Annotation of imach/src/imach.c, revision 1.261
1.261 ! brouard 1: /* $Id: imach.c,v 1.260 2017/04/04 17:46:59 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.261 ! brouard 4: Revision 1.260 2017/04/04 17:46:59 brouard
! 5: Summary: Gnuplot indexations fixed (humm)
! 6:
1.260 brouard 7: Revision 1.259 2017/04/04 13:01:16 brouard
8: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
9:
1.259 brouard 10: Revision 1.258 2017/04/03 10:17:47 brouard
11: Summary: Version 0.99r12
12:
13: Some cleanings, conformed with updated documentation.
14:
1.258 brouard 15: Revision 1.257 2017/03/29 16:53:30 brouard
16: Summary: Temp
17:
1.257 brouard 18: Revision 1.256 2017/03/27 05:50:23 brouard
19: Summary: Temporary
20:
1.256 brouard 21: Revision 1.255 2017/03/08 16:02:28 brouard
22: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
23:
1.255 brouard 24: Revision 1.254 2017/03/08 07:13:00 brouard
25: Summary: Fixing data parameter line
26:
1.254 brouard 27: Revision 1.253 2016/12/15 11:59:41 brouard
28: Summary: 0.99 in progress
29:
1.253 brouard 30: Revision 1.252 2016/09/15 21:15:37 brouard
31: *** empty log message ***
32:
1.252 brouard 33: Revision 1.251 2016/09/15 15:01:13 brouard
34: Summary: not working
35:
1.251 brouard 36: Revision 1.250 2016/09/08 16:07:27 brouard
37: Summary: continue
38:
1.250 brouard 39: Revision 1.249 2016/09/07 17:14:18 brouard
40: Summary: Starting values from frequencies
41:
1.249 brouard 42: Revision 1.248 2016/09/07 14:10:18 brouard
43: *** empty log message ***
44:
1.248 brouard 45: Revision 1.247 2016/09/02 11:11:21 brouard
46: *** empty log message ***
47:
1.247 brouard 48: Revision 1.246 2016/09/02 08:49:22 brouard
49: *** empty log message ***
50:
1.246 brouard 51: Revision 1.245 2016/09/02 07:25:01 brouard
52: *** empty log message ***
53:
1.245 brouard 54: Revision 1.244 2016/09/02 07:17:34 brouard
55: *** empty log message ***
56:
1.244 brouard 57: Revision 1.243 2016/09/02 06:45:35 brouard
58: *** empty log message ***
59:
1.243 brouard 60: Revision 1.242 2016/08/30 15:01:20 brouard
61: Summary: Fixing a lots
62:
1.242 brouard 63: Revision 1.241 2016/08/29 17:17:25 brouard
64: Summary: gnuplot problem in Back projection to fix
65:
1.241 brouard 66: Revision 1.240 2016/08/29 07:53:18 brouard
67: Summary: Better
68:
1.240 brouard 69: Revision 1.239 2016/08/26 15:51:03 brouard
70: Summary: Improvement in Powell output in order to copy and paste
71:
72: Author:
73:
1.239 brouard 74: Revision 1.238 2016/08/26 14:23:35 brouard
75: Summary: Starting tests of 0.99
76:
1.238 brouard 77: Revision 1.237 2016/08/26 09:20:19 brouard
78: Summary: to valgrind
79:
1.237 brouard 80: Revision 1.236 2016/08/25 10:50:18 brouard
81: *** empty log message ***
82:
1.236 brouard 83: Revision 1.235 2016/08/25 06:59:23 brouard
84: *** empty log message ***
85:
1.235 brouard 86: Revision 1.234 2016/08/23 16:51:20 brouard
87: *** empty log message ***
88:
1.234 brouard 89: Revision 1.233 2016/08/23 07:40:50 brouard
90: Summary: not working
91:
1.233 brouard 92: Revision 1.232 2016/08/22 14:20:21 brouard
93: Summary: not working
94:
1.232 brouard 95: Revision 1.231 2016/08/22 07:17:15 brouard
96: Summary: not working
97:
1.231 brouard 98: Revision 1.230 2016/08/22 06:55:53 brouard
99: Summary: Not working
100:
1.230 brouard 101: Revision 1.229 2016/07/23 09:45:53 brouard
102: Summary: Completing for func too
103:
1.229 brouard 104: Revision 1.228 2016/07/22 17:45:30 brouard
105: Summary: Fixing some arrays, still debugging
106:
1.227 brouard 107: Revision 1.226 2016/07/12 18:42:34 brouard
108: Summary: temp
109:
1.226 brouard 110: Revision 1.225 2016/07/12 08:40:03 brouard
111: Summary: saving but not running
112:
1.225 brouard 113: Revision 1.224 2016/07/01 13:16:01 brouard
114: Summary: Fixes
115:
1.224 brouard 116: Revision 1.223 2016/02/19 09:23:35 brouard
117: Summary: temporary
118:
1.223 brouard 119: Revision 1.222 2016/02/17 08:14:50 brouard
120: Summary: Probably last 0.98 stable version 0.98r6
121:
1.222 brouard 122: Revision 1.221 2016/02/15 23:35:36 brouard
123: Summary: minor bug
124:
1.220 brouard 125: Revision 1.219 2016/02/15 00:48:12 brouard
126: *** empty log message ***
127:
1.219 brouard 128: Revision 1.218 2016/02/12 11:29:23 brouard
129: Summary: 0.99 Back projections
130:
1.218 brouard 131: Revision 1.217 2015/12/23 17:18:31 brouard
132: Summary: Experimental backcast
133:
1.217 brouard 134: Revision 1.216 2015/12/18 17:32:11 brouard
135: Summary: 0.98r4 Warning and status=-2
136:
137: Version 0.98r4 is now:
138: - displaying an error when status is -1, date of interview unknown and date of death known;
139: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
140: Older changes concerning s=-2, dating from 2005 have been supersed.
141:
1.216 brouard 142: Revision 1.215 2015/12/16 08:52:24 brouard
143: Summary: 0.98r4 working
144:
1.215 brouard 145: Revision 1.214 2015/12/16 06:57:54 brouard
146: Summary: temporary not working
147:
1.214 brouard 148: Revision 1.213 2015/12/11 18:22:17 brouard
149: Summary: 0.98r4
150:
1.213 brouard 151: Revision 1.212 2015/11/21 12:47:24 brouard
152: Summary: minor typo
153:
1.212 brouard 154: Revision 1.211 2015/11/21 12:41:11 brouard
155: Summary: 0.98r3 with some graph of projected cross-sectional
156:
157: Author: Nicolas Brouard
158:
1.211 brouard 159: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 160: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 161: Summary: Adding ftolpl parameter
162: Author: N Brouard
163:
164: We had difficulties to get smoothed confidence intervals. It was due
165: to the period prevalence which wasn't computed accurately. The inner
166: parameter ftolpl is now an outer parameter of the .imach parameter
167: file after estepm. If ftolpl is small 1.e-4 and estepm too,
168: computation are long.
169:
1.209 brouard 170: Revision 1.208 2015/11/17 14:31:57 brouard
171: Summary: temporary
172:
1.208 brouard 173: Revision 1.207 2015/10/27 17:36:57 brouard
174: *** empty log message ***
175:
1.207 brouard 176: Revision 1.206 2015/10/24 07:14:11 brouard
177: *** empty log message ***
178:
1.206 brouard 179: Revision 1.205 2015/10/23 15:50:53 brouard
180: Summary: 0.98r3 some clarification for graphs on likelihood contributions
181:
1.205 brouard 182: Revision 1.204 2015/10/01 16:20:26 brouard
183: Summary: Some new graphs of contribution to likelihood
184:
1.204 brouard 185: Revision 1.203 2015/09/30 17:45:14 brouard
186: Summary: looking at better estimation of the hessian
187:
188: Also a better criteria for convergence to the period prevalence And
189: therefore adding the number of years needed to converge. (The
190: prevalence in any alive state shold sum to one
191:
1.203 brouard 192: Revision 1.202 2015/09/22 19:45:16 brouard
193: Summary: Adding some overall graph on contribution to likelihood. Might change
194:
1.202 brouard 195: Revision 1.201 2015/09/15 17:34:58 brouard
196: Summary: 0.98r0
197:
198: - Some new graphs like suvival functions
199: - Some bugs fixed like model=1+age+V2.
200:
1.201 brouard 201: Revision 1.200 2015/09/09 16:53:55 brouard
202: Summary: Big bug thanks to Flavia
203:
204: Even model=1+age+V2. did not work anymore
205:
1.200 brouard 206: Revision 1.199 2015/09/07 14:09:23 brouard
207: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
208:
1.199 brouard 209: Revision 1.198 2015/09/03 07:14:39 brouard
210: Summary: 0.98q5 Flavia
211:
1.198 brouard 212: Revision 1.197 2015/09/01 18:24:39 brouard
213: *** empty log message ***
214:
1.197 brouard 215: Revision 1.196 2015/08/18 23:17:52 brouard
216: Summary: 0.98q5
217:
1.196 brouard 218: Revision 1.195 2015/08/18 16:28:39 brouard
219: Summary: Adding a hack for testing purpose
220:
221: After reading the title, ftol and model lines, if the comment line has
222: a q, starting with #q, the answer at the end of the run is quit. It
223: permits to run test files in batch with ctest. The former workaround was
224: $ echo q | imach foo.imach
225:
1.195 brouard 226: Revision 1.194 2015/08/18 13:32:00 brouard
227: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
228:
1.194 brouard 229: Revision 1.193 2015/08/04 07:17:42 brouard
230: Summary: 0.98q4
231:
1.193 brouard 232: Revision 1.192 2015/07/16 16:49:02 brouard
233: Summary: Fixing some outputs
234:
1.192 brouard 235: Revision 1.191 2015/07/14 10:00:33 brouard
236: Summary: Some fixes
237:
1.191 brouard 238: Revision 1.190 2015/05/05 08:51:13 brouard
239: Summary: Adding digits in output parameters (7 digits instead of 6)
240:
241: Fix 1+age+.
242:
1.190 brouard 243: Revision 1.189 2015/04/30 14:45:16 brouard
244: Summary: 0.98q2
245:
1.189 brouard 246: Revision 1.188 2015/04/30 08:27:53 brouard
247: *** empty log message ***
248:
1.188 brouard 249: Revision 1.187 2015/04/29 09:11:15 brouard
250: *** empty log message ***
251:
1.187 brouard 252: Revision 1.186 2015/04/23 12:01:52 brouard
253: Summary: V1*age is working now, version 0.98q1
254:
255: Some codes had been disabled in order to simplify and Vn*age was
256: working in the optimization phase, ie, giving correct MLE parameters,
257: but, as usual, outputs were not correct and program core dumped.
258:
1.186 brouard 259: Revision 1.185 2015/03/11 13:26:42 brouard
260: Summary: Inclusion of compile and links command line for Intel Compiler
261:
1.185 brouard 262: Revision 1.184 2015/03/11 11:52:39 brouard
263: Summary: Back from Windows 8. Intel Compiler
264:
1.184 brouard 265: Revision 1.183 2015/03/10 20:34:32 brouard
266: Summary: 0.98q0, trying with directest, mnbrak fixed
267:
268: We use directest instead of original Powell test; probably no
269: incidence on the results, but better justifications;
270: We fixed Numerical Recipes mnbrak routine which was wrong and gave
271: wrong results.
272:
1.183 brouard 273: Revision 1.182 2015/02/12 08:19:57 brouard
274: Summary: Trying to keep directest which seems simpler and more general
275: Author: Nicolas Brouard
276:
1.182 brouard 277: Revision 1.181 2015/02/11 23:22:24 brouard
278: Summary: Comments on Powell added
279:
280: Author:
281:
1.181 brouard 282: Revision 1.180 2015/02/11 17:33:45 brouard
283: Summary: Finishing move from main to function (hpijx and prevalence_limit)
284:
1.180 brouard 285: Revision 1.179 2015/01/04 09:57:06 brouard
286: Summary: back to OS/X
287:
1.179 brouard 288: Revision 1.178 2015/01/04 09:35:48 brouard
289: *** empty log message ***
290:
1.178 brouard 291: Revision 1.177 2015/01/03 18:40:56 brouard
292: Summary: Still testing ilc32 on OSX
293:
1.177 brouard 294: Revision 1.176 2015/01/03 16:45:04 brouard
295: *** empty log message ***
296:
1.176 brouard 297: Revision 1.175 2015/01/03 16:33:42 brouard
298: *** empty log message ***
299:
1.175 brouard 300: Revision 1.174 2015/01/03 16:15:49 brouard
301: Summary: Still in cross-compilation
302:
1.174 brouard 303: Revision 1.173 2015/01/03 12:06:26 brouard
304: Summary: trying to detect cross-compilation
305:
1.173 brouard 306: Revision 1.172 2014/12/27 12:07:47 brouard
307: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
308:
1.172 brouard 309: Revision 1.171 2014/12/23 13:26:59 brouard
310: Summary: Back from Visual C
311:
312: Still problem with utsname.h on Windows
313:
1.171 brouard 314: Revision 1.170 2014/12/23 11:17:12 brouard
315: Summary: Cleaning some \%% back to %%
316:
317: The escape was mandatory for a specific compiler (which one?), but too many warnings.
318:
1.170 brouard 319: Revision 1.169 2014/12/22 23:08:31 brouard
320: Summary: 0.98p
321:
322: Outputs some informations on compiler used, OS etc. Testing on different platforms.
323:
1.169 brouard 324: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 325: Summary: update
1.169 brouard 326:
1.168 brouard 327: Revision 1.167 2014/12/22 13:50:56 brouard
328: Summary: Testing uname and compiler version and if compiled 32 or 64
329:
330: Testing on Linux 64
331:
1.167 brouard 332: Revision 1.166 2014/12/22 11:40:47 brouard
333: *** empty log message ***
334:
1.166 brouard 335: Revision 1.165 2014/12/16 11:20:36 brouard
336: Summary: After compiling on Visual C
337:
338: * imach.c (Module): Merging 1.61 to 1.162
339:
1.165 brouard 340: Revision 1.164 2014/12/16 10:52:11 brouard
341: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
342:
343: * imach.c (Module): Merging 1.61 to 1.162
344:
1.164 brouard 345: Revision 1.163 2014/12/16 10:30:11 brouard
346: * imach.c (Module): Merging 1.61 to 1.162
347:
1.163 brouard 348: Revision 1.162 2014/09/25 11:43:39 brouard
349: Summary: temporary backup 0.99!
350:
1.162 brouard 351: Revision 1.1 2014/09/16 11:06:58 brouard
352: Summary: With some code (wrong) for nlopt
353:
354: Author:
355:
356: Revision 1.161 2014/09/15 20:41:41 brouard
357: Summary: Problem with macro SQR on Intel compiler
358:
1.161 brouard 359: Revision 1.160 2014/09/02 09:24:05 brouard
360: *** empty log message ***
361:
1.160 brouard 362: Revision 1.159 2014/09/01 10:34:10 brouard
363: Summary: WIN32
364: Author: Brouard
365:
1.159 brouard 366: Revision 1.158 2014/08/27 17:11:51 brouard
367: *** empty log message ***
368:
1.158 brouard 369: Revision 1.157 2014/08/27 16:26:55 brouard
370: Summary: Preparing windows Visual studio version
371: Author: Brouard
372:
373: In order to compile on Visual studio, time.h is now correct and time_t
374: and tm struct should be used. difftime should be used but sometimes I
375: just make the differences in raw time format (time(&now).
376: Trying to suppress #ifdef LINUX
377: Add xdg-open for __linux in order to open default browser.
378:
1.157 brouard 379: Revision 1.156 2014/08/25 20:10:10 brouard
380: *** empty log message ***
381:
1.156 brouard 382: Revision 1.155 2014/08/25 18:32:34 brouard
383: Summary: New compile, minor changes
384: Author: Brouard
385:
1.155 brouard 386: Revision 1.154 2014/06/20 17:32:08 brouard
387: Summary: Outputs now all graphs of convergence to period prevalence
388:
1.154 brouard 389: Revision 1.153 2014/06/20 16:45:46 brouard
390: Summary: If 3 live state, convergence to period prevalence on same graph
391: Author: Brouard
392:
1.153 brouard 393: Revision 1.152 2014/06/18 17:54:09 brouard
394: Summary: open browser, use gnuplot on same dir than imach if not found in the path
395:
1.152 brouard 396: Revision 1.151 2014/06/18 16:43:30 brouard
397: *** empty log message ***
398:
1.151 brouard 399: Revision 1.150 2014/06/18 16:42:35 brouard
400: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
401: Author: brouard
402:
1.150 brouard 403: Revision 1.149 2014/06/18 15:51:14 brouard
404: Summary: Some fixes in parameter files errors
405: Author: Nicolas Brouard
406:
1.149 brouard 407: Revision 1.148 2014/06/17 17:38:48 brouard
408: Summary: Nothing new
409: Author: Brouard
410:
411: Just a new packaging for OS/X version 0.98nS
412:
1.148 brouard 413: Revision 1.147 2014/06/16 10:33:11 brouard
414: *** empty log message ***
415:
1.147 brouard 416: Revision 1.146 2014/06/16 10:20:28 brouard
417: Summary: Merge
418: Author: Brouard
419:
420: Merge, before building revised version.
421:
1.146 brouard 422: Revision 1.145 2014/06/10 21:23:15 brouard
423: Summary: Debugging with valgrind
424: Author: Nicolas Brouard
425:
426: Lot of changes in order to output the results with some covariates
427: After the Edimburgh REVES conference 2014, it seems mandatory to
428: improve the code.
429: No more memory valgrind error but a lot has to be done in order to
430: continue the work of splitting the code into subroutines.
431: Also, decodemodel has been improved. Tricode is still not
432: optimal. nbcode should be improved. Documentation has been added in
433: the source code.
434:
1.144 brouard 435: Revision 1.143 2014/01/26 09:45:38 brouard
436: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
437:
438: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
439: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
440:
1.143 brouard 441: Revision 1.142 2014/01/26 03:57:36 brouard
442: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
443:
444: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
445:
1.142 brouard 446: Revision 1.141 2014/01/26 02:42:01 brouard
447: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
448:
1.141 brouard 449: Revision 1.140 2011/09/02 10:37:54 brouard
450: Summary: times.h is ok with mingw32 now.
451:
1.140 brouard 452: Revision 1.139 2010/06/14 07:50:17 brouard
453: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
454: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
455:
1.139 brouard 456: Revision 1.138 2010/04/30 18:19:40 brouard
457: *** empty log message ***
458:
1.138 brouard 459: Revision 1.137 2010/04/29 18:11:38 brouard
460: (Module): Checking covariates for more complex models
461: than V1+V2. A lot of change to be done. Unstable.
462:
1.137 brouard 463: Revision 1.136 2010/04/26 20:30:53 brouard
464: (Module): merging some libgsl code. Fixing computation
465: of likelione (using inter/intrapolation if mle = 0) in order to
466: get same likelihood as if mle=1.
467: Some cleaning of code and comments added.
468:
1.136 brouard 469: Revision 1.135 2009/10/29 15:33:14 brouard
470: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
471:
1.135 brouard 472: Revision 1.134 2009/10/29 13:18:53 brouard
473: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
474:
1.134 brouard 475: Revision 1.133 2009/07/06 10:21:25 brouard
476: just nforces
477:
1.133 brouard 478: Revision 1.132 2009/07/06 08:22:05 brouard
479: Many tings
480:
1.132 brouard 481: Revision 1.131 2009/06/20 16:22:47 brouard
482: Some dimensions resccaled
483:
1.131 brouard 484: Revision 1.130 2009/05/26 06:44:34 brouard
485: (Module): Max Covariate is now set to 20 instead of 8. A
486: lot of cleaning with variables initialized to 0. Trying to make
487: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
488:
1.130 brouard 489: Revision 1.129 2007/08/31 13:49:27 lievre
490: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
491:
1.129 lievre 492: Revision 1.128 2006/06/30 13:02:05 brouard
493: (Module): Clarifications on computing e.j
494:
1.128 brouard 495: Revision 1.127 2006/04/28 18:11:50 brouard
496: (Module): Yes the sum of survivors was wrong since
497: imach-114 because nhstepm was no more computed in the age
498: loop. Now we define nhstepma in the age loop.
499: (Module): In order to speed up (in case of numerous covariates) we
500: compute health expectancies (without variances) in a first step
501: and then all the health expectancies with variances or standard
502: deviation (needs data from the Hessian matrices) which slows the
503: computation.
504: In the future we should be able to stop the program is only health
505: expectancies and graph are needed without standard deviations.
506:
1.127 brouard 507: Revision 1.126 2006/04/28 17:23:28 brouard
508: (Module): Yes the sum of survivors was wrong since
509: imach-114 because nhstepm was no more computed in the age
510: loop. Now we define nhstepma in the age loop.
511: Version 0.98h
512:
1.126 brouard 513: Revision 1.125 2006/04/04 15:20:31 lievre
514: Errors in calculation of health expectancies. Age was not initialized.
515: Forecasting file added.
516:
517: Revision 1.124 2006/03/22 17:13:53 lievre
518: Parameters are printed with %lf instead of %f (more numbers after the comma).
519: The log-likelihood is printed in the log file
520:
521: Revision 1.123 2006/03/20 10:52:43 brouard
522: * imach.c (Module): <title> changed, corresponds to .htm file
523: name. <head> headers where missing.
524:
525: * imach.c (Module): Weights can have a decimal point as for
526: English (a comma might work with a correct LC_NUMERIC environment,
527: otherwise the weight is truncated).
528: Modification of warning when the covariates values are not 0 or
529: 1.
530: Version 0.98g
531:
532: Revision 1.122 2006/03/20 09:45:41 brouard
533: (Module): Weights can have a decimal point as for
534: English (a comma might work with a correct LC_NUMERIC environment,
535: otherwise the weight is truncated).
536: Modification of warning when the covariates values are not 0 or
537: 1.
538: Version 0.98g
539:
540: Revision 1.121 2006/03/16 17:45:01 lievre
541: * imach.c (Module): Comments concerning covariates added
542:
543: * imach.c (Module): refinements in the computation of lli if
544: status=-2 in order to have more reliable computation if stepm is
545: not 1 month. Version 0.98f
546:
547: Revision 1.120 2006/03/16 15:10:38 lievre
548: (Module): refinements in the computation of lli if
549: status=-2 in order to have more reliable computation if stepm is
550: not 1 month. Version 0.98f
551:
552: Revision 1.119 2006/03/15 17:42:26 brouard
553: (Module): Bug if status = -2, the loglikelihood was
554: computed as likelihood omitting the logarithm. Version O.98e
555:
556: Revision 1.118 2006/03/14 18:20:07 brouard
557: (Module): varevsij Comments added explaining the second
558: table of variances if popbased=1 .
559: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
560: (Module): Function pstamp added
561: (Module): Version 0.98d
562:
563: Revision 1.117 2006/03/14 17:16:22 brouard
564: (Module): varevsij Comments added explaining the second
565: table of variances if popbased=1 .
566: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
567: (Module): Function pstamp added
568: (Module): Version 0.98d
569:
570: Revision 1.116 2006/03/06 10:29:27 brouard
571: (Module): Variance-covariance wrong links and
572: varian-covariance of ej. is needed (Saito).
573:
574: Revision 1.115 2006/02/27 12:17:45 brouard
575: (Module): One freematrix added in mlikeli! 0.98c
576:
577: Revision 1.114 2006/02/26 12:57:58 brouard
578: (Module): Some improvements in processing parameter
579: filename with strsep.
580:
581: Revision 1.113 2006/02/24 14:20:24 brouard
582: (Module): Memory leaks checks with valgrind and:
583: datafile was not closed, some imatrix were not freed and on matrix
584: allocation too.
585:
586: Revision 1.112 2006/01/30 09:55:26 brouard
587: (Module): Back to gnuplot.exe instead of wgnuplot.exe
588:
589: Revision 1.111 2006/01/25 20:38:18 brouard
590: (Module): Lots of cleaning and bugs added (Gompertz)
591: (Module): Comments can be added in data file. Missing date values
592: can be a simple dot '.'.
593:
594: Revision 1.110 2006/01/25 00:51:50 brouard
595: (Module): Lots of cleaning and bugs added (Gompertz)
596:
597: Revision 1.109 2006/01/24 19:37:15 brouard
598: (Module): Comments (lines starting with a #) are allowed in data.
599:
600: Revision 1.108 2006/01/19 18:05:42 lievre
601: Gnuplot problem appeared...
602: To be fixed
603:
604: Revision 1.107 2006/01/19 16:20:37 brouard
605: Test existence of gnuplot in imach path
606:
607: Revision 1.106 2006/01/19 13:24:36 brouard
608: Some cleaning and links added in html output
609:
610: Revision 1.105 2006/01/05 20:23:19 lievre
611: *** empty log message ***
612:
613: Revision 1.104 2005/09/30 16:11:43 lievre
614: (Module): sump fixed, loop imx fixed, and simplifications.
615: (Module): If the status is missing at the last wave but we know
616: that the person is alive, then we can code his/her status as -2
617: (instead of missing=-1 in earlier versions) and his/her
618: contributions to the likelihood is 1 - Prob of dying from last
619: health status (= 1-p13= p11+p12 in the easiest case of somebody in
620: the healthy state at last known wave). Version is 0.98
621:
622: Revision 1.103 2005/09/30 15:54:49 lievre
623: (Module): sump fixed, loop imx fixed, and simplifications.
624:
625: Revision 1.102 2004/09/15 17:31:30 brouard
626: Add the possibility to read data file including tab characters.
627:
628: Revision 1.101 2004/09/15 10:38:38 brouard
629: Fix on curr_time
630:
631: Revision 1.100 2004/07/12 18:29:06 brouard
632: Add version for Mac OS X. Just define UNIX in Makefile
633:
634: Revision 1.99 2004/06/05 08:57:40 brouard
635: *** empty log message ***
636:
637: Revision 1.98 2004/05/16 15:05:56 brouard
638: New version 0.97 . First attempt to estimate force of mortality
639: directly from the data i.e. without the need of knowing the health
640: state at each age, but using a Gompertz model: log u =a + b*age .
641: This is the basic analysis of mortality and should be done before any
642: other analysis, in order to test if the mortality estimated from the
643: cross-longitudinal survey is different from the mortality estimated
644: from other sources like vital statistic data.
645:
646: The same imach parameter file can be used but the option for mle should be -3.
647:
1.133 brouard 648: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 649: former routines in order to include the new code within the former code.
650:
651: The output is very simple: only an estimate of the intercept and of
652: the slope with 95% confident intervals.
653:
654: Current limitations:
655: A) Even if you enter covariates, i.e. with the
656: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
657: B) There is no computation of Life Expectancy nor Life Table.
658:
659: Revision 1.97 2004/02/20 13:25:42 lievre
660: Version 0.96d. Population forecasting command line is (temporarily)
661: suppressed.
662:
663: Revision 1.96 2003/07/15 15:38:55 brouard
664: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
665: rewritten within the same printf. Workaround: many printfs.
666:
667: Revision 1.95 2003/07/08 07:54:34 brouard
668: * imach.c (Repository):
669: (Repository): Using imachwizard code to output a more meaningful covariance
670: matrix (cov(a12,c31) instead of numbers.
671:
672: Revision 1.94 2003/06/27 13:00:02 brouard
673: Just cleaning
674:
675: Revision 1.93 2003/06/25 16:33:55 brouard
676: (Module): On windows (cygwin) function asctime_r doesn't
677: exist so I changed back to asctime which exists.
678: (Module): Version 0.96b
679:
680: Revision 1.92 2003/06/25 16:30:45 brouard
681: (Module): On windows (cygwin) function asctime_r doesn't
682: exist so I changed back to asctime which exists.
683:
684: Revision 1.91 2003/06/25 15:30:29 brouard
685: * imach.c (Repository): Duplicated warning errors corrected.
686: (Repository): Elapsed time after each iteration is now output. It
687: helps to forecast when convergence will be reached. Elapsed time
688: is stamped in powell. We created a new html file for the graphs
689: concerning matrix of covariance. It has extension -cov.htm.
690:
691: Revision 1.90 2003/06/24 12:34:15 brouard
692: (Module): Some bugs corrected for windows. Also, when
693: mle=-1 a template is output in file "or"mypar.txt with the design
694: of the covariance matrix to be input.
695:
696: Revision 1.89 2003/06/24 12:30:52 brouard
697: (Module): Some bugs corrected for windows. Also, when
698: mle=-1 a template is output in file "or"mypar.txt with the design
699: of the covariance matrix to be input.
700:
701: Revision 1.88 2003/06/23 17:54:56 brouard
702: * 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.
703:
704: Revision 1.87 2003/06/18 12:26:01 brouard
705: Version 0.96
706:
707: Revision 1.86 2003/06/17 20:04:08 brouard
708: (Module): Change position of html and gnuplot routines and added
709: routine fileappend.
710:
711: Revision 1.85 2003/06/17 13:12:43 brouard
712: * imach.c (Repository): Check when date of death was earlier that
713: current date of interview. It may happen when the death was just
714: prior to the death. In this case, dh was negative and likelihood
715: was wrong (infinity). We still send an "Error" but patch by
716: assuming that the date of death was just one stepm after the
717: interview.
718: (Repository): Because some people have very long ID (first column)
719: we changed int to long in num[] and we added a new lvector for
720: memory allocation. But we also truncated to 8 characters (left
721: truncation)
722: (Repository): No more line truncation errors.
723:
724: Revision 1.84 2003/06/13 21:44:43 brouard
725: * imach.c (Repository): Replace "freqsummary" at a correct
726: place. It differs from routine "prevalence" which may be called
727: many times. Probs is memory consuming and must be used with
728: parcimony.
729: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
730:
731: Revision 1.83 2003/06/10 13:39:11 lievre
732: *** empty log message ***
733:
734: Revision 1.82 2003/06/05 15:57:20 brouard
735: Add log in imach.c and fullversion number is now printed.
