fprintf(ficresf," %.3f", ppij);
7606: }
7607: }/* end j */
7608: } /* end h */
7609: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7610: } /* end agec */
7611: } /* end yearp */
7612: } /* end k */
1.219 brouard 7613:
1.126 brouard 7614: fclose(ficresf);
1.215 brouard 7615: printf("End of Computing forecasting \n");
7616: fprintf(ficlog,"End of Computing forecasting\n");
7617:
1.126 brouard 7618: }
7619:
1.218 brouard 7620: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7621: /* 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 7622: /* /\* back1, year, month, day of starting backection */
7623: /* agemin, agemax range of age */
7624: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7625: /* anback2 year of en of backection (same day and month as back1). */
7626: /* *\/ */
7627: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7628: /* double agec; /\* generic age *\/ */
7629: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7630: /* double *popeffectif,*popcount; */
7631: /* double ***p3mat; */
7632: /* /\* double ***mobaverage; *\/ */
7633: /* char fileresfb[FILENAMELENGTH]; */
7634:
7635: /* agelim=AGESUP; */
7636: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7637: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7638: /* We still use firstpass and lastpass as another selection. */
7639: /* *\/ */
7640: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7641: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7642: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7643:
7644: /* strcpy(fileresfb,"FB_"); */
7645: /* strcat(fileresfb,fileresu); */
7646: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7647: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7648: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7649: /* } */
7650: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7651: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7652:
1.225 brouard 7653: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7654:
7655: /* /\* if (mobilav!=0) { *\/ */
7656: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7657: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7658: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7659: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7660: /* /\* } *\/ */
7661: /* /\* } *\/ */
7662:
7663: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7664: /* if (stepm<=12) stepsize=1; */
7665: /* if(estepm < stepm){ */
7666: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7667: /* } */
7668: /* else hstepm=estepm; */
7669:
7670: /* hstepm=hstepm/stepm; */
7671: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7672: /* fractional in yp1 *\/ */
7673: /* anprojmean=yp; */
7674: /* yp2=modf((yp1*12),&yp); */
7675: /* mprojmean=yp; */
7676: /* yp1=modf((yp2*30.5),&yp); */
7677: /* jprojmean=yp; */
7678: /* if(jprojmean==0) jprojmean=1; */
7679: /* if(mprojmean==0) jprojmean=1; */
7680:
1.225 brouard 7681: /* i1=cptcoveff; */
1.218 brouard 7682: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7683:
1.218 brouard 7684: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7685:
1.218 brouard 7686: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7687:
7688: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7689: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7690: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7691: /* k=k+1; */
7692: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7693: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7694: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7695: /* } */
7696: /* fprintf(ficresfb," yearbproj age"); */
7697: /* for(j=1; j<=nlstate+ndeath;j++){ */
7698: /* for(i=1; i<=nlstate;i++) */
7699: /* fprintf(ficresfb," p%d%d",i,j); */
7700: /* fprintf(ficresfb," p.%d",j); */
7701: /* } */
7702: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7703: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7704: /* fprintf(ficresfb,"\n"); */
7705: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7706: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7707: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7708: /* nhstepm = nhstepm/hstepm; */
7709: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7710: /* oldm=oldms;savm=savms; */
7711: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7712: /* for (h=0; h<=nhstepm; h++){ */
7713: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7714: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7715: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7716: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7717: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7718: /* } */
7719: /* for(j=1; j<=nlstate+ndeath;j++) { */
7720: /* ppij=0.; */
7721: /* for(i=1; i<=nlstate;i++) { */
7722: /* if (mobilav==1) */
7723: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7724: /* else { */
7725: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7726: /* } */
7727: /* if (h*hstepm/YEARM*stepm== yearp) { */
7728: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7729: /* } */
7730: /* } /\* end i *\/ */
7731: /* if (h*hstepm/YEARM*stepm==yearp) { */
7732: /* fprintf(ficresfb," %.3f", ppij); */
7733: /* } */
7734: /* }/\* end j *\/ */
7735: /* } /\* end h *\/ */
7736: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7737: /* } /\* end agec *\/ */
7738: /* } /\* end yearp *\/ */
7739: /* } /\* end cptcod *\/ */
7740: /* } /\* end cptcov *\/ */
7741:
7742: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7743:
7744: /* fclose(ficresfb); */
7745: /* printf("End of Computing Back forecasting \n"); */
7746: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7747:
1.218 brouard 7748: /* } */
1.217 brouard 7749:
1.126 brouard 7750: /************** Forecasting *****not tested NB*************/
1.227 brouard 7751: /* 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 7752:
1.227 brouard 7753: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7754: /* int *popage; */
7755: /* double calagedatem, agelim, kk1, kk2; */
7756: /* double *popeffectif,*popcount; */
7757: /* double ***p3mat,***tabpop,***tabpopprev; */
7758: /* /\* double ***mobaverage; *\/ */
7759: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7760:
1.227 brouard 7761: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7762: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7763: /* agelim=AGESUP; */
7764: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7765:
1.227 brouard 7766: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7767:
7768:
1.227 brouard 7769: /* strcpy(filerespop,"POP_"); */
7770: /* strcat(filerespop,fileresu); */
7771: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7772: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7773: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7774: /* } */
7775: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7776: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7777:
1.227 brouard 7778: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7779:
1.227 brouard 7780: /* /\* if (mobilav!=0) { *\/ */
7781: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7782: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7783: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7784: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7785: /* /\* } *\/ */
7786: /* /\* } *\/ */
1.126 brouard 7787:
1.227 brouard 7788: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7789: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7790:
1.227 brouard 7791: /* agelim=AGESUP; */
1.126 brouard 7792:
1.227 brouard 7793: /* hstepm=1; */
7794: /* hstepm=hstepm/stepm; */
1.218 brouard 7795:
1.227 brouard 7796: /* if (popforecast==1) { */
7797: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7798: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7799: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7800: /* } */
7801: /* popage=ivector(0,AGESUP); */
7802: /* popeffectif=vector(0,AGESUP); */
7803: /* popcount=vector(0,AGESUP); */
1.126 brouard 7804:
1.227 brouard 7805: /* i=1; */
7806: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7807:
1.227 brouard 7808: /* imx=i; */
7809: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7810: /* } */
1.218 brouard 7811:
1.227 brouard 7812: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7813: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7814: /* k=k+1; */
7815: /* fprintf(ficrespop,"\n#******"); */
7816: /* for(j=1;j<=cptcoveff;j++) { */
7817: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7818: /* } */
7819: /* fprintf(ficrespop,"******\n"); */
7820: /* fprintf(ficrespop,"# Age"); */
7821: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7822: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7823:
1.227 brouard 7824: /* for (cpt=0; cpt<=0;cpt++) { */
7825: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7826:
1.227 brouard 7827: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7828: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7829: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7830:
1.227 brouard 7831: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7832: /* oldm=oldms;savm=savms; */
7833: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7834:
1.227 brouard 7835: /* for (h=0; h<=nhstepm; h++){ */
7836: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7837: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7838: /* } */
7839: /* for(j=1; j<=nlstate+ndeath;j++) { */
7840: /* kk1=0.;kk2=0; */
7841: /* for(i=1; i<=nlstate;i++) { */
7842: /* if (mobilav==1) */
7843: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7844: /* else { */
7845: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7846: /* } */
7847: /* } */
7848: /* if (h==(int)(calagedatem+12*cpt)){ */
7849: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7850: /* /\*fprintf(ficrespop," %.3f", kk1); */
7851: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7852: /* } */
7853: /* } */
7854: /* for(i=1; i<=nlstate;i++){ */
7855: /* kk1=0.; */
7856: /* for(j=1; j<=nlstate;j++){ */
7857: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7858: /* } */
7859: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7860: /* } */
1.218 brouard 7861:
1.227 brouard 7862: /* if (h==(int)(calagedatem+12*cpt)) */
7863: /* for(j=1; j<=nlstate;j++) */
7864: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7865: /* } */
7866: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7867: /* } */
7868: /* } */
1.218 brouard 7869:
1.227 brouard 7870: /* /\******\/ */
1.218 brouard 7871:
1.227 brouard 7872: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7873: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7874: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7875: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7876: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7877:
1.227 brouard 7878: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7879: /* oldm=oldms;savm=savms; */
7880: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7881: /* for (h=0; h<=nhstepm; h++){ */
7882: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7883: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7884: /* } */
7885: /* for(j=1; j<=nlstate+ndeath;j++) { */
7886: /* kk1=0.;kk2=0; */
7887: /* for(i=1; i<=nlstate;i++) { */
7888: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7889: /* } */
7890: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7891: /* } */
7892: /* } */
7893: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7894: /* } */
7895: /* } */
7896: /* } */
7897: /* } */
1.218 brouard 7898:
1.227 brouard 7899: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7900:
1.227 brouard 7901: /* if (popforecast==1) { */
7902: /* free_ivector(popage,0,AGESUP); */
7903: /* free_vector(popeffectif,0,AGESUP); */
7904: /* free_vector(popcount,0,AGESUP); */
7905: /* } */
7906: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7907: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7908: /* fclose(ficrespop); */
7909: /* } /\* End of popforecast *\/ */
1.218 brouard 7910:
1.126 brouard 7911: int fileappend(FILE *fichier, char *optionfich)
7912: {
7913: if((fichier=fopen(optionfich,"a"))==NULL) {
7914: printf("Problem with file: %s\n", optionfich);
7915: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7916: return (0);
7917: }
7918: fflush(fichier);
7919: return (1);
7920: }
7921:
7922:
7923: /**************** function prwizard **********************/
7924: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7925: {
7926:
7927: /* Wizard to print covariance matrix template */
7928:
1.164 brouard 7929: char ca[32], cb[32];
7930: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7931: int numlinepar;
7932:
7933: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7934: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7935: for(i=1; i <=nlstate; i++){
7936: jj=0;
7937: for(j=1; j <=nlstate+ndeath; j++){
7938: if(j==i) continue;
7939: jj++;
7940: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7941: printf("%1d%1d",i,j);
7942: fprintf(ficparo,"%1d%1d",i,j);
7943: for(k=1; k<=ncovmodel;k++){
7944: /* printf(" %lf",param[i][j][k]); */
7945: /* fprintf(ficparo," %lf",param[i][j][k]); */
7946: printf(" 0.");
7947: fprintf(ficparo," 0.");
7948: }
7949: printf("\n");
7950: fprintf(ficparo,"\n");
7951: }
7952: }
7953: printf("# Scales (for hessian or gradient estimation)\n");
7954: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7955: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7956: for(i=1; i <=nlstate; i++){
7957: jj=0;
7958: for(j=1; j <=nlstate+ndeath; j++){
7959: if(j==i) continue;
7960: jj++;
7961: fprintf(ficparo,"%1d%1d",i,j);
7962: printf("%1d%1d",i,j);
7963: fflush(stdout);
7964: for(k=1; k<=ncovmodel;k++){
7965: /* printf(" %le",delti3[i][j][k]); */
7966: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7967: printf(" 0.");
7968: fprintf(ficparo," 0.");
7969: }
7970: numlinepar++;
7971: printf("\n");
7972: fprintf(ficparo,"\n");
7973: }
7974: }
7975: printf("# Covariance matrix\n");
7976: /* # 121 Var(a12)\n\ */
7977: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7978: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7979: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7980: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7981: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7982: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7983: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7984: fflush(stdout);
7985: fprintf(ficparo,"# Covariance matrix\n");
7986: /* # 121 Var(a12)\n\ */
7987: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7988: /* # ...\n\ */
7989: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7990:
7991: for(itimes=1;itimes<=2;itimes++){
7992: jj=0;
7993: for(i=1; i <=nlstate; i++){
7994: for(j=1; j <=nlstate+ndeath; j++){
7995: if(j==i) continue;
7996: for(k=1; k<=ncovmodel;k++){
7997: jj++;
7998: ca[0]= k+'a'-1;ca[1]='\0';
7999: if(itimes==1){
8000: printf("#%1d%1d%d",i,j,k);
8001: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8002: }else{
8003: printf("%1d%1d%d",i,j,k);
8004: fprintf(ficparo,"%1d%1d%d",i,j,k);
8005: /* printf(" %.5le",matcov[i][j]); */
8006: }
8007: ll=0;
8008: for(li=1;li <=nlstate; li++){
8009: for(lj=1;lj <=nlstate+ndeath; lj++){
8010: if(lj==li) continue;
8011: for(lk=1;lk<=ncovmodel;lk++){
8012: ll++;
8013: if(ll<=jj){
8014: cb[0]= lk +'a'-1;cb[1]='\0';
8015: if(ll<jj){
8016: if(itimes==1){
8017: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8018: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8019: }else{
8020: printf(" 0.");
8021: fprintf(ficparo," 0.");
8022: }
8023: }else{
8024: if(itimes==1){
8025: printf(" Var(%s%1d%1d)",ca,i,j);
8026: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8027: }else{
8028: printf(" 0.");
8029: fprintf(ficparo," 0.");
8030: }
8031: }
8032: }
8033: } /* end lk */
8034: } /* end lj */
8035: } /* end li */
8036: printf("\n");
8037: fprintf(ficparo,"\n");
8038: numlinepar++;
8039: } /* end k*/
8040: } /*end j */
8041: } /* end i */
8042: } /* end itimes */
8043:
8044: } /* end of prwizard */
8045: /******************* Gompertz Likelihood ******************************/
8046: double gompertz(double x[])
8047: {
8048: double A,B,L=0.0,sump=0.,num=0.;
8049: int i,n=0; /* n is the size of the sample */
8050:
1.220 brouard 8051: for (i=1;i<=imx ; i++) {
1.126 brouard 8052: sump=sump+weight[i];
8053: /* sump=sump+1;*/
8054: num=num+1;
8055: }
8056:
8057:
8058: /* for (i=0; i<=imx; i++)
8059: 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]);*/
8060:
8061: for (i=1;i<=imx ; i++)
8062: {
8063: if (cens[i] == 1 && wav[i]>1)
8064: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8065:
8066: if (cens[i] == 0 && wav[i]>1)
8067: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8068: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8069:
8070: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8071: if (wav[i] > 1 ) { /* ??? */
8072: L=L+A*weight[i];
8073: /* 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]);*/
8074: }
8075: }
8076:
8077: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8078:
8079: return -2*L*num/sump;
8080: }
8081:
1.136 brouard 8082: #ifdef GSL
8083: /******************* Gompertz_f Likelihood ******************************/
8084: double gompertz_f(const gsl_vector *v, void *params)
8085: {
8086: double A,B,LL=0.0,sump=0.,num=0.;
8087: double *x= (double *) v->data;
8088: int i,n=0; /* n is the size of the sample */
8089:
8090: for (i=0;i<=imx-1 ; i++) {
8091: sump=sump+weight[i];
8092: /* sump=sump+1;*/
8093: num=num+1;
8094: }
8095:
8096:
8097: /* for (i=0; i<=imx; i++)
8098: 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]);*/
8099: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8100: for (i=1;i<=imx ; i++)
8101: {
8102: if (cens[i] == 1 && wav[i]>1)
8103: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8104:
8105: if (cens[i] == 0 && wav[i]>1)
8106: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8107: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8108:
8109: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8110: if (wav[i] > 1 ) { /* ??? */
8111: LL=LL+A*weight[i];
8112: /* 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]);*/
8113: }
8114: }
8115:
8116: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8117: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8118:
8119: return -2*LL*num/sump;
8120: }
8121: #endif
8122:
1.126 brouard 8123: /******************* Printing html file ***********/
1.201 brouard 8124: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8125: int lastpass, int stepm, int weightopt, char model[],\
8126: int imx, double p[],double **matcov,double agemortsup){
8127: int i,k;
8128:
8129: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8130: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8131: for (i=1;i<=2;i++)
8132: 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 8133: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8134: fprintf(fichtm,"</ul>");
8135:
8136: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8137:
8138: 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>");
8139:
8140: for (k=agegomp;k<(agemortsup-2);k++)
8141: 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]);
8142:
8143:
8144: fflush(fichtm);
8145: }
8146:
8147: /******************* Gnuplot file **************/
1.201 brouard 8148: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8149:
8150: char dirfileres[132],optfileres[132];
1.164 brouard 8151:
1.126 brouard 8152: int ng;
8153:
8154:
8155: /*#ifdef windows */
8156: fprintf(ficgp,"cd \"%s\" \n",pathc);
8157: /*#endif */
8158:
8159:
8160: strcpy(dirfileres,optionfilefiname);
8161: strcpy(optfileres,"vpl");
1.199 brouard 8162: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8163: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8164: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8165: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8166: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8167:
8168: }
8169:
1.136 brouard 8170: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8171: {
1.126 brouard 8172:
1.136 brouard 8173: /*-------- data file ----------*/