736:
737: */
738: /*
739: Interpolated Markov Chain
740:
741: Short summary of the programme:
742:
1.227 brouard 743: This program computes Healthy Life Expectancies or State-specific
744: (if states aren't health statuses) Expectancies from
745: cross-longitudinal data. Cross-longitudinal data consist in:
746:
747: -1- a first survey ("cross") where individuals from different ages
748: are interviewed on their health status or degree of disability (in
749: the case of a health survey which is our main interest)
750:
751: -2- at least a second wave of interviews ("longitudinal") which
752: measure each change (if any) in individual health status. Health
753: expectancies are computed from the time spent in each health state
754: according to a model. More health states you consider, more time is
755: necessary to reach the Maximum Likelihood of the parameters involved
756: in the model. The simplest model is the multinomial logistic model
757: where pij is the probability to be observed in state j at the second
758: wave conditional to be observed in state i at the first
759: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
760: etc , where 'age' is age and 'sex' is a covariate. If you want to
761: have a more complex model than "constant and age", you should modify
762: the program where the markup *Covariates have to be included here
763: again* invites you to do it. More covariates you add, slower the
1.126 brouard 764: convergence.
765:
766: The advantage of this computer programme, compared to a simple
767: multinomial logistic model, is clear when the delay between waves is not
768: identical for each individual. Also, if a individual missed an
769: intermediate interview, the information is lost, but taken into
770: account using an interpolation or extrapolation.
771:
772: hPijx is the probability to be observed in state i at age x+h
773: conditional to the observed state i at age x. The delay 'h' can be
774: split into an exact number (nh*stepm) of unobserved intermediate
775: states. This elementary transition (by month, quarter,
776: semester or year) is modelled as a multinomial logistic. The hPx
777: matrix is simply the matrix product of nh*stepm elementary matrices
778: and the contribution of each individual to the likelihood is simply
779: hPijx.
780:
781: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 782: of the life expectancies. It also computes the period (stable) prevalence.
783:
784: Back prevalence and projections:
1.227 brouard 785:
786: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
787: double agemaxpar, double ftolpl, int *ncvyearp, double
788: dateprev1,double dateprev2, int firstpass, int lastpass, int
789: mobilavproj)
790:
791: Computes the back prevalence limit for any combination of
792: covariate values k at any age between ageminpar and agemaxpar and
793: returns it in **bprlim. In the loops,
794:
795: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
796: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
797:
798: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 799: Computes for any combination of covariates k and any age between bage and fage
800: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
801: oldm=oldms;savm=savms;
1.227 brouard 802:
803: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 804: Computes the transition matrix starting at age 'age' over
805: 'nhstepm*hstepm*stepm' months (i.e. until
806: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 807: nhstepm*hstepm matrices.
808:
809: Returns p3mat[i][j][h] after calling
810: p3mat[i][j][h]=matprod2(newm,
811: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
812: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
813: oldm);
1.226 brouard 814:
815: Important routines
816:
817: - func (or funcone), computes logit (pij) distinguishing
818: o fixed variables (single or product dummies or quantitative);
819: o varying variables by:
820: (1) wave (single, product dummies, quantitative),
821: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
822: % fixed dummy (treated) or quantitative (not done because time-consuming);
823: % varying dummy (not done) or quantitative (not done);
824: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
825: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
826: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
827: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
828: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 829:
1.226 brouard 830:
831:
1.133 brouard 832: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
833: Institut national d'études démographiques, Paris.
1.126 brouard 834: This software have been partly granted by Euro-REVES, a concerted action
835: from the European Union.
836: It is copyrighted identically to a GNU software product, ie programme and
837: software can be distributed freely for non commercial use. Latest version
838: can be accessed at http://euroreves.ined.fr/imach .
839:
840: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
841: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
842:
843: **********************************************************************/
844: /*
845: main
846: read parameterfile
847: read datafile
848: concatwav
849: freqsummary
850: if (mle >= 1)
851: mlikeli
852: print results files
853: if mle==1
854: computes hessian
855: read end of parameter file: agemin, agemax, bage, fage, estepm
856: begin-prev-date,...
857: open gnuplot file
858: open html file
1.145 brouard 859: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
860: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
861: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
862: freexexit2 possible for memory heap.
863:
864: h Pij x | pij_nom ficrestpij
865: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
866: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
867: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
868:
869: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
870: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
871: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
872: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
873: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
874:
1.126 brouard 875: forecasting if prevfcast==1 prevforecast call prevalence()
876: health expectancies
877: Variance-covariance of DFLE
878: prevalence()
879: movingaverage()
880: varevsij()
881: if popbased==1 varevsij(,popbased)
882: total life expectancies
883: Variance of period (stable) prevalence
884: end
885: */
886:
1.187 brouard 887: /* #define DEBUG */
888: /* #define DEBUGBRENT */
1.203 brouard 889: /* #define DEBUGLINMIN */
890: /* #define DEBUGHESS */
891: #define DEBUGHESSIJ
1.224 brouard 892: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 893: #define POWELL /* Instead of NLOPT */
1.224 brouard 894: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 895: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
896: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 897:
898: #include <math.h>
899: #include <stdio.h>
900: #include <stdlib.h>
901: #include <string.h>
1.226 brouard 902: #include <ctype.h>
1.159 brouard 903:
904: #ifdef _WIN32
905: #include <io.h>
1.172 brouard 906: #include <windows.h>
907: #include <tchar.h>
1.159 brouard 908: #else
1.126 brouard 909: #include <unistd.h>
1.159 brouard 910: #endif
1.126 brouard 911:
912: #include <limits.h>
913: #include <sys/types.h>
1.171 brouard 914:
915: #if defined(__GNUC__)
916: #include <sys/utsname.h> /* Doesn't work on Windows */
917: #endif
918:
1.126 brouard 919: #include <sys/stat.h>
920: #include <errno.h>
1.159 brouard 921: /* extern int errno; */
1.126 brouard 922:
1.157 brouard 923: /* #ifdef LINUX */
924: /* #include <time.h> */
925: /* #include "timeval.h" */
926: /* #else */
927: /* #include <sys/time.h> */
928: /* #endif */
929:
1.126 brouard 930: #include <time.h>
931:
1.136 brouard 932: #ifdef GSL
933: #include <gsl/gsl_errno.h>
934: #include <gsl/gsl_multimin.h>
935: #endif
936:
1.167 brouard 937:
1.162 brouard 938: #ifdef NLOPT
939: #include <nlopt.h>
940: typedef struct {
941: double (* function)(double [] );
942: } myfunc_data ;
943: #endif
944:
1.126 brouard 945: /* #include <libintl.h> */
946: /* #define _(String) gettext (String) */
947:
1.251 brouard 948: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 949:
950: #define GNUPLOTPROGRAM "gnuplot"
951: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
952: #define FILENAMELENGTH 132
953:
954: #define GLOCK_ERROR_NOPATH -1 /* empty path */
955: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
956:
1.144 brouard 957: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
958: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 959:
960: #define NINTERVMAX 8
1.144 brouard 961: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
962: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
963: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 964: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 965: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
966: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 967: #define MAXN 20000
1.144 brouard 968: #define YEARM 12. /**< Number of months per year */
1.218 brouard 969: /* #define AGESUP 130 */
970: #define AGESUP 150
971: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 972: #define AGEBASE 40
1.194 brouard 973: #define AGEOVERFLOW 1.e20
1.164 brouard 974: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 975: #ifdef _WIN32
976: #define DIRSEPARATOR '\\'
977: #define CHARSEPARATOR "\\"
978: #define ODIRSEPARATOR '/'
979: #else
1.126 brouard 980: #define DIRSEPARATOR '/'
981: #define CHARSEPARATOR "/"
982: #define ODIRSEPARATOR '\\'
983: #endif
984:
1.261 ! brouard 985: /* $Id: imach.c,v 1.260 2017/04/04 17:46:59 brouard Exp $ */
1.126 brouard 986: /* $State: Exp $ */
1.196 brouard 987: #include "version.h"
988: char version[]=__IMACH_VERSION__;
1.224 brouard 989: 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";
1.261 ! brouard 990: char fullversion[]="$Revision: 1.260 $ $Date: 2017/04/04 17:46:59 $";
1.126 brouard 991: char strstart[80];
992: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 993: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 994: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 995: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
996: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
997: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 998: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
999: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1000: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1001: int cptcovprodnoage=0; /**< Number of covariate products without age */
1002: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1003: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1004: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1005: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1006: int nsd=0; /**< Total number of single dummy variables (output) */
1007: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1008: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1009: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1010: int ntveff=0; /**< ntveff number of effective time varying variables */
1011: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1012: int cptcov=0; /* Working variable */
1.218 brouard 1013: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1014: int npar=NPARMAX;
1015: int nlstate=2; /* Number of live states */
1016: int ndeath=1; /* Number of dead states */
1.130 brouard 1017: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1018: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1019: int popbased=0;
1020:
1021: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1022: int maxwav=0; /* Maxim number of waves */
1023: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1024: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1025: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1026: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1027: int mle=1, weightopt=0;
1.126 brouard 1028: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1029: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1030: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1031: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1032: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1033: int selected(int kvar); /* Is covariate kvar selected for printing results */
1034:
1.130 brouard 1035: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1036: double **matprod2(); /* test */
1.126 brouard 1037: double **oldm, **newm, **savm; /* Working pointers to matrices */
1038: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1039: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1040:
1.136 brouard 1041: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1042: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1043: FILE *ficlog, *ficrespow;
1.130 brouard 1044: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1045: double fretone; /* Only one call to likelihood */
1.130 brouard 1046: long ipmx=0; /* Number of contributions */
1.126 brouard 1047: double sw; /* Sum of weights */
1048: char filerespow[FILENAMELENGTH];
1049: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1050: FILE *ficresilk;
1051: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1052: FILE *ficresprobmorprev;
1053: FILE *fichtm, *fichtmcov; /* Html File */
1054: FILE *ficreseij;
1055: char filerese[FILENAMELENGTH];
1056: FILE *ficresstdeij;
1057: char fileresstde[FILENAMELENGTH];
1058: FILE *ficrescveij;
1059: char filerescve[FILENAMELENGTH];
1060: FILE *ficresvij;
1061: char fileresv[FILENAMELENGTH];
1062: FILE *ficresvpl;
1063: char fileresvpl[FILENAMELENGTH];
1064: char title[MAXLINE];
1.234 brouard 1065: char model[MAXLINE]; /**< The model line */
1.217 brouard 1066: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1067: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1068: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1069: char command[FILENAMELENGTH];
1070: int outcmd=0;
1071:
1.217 brouard 1072: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1073: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1074: char filelog[FILENAMELENGTH]; /* Log file */
1075: char filerest[FILENAMELENGTH];
1076: char fileregp[FILENAMELENGTH];
1077: char popfile[FILENAMELENGTH];
1078:
1079: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1080:
1.157 brouard 1081: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1082: /* struct timezone tzp; */
1083: /* extern int gettimeofday(); */
1084: struct tm tml, *gmtime(), *localtime();
1085:
1086: extern time_t time();
1087:
1088: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1089: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1090: struct tm tm;
1091:
1.126 brouard 1092: char strcurr[80], strfor[80];
1093:
1094: char *endptr;
1095: long lval;
1096: double dval;
1097:
1098: #define NR_END 1
1099: #define FREE_ARG char*
1100: #define FTOL 1.0e-10
1101:
1102: #define NRANSI
1.240 brouard 1103: #define ITMAX 200
1104: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1105:
1106: #define TOL 2.0e-4
1107:
1108: #define CGOLD 0.3819660
1109: #define ZEPS 1.0e-10
1110: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1111:
1112: #define GOLD 1.618034
1113: #define GLIMIT 100.0
1114: #define TINY 1.0e-20
1115:
1116: static double maxarg1,maxarg2;
1117: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1118: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1119:
1120: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1121: #define rint(a) floor(a+0.5)
1.166 brouard 1122: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1123: #define mytinydouble 1.0e-16
1.166 brouard 1124: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1125: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1126: /* static double dsqrarg; */
1127: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1128: static double sqrarg;
1129: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1130: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1131: int agegomp= AGEGOMP;
1132:
1133: int imx;
1134: int stepm=1;
1135: /* Stepm, step in month: minimum step interpolation*/
1136:
1137: int estepm;
1138: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1139:
1140: int m,nb;
1141: long *num;
1.197 brouard 1142: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1143: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1144: covariate for which somebody answered excluding
1145: undefined. Usually 2: 0 and 1. */
1146: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1147: covariate for which somebody answered including
1148: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1149: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1150: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1151: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1152: double *ageexmed,*agecens;
1153: double dateintmean=0;
1154:
1155: double *weight;
1156: int **s; /* Status */
1.141 brouard 1157: double *agedc;
1.145 brouard 1158: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1159: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1160: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1161: double **coqvar; /* Fixed quantitative covariate iqv */
1162: double ***cotvar; /* Time varying covariate itv */
1163: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1164: double idx;
1165: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1166: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1167: /*k 1 2 3 4 5 6 7 8 9 */
1168: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1169: /* Tndvar[k] 1 2 3 4 5 */
1170: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1171: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1172: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1173: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1174: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1175: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1176: /* Tprod[i]=k 4 7 */
1177: /* Tage[i]=k 5 8 */
1178: /* */
1179: /* Type */
1180: /* V 1 2 3 4 5 */
1181: /* F F V V V */
1182: /* D Q D D Q */
1183: /* */
1184: int *TvarsD;
1185: int *TvarsDind;
1186: int *TvarsQ;
1187: int *TvarsQind;
1188:
1.235 brouard 1189: #define MAXRESULTLINES 10
1190: int nresult=0;
1.258 brouard 1191: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1192: int TKresult[MAXRESULTLINES];
1.237 brouard 1193: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1194: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1195: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1196: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1197: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1198: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1199:
1.234 brouard 1200: /* 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 1201: 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 */
1202: 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 */
1203: 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 */
1204: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1205: 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 */
1206: 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 1207: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1208: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1209: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1210: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1211: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1212: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1213: 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 */
1214: 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 */
1215:
1.230 brouard 1216: int *Tvarsel; /**< Selected covariates for output */
1217: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1218: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1219: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1220: 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 1221: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1222: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1223: int *Tage;
1.227 brouard 1224: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1225: 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 1226: 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*/
1227: 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 1228: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1229: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1230: int **Tvard;
1231: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1232: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1233: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1234: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1235: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1236: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1237: double *lsurv, *lpop, *tpop;
1238:
1.231 brouard 1239: #define FD 1; /* Fixed dummy covariate */
1240: #define FQ 2; /* Fixed quantitative covariate */
1241: #define FP 3; /* Fixed product covariate */
1242: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1243: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1244: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1245: #define VD 10; /* Varying dummy covariate */
1246: #define VQ 11; /* Varying quantitative covariate */
1247: #define VP 12; /* Varying product covariate */
1248: #define VPDD 13; /* Varying product dummy*dummy covariate */
1249: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1250: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1251: #define APFD 16; /* Age product * fixed dummy covariate */
1252: #define APFQ 17; /* Age product * fixed quantitative covariate */
1253: #define APVD 18; /* Age product * varying dummy covariate */
1254: #define APVQ 19; /* Age product * varying quantitative covariate */
1255:
1256: #define FTYPE 1; /* Fixed covariate */
1257: #define VTYPE 2; /* Varying covariate (loop in wave) */
1258: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1259:
1260: struct kmodel{
1261: int maintype; /* main type */
1262: int subtype; /* subtype */
1263: };
1264: struct kmodel modell[NCOVMAX];
1265:
1.143 brouard 1266: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1267: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1268:
1269: /**************** split *************************/
1270: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1271: {
1272: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1273: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1274: */
1275: char *ss; /* pointer */
1.186 brouard 1276: int l1=0, l2=0; /* length counters */
1.126 brouard 1277:
1278: l1 = strlen(path ); /* length of path */
1279: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1280: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1281: if ( ss == NULL ) { /* no directory, so determine current directory */
1282: strcpy( name, path ); /* we got the fullname name because no directory */
1283: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1284: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1285: /* get current working directory */
1286: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1287: #ifdef WIN32
1288: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1289: #else
1290: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1291: #endif
1.126 brouard 1292: return( GLOCK_ERROR_GETCWD );
1293: }
1294: /* got dirc from getcwd*/
1295: printf(" DIRC = %s \n",dirc);
1.205 brouard 1296: } else { /* strip directory from path */
1.126 brouard 1297: ss++; /* after this, the filename */
1298: l2 = strlen( ss ); /* length of filename */
1299: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1300: strcpy( name, ss ); /* save file name */
1301: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1302: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1303: printf(" DIRC2 = %s \n",dirc);
1304: }
1305: /* We add a separator at the end of dirc if not exists */
1306: l1 = strlen( dirc ); /* length of directory */
1307: if( dirc[l1-1] != DIRSEPARATOR ){
1308: dirc[l1] = DIRSEPARATOR;
1309: dirc[l1+1] = 0;
1310: printf(" DIRC3 = %s \n",dirc);
1311: }
1312: ss = strrchr( name, '.' ); /* find last / */
1313: if (ss >0){
1314: ss++;
1315: strcpy(ext,ss); /* save extension */
1316: l1= strlen( name);
1317: l2= strlen(ss)+1;
1318: strncpy( finame, name, l1-l2);
1319: finame[l1-l2]= 0;
1320: }
1321:
1322: return( 0 ); /* we're done */
1323: }
1324:
1325:
1326: /******************************************/
1327:
1328: void replace_back_to_slash(char *s, char*t)
1329: {
1330: int i;
1331: int lg=0;
1332: i=0;
1333: lg=strlen(t);
1334: for(i=0; i<= lg; i++) {
1335: (s[i] = t[i]);
1336: if (t[i]== '\\') s[i]='/';
1337: }
1338: }
1339:
1.132 brouard 1340: char *trimbb(char *out, char *in)
1.137 brouard 1341: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1342: char *s;
1343: s=out;
1344: while (*in != '\0'){
1.137 brouard 1345: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1346: in++;
1347: }
1348: *out++ = *in++;
1349: }
1350: *out='\0';
1351: return s;
1352: }
1353:
1.187 brouard 1354: /* char *substrchaine(char *out, char *in, char *chain) */
1355: /* { */
1356: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1357: /* char *s, *t; */
1358: /* t=in;s=out; */
1359: /* while ((*in != *chain) && (*in != '\0')){ */
1360: /* *out++ = *in++; */
1361: /* } */
1362:
1363: /* /\* *in matches *chain *\/ */
1364: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1365: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1366: /* } */
1367: /* in--; chain--; */
1368: /* while ( (*in != '\0')){ */
1369: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1370: /* *out++ = *in++; */
1371: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1372: /* } */
1373: /* *out='\0'; */
1374: /* out=s; */
1375: /* return out; */
1376: /* } */
1377: char *substrchaine(char *out, char *in, char *chain)
1378: {
1379: /* Substract chain 'chain' from 'in', return and output 'out' */
1380: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1381:
1382: char *strloc;
1383:
1384: strcpy (out, in);
1385: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1386: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1387: if(strloc != NULL){
1388: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1389: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1390: /* strcpy (strloc, strloc +strlen(chain));*/
1391: }
1392: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1393: return out;
1394: }
1395:
1396:
1.145 brouard 1397: char *cutl(char *blocc, char *alocc, char *in, char occ)
1398: {
1.187 brouard 1399: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1400: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1401: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1402: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1403: */
1.160 brouard 1404: char *s, *t;
1.145 brouard 1405: t=in;s=in;
1406: while ((*in != occ) && (*in != '\0')){
1407: *alocc++ = *in++;
1408: }
1409: if( *in == occ){
1410: *(alocc)='\0';
1411: s=++in;
1412: }
1413:
1414: if (s == t) {/* occ not found */
1415: *(alocc-(in-s))='\0';
1416: in=s;
1417: }
1418: while ( *in != '\0'){
1419: *blocc++ = *in++;
1420: }
1421:
1422: *blocc='\0';
1423: return t;
1424: }
1.137 brouard 1425: char *cutv(char *blocc, char *alocc, char *in, char occ)
1426: {
1.187 brouard 1427: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1428: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1429: gives blocc="abcdef2ghi" and alocc="j".
1430: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1431: */
1432: char *s, *t;
1433: t=in;s=in;
1434: while (*in != '\0'){
1435: while( *in == occ){
1436: *blocc++ = *in++;
1437: s=in;
1438: }
1439: *blocc++ = *in++;
1440: }
1441: if (s == t) /* occ not found */
1442: *(blocc-(in-s))='\0';
1443: else
1444: *(blocc-(in-s)-1)='\0';
1445: in=s;
1446: while ( *in != '\0'){
1447: *alocc++ = *in++;
1448: }
1449:
1450: *alocc='\0';
1451: return s;
1452: }
1453:
1.126 brouard 1454: int nbocc(char *s, char occ)
1455: {
1456: int i,j=0;
1457: int lg=20;
1458: i=0;
1459: lg=strlen(s);
1460: for(i=0; i<= lg; i++) {
1.234 brouard 1461: if (s[i] == occ ) j++;
1.126 brouard 1462: }
1463: return j;
1464: }
1465:
1.137 brouard 1466: /* void cutv(char *u,char *v, char*t, char occ) */
1467: /* { */
1468: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1469: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1470: /* gives u="abcdef2ghi" and v="j" *\/ */
1471: /* int i,lg,j,p=0; */
1472: /* i=0; */
1473: /* lg=strlen(t); */
1474: /* for(j=0; j<=lg-1; j++) { */
1475: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1476: /* } */
1.126 brouard 1477:
1.137 brouard 1478: /* for(j=0; j<p; j++) { */
1479: /* (u[j] = t[j]); */
1480: /* } */
1481: /* u[p]='\0'; */
1.126 brouard 1482:
1.137 brouard 1483: /* for(j=0; j<= lg; j++) { */
1484: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1485: /* } */
1486: /* } */
1.126 brouard 1487:
1.160 brouard 1488: #ifdef _WIN32
1489: char * strsep(char **pp, const char *delim)
1490: {
1491: char *p, *q;
1492:
1493: if ((p = *pp) == NULL)
1494: return 0;
1495: if ((q = strpbrk (p, delim)) != NULL)
1496: {
1497: *pp = q + 1;
1498: *q = '\0';
1499: }
1500: else
1501: *pp = 0;
1502: return p;
1503: }
1504: #endif
1505:
1.126 brouard 1506: /********************** nrerror ********************/
1507:
1508: void nrerror(char error_text[])
1509: {
1510: fprintf(stderr,"ERREUR ...\n");
1511: fprintf(stderr,"%s\n",error_text);
1512: exit(EXIT_FAILURE);
1513: }
1514: /*********************** vector *******************/
1515: double *vector(int nl, int nh)
1516: {
1517: double *v;
1518: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1519: if (!v) nrerror("allocation failure in vector");
1520: return v-nl+NR_END;
1521: }
1522:
1523: /************************ free vector ******************/
1524: void free_vector(double*v, int nl, int nh)
1525: {
1526: free((FREE_ARG)(v+nl-NR_END));
1527: }
1528:
1529: /************************ivector *******************************/
1530: int *ivector(long nl,long nh)
1531: {
1532: int *v;
1533: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1534: if (!v) nrerror("allocation failure in ivector");
1535: return v-nl+NR_END;
1536: }
1537:
1538: /******************free ivector **************************/
1539: void free_ivector(int *v, long nl, long nh)
1540: {
1541: free((FREE_ARG)(v+nl-NR_END));
1542: }
1543:
1544: /************************lvector *******************************/
1545: long *lvector(long nl,long nh)
1546: {
1547: long *v;
1548: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1549: if (!v) nrerror("allocation failure in ivector");
1550: return v-nl+NR_END;
1551: }
1552:
1553: /******************free lvector **************************/
1554: void free_lvector(long *v, long nl, long nh)
1555: {
1556: free((FREE_ARG)(v+nl-NR_END));
1557: }
1558:
1559: /******************* imatrix *******************************/
1560: int **imatrix(long nrl, long nrh, long ncl, long nch)
1561: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1562: {
1563: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1564: int **m;
1565:
1566: /* allocate pointers to rows */
1567: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1568: if (!m) nrerror("allocation failure 1 in matrix()");
1569: m += NR_END;
1570: m -= nrl;
1571:
1572:
1573: /* allocate rows and set pointers to them */
1574: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1575: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1576: m[nrl] += NR_END;
1577: m[nrl] -= ncl;
1578:
1579: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1580:
1581: /* return pointer to array of pointers to rows */
1582: return m;
1583: }
1584:
1585: /****************** free_imatrix *************************/
1586: void free_imatrix(m,nrl,nrh,ncl,nch)
1587: int **m;
1588: long nch,ncl,nrh,nrl;
1589: /* free an int matrix allocated by imatrix() */
1590: {
1591: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1592: free((FREE_ARG) (m+nrl-NR_END));
1593: }
1594:
1595: /******************* matrix *******************************/
1596: double **matrix(long nrl, long nrh, long ncl, long nch)
1597: {
1598: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1599: double **m;
1600:
1601: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1602: if (!m) nrerror("allocation failure 1 in matrix()");
1603: m += NR_END;
1604: m -= nrl;
1605:
1606: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1607: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1608: m[nrl] += NR_END;
1609: m[nrl] -= ncl;
1610:
1611: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1612: return m;
1.145 brouard 1613: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1614: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1615: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1616: */
1617: }
1618:
1619: /*************************free matrix ************************/
1620: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1621: {
1622: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1623: free((FREE_ARG)(m+nrl-NR_END));
1624: }
1625:
1626: /******************* ma3x *******************************/
1627: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1628: {
1629: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1630: double ***m;
1631:
1632: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1633: if (!m) nrerror("allocation failure 1 in matrix()");
1634: m += NR_END;
1635: m -= nrl;
1636:
1637: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1638: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1639: m[nrl] += NR_END;
1640: m[nrl] -= ncl;
1641:
1642: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1643:
1644: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1645: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1646: m[nrl][ncl] += NR_END;
1647: m[nrl][ncl] -= nll;
1648: for (j=ncl+1; j<=nch; j++)
1649: m[nrl][j]=m[nrl][j-1]+nlay;
1650:
1651: for (i=nrl+1; i<=nrh; i++) {
1652: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1653: for (j=ncl+1; j<=nch; j++)
1654: m[i][j]=m[i][j-1]+nlay;
1655: }
1656: return m;
1657: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1658: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1659: */
1660: }
1661:
1662: /*************************free ma3x ************************/
1663: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1664: {
1665: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1666: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1667: free((FREE_ARG)(m+nrl-NR_END));
1668: }
1669:
1670: /*************** function subdirf ***********/
1671: char *subdirf(char fileres[])
1672: {
1673: /* Caution optionfilefiname is hidden */
1674: strcpy(tmpout,optionfilefiname);
1675: strcat(tmpout,"/"); /* Add to the right */
1676: strcat(tmpout,fileres);
1677: return tmpout;
1678: }
1679:
1680: /*************** function subdirf2 ***********/
1681: char *subdirf2(char fileres[], char *preop)
1682: {
1683:
1684: /* Caution optionfilefiname is hidden */
1685: strcpy(tmpout,optionfilefiname);
1686: strcat(tmpout,"/");
1687: strcat(tmpout,preop);
1688: strcat(tmpout,fileres);
1689: return tmpout;
1690: }
1691:
1692: /*************** function subdirf3 ***********/
1693: char *subdirf3(char fileres[], char *preop, char *preop2)
1694: {
1695:
1696: /* Caution optionfilefiname is hidden */
1697: strcpy(tmpout,optionfilefiname);
1698: strcat(tmpout,"/");
1699: strcat(tmpout,preop);
1700: strcat(tmpout,preop2);
1701: strcat(tmpout,fileres);
1702: return tmpout;
1703: }
1.213 brouard 1704:
1705: /*************** function subdirfext ***********/
1706: char *subdirfext(char fileres[], char *preop, char *postop)
1707: {
1708:
1709: strcpy(tmpout,preop);
1710: strcat(tmpout,fileres);
1711: strcat(tmpout,postop);
1712: return tmpout;
1713: }
1.126 brouard 1714:
1.213 brouard 1715: /*************** function subdirfext3 ***********/
1716: char *subdirfext3(char fileres[], char *preop, char *postop)
1717: {
1718:
1719: /* Caution optionfilefiname is hidden */
1720: strcpy(tmpout,optionfilefiname);
1721: strcat(tmpout,"/");
1722: strcat(tmpout,preop);
1723: strcat(tmpout,fileres);
1724: strcat(tmpout,postop);
1725: return tmpout;
1726: }
1727:
1.162 brouard 1728: char *asc_diff_time(long time_sec, char ascdiff[])
1729: {
1730: long sec_left, days, hours, minutes;
1731: days = (time_sec) / (60*60*24);
1732: sec_left = (time_sec) % (60*60*24);
1733: hours = (sec_left) / (60*60) ;
1734: sec_left = (sec_left) %(60*60);
1735: minutes = (sec_left) /60;
1736: sec_left = (sec_left) % (60);
1737: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1738: return ascdiff;
1739: }
1740:
1.126 brouard 1741: /***************** f1dim *************************/
1742: extern int ncom;
1743: extern double *pcom,*xicom;
1744: extern double (*nrfunc)(double []);
1745:
1746: double f1dim(double x)
1747: {
1748: int j;
1749: double f;
1750: double *xt;
1751:
1752: xt=vector(1,ncom);
1753: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1754: f=(*nrfunc)(xt);
1755: free_vector(xt,1,ncom);
1756: return f;
1757: }
1758:
1759: /*****************brent *************************/
1760: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1761: {
1762: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1763: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1764: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1765: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1766: * returned function value.