8174: FILE *fic;
8175: char dummy[]=" ";
1.240 brouard 8176: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8177: int lstra;
1.136 brouard 8178: int linei, month, year,iout;
8179: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8180: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8181: char *stratrunc;
1.223 brouard 8182:
1.240 brouard 8183: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8184: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8185:
1.240 brouard 8186: for(v=1; v <=ncovcol;v++){
8187: DummyV[v]=0;
8188: FixedV[v]=0;
8189: }
8190: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8191: DummyV[v]=1;
8192: FixedV[v]=0;
8193: }
8194: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8195: DummyV[v]=0;
8196: FixedV[v]=1;
8197: }
8198: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8199: DummyV[v]=1;
8200: FixedV[v]=1;
8201: }
8202: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8203: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8204: 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]);
8205: }
1.126 brouard 8206:
1.136 brouard 8207: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8208: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8209: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8210: }
1.126 brouard 8211:
1.136 brouard 8212: i=1;
8213: linei=0;
8214: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8215: linei=linei+1;
8216: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8217: if(line[j] == '\t')
8218: line[j] = ' ';
8219: }
8220: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8221: ;
8222: };
8223: line[j+1]=0; /* Trims blanks at end of line */
8224: if(line[0]=='#'){
8225: fprintf(ficlog,"Comment line\n%s\n",line);
8226: printf("Comment line\n%s\n",line);
8227: continue;
8228: }
8229: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8230: strcpy(line, linetmp);
1.223 brouard 8231:
8232: /* Loops on waves */
8233: for (j=maxwav;j>=1;j--){
8234: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8235: cutv(stra, strb, line, ' ');
8236: if(strb[0]=='.') { /* Missing value */
8237: lval=-1;
8238: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8239: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8240: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8241: 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);
8242: 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);
8243: return 1;
8244: }
8245: }else{
8246: errno=0;
8247: /* what_kind_of_number(strb); */
8248: dval=strtod(strb,&endptr);
8249: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8250: /* if(strb != endptr && *endptr == '\0') */
8251: /* dval=dlval; */
8252: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8253: if( strb[0]=='\0' || (*endptr != '\0')){
8254: 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);
8255: 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);
8256: return 1;
8257: }
8258: cotqvar[j][iv][i]=dval;
8259: cotvar[j][ntv+iv][i]=dval;
8260: }
8261: strcpy(line,stra);
1.223 brouard 8262: }/* end loop ntqv */
1.225 brouard 8263:
1.223 brouard 8264: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8265: cutv(stra, strb, line, ' ');
8266: if(strb[0]=='.') { /* Missing value */
8267: lval=-1;
8268: }else{
8269: errno=0;
8270: lval=strtol(strb,&endptr,10);
8271: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8272: if( strb[0]=='\0' || (*endptr != '\0')){
8273: 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);
8274: 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);
8275: return 1;
8276: }
8277: }
8278: if(lval <-1 || lval >1){
8279: printf("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);
1.238 brouard 8288: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8289: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8290: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8291: For example, for multinomial values like 1, 2 and 3,\n \
8292: build V1=0 V2=0 for the reference value (1),\n \
8293: V1=1 V2=0 for (2) \n \
1.223 brouard 8294: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8295: output of IMaCh is often meaningless.\n \
1.223 brouard 8296: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8297: return 1;
8298: }
8299: cotvar[j][iv][i]=(double)(lval);
8300: strcpy(line,stra);
1.223 brouard 8301: }/* end loop ntv */
1.225 brouard 8302:
1.223 brouard 8303: /* Statuses at wave */
1.137 brouard 8304: cutv(stra, strb, line, ' ');
1.223 brouard 8305: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8306: lval=-1;
1.136 brouard 8307: }else{
1.238 brouard 8308: errno=0;
8309: lval=strtol(strb,&endptr,10);
8310: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8311: if( strb[0]=='\0' || (*endptr != '\0')){
8312: 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);
8313: 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);
8314: return 1;
8315: }
1.136 brouard 8316: }
1.225 brouard 8317:
1.136 brouard 8318: s[j][i]=lval;
1.225 brouard 8319:
1.223 brouard 8320: /* Date of Interview */
1.136 brouard 8321: strcpy(line,stra);
8322: cutv(stra, strb,line,' ');
1.169 brouard 8323: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8324: }
1.169 brouard 8325: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8326: month=99;
8327: year=9999;
1.136 brouard 8328: }else{
1.225 brouard 8329: 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);
8330: 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);
8331: return 1;
1.136 brouard 8332: }
8333: anint[j][i]= (double) year;
8334: mint[j][i]= (double)month;
8335: strcpy(line,stra);
1.223 brouard 8336: } /* End loop on waves */
1.225 brouard 8337:
1.223 brouard 8338: /* Date of death */
1.136 brouard 8339: cutv(stra, strb,line,' ');
1.169 brouard 8340: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8341: }
1.169 brouard 8342: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8343: month=99;
8344: year=9999;
8345: }else{
1.141 brouard 8346: 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 8347: 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);
8348: return 1;
1.136 brouard 8349: }
8350: andc[i]=(double) year;
8351: moisdc[i]=(double) month;
8352: strcpy(line,stra);
8353:
1.223 brouard 8354: /* Date of birth */
1.136 brouard 8355: cutv(stra, strb,line,' ');
1.169 brouard 8356: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8357: }
1.169 brouard 8358: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8359: month=99;
8360: year=9999;
8361: }else{
1.141 brouard 8362: 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);
8363: 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 8364: return 1;
1.136 brouard 8365: }
8366: if (year==9999) {
1.141 brouard 8367: 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);
8368: 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 8369: return 1;
8370:
1.136 brouard 8371: }
8372: annais[i]=(double)(year);
8373: moisnais[i]=(double)(month);
8374: strcpy(line,stra);
1.225 brouard 8375:
1.223 brouard 8376: /* Sample weight */
1.136 brouard 8377: cutv(stra, strb,line,' ');
8378: errno=0;
8379: dval=strtod(strb,&endptr);
8380: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8381: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8382: 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 8383: fflush(ficlog);
8384: return 1;
8385: }
8386: weight[i]=dval;
8387: strcpy(line,stra);
1.225 brouard 8388:
1.223 brouard 8389: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8390: cutv(stra, strb, line, ' ');
8391: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8392: lval=-1;
1.223 brouard 8393: }else{
1.225 brouard 8394: errno=0;
8395: /* what_kind_of_number(strb); */
8396: dval=strtod(strb,&endptr);
8397: /* if(strb != endptr && *endptr == '\0') */
8398: /* dval=dlval; */
8399: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8400: if( strb[0]=='\0' || (*endptr != '\0')){
8401: 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);
8402: 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);
8403: return 1;
8404: }
8405: coqvar[iv][i]=dval;
1.226 brouard 8406: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8407: }
8408: strcpy(line,stra);
8409: }/* end loop nqv */
1.136 brouard 8410:
1.223 brouard 8411: /* Covariate values */
1.136 brouard 8412: for (j=ncovcol;j>=1;j--){
8413: cutv(stra, strb,line,' ');
1.223 brouard 8414: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8415: lval=-1;
1.136 brouard 8416: }else{
1.225 brouard 8417: errno=0;
8418: lval=strtol(strb,&endptr,10);
8419: if( strb[0]=='\0' || (*endptr != '\0')){
8420: 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);
8421: 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);
8422: return 1;
8423: }
1.136 brouard 8424: }
8425: if(lval <-1 || lval >1){
1.225 brouard 8426: printf("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);
1.225 brouard 8435: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8436: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8437: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8438: For example, for multinomial values like 1, 2 and 3,\n \
8439: build V1=0 V2=0 for the reference value (1),\n \
8440: V1=1 V2=0 for (2) \n \
1.136 brouard 8441: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8442: output of IMaCh is often meaningless.\n \
1.136 brouard 8443: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8444: return 1;
1.136 brouard 8445: }
8446: covar[j][i]=(double)(lval);
8447: strcpy(line,stra);
8448: }
8449: lstra=strlen(stra);
1.225 brouard 8450:
1.136 brouard 8451: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8452: stratrunc = &(stra[lstra-9]);
8453: num[i]=atol(stratrunc);
8454: }
8455: else
8456: num[i]=atol(stra);
8457: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8458: 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;}*/
8459:
8460: i=i+1;
8461: } /* End loop reading data */
1.225 brouard 8462:
1.136 brouard 8463: *imax=i-1; /* Number of individuals */
8464: fclose(fic);
1.225 brouard 8465:
1.136 brouard 8466: return (0);
1.164 brouard 8467: /* endread: */
1.225 brouard 8468: printf("Exiting readdata: ");
8469: fclose(fic);
8470: return (1);
1.223 brouard 8471: }
1.126 brouard 8472:
1.234 brouard 8473: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8474: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8475: while (*p2 == ' ')
1.234 brouard 8476: p2++;
8477: /* while ((*p1++ = *p2++) !=0) */
8478: /* ; */
8479: /* do */
8480: /* while (*p2 == ' ') */
8481: /* p2++; */
8482: /* while (*p1++ == *p2++); */
8483: *stri=p2;
1.145 brouard 8484: }
8485:
1.235 brouard 8486: int decoderesult ( char resultline[], int nres)
1.230 brouard 8487: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8488: {
1.235 brouard 8489: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8490: char resultsav[MAXLINE];
1.234 brouard 8491: int resultmodel[MAXLINE];
8492: int modelresult[MAXLINE];
1.230 brouard 8493: char stra[80], strb[80], strc[80], strd[80],stre[80];
8494:
1.234 brouard 8495: removefirstspace(&resultline);
1.233 brouard 8496: printf("decoderesult:%s\n",resultline);
1.230 brouard 8497:
8498: if (strstr(resultline,"v") !=0){
8499: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8500: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8501: return 1;
8502: }
8503: trimbb(resultsav, resultline);
8504: if (strlen(resultsav) >1){
8505: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8506: }
1.253 brouard 8507: if(j == 0){ /* Resultline but no = */
8508: TKresult[nres]=0; /* Combination for the nresult and the model */
8509: return (0);
8510: }
8511:
1.234 brouard 8512: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8513: 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);
8514: 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);
8515: }
8516: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8517: if(nbocc(resultsav,'=') >1){
8518: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8519: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8520: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8521: }else
8522: cutl(strc,strd,resultsav,'=');
1.230 brouard 8523: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8524:
1.230 brouard 8525: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8526: Tvarsel[k]=atoi(strc);
8527: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8528: /* cptcovsel++; */
8529: if (nbocc(stra,'=') >0)
8530: strcpy(resultsav,stra); /* and analyzes it */
8531: }
1.235 brouard 8532: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8533: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8534: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8535: match=0;
1.236 brouard 8536: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8537: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8538: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8539: match=1;
8540: break;
8541: }
8542: }
8543: if(match == 0){
8544: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8545: }
8546: }
8547: }
1.235 brouard 8548: /* Checking for missing or useless values in comparison of current model needs */
8549: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8550: match=0;
1.235 brouard 8551: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8552: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8553: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8554: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8555: ++match;
8556: }
8557: }
8558: }
8559: if(match == 0){
8560: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8561: }else if(match > 1){
8562: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8563: }
8564: }
1.235 brouard 8565:
1.234 brouard 8566: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8567: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8568: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8569: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8570: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8571: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8572: /* 1 0 0 0 */
8573: /* 2 1 0 0 */
8574: /* 3 0 1 0 */
8575: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8576: /* 5 0 0 1 */
8577: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8578: /* 7 0 1 1 */
8579: /* 8 1 1 1 */
1.237 brouard 8580: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8581: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8582: /* V5*age V5 known which value for nres? */
8583: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8584: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8585: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8586: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8587: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8588: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8589: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8590: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8591: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8592: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8593: k4++;;
8594: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8595: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8596: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8597: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8598: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8599: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8600: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8601: k4q++;;
8602: }
8603: }
1.234 brouard 8604:
1.235 brouard 8605: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8606: return (0);
8607: }
1.235 brouard 8608:
1.230 brouard 8609: int decodemodel( char model[], int lastobs)
8610: /**< This routine decodes the model and returns:
1.224 brouard 8611: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8612: * - nagesqr = 1 if age*age in the model, otherwise 0.
8613: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8614: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8615: * - cptcovage number of covariates with age*products =2
8616: * - cptcovs number of simple covariates
8617: * - 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
8618: * which is a new column after the 9 (ncovcol) variables.
8619: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8620: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8621: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8622: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8623: */
1.136 brouard 8624: {
1.238 brouard 8625: int i, j, k, ks, v;
1.227 brouard 8626: int j1, k1, k2, k3, k4;
1.136 brouard 8627: char modelsav[80];
1.145 brouard 8628: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8629: char *strpt;
1.136 brouard 8630:
1.145 brouard 8631: /*removespace(model);*/
1.136 brouard 8632: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8633: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8634: if (strstr(model,"AGE") !=0){
1.192 brouard 8635: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8636: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8637: return 1;
8638: }
1.141 brouard 8639: if (strstr(model,"v") !=0){
8640: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8641: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8642: return 1;
8643: }
1.187 brouard 8644: strcpy(modelsav,model);
8645: if ((strpt=strstr(model,"age*age")) !=0){
8646: printf(" strpt=%s, model=%s\n",strpt, model);
8647: if(strpt != model){
1.234 brouard 8648: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8649: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8650: corresponding column of parameters.\n",model);
1.234 brouard 8651: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8652: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8653: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8654: return 1;
1.225 brouard 8655: }
1.187 brouard 8656: nagesqr=1;
8657: if (strstr(model,"+age*age") !=0)
1.234 brouard 8658: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8659: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8660: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8661: else
1.234 brouard 8662: substrchaine(modelsav, model, "age*age");
1.187 brouard 8663: }else
8664: nagesqr=0;
8665: if (strlen(modelsav) >1){
8666: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8667: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8668: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8669: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8670: * cst, age and age*age
8671: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8672: /* including age products which are counted in cptcovage.
8673: * but the covariates which are products must be treated
8674: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8675: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8676: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8677:
8678:
1.187 brouard 8679: /* Design
8680: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8681: * < ncovcol=8 >
8682: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8683: * k= 1 2 3 4 5 6 7 8
8684: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8685: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8686: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8687: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8688: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8689: * Tage[++cptcovage]=k
8690: * if products, new covar are created after ncovcol with k1
8691: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8692: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8693: * 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
8694: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8695: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8696: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8697: * < ncovcol=8 >
8698: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8699: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8700: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8701: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8702: * p Tprod[1]@2={ 6, 5}
8703: *p Tvard[1][1]@4= {7, 8, 5, 6}
8704: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8705: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8706: *How to reorganize?