1767: */
1.126 brouard 1768: int iter;
1769: double a,b,d,etemp;
1.159 brouard 1770: double fu=0,fv,fw,fx;
1.164 brouard 1771: double ftemp=0.;
1.126 brouard 1772: double p,q,r,tol1,tol2,u,v,w,x,xm;
1773: double e=0.0;
1774:
1775: a=(ax < cx ? ax : cx);
1776: b=(ax > cx ? ax : cx);
1777: x=w=v=bx;
1778: fw=fv=fx=(*f)(x);
1779: for (iter=1;iter<=ITMAX;iter++) {
1780: xm=0.5*(a+b);
1781: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1782: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1783: printf(".");fflush(stdout);
1784: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1785: #ifdef DEBUGBRENT
1.126 brouard 1786: 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);
1787: 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);
1788: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1789: #endif
1790: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1791: *xmin=x;
1792: return fx;
1793: }
1794: ftemp=fu;
1795: if (fabs(e) > tol1) {
1796: r=(x-w)*(fx-fv);
1797: q=(x-v)*(fx-fw);
1798: p=(x-v)*q-(x-w)*r;
1799: q=2.0*(q-r);
1800: if (q > 0.0) p = -p;
1801: q=fabs(q);
1802: etemp=e;
1803: e=d;
1804: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1805: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1806: else {
1.224 brouard 1807: d=p/q;
1808: u=x+d;
1809: if (u-a < tol2 || b-u < tol2)
1810: d=SIGN(tol1,xm-x);
1.126 brouard 1811: }
1812: } else {
1813: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1814: }
1815: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1816: fu=(*f)(u);
1817: if (fu <= fx) {
1818: if (u >= x) a=x; else b=x;
1819: SHFT(v,w,x,u)
1.183 brouard 1820: SHFT(fv,fw,fx,fu)
1821: } else {
1822: if (u < x) a=u; else b=u;
1823: if (fu <= fw || w == x) {
1.224 brouard 1824: v=w;
1825: w=u;
1826: fv=fw;
1827: fw=fu;
1.183 brouard 1828: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1829: v=u;
1830: fv=fu;
1.183 brouard 1831: }
1832: }
1.126 brouard 1833: }
1834: nrerror("Too many iterations in brent");
1835: *xmin=x;
1836: return fx;
1837: }
1838:
1839: /****************** mnbrak ***********************/
1840:
1841: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1842: double (*func)(double))
1.183 brouard 1843: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1844: the downhill direction (defined by the function as evaluated at the initial points) and returns
1845: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1846: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1847: */
1.126 brouard 1848: double ulim,u,r,q, dum;
1849: double fu;
1.187 brouard 1850:
1851: double scale=10.;
1852: int iterscale=0;
1853:
1854: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1855: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1856:
1857:
1858: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1859: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1860: /* *bx = *ax - (*ax - *bx)/scale; */
1861: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1862: /* } */
1863:
1.126 brouard 1864: if (*fb > *fa) {
1865: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1866: SHFT(dum,*fb,*fa,dum)
1867: }
1.126 brouard 1868: *cx=(*bx)+GOLD*(*bx-*ax);
1869: *fc=(*func)(*cx);
1.183 brouard 1870: #ifdef DEBUG
1.224 brouard 1871: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1872: 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 1873: #endif
1.224 brouard 1874: 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 1875: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1876: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1877: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1878: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1879: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1880: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1881: fu=(*func)(u);
1.163 brouard 1882: #ifdef DEBUG
1883: /* f(x)=A(x-u)**2+f(u) */
1884: double A, fparabu;
1885: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1886: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1887: 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);
1888: 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 1889: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1890: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1891: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1892: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1893: #endif
1.184 brouard 1894: #ifdef MNBRAKORIGINAL
1.183 brouard 1895: #else
1.191 brouard 1896: /* if (fu > *fc) { */
1897: /* #ifdef DEBUG */
1898: /* printf("mnbrak4 fu > fc \n"); */
1899: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1900: /* #endif */
1901: /* /\* 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 *\\/ *\/ */
1902: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1903: /* dum=u; /\* Shifting c and u *\/ */
1904: /* u = *cx; */
1905: /* *cx = dum; */
1906: /* dum = fu; */
1907: /* fu = *fc; */
1908: /* *fc =dum; */
1909: /* } else { /\* end *\/ */
1910: /* #ifdef DEBUG */
1911: /* printf("mnbrak3 fu < fc \n"); */
1912: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1913: /* #endif */
1914: /* dum=u; /\* Shifting c and u *\/ */
1915: /* u = *cx; */
1916: /* *cx = dum; */
1917: /* dum = fu; */
1918: /* fu = *fc; */
1919: /* *fc =dum; */
1920: /* } */
1.224 brouard 1921: #ifdef DEBUGMNBRAK
1922: double A, fparabu;
1923: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1924: fparabu= *fa - A*(*ax-u)*(*ax-u);
1925: 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);
1926: 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 1927: #endif
1.191 brouard 1928: dum=u; /* Shifting c and u */
1929: u = *cx;
1930: *cx = dum;
1931: dum = fu;
1932: fu = *fc;
1933: *fc =dum;
1.183 brouard 1934: #endif
1.162 brouard 1935: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1936: #ifdef DEBUG
1.224 brouard 1937: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1938: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1939: #endif
1.126 brouard 1940: fu=(*func)(u);
1941: if (fu < *fc) {
1.183 brouard 1942: #ifdef DEBUG
1.224 brouard 1943: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1944: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1945: #endif
1946: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1947: SHFT(*fb,*fc,fu,(*func)(u))
1948: #ifdef DEBUG
1949: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1950: #endif
1951: }
1.162 brouard 1952: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1953: #ifdef DEBUG
1.224 brouard 1954: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1955: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1956: #endif
1.126 brouard 1957: u=ulim;
1958: fu=(*func)(u);
1.183 brouard 1959: } else { /* u could be left to b (if r > q parabola has a maximum) */
1960: #ifdef DEBUG
1.224 brouard 1961: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1962: 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 1963: #endif
1.126 brouard 1964: u=(*cx)+GOLD*(*cx-*bx);
1965: fu=(*func)(u);
1.224 brouard 1966: #ifdef DEBUG
1967: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1968: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1969: #endif
1.183 brouard 1970: } /* end tests */
1.126 brouard 1971: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1972: SHFT(*fa,*fb,*fc,fu)
1973: #ifdef DEBUG
1.224 brouard 1974: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1975: 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 1976: #endif
1977: } /* 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 1978: }
1979:
1980: /*************** linmin ************************/
1.162 brouard 1981: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1982: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1983: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1984: the value of func at the returned location p . This is actually all accomplished by calling the
1985: routines mnbrak and brent .*/
1.126 brouard 1986: int ncom;
1987: double *pcom,*xicom;
1988: double (*nrfunc)(double []);
1989:
1.224 brouard 1990: #ifdef LINMINORIGINAL
1.126 brouard 1991: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1992: #else
1993: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1994: #endif
1.126 brouard 1995: {
1996: double brent(double ax, double bx, double cx,
1997: double (*f)(double), double tol, double *xmin);
1998: double f1dim(double x);
1999: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2000: double *fc, double (*func)(double));
2001: int j;
2002: double xx,xmin,bx,ax;
2003: double fx,fb,fa;
1.187 brouard 2004:
1.203 brouard 2005: #ifdef LINMINORIGINAL
2006: #else
2007: double scale=10., axs, xxs; /* Scale added for infinity */
2008: #endif
2009:
1.126 brouard 2010: ncom=n;
2011: pcom=vector(1,n);
2012: xicom=vector(1,n);
2013: nrfunc=func;
2014: for (j=1;j<=n;j++) {
2015: pcom[j]=p[j];
1.202 brouard 2016: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2017: }
1.187 brouard 2018:
1.203 brouard 2019: #ifdef LINMINORIGINAL
2020: xx=1.;
2021: #else
2022: axs=0.0;
2023: xxs=1.;
2024: do{
2025: xx= xxs;
2026: #endif
1.187 brouard 2027: ax=0.;
2028: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2029: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2030: /* 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)) */
2031: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2032: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2033: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2034: /* 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 2035: #ifdef LINMINORIGINAL
2036: #else
2037: if (fx != fx){
1.224 brouard 2038: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2039: printf("|");
2040: fprintf(ficlog,"|");
1.203 brouard 2041: #ifdef DEBUGLINMIN
1.224 brouard 2042: 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 2043: #endif
2044: }
1.224 brouard 2045: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2046: #endif
2047:
1.191 brouard 2048: #ifdef DEBUGLINMIN
2049: 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 2050: 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 2051: #endif
1.224 brouard 2052: #ifdef LINMINORIGINAL
2053: #else
2054: if(fb == fx){ /* Flat function in the direction */
2055: xmin=xx;
2056: *flat=1;
2057: }else{
2058: *flat=0;
2059: #endif
2060: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2061: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2062: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2063: /* fmin = f(p[j] + xmin * xi[j]) */
2064: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2065: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2066: #ifdef DEBUG
1.224 brouard 2067: 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);
2068: 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);
2069: #endif
2070: #ifdef LINMINORIGINAL
2071: #else
2072: }
1.126 brouard 2073: #endif
1.191 brouard 2074: #ifdef DEBUGLINMIN
2075: printf("linmin end ");
1.202 brouard 2076: fprintf(ficlog,"linmin end ");
1.191 brouard 2077: #endif
1.126 brouard 2078: for (j=1;j<=n;j++) {
1.203 brouard 2079: #ifdef LINMINORIGINAL
2080: xi[j] *= xmin;
2081: #else
2082: #ifdef DEBUGLINMIN
2083: if(xxs <1.0)
2084: printf(" before xi[%d]=%12.8f", j,xi[j]);
2085: #endif
2086: 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) */
2087: #ifdef DEBUGLINMIN
2088: if(xxs <1.0)
2089: 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 );
2090: #endif
2091: #endif
1.187 brouard 2092: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2093: }
1.191 brouard 2094: #ifdef DEBUGLINMIN
1.203 brouard 2095: printf("\n");
1.191 brouard 2096: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2097: 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 2098: for (j=1;j<=n;j++) {
1.202 brouard 2099: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2100: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2101: if(j % ncovmodel == 0){
1.191 brouard 2102: printf("\n");
1.202 brouard 2103: fprintf(ficlog,"\n");
2104: }
1.191 brouard 2105: }
1.203 brouard 2106: #else
1.191 brouard 2107: #endif
1.126 brouard 2108: free_vector(xicom,1,n);
2109: free_vector(pcom,1,n);
2110: }
2111:
2112:
2113: /*************** powell ************************/
1.162 brouard 2114: /*
2115: Minimization of a function func of n variables. Input consists of an initial starting point
2116: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2117: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2118: such that failure to decrease by more than this amount on one iteration signals doneness. On
2119: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2120: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2121: */
1.224 brouard 2122: #ifdef LINMINORIGINAL
2123: #else
2124: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2125: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2126: #endif
1.126 brouard 2127: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2128: double (*func)(double []))
2129: {
1.224 brouard 2130: #ifdef LINMINORIGINAL
2131: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2132: double (*func)(double []));
1.224 brouard 2133: #else
1.241 brouard 2134: void linmin(double p[], double xi[], int n, double *fret,
2135: double (*func)(double []),int *flat);
1.224 brouard 2136: #endif
1.239 brouard 2137: int i,ibig,j,jk,k;
1.126 brouard 2138: double del,t,*pt,*ptt,*xit;
1.181 brouard 2139: double directest;
1.126 brouard 2140: double fp,fptt;
2141: double *xits;
2142: int niterf, itmp;
1.224 brouard 2143: #ifdef LINMINORIGINAL
2144: #else
2145:
2146: flatdir=ivector(1,n);
2147: for (j=1;j<=n;j++) flatdir[j]=0;
2148: #endif
1.126 brouard 2149:
2150: pt=vector(1,n);
2151: ptt=vector(1,n);
2152: xit=vector(1,n);
2153: xits=vector(1,n);
2154: *fret=(*func)(p);
2155: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2156: rcurr_time = time(NULL);
1.126 brouard 2157: for (*iter=1;;++(*iter)) {
1.187 brouard 2158: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2159: ibig=0;
2160: del=0.0;
1.157 brouard 2161: rlast_time=rcurr_time;
2162: /* (void) gettimeofday(&curr_time,&tzp); */
2163: rcurr_time = time(NULL);
2164: curr_time = *localtime(&rcurr_time);
2165: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2166: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2167: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2168: for (i=1;i<=n;i++) {
1.126 brouard 2169: fprintf(ficrespow," %.12lf", p[i]);
2170: }
1.239 brouard 2171: fprintf(ficrespow,"\n");fflush(ficrespow);
2172: printf("\n#model= 1 + age ");
2173: fprintf(ficlog,"\n#model= 1 + age ");
2174: if(nagesqr==1){
1.241 brouard 2175: printf(" + age*age ");
2176: fprintf(ficlog," + age*age ");
1.239 brouard 2177: }
2178: for(j=1;j <=ncovmodel-2;j++){
2179: if(Typevar[j]==0) {
2180: printf(" + V%d ",Tvar[j]);
2181: fprintf(ficlog," + V%d ",Tvar[j]);
2182: }else if(Typevar[j]==1) {
2183: printf(" + V%d*age ",Tvar[j]);
2184: fprintf(ficlog," + V%d*age ",Tvar[j]);
2185: }else if(Typevar[j]==2) {
2186: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2187: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2188: }
2189: }
1.126 brouard 2190: printf("\n");
1.239 brouard 2191: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2192: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2193: fprintf(ficlog,"\n");
1.239 brouard 2194: for(i=1,jk=1; i <=nlstate; i++){
2195: for(k=1; k <=(nlstate+ndeath); k++){
2196: if (k != i) {
2197: printf("%d%d ",i,k);
2198: fprintf(ficlog,"%d%d ",i,k);
2199: for(j=1; j <=ncovmodel; j++){
2200: printf("%12.7f ",p[jk]);
2201: fprintf(ficlog,"%12.7f ",p[jk]);
2202: jk++;
2203: }
2204: printf("\n");
2205: fprintf(ficlog,"\n");
2206: }
2207: }
2208: }
1.241 brouard 2209: if(*iter <=3 && *iter >1){
1.157 brouard 2210: tml = *localtime(&rcurr_time);
2211: strcpy(strcurr,asctime(&tml));
2212: rforecast_time=rcurr_time;
1.126 brouard 2213: itmp = strlen(strcurr);
2214: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2215: strcurr[itmp-1]='\0';
1.162 brouard 2216: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2217: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2218: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2219: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2220: forecast_time = *localtime(&rforecast_time);
2221: strcpy(strfor,asctime(&forecast_time));
2222: itmp = strlen(strfor);
2223: if(strfor[itmp-1]=='\n')
2224: strfor[itmp-1]='\0';
2225: 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);
2226: 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 2227: }
2228: }
1.187 brouard 2229: for (i=1;i<=n;i++) { /* For each direction i */
2230: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2231: fptt=(*fret);
2232: #ifdef DEBUG
1.203 brouard 2233: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2234: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2235: #endif
1.203 brouard 2236: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2237: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2238: #ifdef LINMINORIGINAL
1.188 brouard 2239: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2240: #else
2241: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2242: flatdir[i]=flat; /* Function is vanishing in that direction i */
2243: #endif
2244: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2245: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2246: /* because that direction will be replaced unless the gain del is small */
2247: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2248: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2249: /* with the new direction. */
2250: del=fabs(fptt-(*fret));
2251: ibig=i;
1.126 brouard 2252: }
2253: #ifdef DEBUG
2254: printf("%d %.12e",i,(*fret));
2255: fprintf(ficlog,"%d %.12e",i,(*fret));
2256: for (j=1;j<=n;j++) {
1.224 brouard 2257: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2258: printf(" x(%d)=%.12e",j,xit[j]);
2259: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2260: }
2261: for(j=1;j<=n;j++) {
1.225 brouard 2262: printf(" p(%d)=%.12e",j,p[j]);
2263: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2264: }
2265: printf("\n");
2266: fprintf(ficlog,"\n");
2267: #endif
1.187 brouard 2268: } /* end loop on each direction i */
2269: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2270: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2271: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2272: for(j=1;j<=n;j++) {
1.225 brouard 2273: if(flatdir[j] >0){
2274: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2275: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2276: }
2277: /* printf("\n"); */
2278: /* fprintf(ficlog,"\n"); */
2279: }
1.243 brouard 2280: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2281: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2282: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2283: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2284: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2285: /* decreased of more than 3.84 */
2286: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2287: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2288: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2289:
1.188 brouard 2290: /* Starting the program with initial values given by a former maximization will simply change */
2291: /* the scales of the directions and the directions, because the are reset to canonical directions */
2292: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2293: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2294: #ifdef DEBUG
2295: int k[2],l;
2296: k[0]=1;
2297: k[1]=-1;
2298: printf("Max: %.12e",(*func)(p));
2299: fprintf(ficlog,"Max: %.12e",(*func)(p));
2300: for (j=1;j<=n;j++) {
2301: printf(" %.12e",p[j]);
2302: fprintf(ficlog," %.12e",p[j]);
2303: }
2304: printf("\n");
2305: fprintf(ficlog,"\n");
2306: for(l=0;l<=1;l++) {
2307: for (j=1;j<=n;j++) {
2308: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2309: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2310: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2311: }
2312: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2313: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2314: }
2315: #endif
2316:
1.224 brouard 2317: #ifdef LINMINORIGINAL
2318: #else
2319: free_ivector(flatdir,1,n);
2320: #endif
1.126 brouard 2321: free_vector(xit,1,n);
2322: free_vector(xits,1,n);
2323: free_vector(ptt,1,n);
2324: free_vector(pt,1,n);
2325: return;
1.192 brouard 2326: } /* enough precision */
1.240 brouard 2327: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2328: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2329: ptt[j]=2.0*p[j]-pt[j];
2330: xit[j]=p[j]-pt[j];
2331: pt[j]=p[j];
2332: }
1.181 brouard 2333: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2334: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2335: if (*iter <=4) {
1.225 brouard 2336: #else
2337: #endif
1.224 brouard 2338: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2339: #else
1.161 brouard 2340: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2341: #endif
1.162 brouard 2342: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2343: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2344: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2345: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2346: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2347: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2348: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2349: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2350: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2351: /* Even if f3 <f1, directest can be negative and t >0 */
2352: /* mu² and del² are equal when f3=f1 */
2353: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2354: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2355: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2356: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2357: #ifdef NRCORIGINAL
2358: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2359: #else
2360: 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 2361: t= t- del*SQR(fp-fptt);
1.183 brouard 2362: #endif
1.202 brouard 2363: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2364: #ifdef DEBUG
1.181 brouard 2365: 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);
2366: 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 2367: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2368: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2369: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2370: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2371: 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);
2372: 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);
2373: #endif
1.183 brouard 2374: #ifdef POWELLORIGINAL
2375: if (t < 0.0) { /* Then we use it for new direction */
2376: #else
1.182 brouard 2377: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2378: 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 2379: 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 2380: 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 2381: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2382: }
1.181 brouard 2383: if (directest < 0.0) { /* Then we use it for new direction */
2384: #endif
1.191 brouard 2385: #ifdef DEBUGLINMIN
1.234 brouard 2386: printf("Before linmin in direction P%d-P0\n",n);
2387: for (j=1;j<=n;j++) {
2388: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2389: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2390: if(j % ncovmodel == 0){
2391: printf("\n");
2392: fprintf(ficlog,"\n");
2393: }
2394: }
1.224 brouard 2395: #endif
2396: #ifdef LINMINORIGINAL
1.234 brouard 2397: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2398: #else
1.234 brouard 2399: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2400: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2401: #endif
1.234 brouard 2402:
1.191 brouard 2403: #ifdef DEBUGLINMIN
1.234 brouard 2404: for (j=1;j<=n;j++) {
2405: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2406: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2407: if(j % ncovmodel == 0){
2408: printf("\n");
2409: fprintf(ficlog,"\n");
2410: }
2411: }
1.224 brouard 2412: #endif
1.234 brouard 2413: for (j=1;j<=n;j++) {
2414: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2415: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2416: }
1.224 brouard 2417: #ifdef LINMINORIGINAL
2418: #else
1.234 brouard 2419: for (j=1, flatd=0;j<=n;j++) {
2420: if(flatdir[j]>0)
2421: flatd++;
2422: }
2423: if(flatd >0){
1.255 brouard 2424: printf("%d flat directions: ",flatd);
2425: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2426: for (j=1;j<=n;j++) {
2427: if(flatdir[j]>0){
2428: printf("%d ",j);
2429: fprintf(ficlog,"%d ",j);
2430: }
2431: }
2432: printf("\n");
2433: fprintf(ficlog,"\n");
2434: }
1.191 brouard 2435: #endif
1.234 brouard 2436: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2437: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2438:
1.126 brouard 2439: #ifdef DEBUG
1.234 brouard 2440: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2441: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2442: for(j=1;j<=n;j++){
2443: printf(" %lf",xit[j]);
2444: fprintf(ficlog," %lf",xit[j]);
2445: }
2446: printf("\n");
2447: fprintf(ficlog,"\n");
1.126 brouard 2448: #endif
1.192 brouard 2449: } /* end of t or directest negative */
1.224 brouard 2450: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2451: #else
1.234 brouard 2452: } /* end if (fptt < fp) */
1.192 brouard 2453: #endif
1.225 brouard 2454: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2455: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2456: #else
1.224 brouard 2457: #endif
1.234 brouard 2458: } /* loop iteration */
1.126 brouard 2459: }
1.234 brouard 2460:
1.126 brouard 2461: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2462:
1.235 brouard 2463: 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 2464: {
1.235 brouard 2465: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2466: (and selected quantitative values in nres)
2467: by left multiplying the unit
1.234 brouard 2468: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2469: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2470: /* Wx is row vector: population in state 1, population in state 2, population dead */
2471: /* or prevalence in state 1, prevalence in state 2, 0 */
2472: /* newm is the matrix after multiplications, its rows are identical at a factor */
2473: /* Initial matrix pimij */
2474: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2475: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2476: /* 0, 0 , 1} */
2477: /*
2478: * and after some iteration: */
2479: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2480: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2481: /* 0, 0 , 1} */
2482: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2483: /* {0.51571254859325999, 0.4842874514067399, */
2484: /* 0.51326036147820708, 0.48673963852179264} */
2485: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2486:
1.126 brouard 2487: int i, ii,j,k;
1.209 brouard 2488: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2489: /* double **matprod2(); */ /* test */
1.218 brouard 2490: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2491: double **newm;
1.209 brouard 2492: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2493: int ncvloop=0;
1.169 brouard 2494:
1.209 brouard 2495: min=vector(1,nlstate);
2496: max=vector(1,nlstate);
2497: meandiff=vector(1,nlstate);
2498:
1.218 brouard 2499: /* Starting with matrix unity */
1.126 brouard 2500: for (ii=1;ii<=nlstate+ndeath;ii++)
2501: for (j=1;j<=nlstate+ndeath;j++){
2502: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2503: }
1.169 brouard 2504:
2505: cov[1]=1.;
2506:
2507: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2508: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2509: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2510: ncvloop++;
1.126 brouard 2511: newm=savm;
2512: /* Covariates have to be included here again */
1.138 brouard 2513: cov[2]=agefin;
1.187 brouard 2514: if(nagesqr==1)
2515: cov[3]= agefin*agefin;;
1.234 brouard 2516: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2517: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2518: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2519: /* 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 2520: }
2521: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2522: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2523: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2524: /* 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 2525: }
1.237 brouard 2526: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2527: if(Dummy[Tvar[Tage[k]]]){
2528: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2529: } else{
1.235 brouard 2530: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2531: }
1.235 brouard 2532: /* 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 2533: }
1.237 brouard 2534: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2535: /* 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 2536: if(Dummy[Tvard[k][1]==0]){
2537: if(Dummy[Tvard[k][2]==0]){
2538: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2539: }else{
2540: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2541: }
2542: }else{
2543: if(Dummy[Tvard[k][2]==0]){
2544: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2545: }else{
2546: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2547: }
2548: }
1.234 brouard 2549: }
1.138 brouard 2550: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2551: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2552: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2553: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2554: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2555: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2556: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2557:
1.126 brouard 2558: savm=oldm;
2559: oldm=newm;
1.209 brouard 2560:
2561: for(j=1; j<=nlstate; j++){
2562: max[j]=0.;
2563: min[j]=1.;
2564: }
2565: for(i=1;i<=nlstate;i++){
2566: sumnew=0;
2567: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2568: for(j=1; j<=nlstate; j++){
2569: prlim[i][j]= newm[i][j]/(1-sumnew);
2570: max[j]=FMAX(max[j],prlim[i][j]);
2571: min[j]=FMIN(min[j],prlim[i][j]);
2572: }
2573: }
2574:
1.126 brouard 2575: maxmax=0.;
1.209 brouard 2576: for(j=1; j<=nlstate; j++){
2577: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2578: maxmax=FMAX(maxmax,meandiff[j]);
2579: /* 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 2580: } /* j loop */
1.203 brouard 2581: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2582: /* 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 2583: if(maxmax < ftolpl){
1.209 brouard 2584: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2585: free_vector(min,1,nlstate);
2586: free_vector(max,1,nlstate);
2587: free_vector(meandiff,1,nlstate);
1.126 brouard 2588: return prlim;
2589: }
1.169 brouard 2590: } /* age loop */
1.208 brouard 2591: /* After some age loop it doesn't converge */
1.209 brouard 2592: 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 2593: 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 2594: /* 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); */
2595: free_vector(min,1,nlstate);
2596: free_vector(max,1,nlstate);
2597: free_vector(meandiff,1,nlstate);
1.208 brouard 2598:
1.169 brouard 2599: return prlim; /* should not reach here */
1.126 brouard 2600: }
2601:
1.217 brouard 2602:
2603: /**** Back Prevalence limit (stable or period prevalence) ****************/
2604:
1.218 brouard 2605: /* 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) */
2606: /* 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 2607: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2608: {
1.218 brouard 2609: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2610: matrix by transitions matrix until convergence is reached with precision ftolpl */
2611: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2612: /* Wx is row vector: population in state 1, population in state 2, population dead */
2613: /* or prevalence in state 1, prevalence in state 2, 0 */
2614: /* newm is the matrix after multiplications, its rows are identical at a factor */
2615: /* Initial matrix pimij */
2616: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2617: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2618: /* 0, 0 , 1} */
2619: /*
2620: * and after some iteration: */
2621: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2622: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2623: /* 0, 0 , 1} */
2624: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2625: /* {0.51571254859325999, 0.4842874514067399, */
2626: /* 0.51326036147820708, 0.48673963852179264} */
2627: /* If we start from prlim again, prlim tends to a constant matrix */
2628:
2629: int i, ii,j,k;
1.247 brouard 2630: int first=0;
1.217 brouard 2631: double *min, *max, *meandiff, maxmax,sumnew=0.;
2632: /* double **matprod2(); */ /* test */
2633: double **out, cov[NCOVMAX+1], **bmij();
2634: double **newm;
1.218 brouard 2635: double **dnewm, **doldm, **dsavm; /* for use */
2636: double **oldm, **savm; /* for use */
2637:
1.217 brouard 2638: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2639: int ncvloop=0;
2640:
2641: min=vector(1,nlstate);
2642: max=vector(1,nlstate);
2643: meandiff=vector(1,nlstate);
2644:
1.218 brouard 2645: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2646: oldm=oldms; savm=savms;
2647:
2648: /* Starting with matrix unity */
2649: for (ii=1;ii<=nlstate+ndeath;ii++)
2650: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2651: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2652: }
2653:
2654: cov[1]=1.;
2655:
2656: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2657: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2658: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2659: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2660: ncvloop++;
1.218 brouard 2661: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2662: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2663: /* Covariates have to be included here again */
2664: cov[2]=agefin;
2665: if(nagesqr==1)
2666: cov[3]= agefin*agefin;;
1.242 brouard 2667: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2668: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2669: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2670: /* 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)); */
2671: }
2672: /* for (k=1; k<=cptcovn;k++) { */
2673: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2674: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2675: /* /\* 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])]); *\/ */
2676: /* } */
2677: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2678: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2679: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2680: /* 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]); */
2681: }
2682: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2683: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2684: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2685: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2686: for (k=1; k<=cptcovage;k++){ /* For product with age */
2687: if(Dummy[Tvar[Tage[k]]]){
2688: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2689: } else{
2690: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2691: }
2692: /* 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]); */
2693: }
2694: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2695: /* 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]); */
2696: if(Dummy[Tvard[k][1]==0]){
2697: if(Dummy[Tvard[k][2]==0]){
2698: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2699: }else{
2700: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2701: }
2702: }else{
2703: if(Dummy[Tvard[k][2]==0]){
2704: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2705: }else{
2706: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2707: }
2708: }
1.217 brouard 2709: }
2710:
2711: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2712: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2713: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2714: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2715: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2716: /* ij should be linked to the correct index of cov */
2717: /* age and covariate values ij are in 'cov', but we need to pass
2718: * ij for the observed prevalence at age and status and covariate
2719: * number: prevacurrent[(int)agefin][ii][ij]
2720: */
2721: /* 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 *\/ */
2722: /* 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 *\/ */
2723: 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 2724: savm=oldm;
2725: oldm=newm;
2726: for(j=1; j<=nlstate; j++){
2727: max[j]=0.;
2728: min[j]=1.;
2729: }
2730: for(j=1; j<=nlstate; j++){
2731: for(i=1;i<=nlstate;i++){
1.234 brouard 2732: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2733: bprlim[i][j]= newm[i][j];
2734: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2735: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2736: }
2737: }
1.218 brouard 2738:
1.217 brouard 2739: maxmax=0.;
2740: for(i=1; i<=nlstate; i++){
2741: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2742: maxmax=FMAX(maxmax,meandiff[i]);
2743: /* 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); */
2744: } /* j loop */
2745: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2746: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2747: if(maxmax < ftolpl){
1.220 brouard 2748: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2749: free_vector(min,1,nlstate);
2750: free_vector(max,1,nlstate);
2751: free_vector(meandiff,1,nlstate);
2752: return bprlim;
2753: }
2754: } /* age loop */
2755: /* After some age loop it doesn't converge */
1.247 brouard 2756: if(first){
2757: first=1;
2758: 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\
2759: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
2760: }
2761: 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 2762: 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);
2763: /* 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); */
2764: free_vector(min,1,nlstate);
2765: free_vector(max,1,nlstate);
2766: free_vector(meandiff,1,nlstate);
2767:
2768: return bprlim; /* should not reach here */
2769: }
2770:
1.126 brouard 2771: /*************** transition probabilities ***************/
2772:
2773: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2774: {
1.138 brouard 2775: /* According to parameters values stored in x and the covariate's values stored in cov,
2776: computes the probability to be observed in state j being in state i by appying the
2777: model to the ncovmodel covariates (including constant and age).
2778: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2779: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2780: ncth covariate in the global vector x is given by the formula:
2781: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2782: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2783: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2784: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2785: Outputs ps[i][j] the probability to be observed in j being in j according to
2786: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2787: */
2788: double s1, lnpijopii;
1.126 brouard 2789: /*double t34;*/
1.164 brouard 2790: int i,j, nc, ii, jj;
1.126 brouard 2791:
1.223 brouard 2792: for(i=1; i<= nlstate; i++){
2793: for(j=1; j<i;j++){
2794: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2795: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2796: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2797: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2798: }
2799: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2800: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2801: }
2802: for(j=i+1; j<=nlstate+ndeath;j++){
2803: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2804: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2805: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2806: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2807: }
2808: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2809: }
2810: }
1.218 brouard 2811:
1.223 brouard 2812: for(i=1; i<= nlstate; i++){
2813: s1=0;
2814: for(j=1; j<i; j++){
2815: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2816: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2817: }
2818: for(j=i+1; j<=nlstate+ndeath; j++){
2819: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2820: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2821: }
2822: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2823: ps[i][i]=1./(s1+1.);
2824: /* Computing other pijs */
2825: for(j=1; j<i; j++)
2826: ps[i][j]= exp(ps[i][j])*ps[i][i];
2827: for(j=i+1; j<=nlstate+ndeath; j++)
2828: ps[i][j]= exp(ps[i][j])*ps[i][i];
2829: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2830: } /* end i */
1.218 brouard 2831:
1.223 brouard 2832: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2833: for(jj=1; jj<= nlstate+ndeath; jj++){
2834: ps[ii][jj]=0;
2835: ps[ii][ii]=1;
2836: }
2837: }
1.218 brouard 2838:
2839:
1.223 brouard 2840: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2841: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2842: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2843: /* } */
2844: /* printf("\n "); */
2845: /* } */
2846: /* printf("\n ");printf("%lf ",cov[2]);*/
2847: /*
2848: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2849: goto end;*/
1.223 brouard 2850: return ps;
1.126 brouard 2851: }
2852:
1.218 brouard 2853: /*************** backward transition probabilities ***************/
2854:
2855: /* 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 ) */
2856: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2857: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2858: {
1.222 brouard 2859: /* Computes the backward probability at age agefin and covariate ij
2860: * and returns in **ps as well as **bmij.
2861: */
1.218 brouard 2862: int i, ii, j,k;
1.222 brouard 2863:
2864: double **out, **pmij();
2865: double sumnew=0.;
1.218 brouard 2866: double agefin;
1.222 brouard 2867:
2868: double **dnewm, **dsavm, **doldm;
2869: double **bbmij;
2870:
1.218 brouard 2871: doldm=ddoldms; /* global pointers */
1.222 brouard 2872: dnewm=ddnewms;
2873: dsavm=ddsavms;
2874:
2875: agefin=cov[2];
2876: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2877: the observed prevalence (with this covariate ij) */
2878: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2879: /* We do have the matrix Px in savm and we need pij */
2880: for (j=1;j<=nlstate+ndeath;j++){
2881: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2882: for (ii=1;ii<=nlstate;ii++){
2883: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2884: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2885: for (ii=1;ii<=nlstate+ndeath;ii++){
2886: if(sumnew >= 1.e-10){
2887: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2888: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2889: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2890: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2891: /* }else */
2892: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2893: }else{
1.242 brouard 2894: ;
2895: /* 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 2896: }
2897: } /*End ii */
2898: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2899: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2900: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2901: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2902: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2903: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2904: /* left Product of this matrix by diag matrix of prevalences (savm) */
2905: for (j=1;j<=nlstate+ndeath;j++){
2906: for (ii=1;ii<=nlstate+ndeath;ii++){
2907: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2908: }
2909: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2910: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2911: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2912: /* end bmij */
2913: return ps;
1.218 brouard 2914: }
1.217 brouard 2915: /*************** transition probabilities ***************/
2916:
1.218 brouard 2917: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2918: {
2919: /* According to parameters values stored in x and the covariate's values stored in cov,
2920: computes the probability to be observed in state j being in state i by appying the
2921: model to the ncovmodel covariates (including constant and age).
2922: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2923: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2924: ncth covariate in the global vector x is given by the formula:
2925: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2926: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2927: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2928: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2929: Outputs ps[i][j] the probability to be observed in j being in j according to
2930: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2931: */
2932: double s1, lnpijopii;
2933: /*double t34;*/
2934: int i,j, nc, ii, jj;
2935:
1.234 brouard 2936: for(i=1; i<= nlstate; i++){
2937: for(j=1; j<i;j++){
2938: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2939: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2940: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2941: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2942: }
2943: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2944: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2945: }
2946: for(j=i+1; j<=nlstate+ndeath;j++){
2947: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2948: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2949: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2950: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2951: }
2952: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2953: }
2954: }
2955:
2956: for(i=1; i<= nlstate; i++){
2957: s1=0;
2958: for(j=1; j<i; j++){
2959: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2960: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2961: }
2962: for(j=i+1; j<=nlstate+ndeath; j++){
2963: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2964: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2965: }
2966: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2967: ps[i][i]=1./(s1+1.);
2968: /* Computing other pijs */
2969: for(j=1; j<i; j++)
2970: ps[i][j]= exp(ps[i][j])*ps[i][i];
2971: for(j=i+1; j<=nlstate+ndeath; j++)
2972: ps[i][j]= exp(ps[i][j])*ps[i][i];
2973: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2974: } /* end i */
2975:
2976: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2977: for(jj=1; jj<= nlstate+ndeath; jj++){
2978: ps[ii][jj]=0;
2979: ps[ii][ii]=1;
2980: }
2981: }
2982: /* Added for backcast */ /* Transposed matrix too */
2983: for(jj=1; jj<= nlstate+ndeath; jj++){
2984: s1=0.;
2985: for(ii=1; ii<= nlstate+ndeath; ii++){
2986: s1+=ps[ii][jj];
2987: }
2988: for(ii=1; ii<= nlstate; ii++){
2989: ps[ii][jj]=ps[ii][jj]/s1;
2990: }
2991: }
2992: /* Transposition */
2993: for(jj=1; jj<= nlstate+ndeath; jj++){
2994: for(ii=jj; ii<= nlstate+ndeath; ii++){
2995: s1=ps[ii][jj];
2996: ps[ii][jj]=ps[jj][ii];
2997: ps[jj][ii]=s1;
2998: }
2999: }
3000: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3001: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3002: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3003: /* } */
3004: /* printf("\n "); */
3005: /* } */
3006: /* printf("\n ");printf("%lf ",cov[2]);*/
3007: /*
3008: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3009: goto end;*/
3010: return ps;
1.217 brouard 3011: }
3012:
3013:
1.126 brouard 3014: /**************** Product of 2 matrices ******************/
3015:
1.145 brouard 3016: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3017: {
3018: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3019: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3020: /* in, b, out are matrice of pointers which should have been initialized
3021: before: only the contents of out is modified. The function returns
3022: a pointer to pointers identical to out */
1.145 brouard 3023: int i, j, k;
1.126 brouard 3024: for(i=nrl; i<= nrh; i++)
1.145 brouard 3025: for(k=ncolol; k<=ncoloh; k++){
3026: out[i][k]=0.;
3027: for(j=ncl; j<=nch; j++)
3028: out[i][k] +=in[i][j]*b[j][k];
3029: }
1.126 brouard 3030: return out;
3031: }
3032:
3033:
3034: /************* Higher Matrix Product ***************/
3035:
1.235 brouard 3036: 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 3037: {
1.218 brouard 3038: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3039: 'nhstepm*hstepm*stepm' months (i.e. until
3040: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3041: nhstepm*hstepm matrices.
3042: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3043: (typically every 2 years instead of every month which is too big
3044: for the memory).
3045: Model is determined by parameters x and covariates have to be
3046: included manually here.
3047:
3048: */
3049:
3050: int i, j, d, h, k;
1.131 brouard 3051: double **out, cov[NCOVMAX+1];
1.126 brouard 3052: double **newm;
1.187 brouard 3053: double agexact;
1.214 brouard 3054: double agebegin, ageend;
1.126 brouard 3055:
3056: /* Hstepm could be zero and should return the unit matrix */
3057: for (i=1;i<=nlstate+ndeath;i++)
3058: for (j=1;j<=nlstate+ndeath;j++){
3059: oldm[i][j]=(i==j ? 1.0 : 0.0);
3060: po[i][j][0]=(i==j ? 1.0 : 0.0);
3061: }
3062: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3063: for(h=1; h <=nhstepm; h++){
3064: for(d=1; d <=hstepm; d++){
3065: newm=savm;
3066: /* Covariates have to be included here again */
3067: cov[1]=1.;
1.214 brouard 3068: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3069: cov[2]=agexact;
3070: if(nagesqr==1)
1.227 brouard 3071: cov[3]= agexact*agexact;
1.235 brouard 3072: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3073: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3074: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3075: /* 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)); */
3076: }
3077: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3078: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3079: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3080: /* 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]); */
3081: }
3082: for (k=1; k<=cptcovage;k++){
3083: if(Dummy[Tvar[Tage[k]]]){
3084: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3085: } else{
3086: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3087: }
3088: /* 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]); */
3089: }
3090: for (k=1; k<=cptcovprod;k++){ /* */
3091: /* 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]); */
3092: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3093: }
3094: /* for (k=1; k<=cptcovn;k++) */
3095: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3096: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3097: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3098: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3099: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3100:
3101:
1.126 brouard 3102: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3103: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3104: /* right multiplication of oldm by the current matrix */
1.126 brouard 3105: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3106: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3107: /* if((int)age == 70){ */
3108: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3109: /* for(i=1; i<=nlstate+ndeath; i++) { */
3110: /* printf("%d pmmij ",i); */
3111: /* for(j=1;j<=nlstate+ndeath;j++) { */
3112: /* printf("%f ",pmmij[i][j]); */
3113: /* } */
3114: /* printf(" oldm "); */
3115: /* for(j=1;j<=nlstate+ndeath;j++) { */
3116: /* printf("%f ",oldm[i][j]); */
3117: /* } */
3118: /* printf("\n"); */
3119: /* } */
3120: /* } */
1.126 brouard 3121: savm=oldm;
3122: oldm=newm;
3123: }
3124: for(i=1; i<=nlstate+ndeath; i++)
3125: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3126: po[i][j][h]=newm[i][j];
3127: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3128: }
1.128 brouard 3129: /*printf("h=%d ",h);*/
1.126 brouard 3130: } /* end h */
1.218 brouard 3131: /* printf("\n H=%d \n",h); */
1.126 brouard 3132: return po;
3133: }
3134:
1.217 brouard 3135: /************* Higher Back Matrix Product ***************/
1.218 brouard 3136: /* 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 3137: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3138: {
1.218 brouard 3139: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3140: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3141: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3142: nhstepm*hstepm matrices.
3143: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3144: (typically every 2 years instead of every month which is too big
1.217 brouard 3145: for the memory).
1.218 brouard 3146: Model is determined by parameters x and covariates have to be
3147: included manually here.
1.217 brouard 3148:
1.222 brouard 3149: */
1.217 brouard 3150:
3151: int i, j, d, h, k;
3152: double **out, cov[NCOVMAX+1];
3153: double **newm;
3154: double agexact;
3155: double agebegin, ageend;
1.222 brouard 3156: double **oldm, **savm;
1.217 brouard 3157:
1.222 brouard 3158: oldm=oldms;savm=savms;
1.217 brouard 3159: /* Hstepm could be zero and should return the unit matrix */
3160: for (i=1;i<=nlstate+ndeath;i++)
3161: for (j=1;j<=nlstate+ndeath;j++){
3162: oldm[i][j]=(i==j ? 1.0 : 0.0);
3163: po[i][j][0]=(i==j ? 1.0 : 0.0);
3164: }
3165: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3166: for(h=1; h <=nhstepm; h++){
3167: for(d=1; d <=hstepm; d++){
3168: newm=savm;
3169: /* Covariates have to be included here again */
3170: cov[1]=1.;
3171: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3172: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3173: cov[2]=agexact;
3174: if(nagesqr==1)
1.222 brouard 3175: cov[3]= agexact*agexact;
1.218 brouard 3176: for (k=1; k<=cptcovn;k++)
1.222 brouard 3177: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3178: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3179: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3180: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3181: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3182: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3183: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3184: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3185: /* 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 3186:
3187:
1.217 brouard 3188: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3189: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3190: /* Careful transposed matrix */
1.222 brouard 3191: /* age is in cov[2] */
1.218 brouard 3192: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3193: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3194: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3195: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3196: /* if((int)age == 70){ */
3197: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3198: /* for(i=1; i<=nlstate+ndeath; i++) { */
3199: /* printf("%d pmmij ",i); */
3200: /* for(j=1;j<=nlstate+ndeath;j++) { */
3201: /* printf("%f ",pmmij[i][j]); */
3202: /* } */
3203: /* printf(" oldm "); */
3204: /* for(j=1;j<=nlstate+ndeath;j++) { */
3205: /* printf("%f ",oldm[i][j]); */
3206: /* } */
3207: /* printf("\n"); */
3208: /* } */
3209: /* } */
3210: savm=oldm;
3211: oldm=newm;
3212: }
3213: for(i=1; i<=nlstate+ndeath; i++)
3214: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3215: po[i][j][h]=newm[i][j];
3216: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3217: }
3218: /*printf("h=%d ",h);*/
3219: } /* end h */
1.222 brouard 3220: /* printf("\n H=%d \n",h); */
1.217 brouard 3221: return po;
3222: }
3223:
3224:
1.162 brouard 3225: #ifdef NLOPT
3226: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3227: double fret;
3228: double *xt;
3229: int j;
3230: myfunc_data *d2 = (myfunc_data *) pd;
3231: /* xt = (p1-1); */
3232: xt=vector(1,n);
3233: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3234:
3235: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3236: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3237: printf("Function = %.12lf ",fret);
3238: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3239: printf("\n");
3240: free_vector(xt,1,n);
3241: return fret;
3242: }
3243: #endif
1.126 brouard 3244:
3245: /*************** log-likelihood *************/
3246: double func( double *x)
3247: {
1.226 brouard 3248: int i, ii, j, k, mi, d, kk;
3249: int ioffset=0;
3250: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3251: double **out;
3252: double lli; /* Individual log likelihood */
3253: int s1, s2;
1.228 brouard 3254: 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 3255: double bbh, survp;
3256: long ipmx;
3257: double agexact;
3258: /*extern weight */
3259: /* We are differentiating ll according to initial status */
3260: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3261: /*for(i=1;i<imx;i++)
3262: printf(" %d\n",s[4][i]);
3263: */
1.162 brouard 3264:
1.226 brouard 3265: ++countcallfunc;
1.162 brouard 3266:
1.226 brouard 3267: cov[1]=1.;
1.126 brouard 3268:
1.226 brouard 3269: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3270: ioffset=0;
1.226 brouard 3271: if(mle==1){
3272: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3273: /* Computes the values of the ncovmodel covariates of the model
3274: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3275: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3276: to be observed in j being in i according to the model.
3277: */
1.243 brouard 3278: ioffset=2+nagesqr ;
1.233 brouard 3279: /* Fixed */
1.234 brouard 3280: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3281: 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)*/
3282: }
1.226 brouard 3283: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3284: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3285: has been calculated etc */
3286: /* For an individual i, wav[i] gives the number of effective waves */
3287: /* We compute the contribution to Likelihood of each effective transition
3288: mw[mi][i] is real wave of the mi th effectve wave */
3289: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3290: s2=s[mw[mi+1][i]][i];
3291: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3292: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3293: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3294: */
3295: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3296: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3297: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3298: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3299: }
3300: for (ii=1;ii<=nlstate+ndeath;ii++)
3301: for (j=1;j<=nlstate+ndeath;j++){
3302: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3303: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3304: }
3305: for(d=0; d<dh[mi][i]; d++){
3306: newm=savm;
3307: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3308: cov[2]=agexact;
3309: if(nagesqr==1)
3310: cov[3]= agexact*agexact; /* Should be changed here */
3311: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3312: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3313: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3314: else
3315: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3316: }
3317: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3318: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3319: savm=oldm;
3320: oldm=newm;
3321: } /* end mult */
3322:
3323: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3324: /* But now since version 0.9 we anticipate for bias at large stepm.
3325: * If stepm is larger than one month (smallest stepm) and if the exact delay
3326: * (in months) between two waves is not a multiple of stepm, we rounded to
3327: * the nearest (and in case of equal distance, to the lowest) interval but now
3328: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3329: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3330: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3331: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3332: * -stepm/2 to stepm/2 .
3333: * For stepm=1 the results are the same as for previous versions of Imach.
3334: * For stepm > 1 the results are less biased than in previous versions.
3335: */
1.234 brouard 3336: s1=s[mw[mi][i]][i];
3337: s2=s[mw[mi+1][i]][i];
3338: bbh=(double)bh[mi][i]/(double)stepm;
3339: /* bias bh is positive if real duration
3340: * is higher than the multiple of stepm and negative otherwise.
3341: */
3342: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3343: if( s2 > nlstate){
3344: /* i.e. if s2 is a death state and if the date of death is known
3345: then the contribution to the likelihood is the probability to
3346: die between last step unit time and current step unit time,
3347: which is also equal to probability to die before dh
3348: minus probability to die before dh-stepm .
3349: In version up to 0.92 likelihood was computed
3350: as if date of death was unknown. Death was treated as any other
3351: health state: the date of the interview describes the actual state
3352: and not the date of a change in health state. The former idea was
3353: to consider that at each interview the state was recorded
3354: (healthy, disable or death) and IMaCh was corrected; but when we
3355: introduced the exact date of death then we should have modified
3356: the contribution of an exact death to the likelihood. This new
3357: contribution is smaller and very dependent of the step unit
3358: stepm. It is no more the probability to die between last interview
3359: and month of death but the probability to survive from last
3360: interview up to one month before death multiplied by the
3361: probability to die within a month. Thanks to Chris
3362: Jackson for correcting this bug. Former versions increased
3363: mortality artificially. The bad side is that we add another loop
3364: which slows down the processing. The difference can be up to 10%
3365: lower mortality.
3366: */
3367: /* If, at the beginning of the maximization mostly, the
3368: cumulative probability or probability to be dead is
3369: constant (ie = 1) over time d, the difference is equal to
3370: 0. out[s1][3] = savm[s1][3]: probability, being at state
3371: s1 at precedent wave, to be dead a month before current
3372: wave is equal to probability, being at state s1 at
3373: precedent wave, to be dead at mont of the current
3374: wave. Then the observed probability (that this person died)
3375: is null according to current estimated parameter. In fact,
3376: it should be very low but not zero otherwise the log go to
3377: infinity.
3378: */
1.183 brouard 3379: /* #ifdef INFINITYORIGINAL */
3380: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3381: /* #else */
3382: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3383: /* lli=log(mytinydouble); */
3384: /* else */
3385: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3386: /* #endif */
1.226 brouard 3387: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3388:
1.226 brouard 3389: } else if ( s2==-1 ) { /* alive */
3390: for (j=1,survp=0. ; j<=nlstate; j++)
3391: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3392: /*survp += out[s1][j]; */
3393: lli= log(survp);
3394: }
3395: else if (s2==-4) {
3396: for (j=3,survp=0. ; j<=nlstate; j++)
3397: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3398: lli= log(survp);
3399: }
3400: else if (s2==-5) {
3401: for (j=1,survp=0. ; j<=2; j++)
3402: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3403: lli= log(survp);
3404: }
3405: else{
3406: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3407: /* 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 */
3408: }
3409: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3410: /*if(lli ==000.0)*/
3411: /*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); */
3412: ipmx +=1;
3413: sw += weight[i];
3414: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3415: /* if (lli < log(mytinydouble)){ */
3416: /* 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); */
3417: /* 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]); */
3418: /* } */
3419: } /* end of wave */
3420: } /* end of individual */
3421: } else if(mle==2){
3422: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3423: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3424: for(mi=1; mi<= wav[i]-1; mi++){
3425: for (ii=1;ii<=nlstate+ndeath;ii++)
3426: for (j=1;j<=nlstate+ndeath;j++){
3427: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3428: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3429: }
3430: for(d=0; d<=dh[mi][i]; d++){
3431: newm=savm;
3432: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3433: cov[2]=agexact;
3434: if(nagesqr==1)
3435: cov[3]= agexact*agexact;
3436: for (kk=1; kk<=cptcovage;kk++) {
3437: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3438: }
3439: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3440: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3441: savm=oldm;
3442: oldm=newm;
3443: } /* end mult */
3444:
3445: s1=s[mw[mi][i]][i];
3446: s2=s[mw[mi+1][i]][i];
3447: bbh=(double)bh[mi][i]/(double)stepm;
3448: 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 */
3449: ipmx +=1;
3450: sw += weight[i];
3451: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3452: } /* end of wave */
3453: } /* end of individual */
3454: } else if(mle==3){ /* exponential inter-extrapolation */
3455: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3456: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3457: for(mi=1; mi<= wav[i]-1; mi++){
3458: for (ii=1;ii<=nlstate+ndeath;ii++)
3459: for (j=1;j<=nlstate+ndeath;j++){
3460: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3461: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3462: }
3463: for(d=0; d<dh[mi][i]; d++){
3464: newm=savm;
3465: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3466: cov[2]=agexact;
3467: if(nagesqr==1)
3468: cov[3]= agexact*agexact;
3469: for (kk=1; kk<=cptcovage;kk++) {
3470: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3471: }
3472: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3473: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3474: savm=oldm;
3475: oldm=newm;
3476: } /* end mult */
3477:
3478: s1=s[mw[mi][i]][i];
3479: s2=s[mw[mi+1][i]][i];
3480: bbh=(double)bh[mi][i]/(double)stepm;
3481: 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 */
3482: ipmx +=1;
3483: sw += weight[i];
3484: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3485: } /* end of wave */
3486: } /* end of individual */
3487: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3488: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3489: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3490: for(mi=1; mi<= wav[i]-1; mi++){
3491: for (ii=1;ii<=nlstate+ndeath;ii++)
3492: for (j=1;j<=nlstate+ndeath;j++){
3493: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3494: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3495: }
3496: for(d=0; d<dh[mi][i]; d++){
3497: newm=savm;
3498: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3499: cov[2]=agexact;
3500: if(nagesqr==1)
3501: cov[3]= agexact*agexact;
3502: for (kk=1; kk<=cptcovage;kk++) {
3503: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3504: }
1.126 brouard 3505:
1.226 brouard 3506: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3507: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3508: savm=oldm;
3509: oldm=newm;
3510: } /* end mult */
3511:
3512: s1=s[mw[mi][i]][i];
3513: s2=s[mw[mi+1][i]][i];
3514: if( s2 > nlstate){
3515: lli=log(out[s1][s2] - savm[s1][s2]);
3516: } else if ( s2==-1 ) { /* alive */
3517: for (j=1,survp=0. ; j<=nlstate; j++)
3518: survp += out[s1][j];
3519: lli= log(survp);
3520: }else{
3521: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3522: }
3523: ipmx +=1;
3524: sw += weight[i];
3525: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3526: /* 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 3527: } /* end of wave */
3528: } /* end of individual */
3529: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3530: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3531: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3532: for(mi=1; mi<= wav[i]-1; mi++){
3533: for (ii=1;ii<=nlstate+ndeath;ii++)
3534: for (j=1;j<=nlstate+ndeath;j++){
3535: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3536: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3537: }
3538: for(d=0; d<dh[mi][i]; d++){
3539: newm=savm;
3540: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3541: cov[2]=agexact;
3542: if(nagesqr==1)
3543: cov[3]= agexact*agexact;
3544: for (kk=1; kk<=cptcovage;kk++) {
3545: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3546: }
1.126 brouard 3547:
1.226 brouard 3548: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3549: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3550: savm=oldm;
3551: oldm=newm;
3552: } /* end mult */
3553:
3554: s1=s[mw[mi][i]][i];
3555: s2=s[mw[mi+1][i]][i];
3556: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3557: ipmx +=1;
3558: sw += weight[i];
3559: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3560: /*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]);*/
3561: } /* end of wave */
3562: } /* end of individual */
3563: } /* End of if */
3564: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3565: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3566: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3567: return -l;
1.126 brouard 3568: }
3569:
3570: /*************** log-likelihood *************/
3571: double funcone( double *x)
3572: {
1.228 brouard 3573: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3574: int i, ii, j, k, mi, d, kk;
1.228 brouard 3575: int ioffset=0;
1.131 brouard 3576: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3577: double **out;
3578: double lli; /* Individual log likelihood */
3579: double llt;
3580: int s1, s2;
1.228 brouard 3581: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3582:
1.126 brouard 3583: double bbh, survp;
1.187 brouard 3584: double agexact;
1.214 brouard 3585: double agebegin, ageend;
1.126 brouard 3586: /*extern weight */
3587: /* We are differentiating ll according to initial status */
3588: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3589: /*for(i=1;i<imx;i++)
3590: printf(" %d\n",s[4][i]);
3591: */
3592: cov[1]=1.;
3593:
3594: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3595: ioffset=0;
3596: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3597: /* ioffset=2+nagesqr+cptcovage; */
3598: ioffset=2+nagesqr;
1.232 brouard 3599: /* Fixed */
1.224 brouard 3600: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3601: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3602: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3603: 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)*/
3604: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3605: /* cov[2+6]=covar[Tvar[6]][i]; */
3606: /* cov[2+6]=covar[2][i]; V2 */
3607: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3608: /* cov[2+7]=covar[Tvar[7]][i]; */
3609: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3610: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3611: /* cov[2+9]=covar[Tvar[9]][i]; */
3612: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3613: }
1.232 brouard 3614: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3615: /* 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?)*\/ */
3616: /* } */
1.231 brouard 3617: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3618: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3619: /* } */
1.225 brouard 3620:
1.233 brouard 3621:
3622: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3623: /* Wave varying (but not age varying) */
3624: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3625: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3626: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3627: }
1.232 brouard 3628: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3629: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3630: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3631: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3632: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3633: /* 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 3634: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3635: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3636: /* /\* 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]); *\/ */
3637: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3638: /* } */
1.126 brouard 3639: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3640: for (j=1;j<=nlstate+ndeath;j++){
3641: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3642: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3643: }
1.214 brouard 3644:
3645: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3646: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3647: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3648: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3649: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3650: and mw[mi+1][i]. dh depends on stepm.*/
3651: newm=savm;
1.247 brouard 3652: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3653: cov[2]=agexact;
3654: if(nagesqr==1)
3655: cov[3]= agexact*agexact;
3656: for (kk=1; kk<=cptcovage;kk++) {
3657: if(!FixedV[Tvar[Tage[kk]]])
3658: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3659: else
3660: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3661: }
3662: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3663: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3664: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3665: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3666: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3667: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3668: savm=oldm;
3669: oldm=newm;
1.126 brouard 3670: } /* end mult */
3671:
3672: s1=s[mw[mi][i]][i];
3673: s2=s[mw[mi+1][i]][i];
1.217 brouard 3674: /* if(s2==-1){ */
3675: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3676: /* /\* exit(1); *\/ */
3677: /* } */
1.126 brouard 3678: bbh=(double)bh[mi][i]/(double)stepm;
3679: /* bias is positive if real duration
3680: * is higher than the multiple of stepm and negative otherwise.
3681: */
3682: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3683: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3684: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3685: for (j=1,survp=0. ; j<=nlstate; j++)
3686: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3687: lli= log(survp);
1.126 brouard 3688: }else if (mle==1){
1.242 brouard 3689: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3690: } else if(mle==2){
1.242 brouard 3691: 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 3692: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3693: 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 3694: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3695: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3696: } else{ /* mle=0 back to 1 */
1.242 brouard 3697: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3698: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3699: } /* End of if */
3700: ipmx +=1;
3701: sw += weight[i];
3702: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3703: /*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 3704: if(globpr){
1.246 brouard 3705: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3706: %11.6f %11.6f %11.6f ", \
1.242 brouard 3707: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3708: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3709: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3710: llt +=ll[k]*gipmx/gsw;
3711: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3712: }
3713: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3714: }
1.232 brouard 3715: } /* end of wave */
3716: } /* end of individual */
3717: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3718: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3719: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3720: if(globpr==0){ /* First time we count the contributions and weights */
3721: gipmx=ipmx;
3722: gsw=sw;
3723: }
3724: return -l;
1.126 brouard 3725: }
3726:
3727:
3728: /*************** function likelione ***********/
3729: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3730: {
3731: /* This routine should help understanding what is done with
3732: the selection of individuals/waves and
3733: to check the exact contribution to the likelihood.
3734: Plotting could be done.