8707: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8708: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8709: * {2, 1, 4, 8, 5, 6, 3, 7}
8710: * Struct []
8711: */
1.225 brouard 8712:
1.187 brouard 8713: /* This loop fills the array Tvar from the string 'model'.*/
8714: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8715: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8716: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8717: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8718: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8719: /* k=1 Tvar[1]=2 (from V2) */
8720: /* k=5 Tvar[5] */
8721: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8722: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8723: /* } */
1.198 brouard 8724: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8725: /*
8726: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8727: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8728: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8729: }
1.187 brouard 8730: cptcovage=0;
8731: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8732: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8733: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8734: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8735: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8736: /*scanf("%d",i);*/
8737: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8738: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8739: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8740: /* covar is not filled and then is empty */
8741: cptcovprod--;
8742: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8743: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8744: Typevar[k]=1; /* 1 for age product */
8745: cptcovage++; /* Sums the number of covariates which include age as a product */
8746: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8747: /*printf("stre=%s ", stre);*/
8748: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8749: cptcovprod--;
8750: cutl(stre,strb,strc,'V');
8751: Tvar[k]=atoi(stre);
8752: Typevar[k]=1; /* 1 for age product */
8753: cptcovage++;
8754: Tage[cptcovage]=k;
8755: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8756: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8757: cptcovn++;
8758: cptcovprodnoage++;k1++;
8759: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8760: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8761: because this model-covariate is a construction we invent a new column
8762: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8763: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8764: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8765: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8766: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8767: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8768: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8769: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8770: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8771: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8772: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8773: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8774: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8775: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8776: for (i=1; i<=lastobs;i++){
8777: /* Computes the new covariate which is a product of
8778: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8779: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8780: }
8781: } /* End age is not in the model */
8782: } /* End if model includes a product */
8783: else { /* no more sum */
8784: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8785: /* scanf("%d",i);*/
8786: cutl(strd,strc,strb,'V');
8787: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8788: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8789: Tvar[k]=atoi(strd);
8790: Typevar[k]=0; /* 0 for simple covariates */
8791: }
8792: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8793: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8794: scanf("%d",i);*/
1.187 brouard 8795: } /* end of loop + on total covariates */
8796: } /* end if strlen(modelsave == 0) age*age might exist */
8797: } /* end if strlen(model == 0) */
1.136 brouard 8798:
8799: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8800: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8801:
1.136 brouard 8802: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8803: printf("cptcovprod=%d ", cptcovprod);
8804: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8805: scanf("%d ",i);*/
8806:
8807:
1.230 brouard 8808: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8809: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8810: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8811: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8812: k = 1 2 3 4 5 6 7 8 9
8813: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8814: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8815: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8816: Dummy[k] 1 0 0 0 3 1 1 2 3
8817: Tmodelind[combination of covar]=k;
1.225 brouard 8818: */
8819: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8820: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8821: /* 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 8822: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8823: printf("Model=%s\n\
8824: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8825: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8826: 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);
8827: fprintf(ficlog,"Model=%s\n\
8828: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8829: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8830: 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 8831: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8832: 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 */
8833: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8834: Fixed[k]= 0;
8835: Dummy[k]= 0;
1.225 brouard 8836: ncoveff++;
1.232 brouard 8837: ncovf++;
1.234 brouard 8838: nsd++;
8839: modell[k].maintype= FTYPE;
8840: TvarsD[nsd]=Tvar[k];
8841: TvarsDind[nsd]=k;
8842: TvarF[ncovf]=Tvar[k];
8843: TvarFind[ncovf]=k;
8844: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8845: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8846: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8847: Fixed[k]= 0;
8848: Dummy[k]= 0;
8849: ncoveff++;
8850: ncovf++;
8851: modell[k].maintype= FTYPE;
8852: TvarF[ncovf]=Tvar[k];
8853: TvarFind[ncovf]=k;
1.230 brouard 8854: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8855: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8856: }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 8857: Fixed[k]= 0;
8858: Dummy[k]= 1;
1.230 brouard 8859: nqfveff++;
1.234 brouard 8860: modell[k].maintype= FTYPE;
8861: modell[k].subtype= FQ;
8862: nsq++;
8863: TvarsQ[nsq]=Tvar[k];
8864: TvarsQind[nsq]=k;
1.232 brouard 8865: ncovf++;
1.234 brouard 8866: TvarF[ncovf]=Tvar[k];
8867: TvarFind[ncovf]=k;
1.231 brouard 8868: 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 8869: 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 8870: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8871: Fixed[k]= 1;
8872: Dummy[k]= 0;
1.225 brouard 8873: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8874: modell[k].maintype= VTYPE;
8875: modell[k].subtype= VD;
8876: nsd++;
8877: TvarsD[nsd]=Tvar[k];
8878: TvarsDind[nsd]=k;
8879: ncovv++; /* Only simple time varying variables */
8880: TvarV[ncovv]=Tvar[k];
1.242 brouard 8881: 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 8882: 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 */
8883: 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 8884: 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);
8885: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8886: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8887: Fixed[k]= 1;
8888: Dummy[k]= 1;
8889: nqtveff++;
8890: modell[k].maintype= VTYPE;
8891: modell[k].subtype= VQ;
8892: ncovv++; /* Only simple time varying variables */
8893: nsq++;
8894: TvarsQ[nsq]=Tvar[k];
8895: TvarsQind[nsq]=k;
8896: TvarV[ncovv]=Tvar[k];
1.242 brouard 8897: 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 8898: 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 */
8899: 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 8900: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8901: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8902: 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 8903: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8904: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8905: ncova++;
8906: TvarA[ncova]=Tvar[k];
8907: TvarAind[ncova]=k;
1.231 brouard 8908: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8909: Fixed[k]= 2;
8910: Dummy[k]= 2;
8911: modell[k].maintype= ATYPE;
8912: modell[k].subtype= APFD;
8913: /* ncoveff++; */
1.227 brouard 8914: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8915: Fixed[k]= 2;
8916: Dummy[k]= 3;
8917: modell[k].maintype= ATYPE;
8918: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8919: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8920: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8921: Fixed[k]= 3;
8922: Dummy[k]= 2;
8923: modell[k].maintype= ATYPE;
8924: modell[k].subtype= APVD; /* Product age * varying dummy */
8925: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8926: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8927: Fixed[k]= 3;
8928: Dummy[k]= 3;
8929: modell[k].maintype= ATYPE;
8930: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8931: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8932: }
8933: }else if (Typevar[k] == 2) { /* product without age */
8934: k1=Tposprod[k];
8935: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8936: if(Tvard[k1][2] <=ncovcol){
8937: Fixed[k]= 1;
8938: Dummy[k]= 0;
8939: modell[k].maintype= FTYPE;
8940: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8941: ncovf++; /* Fixed variables without age */
8942: TvarF[ncovf]=Tvar[k];
8943: TvarFind[ncovf]=k;
8944: }else if(Tvard[k1][2] <=ncovcol+nqv){
8945: Fixed[k]= 0; /* or 2 ?*/
8946: Dummy[k]= 1;
8947: modell[k].maintype= FTYPE;
8948: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8949: ncovf++; /* Varying variables without age */
8950: TvarF[ncovf]=Tvar[k];
8951: TvarFind[ncovf]=k;
8952: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8953: Fixed[k]= 1;
8954: Dummy[k]= 0;
8955: modell[k].maintype= VTYPE;
8956: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8957: ncovv++; /* Varying variables without age */
8958: TvarV[ncovv]=Tvar[k];
8959: TvarVind[ncovv]=k;
8960: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8961: Fixed[k]= 1;
8962: Dummy[k]= 1;
8963: modell[k].maintype= VTYPE;
8964: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8965: ncovv++; /* Varying variables without age */
8966: TvarV[ncovv]=Tvar[k];
8967: TvarVind[ncovv]=k;
8968: }
1.227 brouard 8969: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8970: if(Tvard[k1][2] <=ncovcol){
8971: Fixed[k]= 0; /* or 2 ?*/
8972: Dummy[k]= 1;
8973: modell[k].maintype= FTYPE;
8974: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8975: ncovf++; /* Fixed variables without age */
8976: TvarF[ncovf]=Tvar[k];
8977: TvarFind[ncovf]=k;
8978: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8979: Fixed[k]= 1;
8980: Dummy[k]= 1;
8981: modell[k].maintype= VTYPE;
8982: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8983: ncovv++; /* Varying variables without age */
8984: TvarV[ncovv]=Tvar[k];
8985: TvarVind[ncovv]=k;
8986: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8987: Fixed[k]= 1;
8988: Dummy[k]= 1;
8989: modell[k].maintype= VTYPE;
8990: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8991: ncovv++; /* Varying variables without age */
8992: TvarV[ncovv]=Tvar[k];
8993: TvarVind[ncovv]=k;
8994: ncovv++; /* Varying variables without age */
8995: TvarV[ncovv]=Tvar[k];
8996: TvarVind[ncovv]=k;
8997: }
1.227 brouard 8998: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8999: if(Tvard[k1][2] <=ncovcol){
9000: Fixed[k]= 1;
9001: Dummy[k]= 1;
9002: modell[k].maintype= VTYPE;
9003: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9004: ncovv++; /* Varying variables without age */
9005: TvarV[ncovv]=Tvar[k];
9006: TvarVind[ncovv]=k;
9007: }else if(Tvard[k1][2] <=ncovcol+nqv){
9008: Fixed[k]= 1;
9009: Dummy[k]= 1;
9010: modell[k].maintype= VTYPE;
9011: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9012: ncovv++; /* Varying variables without age */
9013: TvarV[ncovv]=Tvar[k];
9014: TvarVind[ncovv]=k;
9015: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9016: Fixed[k]= 1;
9017: Dummy[k]= 0;
9018: modell[k].maintype= VTYPE;
9019: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9020: ncovv++; /* Varying variables without age */
9021: TvarV[ncovv]=Tvar[k];
9022: TvarVind[ncovv]=k;
9023: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9024: Fixed[k]= 1;
9025: Dummy[k]= 1;
9026: modell[k].maintype= VTYPE;
9027: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9028: ncovv++; /* Varying variables without age */
9029: TvarV[ncovv]=Tvar[k];
9030: TvarVind[ncovv]=k;
9031: }
1.227 brouard 9032: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9033: if(Tvard[k1][2] <=ncovcol){
9034: Fixed[k]= 1;
9035: Dummy[k]= 1;
9036: modell[k].maintype= VTYPE;
9037: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9038: ncovv++; /* Varying variables without age */
9039: TvarV[ncovv]=Tvar[k];
9040: TvarVind[ncovv]=k;
9041: }else if(Tvard[k1][2] <=ncovcol+nqv){
9042: Fixed[k]= 1;
9043: Dummy[k]= 1;
9044: modell[k].maintype= VTYPE;
9045: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9046: ncovv++; /* Varying variables without age */
9047: TvarV[ncovv]=Tvar[k];
9048: TvarVind[ncovv]=k;
9049: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9050: Fixed[k]= 1;
9051: Dummy[k]= 1;
9052: modell[k].maintype= VTYPE;
9053: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9054: ncovv++; /* Varying variables without age */
9055: TvarV[ncovv]=Tvar[k];
9056: TvarVind[ncovv]=k;
9057: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9058: Fixed[k]= 1;
9059: Dummy[k]= 1;
9060: modell[k].maintype= VTYPE;
9061: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9062: ncovv++; /* Varying variables without age */
9063: TvarV[ncovv]=Tvar[k];
9064: TvarVind[ncovv]=k;
9065: }
1.227 brouard 9066: }else{
1.240 brouard 9067: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9068: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9069: } /*end k1*/
1.225 brouard 9070: }else{
1.226 brouard 9071: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9072: 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 9073: }
1.227 brouard 9074: 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 9075: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9076: 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]);
9077: }
9078: /* Searching for doublons in the model */
9079: for(k1=1; k1<= cptcovt;k1++){
9080: for(k2=1; k2 <k1;k2++){
9081: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9082: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9083: if(Tvar[k1]==Tvar[k2]){
9084: 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]]);
9085: 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);
9086: return(1);
9087: }
9088: }else if (Typevar[k1] ==2){
9089: k3=Tposprod[k1];
9090: k4=Tposprod[k2];
9091: 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])) ){
9092: 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]]);
9093: 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);
9094: return(1);
9095: }
9096: }
1.227 brouard 9097: }
9098: }
1.225 brouard 9099: }
9100: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9101: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9102: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9103: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9104: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9105: /*endread:*/
1.225 brouard 9106: printf("Exiting decodemodel: ");
9107: return (1);
1.136 brouard 9108: }
9109:
1.169 brouard 9110: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9111: {/* Check ages at death */
1.136 brouard 9112: int i, m;
1.218 brouard 9113: int firstone=0;
9114:
1.136 brouard 9115: for (i=1; i<=imx; i++) {
9116: for(m=2; (m<= maxwav); m++) {
9117: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9118: anint[m][i]=9999;
1.216 brouard 9119: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9120: s[m][i]=-1;
1.136 brouard 9121: }
9122: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9123: *nberr = *nberr + 1;
1.218 brouard 9124: if(firstone == 0){
9125: firstone=1;
1.260 brouard 9126: 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 9127: }
1.262 brouard 9128: 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.\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.260 brouard 9129: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9130: }
9131: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9132: (*nberr)++;
1.259 brouard 9133: 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);
1.262 brouard 9134: 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).\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.259 brouard 9135: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9136: }
9137: }
9138: }
9139:
9140: for (i=1; i<=imx; i++) {
9141: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9142: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9143: 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 9144: if (s[m][i] >= nlstate+1) {
1.169 brouard 9145: if(agedc[i]>0){
9146: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9147: agev[m][i]=agedc[i];
1.214 brouard 9148: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9149: }else {
1.136 brouard 9150: if ((int)andc[i]!=9999){
9151: nbwarn++;
9152: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9153: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9154: agev[m][i]=-1;
9155: }
9156: }
1.169 brouard 9157: } /* agedc > 0 */
1.214 brouard 9158: } /* end if */
1.136 brouard 9159: else if(s[m][i] !=9){ /* Standard case, age in fractional
9160: years but with the precision of a month */
9161: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9162: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9163: agev[m][i]=1;
9164: else if(agev[m][i] < *agemin){
9165: *agemin=agev[m][i];
9166: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9167: }
9168: else if(agev[m][i] >*agemax){
9169: *agemax=agev[m][i];
1.156 brouard 9170: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9171: }
9172: /*agev[m][i]=anint[m][i]-annais[i];*/
9173: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9174: } /* en if 9*/
1.136 brouard 9175: else { /* =9 */
1.214 brouard 9176: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9177: agev[m][i]=1;
9178: s[m][i]=-1;
9179: }
9180: }
1.214 brouard 9181: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9182: agev[m][i]=1;
1.214 brouard 9183: else{
9184: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9185: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9186: agev[m][i]=0;
9187: }
9188: } /* End for lastpass */
9189: }
1.136 brouard 9190:
9191: for (i=1; i<=imx; i++) {
9192: for(m=firstpass; (m<=lastpass); m++){
9193: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9194: (*nberr)++;
1.136 brouard 9195: 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);
9196: 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);
9197: return 1;
9198: }
9199: }
9200: }
9201:
9202: /*for (i=1; i<=imx; i++){
9203: for (m=firstpass; (m<lastpass); m++){
9204: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9205: }
9206:
9207: }*/
9208:
9209:
1.139 brouard 9210: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9211: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9212:
9213: return (0);
1.164 brouard 9214: /* endread:*/
1.136 brouard 9215: printf("Exiting calandcheckages: ");
9216: return (1);
9217: }
9218:
1.172 brouard 9219: #if defined(_MSC_VER)
9220: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9221: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9222: //#include "stdafx.h"
9223: //#include <stdio.h>
9224: //#include <tchar.h>
9225: //#include <windows.h>
9226: //#include <iostream>
9227: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9228:
9229: LPFN_ISWOW64PROCESS fnIsWow64Process;
9230:
9231: BOOL IsWow64()
9232: {
9233: BOOL bIsWow64 = FALSE;
9234:
9235: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9236: // (HANDLE, PBOOL);
9237:
9238: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9239:
9240: HMODULE module = GetModuleHandle(_T("kernel32"));
9241: const char funcName[] = "IsWow64Process";
9242: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9243: GetProcAddress(module, funcName);
9244:
9245: if (NULL != fnIsWow64Process)
9246: {
9247: if (!fnIsWow64Process(GetCurrentProcess(),
9248: &bIsWow64))
9249: //throw std::exception("Unknown error");
9250: printf("Unknown error\n");
9251: }
9252: return bIsWow64 != FALSE;
9253: }
9254: #endif
1.177 brouard 9255:
1.191 brouard 9256: void syscompilerinfo(int logged)
1.167 brouard 9257: {
9258: /* #include "syscompilerinfo.h"*/
1.185 brouard 9259: /* command line Intel compiler 32bit windows, XP compatible:*/
9260: /* /GS /W3 /Gy
9261: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9262: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9263: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9264: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9265: */
9266: /* 64 bits */
1.185 brouard 9267: /*
9268: /GS /W3 /Gy
9269: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9270: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9271: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9272: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9273: /* Optimization are useless and O3 is slower than O2 */
9274: /*
9275: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9276: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9277: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9278: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9279: */
1.186 brouard 9280: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9281: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9282: /PDB:"visual studio
9283: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9284: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9285: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9286: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9287: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9288: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9289: uiAccess='false'"
9290: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9291: /NOLOGO /TLBID:1
9292: */
1.177 brouard 9293: #if defined __INTEL_COMPILER
1.178 brouard 9294: #if defined(__GNUC__)
9295: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9296: #endif
1.177 brouard 9297: #elif defined(__GNUC__)
1.179 brouard 9298: #ifndef __APPLE__
1.174 brouard 9299: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9300: #endif
1.177 brouard 9301: struct utsname sysInfo;
1.178 brouard 9302: int cross = CROSS;
9303: if (cross){
9304: printf("Cross-");
1.191 brouard 9305: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9306: }
1.174 brouard 9307: #endif
9308:
1.171 brouard 9309: #include <stdint.h>
1.178 brouard 9310:
1.191 brouard 9311: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9312: #if defined(__clang__)
1.191 brouard 9313: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9314: #endif
9315: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9316: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9317: #endif
9318: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9319: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9320: #endif
9321: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9322: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9323: #endif
9324: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9325: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9326: #endif
9327: #if defined(_MSC_VER)
1.191 brouard 9328: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9329: #endif
9330: #if defined(__PGI)
1.191 brouard 9331: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9332: #endif
9333: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9334: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9335: #endif
1.191 brouard 9336: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9337:
1.167 brouard 9338: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9339: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9340: // Windows (x64 and x86)
1.191 brouard 9341: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9342: #elif __unix__ // all unices, not all compilers
9343: // Unix
1.191 brouard 9344: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9345: #elif __linux__
9346: // linux
1.191 brouard 9347: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9348: #elif __APPLE__
1.174 brouard 9349: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9350: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9351: #endif
9352:
9353: /* __MINGW32__ */
9354: /* __CYGWIN__ */
9355: /* __MINGW64__ */
9356: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9357: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9358: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9359: /* _WIN64 // Defined for applications for Win64. */
9360: /* _M_X64 // Defined for compilations that target x64 processors. */
9361: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9362:
1.167 brouard 9363: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9364: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9365: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9366: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9367: #else
1.191 brouard 9368: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9369: #endif
9370:
1.169 brouard 9371: #if defined(__GNUC__)
9372: # if defined(__GNUC_PATCHLEVEL__)
9373: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9374: + __GNUC_MINOR__ * 100 \
9375: + __GNUC_PATCHLEVEL__)
9376: # else
9377: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9378: + __GNUC_MINOR__ * 100)
9379: # endif
1.174 brouard 9380: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9381: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9382:
9383: if (uname(&sysInfo) != -1) {
9384: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9385: 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 9386: }
9387: else
9388: perror("uname() error");
1.179 brouard 9389: //#ifndef __INTEL_COMPILER
9390: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9391: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9392: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9393: #endif
1.169 brouard 9394: #endif
1.172 brouard 9395:
9396: // void main()
9397: // {
1.169 brouard 9398: #if defined(_MSC_VER)
1.174 brouard 9399: if (IsWow64()){
1.191 brouard 9400: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9401: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9402: }
9403: else{
1.191 brouard 9404: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9405: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9406: }
1.172 brouard 9407: // printf("\nPress Enter to continue...");
9408: // getchar();
9409: // }
9410:
1.169 brouard 9411: #endif
9412:
1.167 brouard 9413:
1.219 brouard 9414: }
1.136 brouard 9415:
1.219 brouard 9416: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9417: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9418: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9419: /* double ftolpl = 1.e-10; */
1.180 brouard 9420: double age, agebase, agelim;
1.203 brouard 9421: double tot;
1.180 brouard 9422:
1.202 brouard 9423: strcpy(filerespl,"PL_");
9424: strcat(filerespl,fileresu);
9425: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9426: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9427: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9428: }
1.227 brouard 9429: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9430: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9431: pstamp(ficrespl);
1.203 brouard 9432: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9433: fprintf(ficrespl,"#Age ");
9434: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9435: fprintf(ficrespl,"\n");
1.180 brouard 9436:
1.219 brouard 9437: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9438:
1.219 brouard 9439: agebase=ageminpar;
9440: agelim=agemaxpar;
1.180 brouard 9441:
1.227 brouard 9442: /* i1=pow(2,ncoveff); */
1.234 brouard 9443: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9444: if (cptcovn < 1){i1=1;}
1.180 brouard 9445:
1.238 brouard 9446: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9447: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9448: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9449: continue;
1.235 brouard 9450:
1.238 brouard 9451: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9452: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9453: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9454: /* k=k+1; */
9455: /* to clean */
9456: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9457: fprintf(ficrespl,"#******");
9458: printf("#******");
9459: fprintf(ficlog,"#******");
9460: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9461: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9462: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9463: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9464: }
9465: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9466: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9467: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9468: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9469: }
9470: fprintf(ficrespl,"******\n");
9471: printf("******\n");
9472: fprintf(ficlog,"******\n");
9473: if(invalidvarcomb[k]){
9474: printf("\nCombination (%d) ignored because no case \n",k);
9475: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9476: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9477: continue;
9478: }
1.219 brouard 9479:
1.238 brouard 9480: fprintf(ficrespl,"#Age ");
9481: for(j=1;j<=cptcoveff;j++) {
9482: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9483: }
9484: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9485: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9486:
1.238 brouard 9487: for (age=agebase; age<=agelim; age++){
9488: /* for (age=agebase; age<=agebase; age++){ */
9489: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9490: fprintf(ficrespl,"%.0f ",age );
9491: for(j=1;j<=cptcoveff;j++)
9492: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9493: tot=0.;
9494: for(i=1; i<=nlstate;i++){
9495: tot += prlim[i][i];
9496: fprintf(ficrespl," %.5f", prlim[i][i]);
9497: }
9498: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9499: } /* Age */
9500: /* was end of cptcod */
9501: } /* cptcov */
9502: } /* nres */
1.219 brouard 9503: return 0;
1.180 brouard 9504: }
9505:
1.218 brouard 9506: 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){
9507: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9508:
9509: /* Computes the back prevalence limit for any combination of covariate values
9510: * at any age between ageminpar and agemaxpar
9511: */
1.235 brouard 9512: int i, j, k, i1, nres=0 ;
1.217 brouard 9513: /* double ftolpl = 1.e-10; */
9514: double age, agebase, agelim;
9515: double tot;
1.218 brouard 9516: /* double ***mobaverage; */
9517: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9518:
9519: strcpy(fileresplb,"PLB_");
9520: strcat(fileresplb,fileresu);
9521: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9522: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9523: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9524: }
9525: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9526: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9527: pstamp(ficresplb);
9528: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9529: fprintf(ficresplb,"#Age ");
9530: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9531: fprintf(ficresplb,"\n");
9532:
1.218 brouard 9533:
9534: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9535:
9536: agebase=ageminpar;
9537: agelim=agemaxpar;
9538:
9539:
1.227 brouard 9540: i1=pow(2,cptcoveff);
1.218 brouard 9541: if (cptcovn < 1){i1=1;}
1.227 brouard 9542:
1.238 brouard 9543: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9544: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9545: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9546: continue;
9547: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9548: fprintf(ficresplb,"#******");
9549: printf("#******");
9550: fprintf(ficlog,"#******");
9551: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9552: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9553: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9554: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9555: }
9556: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9557: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9558: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9559: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9560: }
9561: fprintf(ficresplb,"******\n");
9562: printf("******\n");
9563: fprintf(ficlog,"******\n");
9564: if(invalidvarcomb[k]){
9565: printf("\nCombination (%d) ignored because no cases \n",k);
9566: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9567: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9568: continue;
9569: }
1.218 brouard 9570:
1.238 brouard 9571: fprintf(ficresplb,"#Age ");
9572: for(j=1;j<=cptcoveff;j++) {
9573: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9574: }
9575: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9576: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9577:
9578:
1.238 brouard 9579: for (age=agebase; age<=agelim; age++){
9580: /* for (age=agebase; age<=agebase; age++){ */
9581: if(mobilavproj > 0){
9582: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9583: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9584: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9585: }else if (mobilavproj == 0){
9586: 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);
9587: 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);
9588: exit(1);
9589: }else{
9590: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9591: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9592: }
9593: fprintf(ficresplb,"%.0f ",age );
9594: for(j=1;j<=cptcoveff;j++)
9595: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9596: tot=0.;
9597: for(i=1; i<=nlstate;i++){
9598: tot += bprlim[i][i];
9599: fprintf(ficresplb," %.5f", bprlim[i][i]);
9600: }
9601: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9602: } /* Age */
9603: /* was end of cptcod */
1.255 brouard 9604: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9605: } /* end of any combination */
9606: } /* end of nres */
1.218 brouard 9607: /* hBijx(p, bage, fage); */
9608: /* fclose(ficrespijb); */
9609:
9610: return 0;
1.217 brouard 9611: }
1.218 brouard 9612:
1.180 brouard 9613: int hPijx(double *p, int bage, int fage){
9614: /*------------- h Pij x at various ages ------------*/
9615:
9616: int stepsize;
9617: int agelim;
9618: int hstepm;
9619: int nhstepm;
1.235 brouard 9620: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9621:
9622: double agedeb;
9623: double ***p3mat;
9624:
1.201 brouard 9625: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9626: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9627: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9628: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9629: }
9630: printf("Computing pij: result on file '%s' \n", filerespij);
9631: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9632:
9633: stepsize=(int) (stepm+YEARM-1)/YEARM;
9634: /*if (stepm<=24) stepsize=2;*/
9635:
9636: agelim=AGESUP;
9637: hstepm=stepsize*YEARM; /* Every year of age */
9638: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9639:
1.180 brouard 9640: /* hstepm=1; aff par mois*/
9641: pstamp(ficrespij);
9642: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9643: i1= pow(2,cptcoveff);
1.218 brouard 9644: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9645: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9646: /* k=k+1; */
1.235 brouard 9647: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9648: for(k=1; k<=i1;k++){
1.253 brouard 9649: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9650: continue;
1.183 brouard 9651: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9652: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9653: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9654: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9655: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9656: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9657: }
1.183 brouard 9658: fprintf(ficrespij,"******\n");
9659:
9660: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9661: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9662: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9663:
9664: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9665:
1.183 brouard 9666: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9667: oldm=oldms;savm=savms;
1.235 brouard 9668: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9669: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9670: for(i=1; i<=nlstate;i++)
9671: for(j=1; j<=nlstate+ndeath;j++)
9672: fprintf(ficrespij," %1d-%1d",i,j);
9673: fprintf(ficrespij,"\n");
9674: for (h=0; h<=nhstepm; h++){
9675: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9676: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9677: for(i=1; i<=nlstate;i++)
9678: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9679: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9680: fprintf(ficrespij,"\n");
9681: }
1.183 brouard 9682: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9683: fprintf(ficrespij,"\n");
9684: }
1.180 brouard 9685: /*}*/
9686: }
1.218 brouard 9687: return 0;
1.180 brouard 9688: }
1.218 brouard 9689:
9690: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9691: /*------------- h Bij x at various ages ------------*/
9692:
9693: int stepsize;
1.218 brouard 9694: /* int agelim; */
9695: int ageminl;
1.217 brouard 9696: int hstepm;
9697: int nhstepm;
1.238 brouard 9698: int h, i, i1, j, k, nres;
1.218 brouard 9699:
1.217 brouard 9700: double agedeb;
9701: double ***p3mat;
1.218 brouard 9702:
9703: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9704: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9705: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9706: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9707: }
9708: printf("Computing pij back: result on file '%s' \n", filerespijb);
9709: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9710:
9711: stepsize=(int) (stepm+YEARM-1)/YEARM;
9712: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9713:
1.218 brouard 9714: /* agelim=AGESUP; */
9715: ageminl=30;
9716: hstepm=stepsize*YEARM; /* Every year of age */
9717: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9718:
9719: /* hstepm=1; aff par mois*/
9720: pstamp(ficrespijb);
1.255 brouard 9721: 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 9722: i1= pow(2,cptcoveff);
1.218 brouard 9723: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9724: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9725: /* k=k+1; */
1.238 brouard 9726: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9727: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9728: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9729: continue;
9730: fprintf(ficrespijb,"\n#****** ");
9731: for(j=1;j<=cptcoveff;j++)
9732: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9733: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9734: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9735: }
9736: fprintf(ficrespijb,"******\n");
9737: if(invalidvarcomb[k]){
9738: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9739: continue;
9740: }
9741:
9742: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9743: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9744: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9745: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9746: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9747:
9748: /* nhstepm=nhstepm*YEARM; aff par mois*/
9749:
9750: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9751: /* oldm=oldms;savm=savms; */
9752: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9753: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9754: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9755: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9756: for(i=1; i<=nlstate;i++)