3735: */
3736: int k;
3737:
3738: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3739: strcpy(fileresilk,"ILK_");
1.202 brouard 3740: strcat(fileresilk,fileresu);
1.126 brouard 3741: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3742: printf("Problem with resultfile: %s\n", fileresilk);
3743: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3744: }
1.214 brouard 3745: 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");
3746: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3747: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3748: for(k=1; k<=nlstate; k++)
3749: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3750: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3751: }
3752:
3753: *fretone=(*funcone)(p);
3754: if(*globpri !=0){
3755: fclose(ficresilk);
1.205 brouard 3756: if (mle ==0)
3757: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3758: else if(mle >=1)
3759: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3760: 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 3761:
1.208 brouard 3762:
3763: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3764: 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 3765: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3766: }
1.207 brouard 3767: 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 3768: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3769: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3770: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3771: fflush(fichtm);
1.205 brouard 3772: }
1.126 brouard 3773: return;
3774: }
3775:
3776:
3777: /*********** Maximum Likelihood Estimation ***************/
3778:
3779: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3780: {
1.165 brouard 3781: int i,j, iter=0;
1.126 brouard 3782: double **xi;
3783: double fret;
3784: double fretone; /* Only one call to likelihood */
3785: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3786:
3787: #ifdef NLOPT
3788: int creturn;
3789: nlopt_opt opt;
3790: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3791: double *lb;
3792: double minf; /* the minimum objective value, upon return */
3793: double * p1; /* Shifted parameters from 0 instead of 1 */
3794: myfunc_data dinst, *d = &dinst;
3795: #endif
3796:
3797:
1.126 brouard 3798: xi=matrix(1,npar,1,npar);
3799: for (i=1;i<=npar;i++)
3800: for (j=1;j<=npar;j++)
3801: xi[i][j]=(i==j ? 1.0 : 0.0);
3802: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3803: strcpy(filerespow,"POW_");
1.126 brouard 3804: strcat(filerespow,fileres);
3805: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3806: printf("Problem with resultfile: %s\n", filerespow);
3807: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3808: }
3809: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3810: for (i=1;i<=nlstate;i++)
3811: for(j=1;j<=nlstate+ndeath;j++)
3812: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3813: fprintf(ficrespow,"\n");
1.162 brouard 3814: #ifdef POWELL
1.126 brouard 3815: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3816: #endif
1.126 brouard 3817:
1.162 brouard 3818: #ifdef NLOPT
3819: #ifdef NEWUOA
3820: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3821: #else
3822: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3823: #endif
3824: lb=vector(0,npar-1);
3825: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3826: nlopt_set_lower_bounds(opt, lb);
3827: nlopt_set_initial_step1(opt, 0.1);
3828:
3829: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3830: d->function = func;
3831: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3832: nlopt_set_min_objective(opt, myfunc, d);
3833: nlopt_set_xtol_rel(opt, ftol);
3834: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3835: printf("nlopt failed! %d\n",creturn);
3836: }
3837: else {
3838: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3839: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3840: iter=1; /* not equal */
3841: }
3842: nlopt_destroy(opt);
3843: #endif
1.126 brouard 3844: free_matrix(xi,1,npar,1,npar);
3845: fclose(ficrespow);
1.203 brouard 3846: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3847: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3848: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3849:
3850: }
3851:
3852: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3853: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3854: {
3855: double **a,**y,*x,pd;
1.203 brouard 3856: /* double **hess; */
1.164 brouard 3857: int i, j;
1.126 brouard 3858: int *indx;
3859:
3860: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3861: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3862: void lubksb(double **a, int npar, int *indx, double b[]) ;
3863: void ludcmp(double **a, int npar, int *indx, double *d) ;
3864: double gompertz(double p[]);
1.203 brouard 3865: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3866:
3867: printf("\nCalculation of the hessian matrix. Wait...\n");
3868: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3869: for (i=1;i<=npar;i++){
1.203 brouard 3870: printf("%d-",i);fflush(stdout);
3871: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3872:
3873: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3874:
3875: /* printf(" %f ",p[i]);
3876: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3877: }
3878:
3879: for (i=1;i<=npar;i++) {
3880: for (j=1;j<=npar;j++) {
3881: if (j>i) {
1.203 brouard 3882: printf(".%d-%d",i,j);fflush(stdout);
3883: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3884: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3885:
3886: hess[j][i]=hess[i][j];
3887: /*printf(" %lf ",hess[i][j]);*/
3888: }
3889: }
3890: }
3891: printf("\n");
3892: fprintf(ficlog,"\n");
3893:
3894: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3895: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3896:
3897: a=matrix(1,npar,1,npar);
3898: y=matrix(1,npar,1,npar);
3899: x=vector(1,npar);
3900: indx=ivector(1,npar);
3901: for (i=1;i<=npar;i++)
3902: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3903: ludcmp(a,npar,indx,&pd);
3904:
3905: for (j=1;j<=npar;j++) {
3906: for (i=1;i<=npar;i++) x[i]=0;
3907: x[j]=1;
3908: lubksb(a,npar,indx,x);
3909: for (i=1;i<=npar;i++){
3910: matcov[i][j]=x[i];
3911: }
3912: }
3913:
3914: printf("\n#Hessian matrix#\n");
3915: fprintf(ficlog,"\n#Hessian matrix#\n");
3916: for (i=1;i<=npar;i++) {
3917: for (j=1;j<=npar;j++) {
1.203 brouard 3918: printf("%.6e ",hess[i][j]);
3919: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3920: }
3921: printf("\n");
3922: fprintf(ficlog,"\n");
3923: }
3924:
1.203 brouard 3925: /* printf("\n#Covariance matrix#\n"); */
3926: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3927: /* for (i=1;i<=npar;i++) { */
3928: /* for (j=1;j<=npar;j++) { */
3929: /* printf("%.6e ",matcov[i][j]); */
3930: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3931: /* } */
3932: /* printf("\n"); */
3933: /* fprintf(ficlog,"\n"); */
3934: /* } */
3935:
1.126 brouard 3936: /* Recompute Inverse */
1.203 brouard 3937: /* for (i=1;i<=npar;i++) */
3938: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3939: /* ludcmp(a,npar,indx,&pd); */
3940:
3941: /* printf("\n#Hessian matrix recomputed#\n"); */
3942:
3943: /* for (j=1;j<=npar;j++) { */
3944: /* for (i=1;i<=npar;i++) x[i]=0; */
3945: /* x[j]=1; */
3946: /* lubksb(a,npar,indx,x); */
3947: /* for (i=1;i<=npar;i++){ */
3948: /* y[i][j]=x[i]; */
3949: /* printf("%.3e ",y[i][j]); */
3950: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3951: /* } */
3952: /* printf("\n"); */
3953: /* fprintf(ficlog,"\n"); */
3954: /* } */
3955:
3956: /* Verifying the inverse matrix */
3957: #ifdef DEBUGHESS
3958: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3959:
1.203 brouard 3960: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3961: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3962:
3963: for (j=1;j<=npar;j++) {
3964: for (i=1;i<=npar;i++){
1.203 brouard 3965: printf("%.2f ",y[i][j]);
3966: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3967: }
3968: printf("\n");
3969: fprintf(ficlog,"\n");
3970: }
1.203 brouard 3971: #endif
1.126 brouard 3972:
3973: free_matrix(a,1,npar,1,npar);
3974: free_matrix(y,1,npar,1,npar);
3975: free_vector(x,1,npar);
3976: free_ivector(indx,1,npar);
1.203 brouard 3977: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3978:
3979:
3980: }
3981:
3982: /*************** hessian matrix ****************/
3983: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3984: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3985: int i;
3986: int l=1, lmax=20;
1.203 brouard 3987: double k1,k2, res, fx;
1.132 brouard 3988: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3989: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3990: int k=0,kmax=10;
3991: double l1;
3992:
3993: fx=func(x);
3994: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3995: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3996: l1=pow(10,l);
3997: delts=delt;
3998: for(k=1 ; k <kmax; k=k+1){
3999: delt = delta*(l1*k);
4000: p2[theta]=x[theta] +delt;
1.145 brouard 4001: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4002: p2[theta]=x[theta]-delt;
4003: k2=func(p2)-fx;
4004: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4005: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4006:
1.203 brouard 4007: #ifdef DEBUGHESSII
1.126 brouard 4008: 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);
4009: 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);
4010: #endif
4011: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4012: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4013: k=kmax;
4014: }
4015: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4016: k=kmax; l=lmax*10;
1.126 brouard 4017: }
4018: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4019: delts=delt;
4020: }
1.203 brouard 4021: } /* End loop k */
1.126 brouard 4022: }
4023: delti[theta]=delts;
4024: return res;
4025:
4026: }
4027:
1.203 brouard 4028: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4029: {
4030: int i;
1.164 brouard 4031: int l=1, lmax=20;
1.126 brouard 4032: double k1,k2,k3,k4,res,fx;
1.132 brouard 4033: double p2[MAXPARM+1];
1.203 brouard 4034: int k, kmax=1;
4035: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4036:
4037: int firstime=0;
1.203 brouard 4038:
1.126 brouard 4039: fx=func(x);
1.203 brouard 4040: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4041: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4042: p2[thetai]=x[thetai]+delti[thetai]*k;
4043: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4044: k1=func(p2)-fx;
4045:
1.203 brouard 4046: p2[thetai]=x[thetai]+delti[thetai]*k;
4047: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4048: k2=func(p2)-fx;
4049:
1.203 brouard 4050: p2[thetai]=x[thetai]-delti[thetai]*k;
4051: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4052: k3=func(p2)-fx;
4053:
1.203 brouard 4054: p2[thetai]=x[thetai]-delti[thetai]*k;
4055: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4056: k4=func(p2)-fx;
1.203 brouard 4057: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4058: if(k1*k2*k3*k4 <0.){
1.208 brouard 4059: firstime=1;
1.203 brouard 4060: kmax=kmax+10;
1.208 brouard 4061: }
4062: if(kmax >=10 || firstime ==1){
1.246 brouard 4063: 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);
4064: 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 4065: 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);
4066: 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);
4067: }
4068: #ifdef DEBUGHESSIJ
4069: v1=hess[thetai][thetai];
4070: v2=hess[thetaj][thetaj];
4071: cv12=res;
4072: /* Computing eigen value of Hessian matrix */
4073: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4074: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4075: if ((lc2 <0) || (lc1 <0) ){
4076: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4077: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4078: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4079: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4080: }
1.126 brouard 4081: #endif
4082: }
4083: return res;
4084: }
4085:
1.203 brouard 4086: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4087: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4088: /* { */
4089: /* int i; */
4090: /* int l=1, lmax=20; */
4091: /* double k1,k2,k3,k4,res,fx; */
4092: /* double p2[MAXPARM+1]; */
4093: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4094: /* int k=0,kmax=10; */
4095: /* double l1; */
4096:
4097: /* fx=func(x); */
4098: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4099: /* l1=pow(10,l); */
4100: /* delts=delt; */
4101: /* for(k=1 ; k <kmax; k=k+1){ */
4102: /* delt = delti*(l1*k); */
4103: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4104: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4105: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4106: /* k1=func(p2)-fx; */
4107:
4108: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4109: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4110: /* k2=func(p2)-fx; */
4111:
4112: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4113: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4114: /* k3=func(p2)-fx; */
4115:
4116: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4117: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4118: /* k4=func(p2)-fx; */
4119: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4120: /* #ifdef DEBUGHESSIJ */
4121: /* 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); */
4122: /* 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); */
4123: /* #endif */
4124: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4125: /* k=kmax; */
4126: /* } */
4127: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4128: /* k=kmax; l=lmax*10; */
4129: /* } */
4130: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4131: /* delts=delt; */
4132: /* } */
4133: /* } /\* End loop k *\/ */
4134: /* } */
4135: /* delti[theta]=delts; */
4136: /* return res; */
4137: /* } */
4138:
4139:
1.126 brouard 4140: /************** Inverse of matrix **************/
4141: void ludcmp(double **a, int n, int *indx, double *d)
4142: {
4143: int i,imax,j,k;
4144: double big,dum,sum,temp;
4145: double *vv;
4146:
4147: vv=vector(1,n);
4148: *d=1.0;
4149: for (i=1;i<=n;i++) {
4150: big=0.0;
4151: for (j=1;j<=n;j++)
4152: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4153: if (big == 0.0){
4154: printf(" Singular Hessian matrix at row %d:\n",i);
4155: for (j=1;j<=n;j++) {
4156: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4157: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4158: }
4159: fflush(ficlog);
4160: fclose(ficlog);
4161: nrerror("Singular matrix in routine ludcmp");
4162: }
1.126 brouard 4163: vv[i]=1.0/big;
4164: }
4165: for (j=1;j<=n;j++) {
4166: for (i=1;i<j;i++) {
4167: sum=a[i][j];
4168: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4169: a[i][j]=sum;
4170: }
4171: big=0.0;
4172: for (i=j;i<=n;i++) {
4173: sum=a[i][j];
4174: for (k=1;k<j;k++)
4175: sum -= a[i][k]*a[k][j];
4176: a[i][j]=sum;
4177: if ( (dum=vv[i]*fabs(sum)) >= big) {
4178: big=dum;
4179: imax=i;
4180: }
4181: }
4182: if (j != imax) {
4183: for (k=1;k<=n;k++) {
4184: dum=a[imax][k];
4185: a[imax][k]=a[j][k];
4186: a[j][k]=dum;
4187: }
4188: *d = -(*d);
4189: vv[imax]=vv[j];
4190: }
4191: indx[j]=imax;
4192: if (a[j][j] == 0.0) a[j][j]=TINY;
4193: if (j != n) {
4194: dum=1.0/(a[j][j]);
4195: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4196: }
4197: }
4198: free_vector(vv,1,n); /* Doesn't work */
4199: ;
4200: }
4201:
4202: void lubksb(double **a, int n, int *indx, double b[])
4203: {
4204: int i,ii=0,ip,j;
4205: double sum;
4206:
4207: for (i=1;i<=n;i++) {
4208: ip=indx[i];
4209: sum=b[ip];
4210: b[ip]=b[i];
4211: if (ii)
4212: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4213: else if (sum) ii=i;
4214: b[i]=sum;
4215: }
4216: for (i=n;i>=1;i--) {
4217: sum=b[i];
4218: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4219: b[i]=sum/a[i][i];
4220: }
4221: }
4222:
4223: void pstamp(FILE *fichier)
4224: {
1.196 brouard 4225: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4226: }
4227:
1.253 brouard 4228: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4229:
4230: /* y=a+bx regression */
4231: double sumx = 0.0; /* sum of x */
4232: double sumx2 = 0.0; /* sum of x**2 */
4233: double sumxy = 0.0; /* sum of x * y */
4234: double sumy = 0.0; /* sum of y */
4235: double sumy2 = 0.0; /* sum of y**2 */
4236: double sume2; /* sum of square or residuals */
4237: double yhat;
4238:
4239: double denom=0;
4240: int i;
4241: int ne=*no;
4242:
4243: for ( i=ifi, ne=0;i<=ila;i++) {
4244: if(!isfinite(x[i]) || !isfinite(y[i])){
4245: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4246: continue;
4247: }
4248: ne=ne+1;
4249: sumx += x[i];
4250: sumx2 += x[i]*x[i];
4251: sumxy += x[i] * y[i];
4252: sumy += y[i];
4253: sumy2 += y[i]*y[i];
4254: denom = (ne * sumx2 - sumx*sumx);
4255: /* 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); */
4256: }
4257:
4258: denom = (ne * sumx2 - sumx*sumx);
4259: if (denom == 0) {
4260: // vertical, slope m is infinity
4261: *b = INFINITY;
4262: *a = 0;
4263: if (r) *r = 0;
4264: return 1;
4265: }
4266:
4267: *b = (ne * sumxy - sumx * sumy) / denom;
4268: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4269: if (r!=NULL) {
4270: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4271: sqrt((sumx2 - sumx*sumx/ne) *
4272: (sumy2 - sumy*sumy/ne));
4273: }
4274: *no=ne;
4275: for ( i=ifi, ne=0;i<=ila;i++) {
4276: if(!isfinite(x[i]) || !isfinite(y[i])){
4277: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4278: continue;
4279: }
4280: ne=ne+1;
4281: yhat = y[i] - *a -*b* x[i];
4282: sume2 += yhat * yhat ;
4283:
4284: denom = (ne * sumx2 - sumx*sumx);
4285: /* 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); */
4286: }
4287: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4288: *sa= *sb * sqrt(sumx2/ne);
4289:
4290: return 0;
4291: }
4292:
1.126 brouard 4293: /************ Frequencies ********************/
1.251 brouard 4294: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4295: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4296: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4297: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4298:
1.253 brouard 4299: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4300: int iind=0, iage=0;
4301: int mi; /* Effective wave */
4302: int first;
4303: double ***freq; /* Frequencies */
1.253 brouard 4304: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4305: int no;
1.226 brouard 4306: double *meanq;
4307: double **meanqt;
4308: double *pp, **prop, *posprop, *pospropt;
4309: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4310: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4311: double agebegin, ageend;
4312:
4313: pp=vector(1,nlstate);
1.251 brouard 4314: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4315: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4316: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4317: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4318: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4319: meanqt=matrix(1,lastpass,1,nqtveff);
4320: strcpy(fileresp,"P_");
4321: strcat(fileresp,fileresu);
4322: /*strcat(fileresphtm,fileresu);*/
4323: if((ficresp=fopen(fileresp,"w"))==NULL) {
4324: printf("Problem with prevalence resultfile: %s\n", fileresp);
4325: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4326: exit(0);
4327: }
1.240 brouard 4328:
1.226 brouard 4329: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4330: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4331: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4332: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4333: fflush(ficlog);
4334: exit(70);
4335: }
4336: else{
4337: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4338: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4339: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4340: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4341: }
1.237 brouard 4342: 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 4343:
1.226 brouard 4344: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4345: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4346: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4347: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4348: fflush(ficlog);
4349: exit(70);
1.240 brouard 4350: } else{
1.226 brouard 4351: 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 4352: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4353: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4354: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4355: }
1.240 brouard 4356: 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);
4357:
1.253 brouard 4358: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4359: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4360: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4361: j1=0;
1.126 brouard 4362:
1.227 brouard 4363: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4364: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4365: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4366:
4367:
1.226 brouard 4368: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4369: reference=low_education V1=0,V2=0
4370: med_educ V1=1 V2=0,
4371: high_educ V1=0 V2=1
4372: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4373: */
1.249 brouard 4374: dateintsum=0;
4375: k2cpt=0;
4376:
1.253 brouard 4377: if(cptcoveff == 0 )
4378: nl=1; /* Constant model only */
4379: else
4380: nl=2;
4381: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4382: if(nj==1)
4383: j=0; /* First pass for the constant */
4384: else
4385: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4386: first=1;
4387: 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 */
4388: posproptt=0.;
4389: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4390: scanf("%d", i);*/
4391: for (i=-5; i<=nlstate+ndeath; i++)
4392: for (jk=-5; jk<=nlstate+ndeath; jk++)
4393: for(m=iagemin; m <= iagemax+3; m++)
4394: freq[i][jk][m]=0;
4395:
4396: for (i=1; i<=nlstate; i++) {
1.240 brouard 4397: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4398: prop[i][m]=0;
4399: posprop[i]=0;
4400: pospropt[i]=0;
4401: }
4402: /* for (z1=1; z1<= nqfveff; z1++) { */
4403: /* meanq[z1]+=0.; */
4404: /* for(m=1;m<=lastpass;m++){ */
4405: /* meanqt[m][z1]=0.; */
4406: /* } */
4407: /* } */
4408:
4409: /* dateintsum=0; */
4410: /* k2cpt=0; */
4411:
4412: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4413: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4414: bool=1;
4415: if(j !=0){
4416: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4417: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4418: /* for (z1=1; z1<= nqfveff; z1++) { */
4419: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4420: /* } */
4421: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4422: /* if(Tvaraff[z1] ==-20){ */
4423: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4424: /* }else if(Tvaraff[z1] ==-10){ */
4425: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4426: /* }else */
4427: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4428: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4429: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4430: /* 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",
4431: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4432: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4433: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4434: } /* Onlyf fixed */
4435: } /* end z1 */
4436: } /* cptcovn > 0 */
4437: } /* end any */
4438: }/* end j==0 */
4439: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4440: /* for(m=firstpass; m<=lastpass; m++){ */
4441: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4442: m=mw[mi][iind];
4443: if(j!=0){
4444: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4445: for (z1=1; z1<=cptcoveff; z1++) {
4446: if( Fixed[Tmodelind[z1]]==1){
4447: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4448: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4449: value is -1, we don't select. It differs from the
4450: constant and age model which counts them. */
4451: bool=0; /* not selected */
4452: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4453: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4454: bool=0;
4455: }
4456: }
4457: }
4458: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4459: } /* end j==0 */
4460: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4461: if(bool==1){
4462: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4463: and mw[mi+1][iind]. dh depends on stepm. */
4464: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4465: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4466: if(m >=firstpass && m <=lastpass){
4467: k2=anint[m][iind]+(mint[m][iind]/12.);
4468: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4469: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4470: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4471: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4472: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4473: if (m<lastpass) {
4474: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4475: /* 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]); */
4476: if(s[m][iind]==-1)
4477: 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.));
4478: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4479: /* if((int)agev[m][iind] == 55) */
4480: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4481: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4482: 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 4483: }
1.251 brouard 4484: } /* end if between passes */
4485: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4486: dateintsum=dateintsum+k2; /* on all covariates ?*/
4487: k2cpt++;
4488: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4489: }
1.251 brouard 4490: }else{
4491: bool=1;
4492: }/* end bool 2 */
4493: } /* end m */
4494: } /* end bool */
4495: } /* end iind = 1 to imx */
4496: /* prop[s][age] is feeded for any initial and valid live state as well as
4497: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4498:
4499:
4500: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4501: pstamp(ficresp);
4502: if (cptcoveff>0 && j!=0){
4503: printf( "\n#********** Variable ");
4504: fprintf(ficresp, "\n#********** Variable ");
4505: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4506: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4507: fprintf(ficlog, "\n#********** Variable ");
4508: for (z1=1; z1<=cptcoveff; z1++){
4509: if(!FixedV[Tvaraff[z1]]){
4510: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4511: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4512: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4513: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4514: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4515: }else{
1.251 brouard 4516: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4517: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4518: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4519: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4520: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4521: }
4522: }
4523: printf( "**********\n#");
4524: fprintf(ficresp, "**********\n#");
4525: fprintf(ficresphtm, "**********</h3>\n");
4526: fprintf(ficresphtmfr, "**********</h3>\n");
4527: fprintf(ficlog, "**********\n");
4528: }
4529: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4530: for(i=1; i<=nlstate;i++) {
4531: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4532: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4533: }
4534: fprintf(ficresp, "\n");
4535: fprintf(ficresphtm, "\n");
4536:
4537: /* Header of frequency table by age */
4538: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4539: fprintf(ficresphtmfr,"<th>Age</th> ");
4540: for(jk=-1; jk <=nlstate+ndeath; jk++){
4541: for(m=-1; m <=nlstate+ndeath; m++){
4542: if(jk!=0 && m!=0)
4543: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4544: }
1.226 brouard 4545: }
1.251 brouard 4546: fprintf(ficresphtmfr, "\n");
4547:
4548: /* For each age */
4549: for(iage=iagemin; iage <= iagemax+3; iage++){
4550: fprintf(ficresphtm,"<tr>");
4551: if(iage==iagemax+1){
4552: fprintf(ficlog,"1");
4553: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4554: }else if(iage==iagemax+2){
4555: fprintf(ficlog,"0");
4556: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4557: }else if(iage==iagemax+3){
4558: fprintf(ficlog,"Total");
4559: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4560: }else{
1.240 brouard 4561: if(first==1){
1.251 brouard 4562: first=0;
4563: printf("See log file for details...\n");
4564: }
4565: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4566: fprintf(ficlog,"Age %d", iage);
4567: }
4568: for(jk=1; jk <=nlstate ; jk++){
4569: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4570: pp[jk] += freq[jk][m][iage];
4571: }
4572: for(jk=1; jk <=nlstate ; jk++){
4573: for(m=-1, pos=0; m <=0 ; m++)
4574: pos += freq[jk][m][iage];
4575: if(pp[jk]>=1.e-10){
4576: if(first==1){
4577: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4578: }
4579: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4580: }else{
4581: if(first==1)
4582: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4583: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4584: }
4585: }
4586:
1.251 brouard 4587: for(jk=1; jk <=nlstate ; jk++){
4588: /* posprop[jk]=0; */
4589: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4590: pp[jk] += freq[jk][m][iage];
4591: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4592:
4593: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4594: pos += pp[jk]; /* pos is the total number of transitions until this age */
4595: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4596: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4597: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4598: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4599: }
1.251 brouard 4600: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4601: if(pos>=1.e-5){
1.251 brouard 4602: if(first==1)
4603: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4604: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4605: }else{
4606: if(first==1)
4607: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4608: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4609: }
4610: if( iage <= iagemax){
4611: if(pos>=1.e-5){
4612: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4613: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4614: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4615: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4616: }
4617: else{
4618: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4619: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4620: }
1.240 brouard 4621: }
1.251 brouard 4622: pospropt[jk] +=posprop[jk];
4623: } /* end loop jk */
4624: /* pospropt=0.; */
4625: for(jk=-1; jk <=nlstate+ndeath; jk++){
4626: for(m=-1; m <=nlstate+ndeath; m++){
4627: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4628: if(first==1){
4629: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4630: }
1.253 brouard 4631: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4632: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4633: }
4634: if(jk!=0 && m!=0)
4635: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4636: }
1.251 brouard 4637: } /* end loop jk */
4638: posproptt=0.;
4639: for(jk=1; jk <=nlstate; jk++){
4640: posproptt += pospropt[jk];
4641: }
4642: fprintf(ficresphtmfr,"</tr>\n ");
4643: if(iage <= iagemax){
4644: fprintf(ficresp,"\n");
4645: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4646: }
1.251 brouard 4647: if(first==1)
4648: printf("Others in log...\n");
4649: fprintf(ficlog,"\n");
4650: } /* end loop age iage */
4651: fprintf(ficresphtm,"<tr><th>Tot</th>");
4652: for(jk=1; jk <=nlstate ; jk++){
4653: if(posproptt < 1.e-5){
4654: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4655: }else{
4656: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4657: }
1.226 brouard 4658: }
1.251 brouard 4659: fprintf(ficresphtm,"</tr>\n");
4660: fprintf(ficresphtm,"</table>\n");
4661: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4662: if(posproptt < 1.e-5){
1.251 brouard 4663: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4664: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4665: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4666: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4667: invalidvarcomb[j1]=1;
1.226 brouard 4668: }else{
1.251 brouard 4669: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4670: invalidvarcomb[j1]=0;
1.226 brouard 4671: }
1.251 brouard 4672: fprintf(ficresphtmfr,"</table>\n");
4673: fprintf(ficlog,"\n");
4674: if(j!=0){
4675: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4676: for(i=1,jk=1; i <=nlstate; i++){
4677: for(k=1; k <=(nlstate+ndeath); k++){
4678: if (k != i) {
4679: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4680: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4681: if(j1==1){ /* All dummy covariates to zero */
4682: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4683: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4684: printf("%d%d ",i,k);
4685: fprintf(ficlog,"%d%d ",i,k);
4686: 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]));
4687: 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]));
4688: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4689: }
1.253 brouard 4690: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4691: for(iage=iagemin; iage <= iagemax+3; iage++){
4692: x[iage]= (double)iage;
4693: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4694: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4695: }
4696: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4697: pstart[jk]=b;
4698: pstart[jk-1]=a;
1.252 brouard 4699: }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 */
4700: 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]);
4701: 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 4702: 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 4703: printf("%d%d ",i,k);
4704: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4705: 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]));
4706: }else{ /* Other cases, like quantitative fixed or varying covariates */
4707: ;
4708: }
4709: /* printf("%12.7f )", param[i][jj][k]); */
4710: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4711: jk++;
4712: } /* end jj */
4713: } /* end k!= i */
4714: } /* end k */
4715: } /* end i, jk */
4716: } /* end j !=0 */
4717: } /* end selected combination of covariate j1 */
4718: if(j==0){ /* We can estimate starting values from the occurences in each case */
4719: printf("#Freqsummary: Starting values for the constants:\n");
4720: fprintf(ficlog,"\n");
4721: for(i=1,jk=1; i <=nlstate; i++){
4722: for(k=1; k <=(nlstate+ndeath); k++){
4723: if (k != i) {
4724: printf("%d%d ",i,k);
4725: fprintf(ficlog,"%d%d ",i,k);
4726: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4727: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4728: if(jj==1){ /* Age has to be done */
4729: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4730: 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]));
4731: 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]));
4732: }
4733: /* printf("%12.7f )", param[i][jj][k]); */
4734: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4735: jk++;
1.250 brouard 4736: }
1.251 brouard 4737: printf("\n");
4738: fprintf(ficlog,"\n");
1.250 brouard 4739: }
4740: }
4741: }
1.251 brouard 4742: printf("#Freqsummary\n");
4743: fprintf(ficlog,"\n");
4744: for(jk=-1; jk <=nlstate+ndeath; jk++){
4745: for(m=-1; m <=nlstate+ndeath; m++){
4746: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4747: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4748: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4749: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4750: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4751: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4752: /* } */
4753: }
4754: } /* end loop jk */
4755:
4756: printf("\n");
4757: fprintf(ficlog,"\n");
4758: } /* end j=0 */
1.249 brouard 4759: } /* end j */
1.252 brouard 4760:
1.253 brouard 4761: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4762: for(i=1, jk=1; i <=nlstate; i++){
4763: for(j=1; j <=nlstate+ndeath; j++){
4764: if(j!=i){
4765: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4766: printf("%1d%1d",i,j);
4767: fprintf(ficparo,"%1d%1d",i,j);
4768: for(k=1; k<=ncovmodel;k++){
4769: /* printf(" %lf",param[i][j][k]); */
4770: /* fprintf(ficparo," %lf",param[i][j][k]); */
4771: p[jk]=pstart[jk];
4772: printf(" %f ",pstart[jk]);
4773: fprintf(ficparo," %f ",pstart[jk]);
4774: jk++;
4775: }
4776: printf("\n");
4777: fprintf(ficparo,"\n");
4778: }
4779: }
4780: }
4781: } /* end mle=-2 */
1.226 brouard 4782: dateintmean=dateintsum/k2cpt;
1.240 brouard 4783:
1.226 brouard 4784: fclose(ficresp);
4785: fclose(ficresphtm);
4786: fclose(ficresphtmfr);
4787: free_vector(meanq,1,nqfveff);
4788: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4789: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4790: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4791: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4792: free_vector(pospropt,1,nlstate);
4793: free_vector(posprop,1,nlstate);
1.251 brouard 4794: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4795: free_vector(pp,1,nlstate);
4796: /* End of freqsummary */
4797: }
1.126 brouard 4798:
4799: /************ Prevalence ********************/
1.227 brouard 4800: 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)
4801: {
4802: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4803: in each health status at the date of interview (if between dateprev1 and dateprev2).
4804: We still use firstpass and lastpass as another selection.