9757: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9758: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9759: fprintf(ficrespijb,"\n");
1.238 brouard 9760: for (h=0; h<=nhstepm; h++){
9761: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9762: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9763: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9764: for(i=1; i<=nlstate;i++)
9765: for(j=1; j<=nlstate+ndeath;j++)
9766: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9767: fprintf(ficrespijb,"\n");
9768: }
9769: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9770: fprintf(ficrespijb,"\n");
9771: } /* end age deb */
9772: } /* end combination */
9773: } /* end nres */
1.218 brouard 9774: return 0;
9775: } /* hBijx */
1.217 brouard 9776:
1.180 brouard 9777:
1.136 brouard 9778: /***********************************************/
9779: /**************** Main Program *****************/
9780: /***********************************************/
9781:
9782: int main(int argc, char *argv[])
9783: {
9784: #ifdef GSL
9785: const gsl_multimin_fminimizer_type *T;
9786: size_t iteri = 0, it;
9787: int rval = GSL_CONTINUE;
9788: int status = GSL_SUCCESS;
9789: double ssval;
9790: #endif
9791: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9792: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9793: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9794: int jj, ll, li, lj, lk;
1.136 brouard 9795: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9796: int num_filled;
1.136 brouard 9797: int itimes;
9798: int NDIM=2;
9799: int vpopbased=0;
1.235 brouard 9800: int nres=0;
1.258 brouard 9801: int endishere=0;
1.136 brouard 9802:
1.164 brouard 9803: char ca[32], cb[32];
1.136 brouard 9804: /* FILE *fichtm; *//* Html File */
9805: /* FILE *ficgp;*/ /*Gnuplot File */
9806: struct stat info;
1.191 brouard 9807: double agedeb=0.;
1.194 brouard 9808:
9809: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9810: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9811:
1.165 brouard 9812: double fret;
1.191 brouard 9813: double dum=0.; /* Dummy variable */
1.136 brouard 9814: double ***p3mat;
1.218 brouard 9815: /* double ***mobaverage; */
1.164 brouard 9816:
9817: char line[MAXLINE];
1.197 brouard 9818: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9819:
1.234 brouard 9820: char modeltemp[MAXLINE];
1.230 brouard 9821: char resultline[MAXLINE];
9822:
1.136 brouard 9823: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9824: char *tok, *val; /* pathtot */
1.136 brouard 9825: int firstobs=1, lastobs=10;
1.195 brouard 9826: int c, h , cpt, c2;
1.191 brouard 9827: int jl=0;
9828: int i1, j1, jk, stepsize=0;
1.194 brouard 9829: int count=0;
9830:
1.164 brouard 9831: int *tab;
1.136 brouard 9832: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9833: int backcast=0;
1.136 brouard 9834: int mobilav=0,popforecast=0;
1.191 brouard 9835: int hstepm=0, nhstepm=0;
1.136 brouard 9836: int agemortsup;
9837: float sumlpop=0.;
9838: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9839: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9840:
1.191 brouard 9841: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9842: double ftolpl=FTOL;
9843: double **prlim;
1.217 brouard 9844: double **bprlim;
1.136 brouard 9845: double ***param; /* Matrix of parameters */
1.251 brouard 9846: double ***paramstart; /* Matrix of starting parameter values */
9847: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9848: double **matcov; /* Matrix of covariance */
1.203 brouard 9849: double **hess; /* Hessian matrix */
1.136 brouard 9850: double ***delti3; /* Scale */
9851: double *delti; /* Scale */
9852: double ***eij, ***vareij;
9853: double **varpl; /* Variances of prevalence limits by age */
9854: double *epj, vepp;
1.164 brouard 9855:
1.136 brouard 9856: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9857: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9858:
1.136 brouard 9859: double **ximort;
1.145 brouard 9860: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9861: int *dcwave;
9862:
1.164 brouard 9863: char z[1]="c";
1.136 brouard 9864:
9865: /*char *strt;*/
9866: char strtend[80];
1.126 brouard 9867:
1.164 brouard 9868:
1.126 brouard 9869: /* setlocale (LC_ALL, ""); */
9870: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9871: /* textdomain (PACKAGE); */
9872: /* setlocale (LC_CTYPE, ""); */
9873: /* setlocale (LC_MESSAGES, ""); */
9874:
9875: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9876: rstart_time = time(NULL);
9877: /* (void) gettimeofday(&start_time,&tzp);*/
9878: start_time = *localtime(&rstart_time);
1.126 brouard 9879: curr_time=start_time;
1.157 brouard 9880: /*tml = *localtime(&start_time.tm_sec);*/
9881: /* strcpy(strstart,asctime(&tml)); */
9882: strcpy(strstart,asctime(&start_time));
1.126 brouard 9883:
9884: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9885: /* tp.tm_sec = tp.tm_sec +86400; */
9886: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9887: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9888: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9889: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9890: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9891: /* strt=asctime(&tmg); */
9892: /* printf("Time(after) =%s",strstart); */
9893: /* (void) time (&time_value);
9894: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9895: * tm = *localtime(&time_value);
9896: * strstart=asctime(&tm);
9897: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9898: */
9899:
9900: nberr=0; /* Number of errors and warnings */
9901: nbwarn=0;
1.184 brouard 9902: #ifdef WIN32
9903: _getcwd(pathcd, size);
9904: #else
1.126 brouard 9905: getcwd(pathcd, size);
1.184 brouard 9906: #endif
1.191 brouard 9907: syscompilerinfo(0);
1.196 brouard 9908: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9909: if(argc <=1){
9910: printf("\nEnter the parameter file name: ");
1.205 brouard 9911: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9912: printf("ERROR Empty parameter file name\n");
9913: goto end;
9914: }
1.126 brouard 9915: i=strlen(pathr);
9916: if(pathr[i-1]=='\n')
9917: pathr[i-1]='\0';
1.156 brouard 9918: i=strlen(pathr);
1.205 brouard 9919: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9920: pathr[i-1]='\0';
1.205 brouard 9921: }
9922: i=strlen(pathr);
9923: if( i==0 ){
9924: printf("ERROR Empty parameter file name\n");
9925: goto end;
9926: }
9927: for (tok = pathr; tok != NULL; ){
1.126 brouard 9928: printf("Pathr |%s|\n",pathr);
9929: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9930: printf("val= |%s| pathr=%s\n",val,pathr);
9931: strcpy (pathtot, val);
9932: if(pathr[0] == '\0') break; /* Dirty */
9933: }
9934: }
9935: else{
9936: strcpy(pathtot,argv[1]);
9937: }
9938: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9939: /*cygwin_split_path(pathtot,path,optionfile);
9940: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9941: /* cutv(path,optionfile,pathtot,'\\');*/
9942:
9943: /* Split argv[0], imach program to get pathimach */
9944: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9945: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9946: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9947: /* strcpy(pathimach,argv[0]); */
9948: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9949: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9950: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9951: #ifdef WIN32
9952: _chdir(path); /* Can be a relative path */
9953: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9954: #else
1.126 brouard 9955: chdir(path); /* Can be a relative path */
1.184 brouard 9956: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9957: #endif
9958: printf("Current directory %s!\n",pathcd);
1.126 brouard 9959: strcpy(command,"mkdir ");
9960: strcat(command,optionfilefiname);
9961: if((outcmd=system(command)) != 0){
1.169 brouard 9962: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9963: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9964: /* fclose(ficlog); */
9965: /* exit(1); */
9966: }
9967: /* if((imk=mkdir(optionfilefiname))<0){ */
9968: /* perror("mkdir"); */
9969: /* } */
9970:
9971: /*-------- arguments in the command line --------*/
9972:
1.186 brouard 9973: /* Main Log file */
1.126 brouard 9974: strcat(filelog, optionfilefiname);
9975: strcat(filelog,".log"); /* */
9976: if((ficlog=fopen(filelog,"w"))==NULL) {
9977: printf("Problem with logfile %s\n",filelog);
9978: goto end;
9979: }
9980: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9981: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9982: fprintf(ficlog,"\nEnter the parameter file name: \n");
9983: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9984: path=%s \n\
9985: optionfile=%s\n\
9986: optionfilext=%s\n\
1.156 brouard 9987: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9988:
1.197 brouard 9989: syscompilerinfo(1);
1.167 brouard 9990:
1.126 brouard 9991: printf("Local time (at start):%s",strstart);
9992: fprintf(ficlog,"Local time (at start): %s",strstart);
9993: fflush(ficlog);
9994: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9995: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9996:
9997: /* */
9998: strcpy(fileres,"r");
9999: strcat(fileres, optionfilefiname);
1.201 brouard 10000: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10001: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10002: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10003:
1.186 brouard 10004: /* Main ---------arguments file --------*/
1.126 brouard 10005:
10006: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10007: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10008: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10009: fflush(ficlog);
1.149 brouard 10010: /* goto end; */
10011: exit(70);
1.126 brouard 10012: }
10013:
10014:
10015:
10016: strcpy(filereso,"o");
1.201 brouard 10017: strcat(filereso,fileresu);
1.126 brouard 10018: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10019: printf("Problem with Output resultfile: %s\n", filereso);
10020: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10021: fflush(ficlog);
10022: goto end;
10023: }
10024:
10025: /* Reads comments: lines beginning with '#' */
10026: numlinepar=0;
1.197 brouard 10027:
10028: /* First parameter line */
10029: while(fgets(line, MAXLINE, ficpar)) {
10030: /* If line starts with a # it is a comment */
10031: if (line[0] == '#') {
10032: numlinepar++;
10033: fputs(line,stdout);
10034: fputs(line,ficparo);
10035: fputs(line,ficlog);
10036: continue;
10037: }else
10038: break;
10039: }
10040: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10041: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10042: if (num_filled != 5) {
10043: printf("Should be 5 parameters\n");
10044: }
1.126 brouard 10045: numlinepar++;
1.197 brouard 10046: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10047: }
10048: /* Second parameter line */
10049: while(fgets(line, MAXLINE, ficpar)) {
10050: /* If line starts with a # it is a comment */
10051: if (line[0] == '#') {
10052: numlinepar++;
10053: fputs(line,stdout);
10054: fputs(line,ficparo);
10055: fputs(line,ficlog);
10056: continue;
10057: }else
10058: break;
10059: }
1.223 brouard 10060: 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", \
10061: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10062: if (num_filled != 11) {
10063: 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 10064: printf("but line=%s\n",line);
1.197 brouard 10065: }
1.223 brouard 10066: 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 10067: }
1.203 brouard 10068: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10069: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10070: /* Third parameter line */
10071: while(fgets(line, MAXLINE, ficpar)) {
10072: /* If line starts with a # it is a comment */
10073: if (line[0] == '#') {
10074: numlinepar++;
10075: fputs(line,stdout);
10076: fputs(line,ficparo);
10077: fputs(line,ficlog);
10078: continue;
10079: }else
10080: break;
10081: }
1.201 brouard 10082: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 ! brouard 10083: if (num_filled == 0){
! 10084: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
! 10085: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
! 10086: model[0]='\0';
! 10087: goto end;
! 10088: } else if (num_filled != 1){
1.197 brouard 10089: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10090: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10091: model[0]='\0';
10092: goto end;
10093: }
10094: else{
10095: if (model[0]=='+'){
10096: for(i=1; i<=strlen(model);i++)
10097: modeltemp[i-1]=model[i];
1.201 brouard 10098: strcpy(model,modeltemp);
1.197 brouard 10099: }
10100: }
1.199 brouard 10101: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10102: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10103: }
10104: /* 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); */
10105: /* numlinepar=numlinepar+3; /\* In general *\/ */
10106: /* 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 10107: 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);
10108: 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 10109: fflush(ficlog);
1.190 brouard 10110: /* if(model[0]=='#'|| model[0]== '\0'){ */
10111: if(model[0]=='#'){
1.187 brouard 10112: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10113: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10114: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10115: if(mle != -1){
10116: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10117: exit(1);
10118: }
10119: }
1.126 brouard 10120: while((c=getc(ficpar))=='#' && c!= EOF){
10121: ungetc(c,ficpar);
10122: fgets(line, MAXLINE, ficpar);
10123: numlinepar++;
1.195 brouard 10124: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10125: z[0]=line[1];
10126: }
10127: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10128: fputs(line, stdout);
10129: //puts(line);
1.126 brouard 10130: fputs(line,ficparo);
10131: fputs(line,ficlog);
10132: }
10133: ungetc(c,ficpar);
10134:
10135:
1.145 brouard 10136: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10137: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10138: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10139: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10140: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10141: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10142: v1+v2*age+v2*v3 makes cptcovn = 3
10143: */
10144: if (strlen(model)>1)
1.187 brouard 10145: 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 10146: else
1.187 brouard 10147: ncovmodel=2; /* Constant and age */
1.133 brouard 10148: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10149: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10150: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10151: 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);
10152: 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);
10153: fflush(stdout);
10154: fclose (ficlog);
10155: goto end;
10156: }
1.126 brouard 10157: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10158: delti=delti3[1][1];
10159: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10160: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10161: /* We could also provide initial parameters values giving by simple logistic regression
10162: * only one way, that is without matrix product. We will have nlstate maximizations */
10163: /* for(i=1;i<nlstate;i++){ */
10164: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10165: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10166: /* } */
1.126 brouard 10167: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10168: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10169: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10170: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10171: fclose (ficparo);
10172: fclose (ficlog);
10173: goto end;
10174: exit(0);
1.220 brouard 10175: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10176: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10177: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10178: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10179: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10180: matcov=matrix(1,npar,1,npar);
1.203 brouard 10181: hess=matrix(1,npar,1,npar);
1.220 brouard 10182: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10183: /* Read guessed parameters */
1.126 brouard 10184: /* Reads comments: lines beginning with '#' */
10185: while((c=getc(ficpar))=='#' && c!= EOF){
10186: ungetc(c,ficpar);
10187: fgets(line, MAXLINE, ficpar);
10188: numlinepar++;
1.141 brouard 10189: fputs(line,stdout);
1.126 brouard 10190: fputs(line,ficparo);
10191: fputs(line,ficlog);
10192: }
10193: ungetc(c,ficpar);
10194:
10195: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10196: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10197: for(i=1; i <=nlstate; i++){
1.234 brouard 10198: j=0;
1.126 brouard 10199: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10200: if(jj==i) continue;
10201: j++;
10202: fscanf(ficpar,"%1d%1d",&i1,&j1);
10203: if ((i1 != i) || (j1 != jj)){
10204: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10205: It might be a problem of design; if ncovcol and the model are correct\n \
10206: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10207: exit(1);
10208: }
10209: fprintf(ficparo,"%1d%1d",i1,j1);
10210: if(mle==1)
10211: printf("%1d%1d",i,jj);
10212: fprintf(ficlog,"%1d%1d",i,jj);
10213: for(k=1; k<=ncovmodel;k++){
10214: fscanf(ficpar," %lf",¶m[i][j][k]);
10215: if(mle==1){
10216: printf(" %lf",param[i][j][k]);
10217: fprintf(ficlog," %lf",param[i][j][k]);
10218: }
10219: else
10220: fprintf(ficlog," %lf",param[i][j][k]);
10221: fprintf(ficparo," %lf",param[i][j][k]);
10222: }
10223: fscanf(ficpar,"\n");
10224: numlinepar++;
10225: if(mle==1)
10226: printf("\n");
10227: fprintf(ficlog,"\n");
10228: fprintf(ficparo,"\n");
1.126 brouard 10229: }
10230: }
10231: fflush(ficlog);
1.234 brouard 10232:
1.251 brouard 10233: /* Reads parameters values */
1.126 brouard 10234: p=param[1][1];
1.251 brouard 10235: pstart=paramstart[1][1];
1.126 brouard 10236:
10237: /* Reads comments: lines beginning with '#' */
10238: while((c=getc(ficpar))=='#' && c!= EOF){
10239: ungetc(c,ficpar);
10240: fgets(line, MAXLINE, ficpar);
10241: numlinepar++;
1.141 brouard 10242: fputs(line,stdout);
1.126 brouard 10243: fputs(line,ficparo);
10244: fputs(line,ficlog);
10245: }
10246: ungetc(c,ficpar);
10247:
10248: for(i=1; i <=nlstate; i++){
10249: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10250: fscanf(ficpar,"%1d%1d",&i1,&j1);
10251: if ( (i1-i) * (j1-j) != 0){
10252: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10253: exit(1);
10254: }
10255: printf("%1d%1d",i,j);
10256: fprintf(ficparo,"%1d%1d",i1,j1);
10257: fprintf(ficlog,"%1d%1d",i1,j1);
10258: for(k=1; k<=ncovmodel;k++){
10259: fscanf(ficpar,"%le",&delti3[i][j][k]);
10260: printf(" %le",delti3[i][j][k]);
10261: fprintf(ficparo," %le",delti3[i][j][k]);
10262: fprintf(ficlog," %le",delti3[i][j][k]);
10263: }
10264: fscanf(ficpar,"\n");
10265: numlinepar++;
10266: printf("\n");
10267: fprintf(ficparo,"\n");
10268: fprintf(ficlog,"\n");
1.126 brouard 10269: }
10270: }
10271: fflush(ficlog);
1.234 brouard 10272:
1.145 brouard 10273: /* Reads covariance matrix */
1.126 brouard 10274: delti=delti3[1][1];
1.220 brouard 10275:
10276:
1.126 brouard 10277: /* 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 10278:
1.126 brouard 10279: /* Reads comments: lines beginning with '#' */
10280: while((c=getc(ficpar))=='#' && c!= EOF){
10281: ungetc(c,ficpar);
10282: fgets(line, MAXLINE, ficpar);
10283: numlinepar++;
1.141 brouard 10284: fputs(line,stdout);
1.126 brouard 10285: fputs(line,ficparo);
10286: fputs(line,ficlog);
10287: }
10288: ungetc(c,ficpar);
1.220 brouard 10289:
1.126 brouard 10290: matcov=matrix(1,npar,1,npar);
1.203 brouard 10291: hess=matrix(1,npar,1,npar);
1.131 brouard 10292: for(i=1; i <=npar; i++)
10293: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10294:
1.194 brouard 10295: /* Scans npar lines */
1.126 brouard 10296: for(i=1; i <=npar; i++){
1.226 brouard 10297: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10298: if(count != 3){
1.226 brouard 10299: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10300: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10301: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10302: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10303: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10304: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10305: exit(1);
1.220 brouard 10306: }else{
1.226 brouard 10307: if(mle==1)
10308: printf("%1d%1d%d",i1,j1,jk);
10309: }
10310: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10311: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10312: for(j=1; j <=i; j++){
1.226 brouard 10313: fscanf(ficpar," %le",&matcov[i][j]);
10314: if(mle==1){
10315: printf(" %.5le",matcov[i][j]);
10316: }
10317: fprintf(ficlog," %.5le",matcov[i][j]);
10318: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10319: }
10320: fscanf(ficpar,"\n");
10321: numlinepar++;
10322: if(mle==1)
1.220 brouard 10323: printf("\n");
1.126 brouard 10324: fprintf(ficlog,"\n");
10325: fprintf(ficparo,"\n");
10326: }
1.194 brouard 10327: /* End of read covariance matrix npar lines */
1.126 brouard 10328: for(i=1; i <=npar; i++)
10329: for(j=i+1;j<=npar;j++)
1.226 brouard 10330: matcov[i][j]=matcov[j][i];
1.126 brouard 10331:
10332: if(mle==1)
10333: printf("\n");
10334: fprintf(ficlog,"\n");
10335:
10336: fflush(ficlog);
10337:
10338: /*-------- Rewriting parameter file ----------*/
10339: strcpy(rfileres,"r"); /* "Rparameterfile */
10340: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10341: strcat(rfileres,"."); /* */
10342: strcat(rfileres,optionfilext); /* Other files have txt extension */
10343: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10344: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10345: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10346: }
10347: fprintf(ficres,"#%s\n",version);
10348: } /* End of mle != -3 */
1.218 brouard 10349:
1.186 brouard 10350: /* Main data
10351: */
1.126 brouard 10352: n= lastobs;
10353: num=lvector(1,n);
10354: moisnais=vector(1,n);
10355: annais=vector(1,n);
10356: moisdc=vector(1,n);
10357: andc=vector(1,n);
1.220 brouard 10358: weight=vector(1,n);
1.126 brouard 10359: agedc=vector(1,n);
10360: cod=ivector(1,n);
1.220 brouard 10361: for(i=1;i<=n;i++){
1.234 brouard 10362: num[i]=0;
10363: moisnais[i]=0;
10364: annais[i]=0;
10365: moisdc[i]=0;
10366: andc[i]=0;
10367: agedc[i]=0;
10368: cod[i]=0;
10369: weight[i]=1.0; /* Equal weights, 1 by default */
10370: }
1.126 brouard 10371: mint=matrix(1,maxwav,1,n);
10372: anint=matrix(1,maxwav,1,n);
1.131 brouard 10373: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10374: tab=ivector(1,NCOVMAX);
1.144 brouard 10375: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10376: 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 10377:
1.136 brouard 10378: /* Reads data from file datafile */
10379: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10380: goto end;
10381:
10382: /* Calculation of the number of parameters from char model */
1.234 brouard 10383: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10384: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10385: k=3 V4 Tvar[k=3]= 4 (from V4)
10386: k=2 V1 Tvar[k=2]= 1 (from V1)
10387: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10388: */
10389:
10390: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10391: TvarsDind=ivector(1,NCOVMAX); /* */
10392: TvarsD=ivector(1,NCOVMAX); /* */
10393: TvarsQind=ivector(1,NCOVMAX); /* */
10394: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10395: TvarF=ivector(1,NCOVMAX); /* */
10396: TvarFind=ivector(1,NCOVMAX); /* */
10397: TvarV=ivector(1,NCOVMAX); /* */
10398: TvarVind=ivector(1,NCOVMAX); /* */
10399: TvarA=ivector(1,NCOVMAX); /* */
10400: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10401: TvarFD=ivector(1,NCOVMAX); /* */
10402: TvarFDind=ivector(1,NCOVMAX); /* */
10403: TvarFQ=ivector(1,NCOVMAX); /* */
10404: TvarFQind=ivector(1,NCOVMAX); /* */
10405: TvarVD=ivector(1,NCOVMAX); /* */
10406: TvarVDind=ivector(1,NCOVMAX); /* */
10407: TvarVQ=ivector(1,NCOVMAX); /* */
10408: TvarVQind=ivector(1,NCOVMAX); /* */
10409:
1.230 brouard 10410: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10411: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10412: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10413: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10414: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10415: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10416: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10417: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10418: */
10419: /* For model-covariate k tells which data-covariate to use but
10420: because this model-covariate is a construction we invent a new column
10421: ncovcol + k1
10422: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10423: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10424: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10425: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10426: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10427: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10428: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10429: */
1.145 brouard 10430: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10431: 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 10432: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10433: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10434: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10435: 4 covariates (3 plus signs)
10436: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10437: */
1.230 brouard 10438: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10439: * individual dummy, fixed or varying:
10440: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10441: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10442: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10443: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10444: * Tmodelind[1]@9={9,0,3,2,}*/
10445: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10446: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10447: * individual quantitative, fixed or varying:
10448: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10449: * 3, 1, 0, 0, 0, 0, 0, 0},
10450: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10451: /* Main decodemodel */
10452:
1.187 brouard 10453:
1.223 brouard 10454: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10455: goto end;
10456:
1.137 brouard 10457: if((double)(lastobs-imx)/(double)imx > 1.10){
10458: nbwarn++;
10459: 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);
10460: 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);
10461: }
1.136 brouard 10462: /* if(mle==1){*/
1.137 brouard 10463: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10464: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10465: }
10466:
10467: /*-calculation of age at interview from date of interview and age at death -*/
10468: agev=matrix(1,maxwav,1,imx);
10469:
10470: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10471: goto end;
10472:
1.126 brouard 10473:
1.136 brouard 10474: agegomp=(int)agemin;
10475: free_vector(moisnais,1,n);
10476: free_vector(annais,1,n);
1.126 brouard 10477: /* free_matrix(mint,1,maxwav,1,n);
10478: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10479: /* free_vector(moisdc,1,n); */
10480: /* free_vector(andc,1,n); */
1.145 brouard 10481: /* */
10482:
1.126 brouard 10483: wav=ivector(1,imx);
1.214 brouard 10484: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10485: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10486: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10487: 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.*/
10488: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10489: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10490:
10491: /* Concatenates waves */
1.214 brouard 10492: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10493: Death is a valid wave (if date is known).
10494: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10495: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10496: and mw[mi+1][i]. dh depends on stepm.
10497: */
10498:
1.126 brouard 10499: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10500: /* Concatenates waves */
1.145 brouard 10501:
1.215 brouard 10502: free_vector(moisdc,1,n);
10503: free_vector(andc,1,n);
10504:
1.126 brouard 10505: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10506: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10507: ncodemax[1]=1;
1.145 brouard 10508: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10509: cptcoveff=0;
1.220 brouard 10510: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10511: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10512: }
10513:
10514: ncovcombmax=pow(2,cptcoveff);
10515: invalidvarcomb=ivector(1, ncovcombmax);
10516: for(i=1;i<ncovcombmax;i++)
10517: invalidvarcomb[i]=0;
10518:
1.211 brouard 10519: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10520: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10521: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10522:
1.200 brouard 10523: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10524: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10525: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10526: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10527: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10528: * (currently 0 or 1) in the data.
10529: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10530: * corresponding modality (h,j).
10531: */
10532:
1.145 brouard 10533: h=0;
10534: /*if (cptcovn > 0) */
1.126 brouard 10535: m=pow(2,cptcoveff);
10536:
1.144 brouard 10537: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10538: * For k=4 covariates, h goes from 1 to m=2**k
10539: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10540: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10541: * h\k 1 2 3 4
1.143 brouard 10542: *______________________________
10543: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10544: * 2 2 1 1 1
10545: * 3 i=2 1 2 1 1
10546: * 4 2 2 1 1
10547: * 5 i=3 1 i=2 1 2 1
10548: * 6 2 1 2 1
10549: * 7 i=4 1 2 2 1
10550: * 8 2 2 2 1
1.197 brouard 10551: * 9 i=5 1 i=3 1 i=2 1 2
10552: * 10 2 1 1 2
10553: * 11 i=6 1 2 1 2
10554: * 12 2 2 1 2
10555: * 13 i=7 1 i=4 1 2 2
10556: * 14 2 1 2 2
10557: * 15 i=8 1 2 2 2
10558: * 16 2 2 2 2
1.143 brouard 10559: */
1.212 brouard 10560: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10561: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10562: * and the value of each covariate?
10563: * V1=1, V2=1, V3=2, V4=1 ?
10564: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10565: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10566: * In order to get the real value in the data, we use nbcode
10567: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10568: * We are keeping this crazy system in order to be able (in the future?)
10569: * to have more than 2 values (0 or 1) for a covariate.
10570: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10571: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10572: * bbbbbbbb
10573: * 76543210
10574: * h-1 00000101 (6-1=5)
1.219 brouard 10575: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10576: * &
10577: * 1 00000001 (1)
1.219 brouard 10578: * 00000000 = 1 & ((h-1) >> (k-1))
10579: * +1= 00000001 =1
1.211 brouard 10580: *
10581: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10582: * h' 1101 =2^3+2^2+0x2^1+2^0
10583: * >>k' 11
10584: * & 00000001
10585: * = 00000001
10586: * +1 = 00000010=2 = codtabm(14,3)
10587: * Reverse h=6 and m=16?
10588: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10589: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10590: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10591: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10592: * V3=decodtabm(14,3,2**4)=2
10593: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10594: *(h-1) >> (j-1) 0011 =13 >> 2
10595: * &1 000000001
10596: * = 000000001
10597: * +1= 000000010 =2
10598: * 2211
10599: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10600: * V3=2
1.220 brouard 10601: * codtabm and decodtabm are identical
1.211 brouard 10602: */
10603:
1.145 brouard 10604:
10605: free_ivector(Ndum,-1,NCOVMAX);
10606:
10607:
1.126 brouard 10608:
1.186 brouard 10609: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10610: strcpy(optionfilegnuplot,optionfilefiname);
10611: if(mle==-3)
1.201 brouard 10612: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10613: strcat(optionfilegnuplot,".gp");
10614:
10615: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10616: printf("Problem with file %s",optionfilegnuplot);
10617: }
10618: else{
1.204 brouard 10619: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10620: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10621: //fprintf(ficgp,"set missing 'NaNq'\n");
10622: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10623: }
10624: /* fclose(ficgp);*/
1.186 brouard 10625:
10626:
10627: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10628:
10629: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10630: if(mle==-3)
1.201 brouard 10631: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10632: strcat(optionfilehtm,".htm");
10633: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10634: printf("Problem with %s \n",optionfilehtm);
10635: exit(0);
1.126 brouard 10636: }
10637:
10638: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10639: strcat(optionfilehtmcov,"-cov.htm");
10640: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10641: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10642: }
10643: else{
10644: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10645: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10646: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10647: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10648: }
10649:
1.213 brouard 10650: 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 10651: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10652: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10653: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10654: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10655: \n\
10656: <hr size=\"2\" color=\"#EC5E5E\">\
10657: <ul><li><h4>Parameter files</h4>\n\
10658: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10659: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10660: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10661: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10662: - Date and time at start: %s</ul>\n",\
10663: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10664: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10665: fileres,fileres,\
10666: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10667: fflush(fichtm);
10668:
10669: strcpy(pathr,path);
10670: strcat(pathr,optionfilefiname);
1.184 brouard 10671: #ifdef WIN32
10672: _chdir(optionfilefiname); /* Move to directory named optionfile */
10673: #else
1.126 brouard 10674: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10675: #endif
10676:
1.126 brouard 10677:
1.220 brouard 10678: /* Calculates basic frequencies. Computes observed prevalence at single age
10679: and for any valid combination of covariates
1.126 brouard 10680: and prints on file fileres'p'. */
1.251 brouard 10681: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10682: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10683:
10684: fprintf(fichtm,"\n");
10685: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10686: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10687: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10688: imx,agemin,agemax,jmin,jmax,jmean);
10689: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10690: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10691: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10692: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10693: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10694:
1.126 brouard 10695: /* For Powell, parameters are in a vector p[] starting at p[1]
10696: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10697: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10698:
10699: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10700: /* For mortality only */
1.126 brouard 10701: if (mle==-3){
1.136 brouard 10702: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10703: for(i=1;i<=NDIM;i++)
10704: for(j=1;j<=NDIM;j++)
10705: ximort[i][j]=0.;
1.186 brouard 10706: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10707: cens=ivector(1,n);
10708: ageexmed=vector(1,n);
10709: agecens=vector(1,n);
10710: dcwave=ivector(1,n);
1.223 brouard 10711:
1.126 brouard 10712: for (i=1; i<=imx; i++){
10713: dcwave[i]=-1;
10714: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10715: if (s[m][i]>nlstate) {
10716: dcwave[i]=m;
10717: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10718: break;
10719: }
1.126 brouard 10720: }
1.226 brouard 10721:
1.126 brouard 10722: for (i=1; i<=imx; i++) {
10723: if (wav[i]>0){
1.226 brouard 10724: ageexmed[i]=agev[mw[1][i]][i];
10725: j=wav[i];
10726: agecens[i]=1.;
10727:
10728: if (ageexmed[i]> 1 && wav[i] > 0){
10729: agecens[i]=agev[mw[j][i]][i];
10730: cens[i]= 1;
10731: }else if (ageexmed[i]< 1)
10732: cens[i]= -1;
10733: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10734: cens[i]=0 ;
1.126 brouard 10735: }
10736: else cens[i]=-1;
10737: }
10738:
10739: for (i=1;i<=NDIM;i++) {
10740: for (j=1;j<=NDIM;j++)
1.226 brouard 10741: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10742: }
10743:
1.145 brouard 10744: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10745: /*printf("%lf %lf", p[1], p[2]);*/
10746:
10747:
1.136 brouard 10748: #ifdef GSL
10749: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10750: #else
1.126 brouard 10751: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10752: #endif
1.201 brouard 10753: strcpy(filerespow,"POW-MORT_");
10754: strcat(filerespow,fileresu);
1.126 brouard 10755: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10756: printf("Problem with resultfile: %s\n", filerespow);
10757: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10758: }
1.136 brouard 10759: #ifdef GSL
10760: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10761: #else
1.126 brouard 10762: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10763: #endif
1.126 brouard 10764: /* for (i=1;i<=nlstate;i++)
10765: for(j=1;j<=nlstate+ndeath;j++)
10766: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10767: */
10768: fprintf(ficrespow,"\n");
1.136 brouard 10769: #ifdef GSL
10770: /* gsl starts here */
10771: T = gsl_multimin_fminimizer_nmsimplex;
10772: gsl_multimin_fminimizer *sfm = NULL;
10773: gsl_vector *ss, *x;
10774: gsl_multimin_function minex_func;
10775:
10776: /* Initial vertex size vector */
10777: ss = gsl_vector_alloc (NDIM);
10778:
10779: if (ss == NULL){
10780: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10781: }
10782: /* Set all step sizes to 1 */
10783: gsl_vector_set_all (ss, 0.001);
10784:
10785: /* Starting point */
1.126 brouard 10786:
1.136 brouard 10787: x = gsl_vector_alloc (NDIM);
10788:
10789: if (x == NULL){
10790: gsl_vector_free(ss);
10791: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10792: }
10793:
10794: /* Initialize method and iterate */
10795: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10796: /* gsl_vector_set(x, 0, 0.0268); */
10797: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10798: gsl_vector_set(x, 0, p[1]);
10799: gsl_vector_set(x, 1, p[2]);
10800:
10801: minex_func.f = &gompertz_f;
10802: minex_func.n = NDIM;
10803: minex_func.params = (void *)&p; /* ??? */
10804:
10805: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10806: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10807:
10808: printf("Iterations beginning .....\n\n");
10809: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10810:
10811: iteri=0;
10812: while (rval == GSL_CONTINUE){
10813: iteri++;
10814: status = gsl_multimin_fminimizer_iterate(sfm);
10815:
10816: if (status) printf("error: %s\n", gsl_strerror (status));
10817: fflush(0);
10818:
10819: if (status)
10820: break;
10821:
10822: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10823: ssval = gsl_multimin_fminimizer_size (sfm);
10824:
10825: if (rval == GSL_SUCCESS)
10826: printf ("converged to a local maximum at\n");