4805: */
1.126 brouard 4806:
1.227 brouard 4807: int i, m, jk, j1, bool, z1,j, iv;
4808: int mi; /* Effective wave */
4809: int iage;
4810: double agebegin, ageend;
4811:
4812: double **prop;
4813: double posprop;
4814: double y2; /* in fractional years */
4815: int iagemin, iagemax;
4816: int first; /** to stop verbosity which is redirected to log file */
4817:
4818: iagemin= (int) agemin;
4819: iagemax= (int) agemax;
4820: /*pp=vector(1,nlstate);*/
1.251 brouard 4821: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4822: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4823: j1=0;
1.222 brouard 4824:
1.227 brouard 4825: /*j=cptcoveff;*/
4826: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4827:
1.227 brouard 4828: first=1;
4829: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4830: for (i=1; i<=nlstate; i++)
1.251 brouard 4831: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4832: prop[i][iage]=0.0;
4833: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4834: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4835: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4836:
4837: for (i=1; i<=imx; i++) { /* Each individual */
4838: bool=1;
4839: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4840: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4841: m=mw[mi][i];
4842: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4843: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4844: for (z1=1; z1<=cptcoveff; z1++){
4845: if( Fixed[Tmodelind[z1]]==1){
4846: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4847: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4848: bool=0;
4849: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4850: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4851: bool=0;
4852: }
4853: }
4854: if(bool==1){ /* Otherwise we skip that wave/person */
4855: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4856: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4857: if(m >=firstpass && m <=lastpass){
4858: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4859: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4860: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4861: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4862: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4863: 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);
4864: exit(1);
4865: }
4866: if (s[m][i]>0 && s[m][i]<=nlstate) {
4867: /*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]]);*/
4868: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4869: prop[s[m][i]][iagemax+3] += weight[i];
4870: } /* end valid statuses */
4871: } /* end selection of dates */
4872: } /* end selection of waves */
4873: } /* end bool */
4874: } /* end wave */
4875: } /* end individual */
4876: for(i=iagemin; i <= iagemax+3; i++){
4877: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4878: posprop += prop[jk][i];
4879: }
4880:
4881: for(jk=1; jk <=nlstate ; jk++){
4882: if( i <= iagemax){
4883: if(posprop>=1.e-5){
4884: probs[i][jk][j1]= prop[jk][i]/posprop;
4885: } else{
4886: if(first==1){
4887: first=0;
4888: 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]);
4889: }
4890: }
4891: }
4892: }/* end jk */
4893: }/* end i */
1.222 brouard 4894: /*} *//* end i1 */
1.227 brouard 4895: } /* end j1 */
1.222 brouard 4896:
1.227 brouard 4897: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4898: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4899: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4900: } /* End of prevalence */
1.126 brouard 4901:
4902: /************* Waves Concatenation ***************/
4903:
4904: 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)
4905: {
4906: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4907: Death is a valid wave (if date is known).
4908: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4909: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4910: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4911: */
1.126 brouard 4912:
1.224 brouard 4913: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4914: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4915: double sum=0., jmean=0.;*/
1.224 brouard 4916: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4917: int j, k=0,jk, ju, jl;
4918: double sum=0.;
4919: first=0;
1.214 brouard 4920: firstwo=0;
1.217 brouard 4921: firsthree=0;
1.218 brouard 4922: firstfour=0;
1.164 brouard 4923: jmin=100000;
1.126 brouard 4924: jmax=-1;
4925: jmean=0.;
1.224 brouard 4926:
4927: /* Treating live states */
1.214 brouard 4928: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4929: mi=0; /* First valid wave */
1.227 brouard 4930: mli=0; /* Last valid wave */
1.126 brouard 4931: m=firstpass;
1.214 brouard 4932: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4933: 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 */
4934: mli=m-1;/* mw[++mi][i]=m-1; */
4935: }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 */
4936: mw[++mi][i]=m;
4937: mli=m;
1.224 brouard 4938: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4939: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4940: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4941: }
1.227 brouard 4942: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4943: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4944: break;
1.224 brouard 4945: #else
1.227 brouard 4946: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4947: if(firsthree == 0){
4948: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as pi. .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m);
4949: firsthree=1;
4950: }
4951: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as pi. .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m);
4952: mw[++mi][i]=m;
4953: mli=m;
4954: }
4955: if(s[m][i]==-2){ /* Vital status is really unknown */
4956: nbwarn++;
4957: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4958: 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);
4959: 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);
4960: }
4961: break;
4962: }
4963: break;
1.224 brouard 4964: #endif
1.227 brouard 4965: }/* End m >= lastpass */
1.126 brouard 4966: }/* end while */
1.224 brouard 4967:
1.227 brouard 4968: /* 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 4969: /* After last pass */
1.224 brouard 4970: /* Treating death states */
1.214 brouard 4971: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4972: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4973: /* } */
1.126 brouard 4974: mi++; /* Death is another wave */
4975: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4976: /* Only death is a correct wave */
1.126 brouard 4977: mw[mi][i]=m;
1.257 brouard 4978: } /* else not in a death state */
1.224 brouard 4979: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 4980: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 4981: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4982: 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 */
4983: nbwarn++;
4984: if(firstfiv==0){
4985: 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 );
4986: firstfiv=1;
4987: }else{
4988: 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 );
4989: }
4990: }else{ /* Death occured afer last wave potential bias */
4991: nberr++;
4992: if(firstwo==0){
1.257 brouard 4993: 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 4994: firstwo=1;
4995: }
1.257 brouard 4996: 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 4997: }
1.257 brouard 4998: }else{ /* if date of interview is unknown */
1.227 brouard 4999: /* death is known but not confirmed by death status at any wave */
5000: if(firstfour==0){
5001: 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 );
5002: firstfour=1;
5003: }
5004: 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 5005: }
1.224 brouard 5006: } /* end if date of death is known */
5007: #endif
5008: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5009: /* wav[i]=mw[mi][i]; */
1.126 brouard 5010: if(mi==0){
5011: nbwarn++;
5012: if(first==0){
1.227 brouard 5013: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5014: first=1;
1.126 brouard 5015: }
5016: if(first==1){
1.227 brouard 5017: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5018: }
5019: } /* end mi==0 */
5020: } /* End individuals */
1.214 brouard 5021: /* wav and mw are no more changed */
1.223 brouard 5022:
1.214 brouard 5023:
1.126 brouard 5024: for(i=1; i<=imx; i++){
5025: for(mi=1; mi<wav[i];mi++){
5026: if (stepm <=0)
1.227 brouard 5027: dh[mi][i]=1;
1.126 brouard 5028: else{
1.260 brouard 5029: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5030: if (agedc[i] < 2*AGESUP) {
5031: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5032: if(j==0) j=1; /* Survives at least one month after exam */
5033: else if(j<0){
5034: nberr++;
5035: 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]);
5036: j=1; /* Temporary Dangerous patch */
5037: 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);
5038: 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]);
5039: 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);
5040: }
5041: k=k+1;
5042: if (j >= jmax){
5043: jmax=j;
5044: ijmax=i;
5045: }
5046: if (j <= jmin){
5047: jmin=j;
5048: ijmin=i;
5049: }
5050: sum=sum+j;
5051: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5052: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5053: }
5054: }
5055: else{
5056: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5057: /* 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 5058:
1.227 brouard 5059: k=k+1;
5060: if (j >= jmax) {
5061: jmax=j;
5062: ijmax=i;
5063: }
5064: else if (j <= jmin){
5065: jmin=j;
5066: ijmin=i;
5067: }
5068: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5069: /*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]);*/
5070: if(j<0){
5071: nberr++;
5072: 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]);
5073: 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]);
5074: }
5075: sum=sum+j;
5076: }
5077: jk= j/stepm;
5078: jl= j -jk*stepm;
5079: ju= j -(jk+1)*stepm;
5080: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5081: if(jl==0){
5082: dh[mi][i]=jk;
5083: bh[mi][i]=0;
5084: }else{ /* We want a negative bias in order to only have interpolation ie
5085: * to avoid the price of an extra matrix product in likelihood */
5086: dh[mi][i]=jk+1;
5087: bh[mi][i]=ju;
5088: }
5089: }else{
5090: if(jl <= -ju){
5091: dh[mi][i]=jk;
5092: bh[mi][i]=jl; /* bias is positive if real duration
5093: * is higher than the multiple of stepm and negative otherwise.
5094: */
5095: }
5096: else{
5097: dh[mi][i]=jk+1;
5098: bh[mi][i]=ju;
5099: }
5100: if(dh[mi][i]==0){
5101: dh[mi][i]=1; /* At least one step */
5102: bh[mi][i]=ju; /* At least one step */
5103: /* 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);*/
5104: }
5105: } /* end if mle */
1.126 brouard 5106: }
5107: } /* end wave */
5108: }
5109: jmean=sum/k;
5110: 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 5111: 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 5112: }
1.126 brouard 5113:
5114: /*********** Tricode ****************************/
1.220 brouard 5115: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5116: {
5117: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5118: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5119: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5120: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5121: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5122: */
1.130 brouard 5123:
1.242 brouard 5124: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5125: int modmaxcovj=0; /* Modality max of covariates j */
5126: int cptcode=0; /* Modality max of covariates j */
5127: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5128:
5129:
1.242 brouard 5130: /* cptcoveff=0; */
5131: /* *cptcov=0; */
1.126 brouard 5132:
1.242 brouard 5133: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5134:
1.242 brouard 5135: /* Loop on covariates without age and products and no quantitative variable */
5136: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5137: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5138: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5139: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5140: switch(Fixed[k]) {
5141: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5142: 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*/
5143: ij=(int)(covar[Tvar[k]][i]);
5144: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5145: * If product of Vn*Vm, still boolean *:
5146: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5147: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5148: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5149: modality of the nth covariate of individual i. */
5150: if (ij > modmaxcovj)
5151: modmaxcovj=ij;
5152: else if (ij < modmincovj)
5153: modmincovj=ij;
5154: if ((ij < -1) && (ij > NCOVMAX)){
5155: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5156: exit(1);
5157: }else
5158: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5159: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5160: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5161: /* getting the maximum value of the modality of the covariate
5162: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5163: female ies 1, then modmaxcovj=1.
5164: */
5165: } /* end for loop on individuals i */
5166: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5167: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5168: cptcode=modmaxcovj;
5169: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5170: /*for (i=0; i<=cptcode; i++) {*/
5171: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5172: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5173: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5174: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5175: if( j != -1){
5176: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5177: covariate for which somebody answered excluding
5178: undefined. Usually 2: 0 and 1. */
5179: }
5180: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5181: covariate for which somebody answered including
5182: undefined. Usually 3: -1, 0 and 1. */
5183: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5184: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5185: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5186:
1.242 brouard 5187: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5188: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5189: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5190: /* modmincovj=3; modmaxcovj = 7; */
5191: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5192: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5193: /* defining two dummy variables: variables V1_1 and V1_2.*/
5194: /* nbcode[Tvar[j]][ij]=k; */
5195: /* nbcode[Tvar[j]][1]=0; */
5196: /* nbcode[Tvar[j]][2]=1; */
5197: /* nbcode[Tvar[j]][3]=2; */
5198: /* To be continued (not working yet). */
5199: ij=0; /* ij is similar to i but can jump over null modalities */
5200: 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*/
5201: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5202: break;
5203: }
5204: ij++;
5205: 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*/
5206: cptcode = ij; /* New max modality for covar j */
5207: } /* end of loop on modality i=-1 to 1 or more */
5208: break;
5209: case 1: /* Testing on varying covariate, could be simple and
5210: * should look at waves or product of fixed *
5211: * varying. No time to test -1, assuming 0 and 1 only */
5212: ij=0;
5213: for(i=0; i<=1;i++){
5214: nbcode[Tvar[k]][++ij]=i;
5215: }
5216: break;
5217: default:
5218: break;
5219: } /* end switch */
5220: } /* end dummy test */
5221:
5222: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5223: /* /\*recode from 0 *\/ */
5224: /* k is a modality. If we have model=V1+V1*sex */
5225: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5226: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5227: /* } */
5228: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5229: /* if (ij > ncodemax[j]) { */
5230: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5231: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5232: /* break; */
5233: /* } */
5234: /* } /\* end of loop on modality k *\/ */
5235: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5236:
5237: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5238: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5239: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5240: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5241: 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 */
5242: 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 */
5243: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5244: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5245:
5246: ij=0;
5247: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5248: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5249: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5250: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5251: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5252: /* If product not in single variable we don't print results */
5253: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5254: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5255: 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*/
5256: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5257: 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 */
5258: if(Fixed[k]!=0)
5259: anyvaryingduminmodel=1;
5260: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5261: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5262: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5263: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5264: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5265: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5266: }
5267: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5268: /* ij--; */
5269: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5270: *cptcov=ij; /*Number of total real effective covariates: effective
5271: * because they can be excluded from the model and real
5272: * if in the model but excluded because missing values, but how to get k from ij?*/
5273: for(j=ij+1; j<= cptcovt; j++){
5274: Tvaraff[j]=0;
5275: Tmodelind[j]=0;
5276: }
5277: for(j=ntveff+1; j<= cptcovt; j++){
5278: TmodelInvind[j]=0;
5279: }
5280: /* To be sorted */
5281: ;
5282: }
1.126 brouard 5283:
1.145 brouard 5284:
1.126 brouard 5285: /*********** Health Expectancies ****************/
5286:
1.235 brouard 5287: 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 5288:
5289: {
5290: /* Health expectancies, no variances */
1.164 brouard 5291: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5292: int nhstepma, nstepma; /* Decreasing with age */
5293: double age, agelim, hf;
5294: double ***p3mat;
5295: double eip;
5296:
1.238 brouard 5297: /* pstamp(ficreseij); */
1.126 brouard 5298: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5299: fprintf(ficreseij,"# Age");
5300: for(i=1; i<=nlstate;i++){
5301: for(j=1; j<=nlstate;j++){
5302: fprintf(ficreseij," e%1d%1d ",i,j);
5303: }
5304: fprintf(ficreseij," e%1d. ",i);
5305: }
5306: fprintf(ficreseij,"\n");
5307:
5308:
5309: if(estepm < stepm){
5310: printf ("Problem %d lower than %d\n",estepm, stepm);
5311: }
5312: else hstepm=estepm;
5313: /* We compute the life expectancy from trapezoids spaced every estepm months
5314: * This is mainly to measure the difference between two models: for example
5315: * if stepm=24 months pijx are given only every 2 years and by summing them
5316: * we are calculating an estimate of the Life Expectancy assuming a linear
5317: * progression in between and thus overestimating or underestimating according
5318: * to the curvature of the survival function. If, for the same date, we
5319: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5320: * to compare the new estimate of Life expectancy with the same linear
5321: * hypothesis. A more precise result, taking into account a more precise
5322: * curvature will be obtained if estepm is as small as stepm. */
5323:
5324: /* For example we decided to compute the life expectancy with the smallest unit */
5325: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5326: nhstepm is the number of hstepm from age to agelim
5327: nstepm is the number of stepm from age to agelin.
5328: Look at hpijx to understand the reason of that which relies in memory size
5329: and note for a fixed period like estepm months */
5330: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5331: survival function given by stepm (the optimization length). Unfortunately it
5332: means that if the survival funtion is printed only each two years of age and if
5333: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5334: results. So we changed our mind and took the option of the best precision.
5335: */
5336: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5337:
5338: agelim=AGESUP;
5339: /* If stepm=6 months */
5340: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5341: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5342:
5343: /* nhstepm age range expressed in number of stepm */
5344: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5345: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5346: /* if (stepm >= YEARM) hstepm=1;*/
5347: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5348: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5349:
5350: for (age=bage; age<=fage; age ++){
5351: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5352: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5353: /* if (stepm >= YEARM) hstepm=1;*/
5354: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5355:
5356: /* If stepm=6 months */
5357: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5358: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5359:
1.235 brouard 5360: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5361:
5362: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5363:
5364: printf("%d|",(int)age);fflush(stdout);
5365: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5366:
5367: /* Computing expectancies */
5368: for(i=1; i<=nlstate;i++)
5369: for(j=1; j<=nlstate;j++)
5370: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5371: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5372:
5373: /* 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]);*/
5374:
5375: }
5376:
5377: fprintf(ficreseij,"%3.0f",age );
5378: for(i=1; i<=nlstate;i++){
5379: eip=0;
5380: for(j=1; j<=nlstate;j++){
5381: eip +=eij[i][j][(int)age];
5382: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5383: }
5384: fprintf(ficreseij,"%9.4f", eip );
5385: }
5386: fprintf(ficreseij,"\n");
5387:
5388: }
5389: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5390: printf("\n");
5391: fprintf(ficlog,"\n");
5392:
5393: }
5394:
1.235 brouard 5395: 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 5396:
5397: {
5398: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5399: to initial status i, ei. .
1.126 brouard 5400: */
5401: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5402: int nhstepma, nstepma; /* Decreasing with age */
5403: double age, agelim, hf;
5404: double ***p3matp, ***p3matm, ***varhe;
5405: double **dnewm,**doldm;
5406: double *xp, *xm;
5407: double **gp, **gm;
5408: double ***gradg, ***trgradg;
5409: int theta;
5410:
5411: double eip, vip;
5412:
5413: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5414: xp=vector(1,npar);
5415: xm=vector(1,npar);
5416: dnewm=matrix(1,nlstate*nlstate,1,npar);
5417: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5418:
5419: pstamp(ficresstdeij);
5420: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5421: fprintf(ficresstdeij,"# Age");
5422: for(i=1; i<=nlstate;i++){
5423: for(j=1; j<=nlstate;j++)
5424: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5425: fprintf(ficresstdeij," e%1d. ",i);
5426: }
5427: fprintf(ficresstdeij,"\n");
5428:
5429: pstamp(ficrescveij);
5430: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5431: fprintf(ficrescveij,"# Age");
5432: for(i=1; i<=nlstate;i++)
5433: for(j=1; j<=nlstate;j++){
5434: cptj= (j-1)*nlstate+i;
5435: for(i2=1; i2<=nlstate;i2++)
5436: for(j2=1; j2<=nlstate;j2++){
5437: cptj2= (j2-1)*nlstate+i2;
5438: if(cptj2 <= cptj)
5439: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5440: }
5441: }
5442: fprintf(ficrescveij,"\n");
5443:
5444: if(estepm < stepm){
5445: printf ("Problem %d lower than %d\n",estepm, stepm);
5446: }
5447: else hstepm=estepm;
5448: /* We compute the life expectancy from trapezoids spaced every estepm months
5449: * This is mainly to measure the difference between two models: for example
5450: * if stepm=24 months pijx are given only every 2 years and by summing them
5451: * we are calculating an estimate of the Life Expectancy assuming a linear
5452: * progression in between and thus overestimating or underestimating according
5453: * to the curvature of the survival function. If, for the same date, we
5454: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5455: * to compare the new estimate of Life expectancy with the same linear
5456: * hypothesis. A more precise result, taking into account a more precise
5457: * curvature will be obtained if estepm is as small as stepm. */
5458:
5459: /* For example we decided to compute the life expectancy with the smallest unit */
5460: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5461: nhstepm is the number of hstepm from age to agelim
5462: nstepm is the number of stepm from age to agelin.
5463: Look at hpijx to understand the reason of that which relies in memory size
5464: and note for a fixed period like estepm months */
5465: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5466: survival function given by stepm (the optimization length). Unfortunately it
5467: means that if the survival funtion is printed only each two years of age and if
5468: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5469: results. So we changed our mind and took the option of the best precision.
5470: */
5471: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5472:
5473: /* If stepm=6 months */
5474: /* nhstepm age range expressed in number of stepm */
5475: agelim=AGESUP;
5476: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5477: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5478: /* if (stepm >= YEARM) hstepm=1;*/
5479: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5480:
5481: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5482: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5483: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5484: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5485: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5486: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5487:
5488: for (age=bage; age<=fage; age ++){
5489: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5490: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5491: /* if (stepm >= YEARM) hstepm=1;*/
5492: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5493:
1.126 brouard 5494: /* If stepm=6 months */
5495: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5496: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5497:
5498: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5499:
1.126 brouard 5500: /* Computing Variances of health expectancies */
5501: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5502: decrease memory allocation */
5503: for(theta=1; theta <=npar; theta++){
5504: for(i=1; i<=npar; i++){
1.222 brouard 5505: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5506: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5507: }
1.235 brouard 5508: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5509: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5510:
1.126 brouard 5511: for(j=1; j<= nlstate; j++){
1.222 brouard 5512: for(i=1; i<=nlstate; i++){
5513: for(h=0; h<=nhstepm-1; h++){
5514: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5515: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5516: }
5517: }
1.126 brouard 5518: }
1.218 brouard 5519:
1.126 brouard 5520: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5521: for(h=0; h<=nhstepm-1; h++){
5522: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5523: }
1.126 brouard 5524: }/* End theta */
5525:
5526:
5527: for(h=0; h<=nhstepm-1; h++)
5528: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5529: for(theta=1; theta <=npar; theta++)
5530: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5531:
1.218 brouard 5532:
1.222 brouard 5533: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5534: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5535: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5536:
1.222 brouard 5537: printf("%d|",(int)age);fflush(stdout);
5538: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5539: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5540: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5541: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5542: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5543: for(ij=1;ij<=nlstate*nlstate;ij++)
5544: for(ji=1;ji<=nlstate*nlstate;ji++)
5545: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5546: }
5547: }
1.218 brouard 5548:
1.126 brouard 5549: /* Computing expectancies */
1.235 brouard 5550: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5551: for(i=1; i<=nlstate;i++)
5552: for(j=1; j<=nlstate;j++)
1.222 brouard 5553: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5554: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5555:
1.222 brouard 5556: /* 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 5557:
1.222 brouard 5558: }
1.218 brouard 5559:
1.126 brouard 5560: fprintf(ficresstdeij,"%3.0f",age );
5561: for(i=1; i<=nlstate;i++){
5562: eip=0.;
5563: vip=0.;
5564: for(j=1; j<=nlstate;j++){
1.222 brouard 5565: eip += eij[i][j][(int)age];
5566: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5567: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5568: 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 5569: }
5570: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5571: }
5572: fprintf(ficresstdeij,"\n");
1.218 brouard 5573:
1.126 brouard 5574: fprintf(ficrescveij,"%3.0f",age );
5575: for(i=1; i<=nlstate;i++)
5576: for(j=1; j<=nlstate;j++){
1.222 brouard 5577: cptj= (j-1)*nlstate+i;
5578: for(i2=1; i2<=nlstate;i2++)
5579: for(j2=1; j2<=nlstate;j2++){
5580: cptj2= (j2-1)*nlstate+i2;
5581: if(cptj2 <= cptj)
5582: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5583: }
1.126 brouard 5584: }
5585: fprintf(ficrescveij,"\n");
1.218 brouard 5586:
1.126 brouard 5587: }
5588: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5589: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5590: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5591: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5592: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5593: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5594: printf("\n");
5595: fprintf(ficlog,"\n");
1.218 brouard 5596:
1.126 brouard 5597: free_vector(xm,1,npar);
5598: free_vector(xp,1,npar);
5599: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5600: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5601: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5602: }
1.218 brouard 5603:
1.126 brouard 5604: /************ Variance ******************/
1.235 brouard 5605: 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 5606: {
5607: /* Variance of health expectancies */
5608: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5609: /* double **newm;*/
5610: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5611:
5612: /* int movingaverage(); */
5613: double **dnewm,**doldm;
5614: double **dnewmp,**doldmp;
5615: int i, j, nhstepm, hstepm, h, nstepm ;
5616: int k;
5617: double *xp;
5618: double **gp, **gm; /* for var eij */
5619: double ***gradg, ***trgradg; /*for var eij */
5620: double **gradgp, **trgradgp; /* for var p point j */
5621: double *gpp, *gmp; /* for var p point j */
5622: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5623: double ***p3mat;
5624: double age,agelim, hf;
5625: /* double ***mobaverage; */
5626: int theta;
5627: char digit[4];
5628: char digitp[25];
5629:
5630: char fileresprobmorprev[FILENAMELENGTH];
5631:
5632: if(popbased==1){
5633: if(mobilav!=0)
5634: strcpy(digitp,"-POPULBASED-MOBILAV_");
5635: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5636: }
5637: else
5638: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5639:
1.218 brouard 5640: /* if (mobilav!=0) { */
5641: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5642: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5643: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5644: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5645: /* } */
5646: /* } */
5647:
5648: strcpy(fileresprobmorprev,"PRMORPREV-");
5649: sprintf(digit,"%-d",ij);
5650: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5651: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5652: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5653: strcat(fileresprobmorprev,fileresu);
5654: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5655: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5656: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5657: }
5658: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5659: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5660: pstamp(ficresprobmorprev);
5661: 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 5662: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5663: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5664: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5665: }
5666: for(j=1;j<=cptcoveff;j++)
5667: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5668: fprintf(ficresprobmorprev,"\n");
5669:
1.218 brouard 5670: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5671: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5672: fprintf(ficresprobmorprev," p.%-d SE",j);
5673: for(i=1; i<=nlstate;i++)
5674: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5675: }
5676: fprintf(ficresprobmorprev,"\n");
5677:
5678: fprintf(ficgp,"\n# Routine varevsij");
5679: fprintf(ficgp,"\nunset title \n");
5680: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5681: 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");
5682: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5683: /* } */
5684: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5685: pstamp(ficresvij);
5686: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5687: if(popbased==1)
5688: 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);
5689: else
5690: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5691: fprintf(ficresvij,"# Age");
5692: for(i=1; i<=nlstate;i++)
5693: for(j=1; j<=nlstate;j++)
5694: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5695: fprintf(ficresvij,"\n");
5696:
5697: xp=vector(1,npar);
5698: dnewm=matrix(1,nlstate,1,npar);
5699: doldm=matrix(1,nlstate,1,nlstate);
5700: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5701: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5702:
5703: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5704: gpp=vector(nlstate+1,nlstate+ndeath);
5705: gmp=vector(nlstate+1,nlstate+ndeath);
5706: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5707:
1.218 brouard 5708: if(estepm < stepm){
5709: printf ("Problem %d lower than %d\n",estepm, stepm);
5710: }
5711: else hstepm=estepm;
5712: /* For example we decided to compute the life expectancy with the smallest unit */
5713: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5714: nhstepm is the number of hstepm from age to agelim
5715: nstepm is the number of stepm from age to agelim.
5716: Look at function hpijx to understand why because of memory size limitations,
5717: we decided (b) to get a life expectancy respecting the most precise curvature of the
5718: survival function given by stepm (the optimization length). Unfortunately it
5719: means that if the survival funtion is printed every two years of age and if
5720: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5721: results. So we changed our mind and took the option of the best precision.
5722: */
5723: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5724: agelim = AGESUP;
5725: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5726: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5727: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5728: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5729: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5730: gp=matrix(0,nhstepm,1,nlstate);
5731: gm=matrix(0,nhstepm,1,nlstate);
5732:
5733:
5734: for(theta=1; theta <=npar; theta++){
5735: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5736: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5737: }
5738:
1.242 brouard 5739: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5740:
5741: if (popbased==1) {
5742: if(mobilav ==0){
5743: for(i=1; i<=nlstate;i++)
5744: prlim[i][i]=probs[(int)age][i][ij];
5745: }else{ /* mobilav */
5746: for(i=1; i<=nlstate;i++)
5747: prlim[i][i]=mobaverage[(int)age][i][ij];
5748: }
5749: }
5750:
1.235 brouard 5751: 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 5752: for(j=1; j<= nlstate; j++){
5753: for(h=0; h<=nhstepm; h++){
5754: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5755: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5756: }
5757: }
5758: /* Next for computing probability of death (h=1 means
5759: computed over hstepm matrices product = hstepm*stepm months)
5760: as a weighted average of prlim.
5761: */
5762: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5763: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5764: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5765: }
5766: /* end probability of death */
5767:
5768: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5769: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5770:
1.242 brouard 5771: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5772:
5773: if (popbased==1) {
5774: if(mobilav ==0){
5775: for(i=1; i<=nlstate;i++)
5776: prlim[i][i]=probs[(int)age][i][ij];
5777: }else{ /* mobilav */
5778: for(i=1; i<=nlstate;i++)
5779: prlim[i][i]=mobaverage[(int)age][i][ij];
5780: }
5781: }
5782:
1.235 brouard 5783: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5784:
5785: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5786: for(h=0; h<=nhstepm; h++){
5787: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5788: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5789: }
5790: }
5791: /* This for computing probability of death (h=1 means
5792: computed over hstepm matrices product = hstepm*stepm months)
5793: as a weighted average of prlim.
5794: */
5795: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5796: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5797: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5798: }
5799: /* end probability of death */
5800:
5801: for(j=1; j<= nlstate; j++) /* vareij */
5802: for(h=0; h<=nhstepm; h++){
5803: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5804: }
5805:
5806: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5807: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5808: }
5809:
5810: } /* End theta */
5811:
5812: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5813:
5814: for(h=0; h<=nhstepm; h++) /* veij */
5815: for(j=1; j<=nlstate;j++)
5816: for(theta=1; theta <=npar; theta++)
5817: trgradg[h][j][theta]=gradg[h][theta][j];
5818:
5819: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5820: for(theta=1; theta <=npar; theta++)
5821: trgradgp[j][theta]=gradgp[theta][j];
5822:
5823:
5824: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5825: for(i=1;i<=nlstate;i++)
5826: for(j=1;j<=nlstate;j++)
5827: vareij[i][j][(int)age] =0.;
5828:
5829: for(h=0;h<=nhstepm;h++){
5830: for(k=0;k<=nhstepm;k++){
5831: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5832: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5833: for(i=1;i<=nlstate;i++)
5834: for(j=1;j<=nlstate;j++)
5835: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5836: }
5837: }
5838:
5839: /* pptj */
5840: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5841: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5842: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5843: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5844: varppt[j][i]=doldmp[j][i];
5845: /* end ppptj */
5846: /* x centered again */
5847:
1.242 brouard 5848: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5849:
5850: if (popbased==1) {
5851: if(mobilav ==0){
5852: for(i=1; i<=nlstate;i++)
5853: prlim[i][i]=probs[(int)age][i][ij];
5854: }else{ /* mobilav */
5855: for(i=1; i<=nlstate;i++)
5856: prlim[i][i]=mobaverage[(int)age][i][ij];
5857: }
5858: }
5859:
5860: /* This for computing probability of death (h=1 means
5861: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5862: as a weighted average of prlim.