10827:
10828: printf("%5d ", iteri);
10829: for (it = 0; it < NDIM; it++){
10830: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10831: }
10832: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10833: }
10834:
10835: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10836:
10837: gsl_vector_free(x); /* initial values */
10838: gsl_vector_free(ss); /* inital step size */
10839: for (it=0; it<NDIM; it++){
10840: p[it+1]=gsl_vector_get(sfm->x,it);
10841: fprintf(ficrespow," %.12lf", p[it]);
10842: }
10843: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10844: #endif
10845: #ifdef POWELL
10846: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10847: #endif
1.126 brouard 10848: fclose(ficrespow);
10849:
1.203 brouard 10850: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10851:
10852: for(i=1; i <=NDIM; i++)
10853: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10854: matcov[i][j]=matcov[j][i];
1.126 brouard 10855:
10856: printf("\nCovariance matrix\n ");
1.203 brouard 10857: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10858: for(i=1; i <=NDIM; i++) {
10859: for(j=1;j<=NDIM;j++){
1.220 brouard 10860: printf("%f ",matcov[i][j]);
10861: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10862: }
1.203 brouard 10863: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10864: }
10865:
10866: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10867: for (i=1;i<=NDIM;i++) {
1.126 brouard 10868: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10869: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10870: }
1.126 brouard 10871: lsurv=vector(1,AGESUP);
10872: lpop=vector(1,AGESUP);
10873: tpop=vector(1,AGESUP);
10874: lsurv[agegomp]=100000;
10875:
10876: for (k=agegomp;k<=AGESUP;k++) {
10877: agemortsup=k;
10878: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10879: }
10880:
10881: for (k=agegomp;k<agemortsup;k++)
10882: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10883:
10884: for (k=agegomp;k<agemortsup;k++){
10885: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10886: sumlpop=sumlpop+lpop[k];
10887: }
10888:
10889: tpop[agegomp]=sumlpop;
10890: for (k=agegomp;k<(agemortsup-3);k++){
10891: /* tpop[k+1]=2;*/
10892: tpop[k+1]=tpop[k]-lpop[k];
10893: }
10894:
10895:
10896: printf("\nAge lx qx dx Lx Tx e(x)\n");
10897: for (k=agegomp;k<(agemortsup-2);k++)
10898: 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]);
10899:
10900:
10901: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10902: ageminpar=50;
10903: agemaxpar=100;
1.194 brouard 10904: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10905: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10906: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10907: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10908: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10909: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10910: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10911: }else{
10912: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10913: 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 10914: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10915: }
1.201 brouard 10916: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10917: stepm, weightopt,\
10918: model,imx,p,matcov,agemortsup);
10919:
10920: free_vector(lsurv,1,AGESUP);
10921: free_vector(lpop,1,AGESUP);
10922: free_vector(tpop,1,AGESUP);
1.220 brouard 10923: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10924: free_ivector(cens,1,n);
10925: free_vector(agecens,1,n);
10926: free_ivector(dcwave,1,n);
1.220 brouard 10927: #ifdef GSL
1.136 brouard 10928: #endif
1.186 brouard 10929: } /* Endof if mle==-3 mortality only */
1.205 brouard 10930: /* Standard */
10931: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10932: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10933: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10934: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10935: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10936: for (k=1; k<=npar;k++)
10937: printf(" %d %8.5f",k,p[k]);
10938: printf("\n");
1.205 brouard 10939: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10940: /* mlikeli uses func not funcone */
1.247 brouard 10941: /* for(i=1;i<nlstate;i++){ */
10942: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10943: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10944: /* } */
1.205 brouard 10945: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10946: }
10947: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10948: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10949: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10950: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10951: }
10952: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10953: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10954: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10955: for (k=1; k<=npar;k++)
10956: printf(" %d %8.5f",k,p[k]);
10957: printf("\n");
10958:
10959: /*--------- results files --------------*/
1.224 brouard 10960: 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 10961:
10962:
10963: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10964: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10965: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10966: for(i=1,jk=1; i <=nlstate; i++){
10967: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10968: if (k != i) {
10969: printf("%d%d ",i,k);
10970: fprintf(ficlog,"%d%d ",i,k);
10971: fprintf(ficres,"%1d%1d ",i,k);
10972: for(j=1; j <=ncovmodel; j++){
10973: printf("%12.7f ",p[jk]);
10974: fprintf(ficlog,"%12.7f ",p[jk]);
10975: fprintf(ficres,"%12.7f ",p[jk]);
10976: jk++;
10977: }
10978: printf("\n");
10979: fprintf(ficlog,"\n");
10980: fprintf(ficres,"\n");
10981: }
1.126 brouard 10982: }
10983: }
1.203 brouard 10984: if(mle != 0){
10985: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10986: ftolhess=ftol; /* Usually correct */
1.203 brouard 10987: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10988: 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");
10989: 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");
10990: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10991: for(k=1; k <=(nlstate+ndeath); k++){
10992: if (k != i) {
10993: printf("%d%d ",i,k);
10994: fprintf(ficlog,"%d%d ",i,k);
10995: for(j=1; j <=ncovmodel; j++){
10996: 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]));
10997: 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]));
10998: jk++;
10999: }
11000: printf("\n");
11001: fprintf(ficlog,"\n");
11002: }
11003: }
1.193 brouard 11004: }
1.203 brouard 11005: } /* end of hesscov and Wald tests */
1.225 brouard 11006:
1.203 brouard 11007: /* */
1.126 brouard 11008: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11009: printf("# Scales (for hessian or gradient estimation)\n");
11010: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11011: for(i=1,jk=1; i <=nlstate; i++){
11012: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11013: if (j!=i) {
11014: fprintf(ficres,"%1d%1d",i,j);
11015: printf("%1d%1d",i,j);
11016: fprintf(ficlog,"%1d%1d",i,j);
11017: for(k=1; k<=ncovmodel;k++){
11018: printf(" %.5e",delti[jk]);
11019: fprintf(ficlog," %.5e",delti[jk]);
11020: fprintf(ficres," %.5e",delti[jk]);
11021: jk++;
11022: }
11023: printf("\n");
11024: fprintf(ficlog,"\n");
11025: fprintf(ficres,"\n");
11026: }
1.126 brouard 11027: }
11028: }
11029:
11030: 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 11031: if(mle >= 1) /* To big for the screen */
1.126 brouard 11032: 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");
11033: 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");
11034: /* # 121 Var(a12)\n\ */
11035: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11036: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11037: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11038: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11039: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11040: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11041: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11042:
11043:
11044: /* Just to have a covariance matrix which will be more understandable
11045: even is we still don't want to manage dictionary of variables
11046: */
11047: for(itimes=1;itimes<=2;itimes++){
11048: jj=0;
11049: for(i=1; i <=nlstate; i++){
1.225 brouard 11050: for(j=1; j <=nlstate+ndeath; j++){
11051: if(j==i) continue;
11052: for(k=1; k<=ncovmodel;k++){
11053: jj++;
11054: ca[0]= k+'a'-1;ca[1]='\0';
11055: if(itimes==1){
11056: if(mle>=1)
11057: printf("#%1d%1d%d",i,j,k);
11058: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11059: fprintf(ficres,"#%1d%1d%d",i,j,k);
11060: }else{
11061: if(mle>=1)
11062: printf("%1d%1d%d",i,j,k);
11063: fprintf(ficlog,"%1d%1d%d",i,j,k);
11064: fprintf(ficres,"%1d%1d%d",i,j,k);
11065: }
11066: ll=0;
11067: for(li=1;li <=nlstate; li++){
11068: for(lj=1;lj <=nlstate+ndeath; lj++){
11069: if(lj==li) continue;
11070: for(lk=1;lk<=ncovmodel;lk++){
11071: ll++;
11072: if(ll<=jj){
11073: cb[0]= lk +'a'-1;cb[1]='\0';
11074: if(ll<jj){
11075: if(itimes==1){
11076: if(mle>=1)
11077: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11078: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11079: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11080: }else{
11081: if(mle>=1)
11082: printf(" %.5e",matcov[jj][ll]);
11083: fprintf(ficlog," %.5e",matcov[jj][ll]);
11084: fprintf(ficres," %.5e",matcov[jj][ll]);
11085: }
11086: }else{
11087: if(itimes==1){
11088: if(mle>=1)
11089: printf(" Var(%s%1d%1d)",ca,i,j);
11090: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11091: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11092: }else{
11093: if(mle>=1)
11094: printf(" %.7e",matcov[jj][ll]);
11095: fprintf(ficlog," %.7e",matcov[jj][ll]);
11096: fprintf(ficres," %.7e",matcov[jj][ll]);
11097: }
11098: }
11099: }
11100: } /* end lk */
11101: } /* end lj */
11102: } /* end li */
11103: if(mle>=1)
11104: printf("\n");
11105: fprintf(ficlog,"\n");
11106: fprintf(ficres,"\n");
11107: numlinepar++;
11108: } /* end k*/
11109: } /*end j */
1.126 brouard 11110: } /* end i */
11111: } /* end itimes */
11112:
11113: fflush(ficlog);
11114: fflush(ficres);
1.225 brouard 11115: while(fgets(line, MAXLINE, ficpar)) {
11116: /* If line starts with a # it is a comment */
11117: if (line[0] == '#') {
11118: numlinepar++;
11119: fputs(line,stdout);
11120: fputs(line,ficparo);
11121: fputs(line,ficlog);
11122: continue;
11123: }else
11124: break;
11125: }
11126:
1.209 brouard 11127: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11128: /* ungetc(c,ficpar); */
11129: /* fgets(line, MAXLINE, ficpar); */
11130: /* fputs(line,stdout); */
11131: /* fputs(line,ficparo); */
11132: /* } */
11133: /* ungetc(c,ficpar); */
1.126 brouard 11134:
11135: estepm=0;
1.209 brouard 11136: 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 11137:
11138: if (num_filled != 6) {
11139: 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);
11140: 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);
11141: goto end;
11142: }
11143: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11144: }
11145: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11146: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11147:
1.209 brouard 11148: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11149: if (estepm==0 || estepm < stepm) estepm=stepm;
11150: if (fage <= 2) {
11151: bage = ageminpar;
11152: fage = agemaxpar;
11153: }
11154:
11155: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11156: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11157: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11158:
1.186 brouard 11159: /* Other stuffs, more or less useful */
1.254 brouard 11160: while(fgets(line, MAXLINE, ficpar)) {
11161: /* If line starts with a # it is a comment */
11162: if (line[0] == '#') {
11163: numlinepar++;
11164: fputs(line,stdout);
11165: fputs(line,ficparo);
11166: fputs(line,ficlog);
11167: continue;
11168: }else
11169: break;
11170: }
11171:
11172: 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){
11173:
11174: if (num_filled != 7) {
11175: 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);
11176: 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);
11177: goto end;
11178: }
11179: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11180: 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);
11181: 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);
11182: 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 11183: }
1.254 brouard 11184:
11185: while(fgets(line, MAXLINE, ficpar)) {
11186: /* If line starts with a # it is a comment */
11187: if (line[0] == '#') {
11188: numlinepar++;
11189: fputs(line,stdout);
11190: fputs(line,ficparo);
11191: fputs(line,ficlog);
11192: continue;
11193: }else
11194: break;
1.126 brouard 11195: }
11196:
11197:
11198: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11199: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11200:
1.254 brouard 11201: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11202: if (num_filled != 1) {
11203: 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);
11204: 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);
11205: goto end;
11206: }
11207: printf("pop_based=%d\n",popbased);
11208: fprintf(ficlog,"pop_based=%d\n",popbased);
11209: fprintf(ficparo,"pop_based=%d\n",popbased);
11210: fprintf(ficres,"pop_based=%d\n",popbased);
11211: }
11212:
1.258 brouard 11213: /* Results */
11214: nresult=0;
11215: do{
11216: if(!fgets(line, MAXLINE, ficpar)){
11217: endishere=1;
11218: parameterline=14;
11219: }else if (line[0] == '#') {
11220: /* If line starts with a # it is a comment */
1.254 brouard 11221: numlinepar++;
11222: fputs(line,stdout);
11223: fputs(line,ficparo);
11224: fputs(line,ficlog);
11225: continue;
1.258 brouard 11226: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11227: parameterline=11;
11228: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11229: parameterline=12;
11230: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11231: parameterline=13;
11232: else{
11233: parameterline=14;
1.254 brouard 11234: }
1.258 brouard 11235: switch (parameterline){
11236: case 11:
11237: 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){
11238: if (num_filled != 8) {
11239: 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);
11240: 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);
11241: goto end;
11242: }
11243: 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);
11244: 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);
11245: 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);
11246: 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);
11247: /* day and month of proj2 are not used but only year anproj2.*/
11248: }
1.254 brouard 11249: break;
1.258 brouard 11250: case 12:
11251: /*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);*/
11252: 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){
11253: if (num_filled != 8) {
1.262 brouard 11254: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11255: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
1.258 brouard 11256: goto end;
11257: }
11258: 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);
11259: 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);
11260: 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);
11261: 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);
11262: /* day and month of proj2 are not used but only year anproj2.*/
11263: }
1.230 brouard 11264: break;
1.258 brouard 11265: case 13:
11266: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11267: if (num_filled == 0){
11268: resultline[0]='\0';
11269: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11270: 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);
11271: break;
11272: } else if (num_filled != 1){
11273: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11274: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11275: }
11276: nresult++; /* Sum of resultlines */
11277: printf("Result %d: result=%s\n",nresult, resultline);
11278: if(nresult > MAXRESULTLINES){
11279: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11280: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11281: goto end;
11282: }
11283: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11284: fprintf(ficparo,"result: %s\n",resultline);
11285: fprintf(ficres,"result: %s\n",resultline);
11286: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11287: break;
1.258 brouard 11288: case 14:
1.259 brouard 11289: if(ncovmodel >2 && nresult==0 ){
11290: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11291: goto end;
11292: }
1.259 brouard 11293: break;
1.258 brouard 11294: default:
11295: nresult=1;
11296: decoderesult(".",nresult ); /* No covariate */
11297: }
11298: } /* End switch parameterline */
11299: }while(endishere==0); /* End do */
1.126 brouard 11300:
1.230 brouard 11301: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11302: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11303:
11304: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11305: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11306: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11307: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11308: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11309: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11310: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11311: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11312: }else{
1.218 brouard 11313: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11314: }
11315: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11316: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11317: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11318:
1.225 brouard 11319: /*------------ free_vector -------------*/
11320: /* chdir(path); */
1.220 brouard 11321:
1.215 brouard 11322: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11323: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11324: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11325: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11326: free_lvector(num,1,n);
11327: free_vector(agedc,1,n);
11328: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11329: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11330: fclose(ficparo);
11331: fclose(ficres);
1.220 brouard 11332:
11333:
1.186 brouard 11334: /* Other results (useful)*/
1.220 brouard 11335:
11336:
1.126 brouard 11337: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11338: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11339: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11340: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11341: fclose(ficrespl);
11342:
11343: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11344: /*#include "hpijx.h"*/
11345: hPijx(p, bage, fage);
1.145 brouard 11346: fclose(ficrespij);
1.227 brouard 11347:
1.220 brouard 11348: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11349: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11350: k=1;
1.126 brouard 11351: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11352:
1.219 brouard 11353: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11354: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11355: for(i=1;i<=AGESUP;i++)
1.219 brouard 11356: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11357: for(k=1;k<=ncovcombmax;k++)
11358: probs[i][j][k]=0.;
1.219 brouard 11359: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11360: if (mobilav!=0 ||mobilavproj !=0 ) {
11361: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11362: for(i=1;i<=AGESUP;i++)
11363: for(j=1;j<=nlstate;j++)
11364: for(k=1;k<=ncovcombmax;k++)
11365: mobaverages[i][j][k]=0.;
1.219 brouard 11366: mobaverage=mobaverages;
11367: if (mobilav!=0) {
1.235 brouard 11368: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11369: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11370: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11371: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11372: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11373: }
1.219 brouard 11374: }
11375: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11376: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11377: else if (mobilavproj !=0) {