5863: */
1.235 brouard 5864: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5865: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5866: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5867: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5868: }
5869: /* end probability of death */
5870:
5871: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5872: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5873: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5874: for(i=1; i<=nlstate;i++){
5875: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5876: }
5877: }
5878: fprintf(ficresprobmorprev,"\n");
5879:
5880: fprintf(ficresvij,"%.0f ",age );
5881: for(i=1; i<=nlstate;i++)
5882: for(j=1; j<=nlstate;j++){
5883: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5884: }
5885: fprintf(ficresvij,"\n");
5886: free_matrix(gp,0,nhstepm,1,nlstate);
5887: free_matrix(gm,0,nhstepm,1,nlstate);
5888: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5889: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5890: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5891: } /* End age */
5892: free_vector(gpp,nlstate+1,nlstate+ndeath);
5893: free_vector(gmp,nlstate+1,nlstate+ndeath);
5894: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5895: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5896: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5897: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5898: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5899: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5900: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5901: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5902: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5903: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5904: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5905: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5906: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5907: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5908: 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);
5909: /* 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 5910: */
1.218 brouard 5911: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5912: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5913:
1.218 brouard 5914: free_vector(xp,1,npar);
5915: free_matrix(doldm,1,nlstate,1,nlstate);
5916: free_matrix(dnewm,1,nlstate,1,npar);
5917: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5918: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5919: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5920: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5921: fclose(ficresprobmorprev);
5922: fflush(ficgp);
5923: fflush(fichtm);
5924: } /* end varevsij */
1.126 brouard 5925:
5926: /************ Variance of prevlim ******************/
1.235 brouard 5927: 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 5928: {
1.205 brouard 5929: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5930: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5931:
1.126 brouard 5932: double **dnewm,**doldm;
5933: int i, j, nhstepm, hstepm;
5934: double *xp;
5935: double *gp, *gm;
5936: double **gradg, **trgradg;
1.208 brouard 5937: double **mgm, **mgp;
1.126 brouard 5938: double age,agelim;
5939: int theta;
5940:
5941: pstamp(ficresvpl);
5942: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5943: fprintf(ficresvpl,"# Age ");
5944: if(nresult >=1)
5945: fprintf(ficresvpl," Result# ");
1.126 brouard 5946: for(i=1; i<=nlstate;i++)
5947: fprintf(ficresvpl," %1d-%1d",i,i);
5948: fprintf(ficresvpl,"\n");
5949:
5950: xp=vector(1,npar);
5951: dnewm=matrix(1,nlstate,1,npar);
5952: doldm=matrix(1,nlstate,1,nlstate);
5953:
5954: hstepm=1*YEARM; /* Every year of age */
5955: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5956: agelim = AGESUP;
5957: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5958: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5959: if (stepm >= YEARM) hstepm=1;
5960: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5961: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5962: mgp=matrix(1,npar,1,nlstate);
5963: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5964: gp=vector(1,nlstate);
5965: gm=vector(1,nlstate);
5966:
5967: for(theta=1; theta <=npar; theta++){
5968: for(i=1; i<=npar; i++){ /* Computes gradient */
5969: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5970: }
1.209 brouard 5971: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5972: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5973: else
1.235 brouard 5974: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5975: for(i=1;i<=nlstate;i++){
1.126 brouard 5976: gp[i] = prlim[i][i];
1.208 brouard 5977: mgp[theta][i] = prlim[i][i];
5978: }
1.126 brouard 5979: for(i=1; i<=npar; i++) /* Computes gradient */
5980: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5981: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5982: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5983: else
1.235 brouard 5984: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5985: for(i=1;i<=nlstate;i++){
1.126 brouard 5986: gm[i] = prlim[i][i];
1.208 brouard 5987: mgm[theta][i] = prlim[i][i];
5988: }
1.126 brouard 5989: for(i=1;i<=nlstate;i++)
5990: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5991: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5992: } /* End theta */
5993:
5994: trgradg =matrix(1,nlstate,1,npar);
5995:
5996: for(j=1; j<=nlstate;j++)
5997: for(theta=1; theta <=npar; theta++)
5998: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5999: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6000: /* printf("\nmgm mgp %d ",(int)age); */
6001: /* for(j=1; j<=nlstate;j++){ */
6002: /* printf(" %d ",j); */
6003: /* for(theta=1; theta <=npar; theta++) */
6004: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6005: /* printf("\n "); */
6006: /* } */
6007: /* } */
6008: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6009: /* printf("\n gradg %d ",(int)age); */
6010: /* for(j=1; j<=nlstate;j++){ */
6011: /* printf("%d ",j); */
6012: /* for(theta=1; theta <=npar; theta++) */
6013: /* printf("%d %lf ",theta,gradg[theta][j]); */
6014: /* printf("\n "); */
6015: /* } */
6016: /* } */
1.126 brouard 6017:
6018: for(i=1;i<=nlstate;i++)
6019: varpl[i][(int)age] =0.;
1.209 brouard 6020: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6021: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6022: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6023: }else{
1.126 brouard 6024: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6025: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6026: }
1.126 brouard 6027: for(i=1;i<=nlstate;i++)
6028: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6029:
6030: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6031: if(nresult >=1)
6032: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6033: for(i=1; i<=nlstate;i++)
6034: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6035: fprintf(ficresvpl,"\n");
6036: free_vector(gp,1,nlstate);
6037: free_vector(gm,1,nlstate);
1.208 brouard 6038: free_matrix(mgm,1,npar,1,nlstate);
6039: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6040: free_matrix(gradg,1,npar,1,nlstate);
6041: free_matrix(trgradg,1,nlstate,1,npar);
6042: } /* End age */
6043:
6044: free_vector(xp,1,npar);
6045: free_matrix(doldm,1,nlstate,1,npar);
6046: free_matrix(dnewm,1,nlstate,1,nlstate);
6047:
6048: }
6049:
6050: /************ Variance of one-step probabilities ******************/
6051: 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 6052: {
6053: int i, j=0, k1, l1, tj;
6054: int k2, l2, j1, z1;
6055: int k=0, l;
6056: int first=1, first1, first2;
6057: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6058: double **dnewm,**doldm;
6059: double *xp;
6060: double *gp, *gm;
6061: double **gradg, **trgradg;
6062: double **mu;
6063: double age, cov[NCOVMAX+1];
6064: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6065: int theta;
6066: char fileresprob[FILENAMELENGTH];
6067: char fileresprobcov[FILENAMELENGTH];
6068: char fileresprobcor[FILENAMELENGTH];
6069: double ***varpij;
6070:
6071: strcpy(fileresprob,"PROB_");
6072: strcat(fileresprob,fileres);
6073: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6074: printf("Problem with resultfile: %s\n", fileresprob);
6075: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6076: }
6077: strcpy(fileresprobcov,"PROBCOV_");
6078: strcat(fileresprobcov,fileresu);
6079: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6080: printf("Problem with resultfile: %s\n", fileresprobcov);
6081: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6082: }
6083: strcpy(fileresprobcor,"PROBCOR_");
6084: strcat(fileresprobcor,fileresu);
6085: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6086: printf("Problem with resultfile: %s\n", fileresprobcor);
6087: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6088: }
6089: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6090: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6091: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6092: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6093: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6094: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6095: pstamp(ficresprob);
6096: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6097: fprintf(ficresprob,"# Age");
6098: pstamp(ficresprobcov);
6099: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6100: fprintf(ficresprobcov,"# Age");
6101: pstamp(ficresprobcor);
6102: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6103: fprintf(ficresprobcor,"# Age");
1.126 brouard 6104:
6105:
1.222 brouard 6106: for(i=1; i<=nlstate;i++)
6107: for(j=1; j<=(nlstate+ndeath);j++){
6108: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6109: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6110: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6111: }
6112: /* fprintf(ficresprob,"\n");
6113: fprintf(ficresprobcov,"\n");
6114: fprintf(ficresprobcor,"\n");
6115: */
6116: xp=vector(1,npar);
6117: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6118: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6119: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6120: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6121: first=1;
6122: fprintf(ficgp,"\n# Routine varprob");
6123: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6124: fprintf(fichtm,"\n");
6125:
6126: 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);
6127: 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);
6128: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6129: and drawn. It helps understanding how is the covariance between two incidences.\
6130: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6131: 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 6132: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6133: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6134: standard deviations wide on each axis. <br>\
6135: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6136: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6137: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6138:
1.222 brouard 6139: cov[1]=1;
6140: /* tj=cptcoveff; */
1.225 brouard 6141: tj = (int) pow(2,cptcoveff);
1.222 brouard 6142: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6143: j1=0;
1.224 brouard 6144: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6145: if (cptcovn>0) {
6146: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6147: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6148: fprintf(ficresprob, "**********\n#\n");
6149: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6150: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6151: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6152:
1.222 brouard 6153: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6154: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6155: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6156:
6157:
1.222 brouard 6158: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6159: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6160: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6161:
1.222 brouard 6162: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6163: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6164: fprintf(ficresprobcor, "**********\n#");
6165: if(invalidvarcomb[j1]){
6166: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6167: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6168: continue;
6169: }
6170: }
6171: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6172: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6173: gp=vector(1,(nlstate)*(nlstate+ndeath));
6174: gm=vector(1,(nlstate)*(nlstate+ndeath));
6175: for (age=bage; age<=fage; age ++){
6176: cov[2]=age;
6177: if(nagesqr==1)
6178: cov[3]= age*age;
6179: for (k=1; k<=cptcovn;k++) {
6180: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6181: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6182: * 1 1 1 1 1
6183: * 2 2 1 1 1
6184: * 3 1 2 1 1
6185: */
6186: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6187: }
6188: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6189: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6190: for (k=1; k<=cptcovprod;k++)
6191: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6192:
6193:
1.222 brouard 6194: for(theta=1; theta <=npar; theta++){
6195: for(i=1; i<=npar; i++)
6196: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6197:
1.222 brouard 6198: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6199:
1.222 brouard 6200: k=0;
6201: for(i=1; i<= (nlstate); i++){
6202: for(j=1; j<=(nlstate+ndeath);j++){
6203: k=k+1;
6204: gp[k]=pmmij[i][j];
6205: }
6206: }
1.220 brouard 6207:
1.222 brouard 6208: for(i=1; i<=npar; i++)
6209: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6210:
1.222 brouard 6211: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6212: k=0;
6213: for(i=1; i<=(nlstate); i++){
6214: for(j=1; j<=(nlstate+ndeath);j++){
6215: k=k+1;
6216: gm[k]=pmmij[i][j];
6217: }
6218: }
1.220 brouard 6219:
1.222 brouard 6220: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6221: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6222: }
1.126 brouard 6223:
1.222 brouard 6224: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6225: for(theta=1; theta <=npar; theta++)
6226: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6227:
1.222 brouard 6228: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6229: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6230:
1.222 brouard 6231: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6232:
1.222 brouard 6233: k=0;
6234: for(i=1; i<=(nlstate); i++){
6235: for(j=1; j<=(nlstate+ndeath);j++){
6236: k=k+1;
6237: mu[k][(int) age]=pmmij[i][j];
6238: }
6239: }
6240: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6241: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6242: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6243:
1.222 brouard 6244: /*printf("\n%d ",(int)age);
6245: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6246: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6247: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6248: }*/
1.220 brouard 6249:
1.222 brouard 6250: fprintf(ficresprob,"\n%d ",(int)age);
6251: fprintf(ficresprobcov,"\n%d ",(int)age);
6252: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6253:
1.222 brouard 6254: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6255: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6256: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6257: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6258: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6259: }
6260: i=0;
6261: for (k=1; k<=(nlstate);k++){
6262: for (l=1; l<=(nlstate+ndeath);l++){
6263: i++;
6264: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6265: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6266: for (j=1; j<=i;j++){
6267: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6268: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6269: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6270: }
6271: }
6272: }/* end of loop for state */
6273: } /* end of loop for age */
6274: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6275: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6276: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6277: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6278:
6279: /* Confidence intervalle of pij */
6280: /*
6281: fprintf(ficgp,"\nunset parametric;unset label");
6282: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6283: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6284: 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);
6285: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6286: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6287: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6288: */
6289:
6290: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6291: first1=1;first2=2;
6292: for (k2=1; k2<=(nlstate);k2++){
6293: for (l2=1; l2<=(nlstate+ndeath);l2++){
6294: if(l2==k2) continue;
6295: j=(k2-1)*(nlstate+ndeath)+l2;
6296: for (k1=1; k1<=(nlstate);k1++){
6297: for (l1=1; l1<=(nlstate+ndeath);l1++){
6298: if(l1==k1) continue;
6299: i=(k1-1)*(nlstate+ndeath)+l1;
6300: if(i<=j) continue;
6301: for (age=bage; age<=fage; age ++){
6302: if ((int)age %5==0){
6303: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6304: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6305: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6306: mu1=mu[i][(int) age]/stepm*YEARM ;
6307: mu2=mu[j][(int) age]/stepm*YEARM;
6308: c12=cv12/sqrt(v1*v2);
6309: /* Computing eigen value of matrix of covariance */
6310: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6311: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6312: if ((lc2 <0) || (lc1 <0) ){
6313: if(first2==1){
6314: first1=0;
6315: 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);
6316: }
6317: 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);
6318: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6319: /* lc2=fabs(lc2); */
6320: }
1.220 brouard 6321:
1.222 brouard 6322: /* Eigen vectors */
6323: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6324: /*v21=sqrt(1.-v11*v11); *//* error */
6325: v21=(lc1-v1)/cv12*v11;
6326: v12=-v21;
6327: v22=v11;
6328: tnalp=v21/v11;
6329: if(first1==1){
6330: first1=0;
6331: 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);
6332: }
6333: 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);
6334: /*printf(fignu*/
6335: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6336: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6337: if(first==1){
6338: first=0;
6339: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6340: fprintf(ficgp,"\nset parametric;unset label");
6341: 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);
6342: fprintf(ficgp,"\nset ter svg size 640, 480");
6343: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6344: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6345: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6346: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6347: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6348: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6349: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6350: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6351: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6352: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6353: 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", \
6354: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6355: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6356: }else{
6357: first=0;
6358: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6359: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6360: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6361: 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", \
6362: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6363: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6364: }/* if first */
6365: } /* age mod 5 */
6366: } /* end loop age */
6367: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6368: first=1;
6369: } /*l12 */
6370: } /* k12 */
6371: } /*l1 */
6372: }/* k1 */
6373: } /* loop on combination of covariates j1 */
6374: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6375: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6376: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6377: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6378: free_vector(xp,1,npar);
6379: fclose(ficresprob);
6380: fclose(ficresprobcov);
6381: fclose(ficresprobcor);
6382: fflush(ficgp);
6383: fflush(fichtmcov);
6384: }
1.126 brouard 6385:
6386:
6387: /******************* Printing html file ***********/
1.201 brouard 6388: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6389: int lastpass, int stepm, int weightopt, char model[],\
6390: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6391: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6392: double jprev1, double mprev1,double anprev1, double dateprev1, \
6393: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6394: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6395:
6396: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6397: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6398: </ul>");
1.237 brouard 6399: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6400: </ul>", model);
1.214 brouard 6401: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6402: 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",
6403: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6404: 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 6405: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6406: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6407: fprintf(fichtm,"\
6408: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6409: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6410: fprintf(fichtm,"\
1.217 brouard 6411: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6412: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6413: fprintf(fichtm,"\
1.126 brouard 6414: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6415: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6416: fprintf(fichtm,"\
1.217 brouard 6417: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6418: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6419: fprintf(fichtm,"\
1.211 brouard 6420: - (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 6421: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6422: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6423: if(prevfcast==1){
6424: fprintf(fichtm,"\
6425: - Prevalence projections by age and states: \
1.201 brouard 6426: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6427: }
1.126 brouard 6428:
1.222 brouard 6429: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6430:
1.225 brouard 6431: m=pow(2,cptcoveff);
1.222 brouard 6432: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6433:
1.222 brouard 6434: jj1=0;
1.237 brouard 6435:
6436: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6437: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6438: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6439: continue;
1.220 brouard 6440:
1.222 brouard 6441: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6442: jj1++;
6443: if (cptcovn > 0) {
6444: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6445: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6446: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6447: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6448: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6449: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6450: }
1.237 brouard 6451: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6452: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6453: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6454: }
6455:
1.230 brouard 6456: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6457: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6458: if(invalidvarcomb[k1]){
6459: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6460: printf("\nCombination (%d) ignored because no cases \n",k1);
6461: continue;
6462: }
6463: }
6464: /* aij, bij */
1.259 brouard 6465: 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 6466: <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 6467: /* Pij */
1.241 brouard 6468: 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> \
6469: <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 6470: /* Quasi-incidences */
6471: 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 6472: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6473: 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 6474: 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> \
6475: <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 6476: /* Survival functions (period) in state j */
6477: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6478: 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> \
6479: <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 6480: }
6481: /* State specific survival functions (period) */
6482: for(cpt=1; cpt<=nlstate;cpt++){
6483: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6484: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6485: <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 6486: }
6487: /* Period (stable) prevalence in each health state */
6488: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6489: 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 6490: <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 6491: }
6492: if(backcast==1){
6493: /* Period (stable) back prevalence in each health state */
6494: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6495: 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 6496: <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 6497: }
1.217 brouard 6498: }
1.222 brouard 6499: if(prevfcast==1){
6500: /* Projection of prevalence up to period (stable) prevalence in each health state */
6501: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6502: 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> \
6503: <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 6504: }
6505: }
1.220 brouard 6506:
1.222 brouard 6507: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6508: 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> \
6509: <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 6510: }
6511: /* } /\* end i1 *\/ */
6512: }/* End k1 */
6513: fprintf(fichtm,"</ul>");
1.126 brouard 6514:
1.222 brouard 6515: fprintf(fichtm,"\
1.126 brouard 6516: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6517: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6518: - 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 6519: But because parameters are usually highly correlated (a higher incidence of disability \
6520: and a higher incidence of recovery can give very close observed transition) it might \
6521: be very useful to look not only at linear confidence intervals estimated from the \
6522: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6523: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6524: covariance matrix of the one-step probabilities. \
6525: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6526:
1.222 brouard 6527: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6528: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6529: fprintf(fichtm,"\
1.126 brouard 6530: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6531: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6532:
1.222 brouard 6533: fprintf(fichtm,"\
1.126 brouard 6534: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6535: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6536: fprintf(fichtm,"\
1.126 brouard 6537: - 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): \
6538: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6539: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6540: fprintf(fichtm,"\
1.126 brouard 6541: - (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): \
6542: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6543: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6544: fprintf(fichtm,"\
1.128 brouard 6545: - 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 6546: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6547: fprintf(fichtm,"\
1.128 brouard 6548: - 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 6549: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6550: fprintf(fichtm,"\
1.126 brouard 6551: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6552: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6553:
6554: /* if(popforecast==1) fprintf(fichtm,"\n */
6555: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6556: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6557: /* <br>",fileres,fileres,fileres,fileres); */
6558: /* else */
6559: /* 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 6560: fflush(fichtm);
6561: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6562:
1.225 brouard 6563: m=pow(2,cptcoveff);
1.222 brouard 6564: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6565:
1.222 brouard 6566: jj1=0;
1.237 brouard 6567:
1.241 brouard 6568: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6569: for(k1=1; k1<=m;k1++){
1.253 brouard 6570: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6571: continue;
1.222 brouard 6572: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6573: jj1++;
1.126 brouard 6574: if (cptcovn > 0) {
6575: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6576: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6577: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6578: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6579: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6580: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6581: }
6582:
1.126 brouard 6583: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6584:
1.222 brouard 6585: if(invalidvarcomb[k1]){
6586: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6587: continue;
6588: }
1.126 brouard 6589: }
6590: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6591: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6592: 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 6593: <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 6594: }
6595: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6596: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6597: true period expectancies (those weighted with period prevalences are also\
6598: drawn in addition to the population based expectancies computed using\
1.241 brouard 6599: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6600: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6601: /* } /\* end i1 *\/ */
6602: }/* End k1 */
1.241 brouard 6603: }/* End nres */
1.222 brouard 6604: fprintf(fichtm,"</ul>");
6605: fflush(fichtm);
1.126 brouard 6606: }
6607:
6608: /******************* Gnuplot file **************/
1.223 brouard 6609: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6610:
6611: char dirfileres[132],optfileres[132];
1.223 brouard 6612: char gplotcondition[132];
1.237 brouard 6613: 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 6614: int lv=0, vlv=0, kl=0;
1.130 brouard 6615: int ng=0;
1.201 brouard 6616: int vpopbased;
1.223 brouard 6617: int ioffset; /* variable offset for columns */
1.235 brouard 6618: int nres=0; /* Index of resultline */
1.219 brouard 6619:
1.126 brouard 6620: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6621: /* printf("Problem with file %s",optionfilegnuplot); */
6622: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6623: /* } */
6624:
6625: /*#ifdef windows */
6626: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6627: /*#endif */
1.225 brouard 6628: m=pow(2,cptcoveff);
1.126 brouard 6629:
1.202 brouard 6630: /* Contribution to likelihood */
6631: /* Plot the probability implied in the likelihood */
1.223 brouard 6632: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6633: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6634: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6635: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6636: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6637: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6638: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6639: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6640: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6641: 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));
6642: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6643: 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));
6644: for (i=1; i<= nlstate ; i ++) {
6645: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6646: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6647: 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);
6648: for (j=2; j<= nlstate+ndeath ; j ++) {
6649: 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);
6650: }
6651: fprintf(ficgp,";\nset out; unset ylabel;\n");
6652: }
6653: /* 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 */
6654: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6655: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6656: fprintf(ficgp,"\nset out;unset log\n");
6657: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6658:
1.126 brouard 6659: strcpy(dirfileres,optionfilefiname);
6660: strcpy(optfileres,"vpl");
1.223 brouard 6661: /* 1eme*/
1.238 brouard 6662: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6663: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6664: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6665: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6666: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6667: continue;
6668: /* We are interested in selected combination by the resultline */
1.246 brouard 6669: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6670: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6671: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6672: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6673: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6674: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6675: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6676: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6677: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6678: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6679: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6680: }
6681: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6682: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6683: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6684: }
1.246 brouard 6685: /* printf("\n#\n"); */
1.238 brouard 6686: fprintf(ficgp,"\n#\n");
6687: if(invalidvarcomb[k1]){
1.260 brouard 6688: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 6689: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6690: continue;
6691: }
1.235 brouard 6692:
1.241 brouard 6693: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6694: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.260 brouard 6695: 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);
6696: /* 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); */
6697: /* k1-1 error should be nres-1*/
1.238 brouard 6698: for (i=1; i<= nlstate ; i ++) {
6699: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6700: else fprintf(ficgp," %%*lf (%%*lf)");
6701: }
1.260 brouard 6702: 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 6703: for (i=1; i<= nlstate ; i ++) {
6704: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6705: else fprintf(ficgp," %%*lf (%%*lf)");
6706: }
1.260 brouard 6707: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6708: for (i=1; i<= nlstate ; i ++) {
6709: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6710: else fprintf(ficgp," %%*lf (%%*lf)");
6711: }
6712: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6713: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6714: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6715: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6716: if(cptcoveff ==0){
1.245 brouard 6717: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6718: }else{
6719: kl=0;
6720: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6721: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6722: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6723: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6724: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6725: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6726: kl++;
1.238 brouard 6727: /* 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 *\/ */
6728: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6729: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6730: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6731: if(k==cptcoveff){
1.245 brouard 6732: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6733: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6734: }else{
6735: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6736: kl++;
6737: }
6738: } /* end covariate */
6739: } /* end if no covariate */
6740: } /* end if backcast */
6741: fprintf(ficgp,"\nset out \n");
6742: } /* nres */
1.201 brouard 6743: } /* k1 */
6744: } /* cpt */
1.235 brouard 6745:
6746:
1.126 brouard 6747: /*2 eme*/
1.238 brouard 6748: for (k1=1; k1<= m ; k1 ++){
6749: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6750: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6751: continue;
6752: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6753: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6754: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6755: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6756: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6757: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6758: vlv= nbcode[Tvaraff[k]][lv];
6759: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6760: }
1.237 brouard 6761: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6762: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6763: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6764: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6765: }
1.211 brouard 6766: fprintf(ficgp,"\n#\n");
1.223 brouard 6767: if(invalidvarcomb[k1]){
6768: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6769: continue;
6770: }
1.219 brouard 6771:
1.241 brouard 6772: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6773: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6774: if(vpopbased==0)
6775: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6776: else
6777: fprintf(ficgp,"\nreplot ");
6778: for (i=1; i<= nlstate+1 ; i ++) {
6779: k=2*i;
1.261 ! brouard 6780: 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 6781: for (j=1; j<= nlstate+1 ; j ++) {
6782: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6783: else fprintf(ficgp," %%*lf (%%*lf)");
6784: }
6785: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6786: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 ! brouard 6787: 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 6788: for (j=1; j<= nlstate+1 ; j ++) {
6789: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6790: else fprintf(ficgp," %%*lf (%%*lf)");
6791: }
6792: fprintf(ficgp,"\" t\"\" w l lt 0,");
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: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6799: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6800: } /* state */
6801: } /* vpopbased */
1.244 brouard 6802: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6803: } /* end nres */
6804: } /* k1 end 2 eme*/
6805:
6806:
6807: /*3eme*/
6808: for (k1=1; k1<= m ; k1 ++){
6809: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6810: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6811: continue;
6812:
6813: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 ! brouard 6814: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.238 brouard 6815: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6816: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6817: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6818: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6819: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6820: vlv= nbcode[Tvaraff[k]][lv];
6821: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6822: }
6823: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6824: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6825: }
6826: fprintf(ficgp,"\n#\n");
6827: if(invalidvarcomb[k1]){
6828: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6829: continue;
6830: }
6831:
6832: /* k=2+nlstate*(2*cpt-2); */
6833: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6834: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6835: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 ! brouard 6836: 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 6837: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6838: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6839: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6840: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6841: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6842: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6843:
1.238 brouard 6844: */
6845: for (i=1; i< nlstate ; i ++) {
1.261 ! brouard 6846: 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 6847: /* 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 6848:
1.238 brouard 6849: }
1.261 ! brouard 6850: 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 6851: }
6852: } /* end nres */
6853: } /* end kl 3eme */
1.126 brouard 6854:
1.223 brouard 6855: /* 4eme */
1.201 brouard 6856: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6857: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6858: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6859: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6860: continue;
1.238 brouard 6861: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6862: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6863: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6864: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6865: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6866: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6867: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6868: vlv= nbcode[Tvaraff[k]][lv];
6869: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6870: }
6871: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6872: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6873: }
6874: fprintf(ficgp,"\n#\n");
6875: if(invalidvarcomb[k1]){
6876: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6877: continue;
1.223 brouard 6878: }
1.238 brouard 6879:
1.241 brouard 6880: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6881: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6882: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6883: k=3;
6884: for (i=1; i<= nlstate ; i ++){
6885: if(i==1){
6886: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6887: }else{
6888: fprintf(ficgp,", '' ");
6889: }
6890: l=(nlstate+ndeath)*(i-1)+1;
6891: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6892: for (j=2; j<= nlstate+ndeath ; j ++)
6893: fprintf(ficgp,"+$%d",k+l+j-1);
6894: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6895: } /* nlstate */
6896: fprintf(ficgp,"\nset out\n");
6897: } /* end cpt state*/
6898: } /* end nres */
6899: } /* end covariate k1 */
6900:
1.220 brouard 6901: /* 5eme */
1.201 brouard 6902: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6903: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6904: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6905: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6906: continue;
1.238 brouard 6907: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6908: 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);