1.235 brouard 11378: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11379: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11380: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11381: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11382: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11383: }
1.219 brouard 11384: }
11385: }/* end if moving average */
1.227 brouard 11386:
1.126 brouard 11387: /*---------- Forecasting ------------------*/
11388: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11389: if(prevfcast==1){
11390: /* if(stepm ==1){*/
1.225 brouard 11391: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11392: }
1.217 brouard 11393: if(backcast==1){
1.219 brouard 11394: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11395: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11396: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11397:
11398: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11399:
11400: bprlim=matrix(1,nlstate,1,nlstate);
11401: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11402: fclose(ficresplb);
11403:
1.222 brouard 11404: hBijx(p, bage, fage, mobaverage);
11405: fclose(ficrespijb);
1.219 brouard 11406: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11407:
11408: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11409: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11410: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11411: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11412: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11413: }
1.217 brouard 11414:
1.186 brouard 11415:
11416: /* ------ Other prevalence ratios------------ */
1.126 brouard 11417:
1.215 brouard 11418: free_ivector(wav,1,imx);
11419: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11420: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11421: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11422:
11423:
1.127 brouard 11424: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11425:
1.201 brouard 11426: strcpy(filerese,"E_");
11427: strcat(filerese,fileresu);
1.126 brouard 11428: if((ficreseij=fopen(filerese,"w"))==NULL) {
11429: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11430: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11431: }
1.208 brouard 11432: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11433: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11434:
11435: pstamp(ficreseij);
1.219 brouard 11436:
1.235 brouard 11437: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11438: if (cptcovn < 1){i1=1;}
11439:
11440: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11441: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11442: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11443: continue;
1.219 brouard 11444: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11445: printf("\n#****** ");
1.225 brouard 11446: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11447: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11448: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11449: }
11450: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11451: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11452: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11453: }
11454: fprintf(ficreseij,"******\n");
1.235 brouard 11455: printf("******\n");
1.219 brouard 11456:
11457: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11458: oldm=oldms;savm=savms;
1.235 brouard 11459: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11460:
1.219 brouard 11461: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11462: }
11463: fclose(ficreseij);
1.208 brouard 11464: printf("done evsij\n");fflush(stdout);
11465: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11466:
1.227 brouard 11467: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11468:
11469:
1.201 brouard 11470: strcpy(filerest,"T_");
11471: strcat(filerest,fileresu);
1.127 brouard 11472: if((ficrest=fopen(filerest,"w"))==NULL) {
11473: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11474: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11475: }
1.208 brouard 11476: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11477: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11478:
1.126 brouard 11479:
1.201 brouard 11480: strcpy(fileresstde,"STDE_");
11481: strcat(fileresstde,fileresu);
1.126 brouard 11482: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11483: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11484: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11485: }
1.227 brouard 11486: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11487: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11488:
1.201 brouard 11489: strcpy(filerescve,"CVE_");
11490: strcat(filerescve,fileresu);
1.126 brouard 11491: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11492: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11493: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11494: }
1.227 brouard 11495: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11496: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11497:
1.201 brouard 11498: strcpy(fileresv,"V_");
11499: strcat(fileresv,fileresu);
1.126 brouard 11500: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11501: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11502: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11503: }
1.227 brouard 11504: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11505: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11506:
1.145 brouard 11507: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11508: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11509:
1.235 brouard 11510: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11511: if (cptcovn < 1){i1=1;}
11512:
11513: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11514: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11515: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11516: continue;
1.242 brouard 11517: printf("\n#****** Result for:");
11518: fprintf(ficrest,"\n#****** Result for:");
11519: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11520: for(j=1;j<=cptcoveff;j++){
11521: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11522: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11523: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11524: }
1.235 brouard 11525: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11526: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11527: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11528: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11529: }
1.208 brouard 11530: fprintf(ficrest,"******\n");
1.227 brouard 11531: fprintf(ficlog,"******\n");
11532: printf("******\n");
1.208 brouard 11533:
11534: fprintf(ficresstdeij,"\n#****** ");
11535: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11536: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11537: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11538: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11539: }
1.235 brouard 11540: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11541: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11542: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11543: }
1.208 brouard 11544: fprintf(ficresstdeij,"******\n");
11545: fprintf(ficrescveij,"******\n");
11546:
11547: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11548: /* pstamp(ficresvij); */
1.225 brouard 11549: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11550: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11551: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11552: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11553: }
1.208 brouard 11554: fprintf(ficresvij,"******\n");
11555:
11556: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11557: oldm=oldms;savm=savms;
1.235 brouard 11558: printf(" cvevsij ");
11559: fprintf(ficlog, " cvevsij ");
11560: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11561: printf(" end cvevsij \n ");
11562: fprintf(ficlog, " end cvevsij \n ");
11563:
11564: /*
11565: */
11566: /* goto endfree; */
11567:
11568: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11569: pstamp(ficrest);
11570:
11571:
11572: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11573: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11574: cptcod= 0; /* To be deleted */
11575: printf("varevsij vpopbased=%d \n",vpopbased);
11576: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11577: 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 11578: 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 ");
11579: if(vpopbased==1)
11580: 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);
11581: else
11582: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11583: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11584: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11585: fprintf(ficrest,"\n");
11586: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11587: epj=vector(1,nlstate+1);
11588: printf("Computing age specific period (stable) prevalences in each health state \n");
11589: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11590: for(age=bage; age <=fage ;age++){
1.235 brouard 11591: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11592: if (vpopbased==1) {
11593: if(mobilav ==0){
11594: for(i=1; i<=nlstate;i++)
11595: prlim[i][i]=probs[(int)age][i][k];
11596: }else{ /* mobilav */
11597: for(i=1; i<=nlstate;i++)
11598: prlim[i][i]=mobaverage[(int)age][i][k];
11599: }
11600: }
1.219 brouard 11601:
1.227 brouard 11602: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11603: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11604: /* printf(" age %4.0f ",age); */
11605: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11606: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11607: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11608: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11609: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11610: }
11611: epj[nlstate+1] +=epj[j];
11612: }
11613: /* printf(" age %4.0f \n",age); */
1.219 brouard 11614:
1.227 brouard 11615: for(i=1, vepp=0.;i <=nlstate;i++)
11616: for(j=1;j <=nlstate;j++)
11617: vepp += vareij[i][j][(int)age];
11618: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11619: for(j=1;j <=nlstate;j++){
11620: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11621: }
11622: fprintf(ficrest,"\n");
11623: }
1.208 brouard 11624: } /* End vpopbased */
11625: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11626: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11627: free_vector(epj,1,nlstate+1);
1.235 brouard 11628: printf("done selection\n");fflush(stdout);
11629: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11630:
1.145 brouard 11631: /*}*/
1.235 brouard 11632: } /* End k selection */
1.227 brouard 11633:
11634: printf("done State-specific expectancies\n");fflush(stdout);
11635: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11636:
1.126 brouard 11637: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11638:
1.201 brouard 11639: strcpy(fileresvpl,"VPL_");
11640: strcat(fileresvpl,fileresu);
1.126 brouard 11641: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11642: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11643: exit(0);
11644: }
1.208 brouard 11645: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11646: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11647:
1.145 brouard 11648: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11649: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11650:
1.235 brouard 11651: i1=pow(2,cptcoveff);
11652: if (cptcovn < 1){i1=1;}
11653:
11654: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11655: for(k=1; k<=i1;k++){
1.253 brouard 11656: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11657: continue;
1.227 brouard 11658: fprintf(ficresvpl,"\n#****** ");
11659: printf("\n#****** ");
11660: fprintf(ficlog,"\n#****** ");
11661: for(j=1;j<=cptcoveff;j++) {
11662: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11663: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11664: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11665: }
1.235 brouard 11666: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11667: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11668: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11669: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11670: }
1.227 brouard 11671: fprintf(ficresvpl,"******\n");
11672: printf("******\n");
11673: fprintf(ficlog,"******\n");
11674:
11675: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11676: oldm=oldms;savm=savms;
1.235 brouard 11677: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11678: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11679: /*}*/
1.126 brouard 11680: }
1.227 brouard 11681:
1.126 brouard 11682: fclose(ficresvpl);
1.208 brouard 11683: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11684: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11685:
11686: free_vector(weight,1,n);
11687: free_imatrix(Tvard,1,NCOVMAX,1,2);
11688: free_imatrix(s,1,maxwav+1,1,n);
11689: free_matrix(anint,1,maxwav,1,n);
11690: free_matrix(mint,1,maxwav,1,n);
11691: free_ivector(cod,1,n);
11692: free_ivector(tab,1,NCOVMAX);
11693: fclose(ficresstdeij);
11694: fclose(ficrescveij);
11695: fclose(ficresvij);
11696: fclose(ficrest);
11697: fclose(ficpar);
11698:
11699:
1.126 brouard 11700: /*---------- End : free ----------------*/
1.219 brouard 11701: if (mobilav!=0 ||mobilavproj !=0)
11702: 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 11703: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11704: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11705: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11706: } /* mle==-3 arrives here for freeing */
1.227 brouard 11707: /* endfree:*/
11708: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11709: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11710: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11711: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11712: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11713: free_matrix(coqvar,1,maxwav,1,n);
11714: free_matrix(covar,0,NCOVMAX,1,n);
11715: free_matrix(matcov,1,npar,1,npar);
11716: free_matrix(hess,1,npar,1,npar);
11717: /*free_vector(delti,1,npar);*/
11718: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11719: free_matrix(agev,1,maxwav,1,imx);
11720: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11721:
11722: free_ivector(ncodemax,1,NCOVMAX);
11723: free_ivector(ncodemaxwundef,1,NCOVMAX);
11724: free_ivector(Dummy,-1,NCOVMAX);
11725: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11726: free_ivector(DummyV,1,NCOVMAX);
11727: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11728: free_ivector(Typevar,-1,NCOVMAX);
11729: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11730: free_ivector(TvarsQ,1,NCOVMAX);
11731: free_ivector(TvarsQind,1,NCOVMAX);
11732: free_ivector(TvarsD,1,NCOVMAX);
11733: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11734: free_ivector(TvarFD,1,NCOVMAX);
11735: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11736: free_ivector(TvarF,1,NCOVMAX);
11737: free_ivector(TvarFind,1,NCOVMAX);
11738: free_ivector(TvarV,1,NCOVMAX);
11739: free_ivector(TvarVind,1,NCOVMAX);
11740: free_ivector(TvarA,1,NCOVMAX);
11741: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11742: free_ivector(TvarFQ,1,NCOVMAX);
11743: free_ivector(TvarFQind,1,NCOVMAX);
11744: free_ivector(TvarVD,1,NCOVMAX);
11745: free_ivector(TvarVDind,1,NCOVMAX);
11746: free_ivector(TvarVQ,1,NCOVMAX);
11747: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11748: free_ivector(Tvarsel,1,NCOVMAX);
11749: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11750: free_ivector(Tposprod,1,NCOVMAX);
11751: free_ivector(Tprod,1,NCOVMAX);
11752: free_ivector(Tvaraff,1,NCOVMAX);
11753: free_ivector(invalidvarcomb,1,ncovcombmax);
11754: free_ivector(Tage,1,NCOVMAX);
11755: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11756: free_ivector(TmodelInvind,1,NCOVMAX);
11757: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11758:
11759: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11760: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11761: fflush(fichtm);
11762: fflush(ficgp);
11763:
1.227 brouard 11764:
1.126 brouard 11765: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11766: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11767: 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 11768: }else{
11769: printf("End of Imach\n");
11770: fprintf(ficlog,"End of Imach\n");
11771: }
11772: printf("See log file on %s\n",filelog);
11773: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11774: /*(void) gettimeofday(&end_time,&tzp);*/
11775: rend_time = time(NULL);
11776: end_time = *localtime(&rend_time);
11777: /* tml = *localtime(&end_time.tm_sec); */
11778: strcpy(strtend,asctime(&end_time));
1.126 brouard 11779: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11780: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11781: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11782:
1.157 brouard 11783: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11784: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11785: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11786: /* printf("Total time was %d uSec.\n", total_usecs);*/
11787: /* if(fileappend(fichtm,optionfilehtm)){ */
11788: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11789: fclose(fichtm);
11790: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11791: fclose(fichtmcov);
11792: fclose(ficgp);
11793: fclose(ficlog);
11794: /*------ End -----------*/
1.227 brouard 11795:
11796:
11797: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11798: #ifdef WIN32
1.227 brouard 11799: if (_chdir(pathcd) != 0)
11800: printf("Can't move to directory %s!\n",path);
11801: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11802: #else
1.227 brouard 11803: if(chdir(pathcd) != 0)
11804: printf("Can't move to directory %s!\n", path);
11805: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11806: #endif
1.126 brouard 11807: printf("Current directory %s!\n",pathcd);
11808: /*strcat(plotcmd,CHARSEPARATOR);*/
11809: sprintf(plotcmd,"gnuplot");
1.157 brouard 11810: #ifdef _WIN32
1.126 brouard 11811: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11812: #endif
11813: if(!stat(plotcmd,&info)){
1.158 brouard 11814: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11815: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11816: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11817: }else
11818: strcpy(pplotcmd,plotcmd);
1.157 brouard 11819: #ifdef __unix
1.126 brouard 11820: strcpy(plotcmd,GNUPLOTPROGRAM);
11821: if(!stat(plotcmd,&info)){
1.158 brouard 11822: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11823: }else
11824: strcpy(pplotcmd,plotcmd);
11825: #endif
11826: }else
11827: strcpy(pplotcmd,plotcmd);
11828:
11829: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11830: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11831:
1.126 brouard 11832: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11833: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11834: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11835: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11836: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11837: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11838: }
1.158 brouard 11839: printf(" Successful, please wait...");
1.126 brouard 11840: while (z[0] != 'q') {
11841: /* chdir(path); */
1.154 brouard 11842: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11843: scanf("%s",z);
11844: /* if (z[0] == 'c') system("./imach"); */
11845: if (z[0] == 'e') {
1.158 brouard 11846: #ifdef __APPLE__
1.152 brouard 11847: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11848: #elif __linux
11849: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11850: #else
1.152 brouard 11851: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11852: #endif
11853: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11854: system(pplotcmd);
1.126 brouard 11855: }
11856: else if (z[0] == 'g') system(plotcmd);
11857: else if (z[0] == 'q') exit(0);
11858: }
1.227 brouard 11859: end:
1.126 brouard 11860: while (z[0] != 'q') {
1.195 brouard 11861: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11862: scanf("%s",z);
11863: }
11864: }
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