6909: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6910: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6911: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6912: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6913: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6914: vlv= nbcode[Tvaraff[k]][lv];
6915: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6916: }
6917: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6918: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6919: }
6920: fprintf(ficgp,"\n#\n");
6921: if(invalidvarcomb[k1]){
6922: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6923: continue;
6924: }
1.227 brouard 6925:
1.241 brouard 6926: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6927: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6928: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6929: k=3;
6930: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6931: if(j==1)
6932: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6933: else
6934: fprintf(ficgp,", '' ");
6935: l=(nlstate+ndeath)*(cpt-1) +j;
6936: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6937: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6938: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6939: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6940: } /* nlstate */
6941: fprintf(ficgp,", '' ");
6942: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6943: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6944: l=(nlstate+ndeath)*(cpt-1) +j;
6945: if(j < nlstate)
6946: fprintf(ficgp,"$%d +",k+l);
6947: else
6948: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6949: }
6950: fprintf(ficgp,"\nset out\n");
6951: } /* end cpt state*/
6952: } /* end covariate */
6953: } /* end nres */
1.227 brouard 6954:
1.220 brouard 6955: /* 6eme */
1.202 brouard 6956: /* CV preval stable (period) for each covariate */
1.237 brouard 6957: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6958: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6959: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6960: continue;
1.255 brouard 6961: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6962:
1.211 brouard 6963: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6964: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6965: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6966: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6967: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6968: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6969: vlv= nbcode[Tvaraff[k]][lv];
6970: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6971: }
1.237 brouard 6972: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6973: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6974: }
1.211 brouard 6975: fprintf(ficgp,"\n#\n");
1.223 brouard 6976: if(invalidvarcomb[k1]){
1.227 brouard 6977: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6978: continue;
1.223 brouard 6979: }
1.227 brouard 6980:
1.241 brouard 6981: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6982: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6983: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6984: k=3; /* Offset */
1.255 brouard 6985: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6986: if(i==1)
6987: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6988: else
6989: fprintf(ficgp,", '' ");
1.255 brouard 6990: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6991: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6992: for (j=2; j<= nlstate ; j ++)
6993: fprintf(ficgp,"+$%d",k+l+j-1);
6994: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6995: } /* nlstate */
1.201 brouard 6996: fprintf(ficgp,"\nset out\n");
1.153 brouard 6997: } /* end cpt state*/
6998: } /* end covariate */
1.227 brouard 6999:
7000:
1.220 brouard 7001: /* 7eme */
1.218 brouard 7002: if(backcast == 1){
1.217 brouard 7003: /* CV back preval stable (period) for each covariate */
1.237 brouard 7004: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7005: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7006: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7007: continue;
1.255 brouard 7008: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
7009: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7010: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7011: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7012: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7013: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7014: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7015: vlv= nbcode[Tvaraff[k]][lv];
7016: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7017: }
1.237 brouard 7018: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7019: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7020: }
1.227 brouard 7021: fprintf(ficgp,"\n#\n");
7022: if(invalidvarcomb[k1]){
7023: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7024: continue;
7025: }
7026:
1.241 brouard 7027: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7028: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7029: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7030: k=3; /* Offset */
1.255 brouard 7031: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7032: if(i==1)
7033: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7034: else
7035: fprintf(ficgp,", '' ");
7036: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7037: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7038: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7039: /* 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 7040: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7041: /* for (j=2; j<= nlstate ; j ++) */
7042: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7043: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7044: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7045: } /* nlstate */
7046: fprintf(ficgp,"\nset out\n");
1.218 brouard 7047: } /* end cpt state*/
7048: } /* end covariate */
7049: } /* End if backcast */
7050:
1.223 brouard 7051: /* 8eme */
1.218 brouard 7052: if(prevfcast==1){
7053: /* Projection from cross-sectional to stable (period) for each covariate */
7054:
1.237 brouard 7055: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7056: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7057: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7058: continue;
1.211 brouard 7059: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7060: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7061: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7062: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7063: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7064: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7065: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7066: vlv= nbcode[Tvaraff[k]][lv];
7067: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7068: }
1.237 brouard 7069: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7070: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7071: }
1.227 brouard 7072: fprintf(ficgp,"\n#\n");
7073: if(invalidvarcomb[k1]){
7074: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7075: continue;
7076: }
7077:
7078: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7079: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7080: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7081: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7082: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7083: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7084: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7085: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7086: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7087: if(i==1){
7088: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7089: }else{
7090: fprintf(ficgp,",\\\n '' ");
7091: }
7092: if(cptcoveff ==0){ /* No covariate */
7093: ioffset=2; /* Age is in 2 */
7094: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7095: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7096: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7097: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7098: fprintf(ficgp," u %d:(", ioffset);
7099: if(i==nlstate+1)
7100: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7101: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7102: else
7103: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7104: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7105: }else{ /* more than 2 covariates */
7106: if(cptcoveff ==1){
7107: ioffset=4; /* Age is in 4 */
7108: }else{
7109: ioffset=6; /* Age is in 6 */
7110: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7111: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7112: }
7113: fprintf(ficgp," u %d:(",ioffset);
7114: kl=0;
7115: strcpy(gplotcondition,"(");
7116: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7117: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7118: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7119: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7120: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7121: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7122: kl++;
7123: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7124: kl++;
7125: if(k <cptcoveff && cptcoveff>1)
7126: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7127: }
7128: strcpy(gplotcondition+strlen(gplotcondition),")");
7129: /* 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 *\/ */
7130: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7131: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7132: /* '' 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*/
7133: if(i==nlstate+1){
7134: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7135: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7136: }else{
7137: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7138: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7139: }
7140: } /* end if covariate */
7141: } /* nlstate */
7142: fprintf(ficgp,"\nset out\n");
1.223 brouard 7143: } /* end cpt state*/
7144: } /* end covariate */
7145: } /* End if prevfcast */
1.227 brouard 7146:
7147:
1.238 brouard 7148: /* 9eme writing MLE parameters */
7149: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7150: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7151: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7152: for(k=1; k <=(nlstate+ndeath); k++){
7153: if (k != i) {
1.227 brouard 7154: fprintf(ficgp,"# current state %d\n",k);
7155: for(j=1; j <=ncovmodel; j++){
7156: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7157: jk++;
7158: }
7159: fprintf(ficgp,"\n");
1.126 brouard 7160: }
7161: }
1.223 brouard 7162: }
1.187 brouard 7163: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7164:
1.145 brouard 7165: /*goto avoid;*/
1.238 brouard 7166: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7167: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7168: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7169: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7170: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7171: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7172: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7173: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7174: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7175: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7176: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7177: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7178: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7179: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7180: fprintf(ficgp,"#\n");
1.223 brouard 7181: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7182: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7183: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7184: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7185: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7186: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7187: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7188: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7189: continue;
7190: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7191: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7192: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7193: }
7194: fprintf(ficgp,"\n#\n");
1.241 brouard 7195: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7196: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7197: if (ng==1){
7198: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7199: fprintf(ficgp,"\nunset log y");
7200: }else if (ng==2){
7201: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7202: fprintf(ficgp,"\nset log y");
7203: }else if (ng==3){
7204: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7205: fprintf(ficgp,"\nset log y");
7206: }else
7207: fprintf(ficgp,"\nunset title ");
7208: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7209: i=1;
7210: for(k2=1; k2<=nlstate; k2++) {
7211: k3=i;
7212: for(k=1; k<=(nlstate+ndeath); k++) {
7213: if (k != k2){
7214: switch( ng) {
7215: case 1:
7216: if(nagesqr==0)
7217: fprintf(ficgp," p%d+p%d*x",i,i+1);
7218: else /* nagesqr =1 */
7219: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7220: break;
7221: case 2: /* ng=2 */
7222: if(nagesqr==0)
7223: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7224: else /* nagesqr =1 */
7225: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7226: break;
7227: case 3:
7228: if(nagesqr==0)
7229: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7230: else /* nagesqr =1 */
7231: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7232: break;
7233: }
7234: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7235: ijp=1; /* product no age */
7236: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7237: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7238: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7239: if(j==Tage[ij]) { /* Product by age */
7240: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7241: if(DummyV[j]==0){
1.237 brouard 7242: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7243: }else{ /* quantitative */
7244: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7245: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7246: }
7247: ij++;
7248: }
7249: }else if(j==Tprod[ijp]) { /* */
7250: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7251: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7252: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7253: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7254: /* 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)]); */
7255: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7256: }else{ /* Vn is dummy and Vm is quanti */
7257: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7258: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7259: }
7260: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7261: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7262: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7263: }else{ /* Both quanti */
7264: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7265: }
7266: }
1.238 brouard 7267: ijp++;
1.237 brouard 7268: }
7269: } else{ /* simple covariate */
7270: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7271: if(Dummy[j]==0){
7272: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7273: }else{ /* quantitative */
7274: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7275: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7276: }
1.237 brouard 7277: } /* end simple */
7278: } /* end j */
1.223 brouard 7279: }else{
7280: i=i-ncovmodel;
7281: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7282: fprintf(ficgp," (1.");
7283: }
1.227 brouard 7284:
1.223 brouard 7285: if(ng != 1){
7286: fprintf(ficgp,")/(1");
1.227 brouard 7287:
1.223 brouard 7288: for(k1=1; k1 <=nlstate; k1++){
7289: if(nagesqr==0)
7290: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7291: else /* nagesqr =1 */
7292: 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 7293:
1.223 brouard 7294: ij=1;
7295: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7296: if((j-2)==Tage[ij]) { /* Bug valgrind */
7297: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7298: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7299: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7300: ij++;
7301: }
7302: }
7303: else
1.225 brouard 7304: 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 7305: }
7306: fprintf(ficgp,")");
7307: }
7308: fprintf(ficgp,")");
7309: if(ng ==2)
7310: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7311: else /* ng= 3 */
7312: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7313: }else{ /* end ng <> 1 */
7314: if( k !=k2) /* logit p11 is hard to draw */
7315: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7316: }
7317: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7318: fprintf(ficgp,",");
7319: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7320: fprintf(ficgp,",");
7321: i=i+ncovmodel;
7322: } /* end k */
7323: } /* end k2 */
7324: fprintf(ficgp,"\n set out\n");
7325: } /* end jk */
7326: } /* end ng */
7327: /* avoid: */
7328: fflush(ficgp);
1.126 brouard 7329: } /* end gnuplot */
7330:
7331:
7332: /*************** Moving average **************/
1.219 brouard 7333: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7334: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7335:
1.222 brouard 7336: int i, cpt, cptcod;
7337: int modcovmax =1;
7338: int mobilavrange, mob;
7339: int iage=0;
7340:
7341: double sum=0.;
7342: double age;
7343: double *sumnewp, *sumnewm;
7344: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7345:
7346:
1.225 brouard 7347: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7348: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7349:
7350: sumnewp = vector(1,ncovcombmax);
7351: sumnewm = vector(1,ncovcombmax);
7352: agemingood = vector(1,ncovcombmax);
7353: agemaxgood = vector(1,ncovcombmax);
7354:
7355: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7356: sumnewm[cptcod]=0.;
7357: sumnewp[cptcod]=0.;
7358: agemingood[cptcod]=0;
7359: agemaxgood[cptcod]=0;
7360: }
7361: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7362:
7363: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7364: if(mobilav==1) mobilavrange=5; /* default */
7365: else mobilavrange=mobilav;
7366: for (age=bage; age<=fage; age++)
7367: for (i=1; i<=nlstate;i++)
7368: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7369: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7370: /* We keep the original values on the extreme ages bage, fage and for
7371: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7372: we use a 5 terms etc. until the borders are no more concerned.
7373: */
7374: for (mob=3;mob <=mobilavrange;mob=mob+2){
7375: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7376: for (i=1; i<=nlstate;i++){
7377: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7378: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7379: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7380: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7381: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7382: }
7383: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7384: }
7385: }
7386: }/* end age */
7387: }/* end mob */
7388: }else
7389: return -1;
7390: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7391: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7392: if(invalidvarcomb[cptcod]){
7393: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7394: continue;
7395: }
1.219 brouard 7396:
1.222 brouard 7397: agemingood[cptcod]=fage-(mob-1)/2;
7398: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7399: sumnewm[cptcod]=0.;
7400: for (i=1; i<=nlstate;i++){
7401: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7402: }
7403: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7404: agemingood[cptcod]=age;
7405: }else{ /* bad */
7406: for (i=1; i<=nlstate;i++){
7407: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7408: } /* i */
7409: } /* end bad */
7410: }/* age */
7411: sum=0.;
7412: for (i=1; i<=nlstate;i++){
7413: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7414: }
7415: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7416: 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);
7417: /* for (i=1; i<=nlstate;i++){ */
7418: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7419: /* } /\* i *\/ */
7420: } /* end bad */
7421: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7422: /* From youngest, finding the oldest wrong */
7423: agemaxgood[cptcod]=bage+(mob-1)/2;
7424: for (age=bage+(mob-1)/2; age<=fage; age++){
7425: sumnewm[cptcod]=0.;
7426: for (i=1; i<=nlstate;i++){
7427: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7428: }
7429: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7430: agemaxgood[cptcod]=age;
7431: }else{ /* bad */
7432: for (i=1; i<=nlstate;i++){
7433: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7434: } /* i */
7435: } /* end bad */
7436: }/* age */
7437: sum=0.;
7438: for (i=1; i<=nlstate;i++){
7439: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7440: }
7441: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7442: 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);
7443: /* for (i=1; i<=nlstate;i++){ */
7444: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7445: /* } /\* i *\/ */
7446: } /* end bad */
7447:
7448: for (age=bage; age<=fage; age++){
1.235 brouard 7449: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7450: sumnewp[cptcod]=0.;
7451: sumnewm[cptcod]=0.;
7452: for (i=1; i<=nlstate;i++){
7453: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7454: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7455: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7456: }
7457: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7458: }
7459: /* printf("\n"); */
7460: /* } */
7461: /* brutal averaging */
7462: for (i=1; i<=nlstate;i++){
7463: for (age=1; age<=bage; age++){
7464: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7465: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7466: }
7467: for (age=fage; age<=AGESUP; age++){
7468: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7469: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7470: }
7471: } /* end i status */
7472: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7473: for (age=1; age<=AGESUP; age++){
7474: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7475: mobaverage[(int)age][i][cptcod]=0.;
7476: }
7477: }
7478: }/* end cptcod */
7479: free_vector(sumnewm,1, ncovcombmax);
7480: free_vector(sumnewp,1, ncovcombmax);
7481: free_vector(agemaxgood,1, ncovcombmax);
7482: free_vector(agemingood,1, ncovcombmax);
7483: return 0;
7484: }/* End movingaverage */
1.218 brouard 7485:
1.126 brouard 7486:
7487: /************** Forecasting ******************/
1.235 brouard 7488: 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 7489: /* proj1, year, month, day of starting projection
7490: agemin, agemax range of age
7491: dateprev1 dateprev2 range of dates during which prevalence is computed
7492: anproj2 year of en of projection (same day and month as proj1).
7493: */
1.235 brouard 7494: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7495: double agec; /* generic age */
7496: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7497: double *popeffectif,*popcount;
7498: double ***p3mat;
1.218 brouard 7499: /* double ***mobaverage; */
1.126 brouard 7500: char fileresf[FILENAMELENGTH];
7501:
7502: agelim=AGESUP;
1.211 brouard 7503: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7504: in each health status at the date of interview (if between dateprev1 and dateprev2).
7505: We still use firstpass and lastpass as another selection.
7506: */
1.214 brouard 7507: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7508: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7509:
1.201 brouard 7510: strcpy(fileresf,"F_");
7511: strcat(fileresf,fileresu);
1.126 brouard 7512: if((ficresf=fopen(fileresf,"w"))==NULL) {
7513: printf("Problem with forecast resultfile: %s\n", fileresf);
7514: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7515: }
1.235 brouard 7516: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7517: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7518:
1.225 brouard 7519: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7520:
7521:
7522: stepsize=(int) (stepm+YEARM-1)/YEARM;
7523: if (stepm<=12) stepsize=1;
7524: if(estepm < stepm){
7525: printf ("Problem %d lower than %d\n",estepm, stepm);
7526: }
7527: else hstepm=estepm;
7528:
7529: hstepm=hstepm/stepm;
7530: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7531: fractional in yp1 */
7532: anprojmean=yp;
7533: yp2=modf((yp1*12),&yp);
7534: mprojmean=yp;
7535: yp1=modf((yp2*30.5),&yp);
7536: jprojmean=yp;
7537: if(jprojmean==0) jprojmean=1;
7538: if(mprojmean==0) jprojmean=1;
7539:
1.227 brouard 7540: i1=pow(2,cptcoveff);
1.126 brouard 7541: if (cptcovn < 1){i1=1;}
7542:
7543: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7544:
7545: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7546:
1.126 brouard 7547: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7548: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7549: for(k=1; k<=i1;k++){
1.253 brouard 7550: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7551: continue;
1.227 brouard 7552: if(invalidvarcomb[k]){
7553: printf("\nCombination (%d) projection ignored because no cases \n",k);
7554: continue;
7555: }
7556: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7557: for(j=1;j<=cptcoveff;j++) {
7558: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7559: }
1.235 brouard 7560: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7561: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7562: }
1.227 brouard 7563: fprintf(ficresf," yearproj age");
7564: for(j=1; j<=nlstate+ndeath;j++){
7565: for(i=1; i<=nlstate;i++)
7566: fprintf(ficresf," p%d%d",i,j);
7567: fprintf(ficresf," wp.%d",j);
7568: }
7569: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7570: fprintf(ficresf,"\n");
7571: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7572: for (agec=fage; agec>=(ageminpar-1); agec--){
7573: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7574: nhstepm = nhstepm/hstepm;
7575: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7576: oldm=oldms;savm=savms;
1.235 brouard 7577: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7578:
7579: for (h=0; h<=nhstepm; h++){
7580: if (h*hstepm/YEARM*stepm ==yearp) {
7581: fprintf(ficresf,"\n");
7582: for(j=1;j<=cptcoveff;j++)
7583: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7584: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7585: }
7586: for(j=1; j<=nlstate+ndeath;j++) {
7587: ppij=0.;
7588: for(i=1; i<=nlstate;i++) {
7589: if (mobilav==1)
7590: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7591: else {
7592: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7593: }
7594: if (h*hstepm/YEARM*stepm== yearp) {
7595: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7596: }
7597: } /* end i */
7598: if (h*hstepm/YEARM*stepm==yearp) {
7599: fprintf(ficresf," %.3f", ppij);
7600: }
7601: }/* end j */
7602: } /* end h */
7603: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7604: } /* end agec */
7605: } /* end yearp */
7606: } /* end k */
1.219 brouard 7607:
1.126 brouard 7608: fclose(ficresf);
1.215 brouard 7609: printf("End of Computing forecasting \n");
7610: fprintf(ficlog,"End of Computing forecasting\n");
7611:
1.126 brouard 7612: }
7613:
1.218 brouard 7614: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7615: /* 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 7616: /* /\* back1, year, month, day of starting backection */
7617: /* agemin, agemax range of age */
7618: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7619: /* anback2 year of en of backection (same day and month as back1). */
7620: /* *\/ */
7621: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7622: /* double agec; /\* generic age *\/ */
7623: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7624: /* double *popeffectif,*popcount; */
7625: /* double ***p3mat; */
7626: /* /\* double ***mobaverage; *\/ */
7627: /* char fileresfb[FILENAMELENGTH]; */
7628:
7629: /* agelim=AGESUP; */
7630: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7631: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7632: /* We still use firstpass and lastpass as another selection. */
7633: /* *\/ */
7634: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7635: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7636: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7637:
7638: /* strcpy(fileresfb,"FB_"); */
7639: /* strcat(fileresfb,fileresu); */
7640: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7641: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7642: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7643: /* } */
7644: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7645: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7646:
1.225 brouard 7647: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7648:
7649: /* /\* if (mobilav!=0) { *\/ */
7650: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7651: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7652: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7653: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7654: /* /\* } *\/ */
7655: /* /\* } *\/ */
7656:
7657: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7658: /* if (stepm<=12) stepsize=1; */
7659: /* if(estepm < stepm){ */
7660: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7661: /* } */
7662: /* else hstepm=estepm; */
7663:
7664: /* hstepm=hstepm/stepm; */
7665: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7666: /* fractional in yp1 *\/ */
7667: /* anprojmean=yp; */
7668: /* yp2=modf((yp1*12),&yp); */
7669: /* mprojmean=yp; */
7670: /* yp1=modf((yp2*30.5),&yp); */
7671: /* jprojmean=yp; */
7672: /* if(jprojmean==0) jprojmean=1; */
7673: /* if(mprojmean==0) jprojmean=1; */
7674:
1.225 brouard 7675: /* i1=cptcoveff; */
1.218 brouard 7676: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7677:
1.218 brouard 7678: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7679:
1.218 brouard 7680: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7681:
7682: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7683: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7684: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7685: /* k=k+1; */
7686: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7687: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7688: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7689: /* } */
7690: /* fprintf(ficresfb," yearbproj age"); */
7691: /* for(j=1; j<=nlstate+ndeath;j++){ */
7692: /* for(i=1; i<=nlstate;i++) */
7693: /* fprintf(ficresfb," p%d%d",i,j); */
7694: /* fprintf(ficresfb," p.%d",j); */
7695: /* } */
7696: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7697: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7698: /* fprintf(ficresfb,"\n"); */
7699: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7700: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7701: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7702: /* nhstepm = nhstepm/hstepm; */
7703: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7704: /* oldm=oldms;savm=savms; */
7705: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7706: /* for (h=0; h<=nhstepm; h++){ */
7707: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7708: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7709: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7710: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7711: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7712: /* } */
7713: /* for(j=1; j<=nlstate+ndeath;j++) { */
7714: /* ppij=0.; */
7715: /* for(i=1; i<=nlstate;i++) { */
7716: /* if (mobilav==1) */
7717: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7718: /* else { */
7719: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7720: /* } */
7721: /* if (h*hstepm/YEARM*stepm== yearp) { */
7722: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7723: /* } */
7724: /* } /\* end i *\/ */
7725: /* if (h*hstepm/YEARM*stepm==yearp) { */
7726: /* fprintf(ficresfb," %.3f", ppij); */
7727: /* } */
7728: /* }/\* end j *\/ */
7729: /* } /\* end h *\/ */
7730: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7731: /* } /\* end agec *\/ */
7732: /* } /\* end yearp *\/ */
7733: /* } /\* end cptcod *\/ */
7734: /* } /\* end cptcov *\/ */
7735:
7736: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7737:
7738: /* fclose(ficresfb); */
7739: /* printf("End of Computing Back forecasting \n"); */
7740: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7741:
1.218 brouard 7742: /* } */
1.217 brouard 7743:
1.126 brouard 7744: /************** Forecasting *****not tested NB*************/
1.227 brouard 7745: /* 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 7746:
1.227 brouard 7747: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7748: /* int *popage; */
7749: /* double calagedatem, agelim, kk1, kk2; */
7750: /* double *popeffectif,*popcount; */
7751: /* double ***p3mat,***tabpop,***tabpopprev; */
7752: /* /\* double ***mobaverage; *\/ */
7753: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7754:
1.227 brouard 7755: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7756: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7757: /* agelim=AGESUP; */
7758: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7759:
1.227 brouard 7760: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7761:
7762:
1.227 brouard 7763: /* strcpy(filerespop,"POP_"); */
7764: /* strcat(filerespop,fileresu); */
7765: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7766: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7767: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7768: /* } */
7769: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7770: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7771:
1.227 brouard 7772: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7773:
1.227 brouard 7774: /* /\* if (mobilav!=0) { *\/ */
7775: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7776: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7777: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7778: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7779: /* /\* } *\/ */
7780: /* /\* } *\/ */
1.126 brouard 7781:
1.227 brouard 7782: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7783: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7784:
1.227 brouard 7785: /* agelim=AGESUP; */
1.126 brouard 7786:
1.227 brouard 7787: /* hstepm=1; */
7788: /* hstepm=hstepm/stepm; */
1.218 brouard 7789:
1.227 brouard 7790: /* if (popforecast==1) { */
7791: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7792: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7793: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7794: /* } */
7795: /* popage=ivector(0,AGESUP); */
7796: /* popeffectif=vector(0,AGESUP); */
7797: /* popcount=vector(0,AGESUP); */
1.126 brouard 7798:
1.227 brouard 7799: /* i=1; */
7800: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7801:
1.227 brouard 7802: /* imx=i; */
7803: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7804: /* } */
1.218 brouard 7805:
1.227 brouard 7806: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7807: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7808: /* k=k+1; */
7809: /* fprintf(ficrespop,"\n#******"); */
7810: /* for(j=1;j<=cptcoveff;j++) { */
7811: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7812: /* } */
7813: /* fprintf(ficrespop,"******\n"); */
7814: /* fprintf(ficrespop,"# Age"); */
7815: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7816: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7817:
1.227 brouard 7818: /* for (cpt=0; cpt<=0;cpt++) { */
7819: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7820:
1.227 brouard 7821: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7822: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7823: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7824:
1.227 brouard 7825: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7826: /* oldm=oldms;savm=savms; */
7827: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7828:
1.227 brouard 7829: /* for (h=0; h<=nhstepm; h++){ */
7830: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7831: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7832: /* } */
7833: /* for(j=1; j<=nlstate+ndeath;j++) { */
7834: /* kk1=0.;kk2=0; */
7835: /* for(i=1; i<=nlstate;i++) { */
7836: /* if (mobilav==1) */
7837: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7838: /* else { */
7839: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7840: /* } */
7841: /* } */
7842: /* if (h==(int)(calagedatem+12*cpt)){ */
7843: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7844: /* /\*fprintf(ficrespop," %.3f", kk1); */
7845: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7846: /* } */
7847: /* } */
7848: /* for(i=1; i<=nlstate;i++){ */
7849: /* kk1=0.; */
7850: /* for(j=1; j<=nlstate;j++){ */
7851: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7852: /* } */
7853: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7854: /* } */
1.218 brouard 7855:
1.227 brouard 7856: /* if (h==(int)(calagedatem+12*cpt)) */
7857: /* for(j=1; j<=nlstate;j++) */
7858: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7859: /* } */
7860: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7861: /* } */
7862: /* } */
1.218 brouard 7863:
1.227 brouard 7864: /* /\******\/ */
1.218 brouard 7865:
1.227 brouard 7866: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7867: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7868: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7869: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7870: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7871:
1.227 brouard 7872: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7873: /* oldm=oldms;savm=savms; */
7874: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7875: /* for (h=0; h<=nhstepm; h++){ */
7876: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7877: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7878: /* } */
7879: /* for(j=1; j<=nlstate+ndeath;j++) { */
7880: /* kk1=0.;kk2=0; */
7881: /* for(i=1; i<=nlstate;i++) { */
7882: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7883: /* } */
7884: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7885: /* } */
7886: /* } */
7887: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7888: /* } */
7889: /* } */
7890: /* } */
7891: /* } */
1.218 brouard 7892:
1.227 brouard 7893: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7894:
1.227 brouard 7895: /* if (popforecast==1) { */
7896: /* free_ivector(popage,0,AGESUP); */
7897: /* free_vector(popeffectif,0,AGESUP); */
7898: /* free_vector(popcount,0,AGESUP); */
7899: /* } */
7900: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7901: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7902: /* fclose(ficrespop); */
7903: /* } /\* End of popforecast *\/ */
1.218 brouard 7904:
1.126 brouard 7905: int fileappend(FILE *fichier, char *optionfich)
7906: {
7907: if((fichier=fopen(optionfich,"a"))==NULL) {
7908: printf("Problem with file: %s\n", optionfich);
7909: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7910: return (0);
7911: }
7912: fflush(fichier);
7913: return (1);
7914: }
7915:
7916:
7917: /**************** function prwizard **********************/
7918: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7919: {
7920:
7921: /* Wizard to print covariance matrix template */
7922:
1.164 brouard 7923: char ca[32], cb[32];
7924: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7925: int numlinepar;
7926:
7927: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7928: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7929: for(i=1; i <=nlstate; i++){
7930: jj=0;
7931: for(j=1; j <=nlstate+ndeath; j++){
7932: if(j==i) continue;
7933: jj++;
7934: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7935: printf("%1d%1d",i,j);
7936: fprintf(ficparo,"%1d%1d",i,j);
7937: for(k=1; k<=ncovmodel;k++){
7938: /* printf(" %lf",param[i][j][k]); */
7939: /* fprintf(ficparo," %lf",param[i][j][k]); */
7940: printf(" 0.");
7941: fprintf(ficparo," 0.");
7942: }
7943: printf("\n");
7944: fprintf(ficparo,"\n");
7945: }
7946: }
7947: printf("# Scales (for hessian or gradient estimation)\n");
7948: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7949: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7950: for(i=1; i <=nlstate; i++){
7951: jj=0;
7952: for(j=1; j <=nlstate+ndeath; j++){
7953: if(j==i) continue;
7954: jj++;
7955: fprintf(ficparo,"%1d%1d",i,j);
7956: printf("%1d%1d",i,j);
7957: fflush(stdout);
7958: for(k=1; k<=ncovmodel;k++){
7959: /* printf(" %le",delti3[i][j][k]); */
7960: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7961: printf(" 0.");
7962: fprintf(ficparo," 0.");
7963: }
7964: numlinepar++;
7965: printf("\n");
7966: fprintf(ficparo,"\n");
7967: }
7968: }
7969: printf("# Covariance matrix\n");
7970: /* # 121 Var(a12)\n\ */
7971: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7972: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7973: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7974: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7975: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7976: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7977: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7978: fflush(stdout);
7979: fprintf(ficparo,"# Covariance matrix\n");
7980: /* # 121 Var(a12)\n\ */
7981: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7982: /* # ...\n\ */
7983: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7984:
7985: for(itimes=1;itimes<=2;itimes++){
7986: jj=0;
7987: for(i=1; i <=nlstate; i++){
7988: for(j=1; j <=nlstate+ndeath; j++){
7989: if(j==i) continue;
7990: for(k=1; k<=ncovmodel;k++){
7991: jj++;
7992: ca[0]= k+'a'-1;ca[1]='\0';
7993: if(itimes==1){
7994: printf("#%1d%1d%d",i,j,k);
7995: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7996: }else{
7997: printf("%1d%1d%d",i,j,k);
7998: fprintf(ficparo,"%1d%1d%d",i,j,k);
7999: /* printf(" %.5le",matcov[i][j]); */
8000: }
8001: ll=0;
8002: for(li=1;li <=nlstate; li++){
8003: for(lj=1;lj <=nlstate+ndeath; lj++){
8004: if(lj==li) continue;
8005: for(lk=1;lk<=ncovmodel;lk++){
8006: ll++;
8007: if(ll<=jj){
8008: cb[0]= lk +'a'-1;cb[1]='\0';
8009: if(ll<jj){
8010: if(itimes==1){
8011: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8012: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8013: }else{
8014: printf(" 0.");
8015: fprintf(ficparo," 0.");
8016: }
8017: }else{
8018: if(itimes==1){
8019: printf(" Var(%s%1d%1d)",ca,i,j);
8020: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8021: }else{
8022: printf(" 0.");
8023: fprintf(ficparo," 0.");
8024: }
8025: }
8026: }
8027: } /* end lk */
8028: } /* end lj */
8029: } /* end li */
8030: printf("\n");
8031: fprintf(ficparo,"\n");
8032: numlinepar++;
8033: } /* end k*/
8034: } /*end j */
8035: } /* end i */
8036: } /* end itimes */
8037:
8038: } /* end of prwizard */
8039: /******************* Gompertz Likelihood ******************************/
8040: double gompertz(double x[])
8041: {
8042: double A,B,L=0.0,sump=0.,num=0.;
8043: int i,n=0; /* n is the size of the sample */
8044:
1.220 brouard 8045: for (i=1;i<=imx ; i++) {
1.126 brouard 8046: sump=sump+weight[i];
8047: /* sump=sump+1;*/
8048: num=num+1;
8049: }
8050:
8051:
8052: /* for (i=0; i<=imx; i++)
8053: 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]);*/
8054:
8055: for (i=1;i<=imx ; i++)
8056: {
8057: if (cens[i] == 1 && wav[i]>1)
8058: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8059:
8060: if (cens[i] == 0 && wav[i]>1)
8061: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8062: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8063:
8064: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8065: if (wav[i] > 1 ) { /* ??? */
8066: L=L+A*weight[i];
8067: /* 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]);*/
8068: }
8069: }
8070:
8071: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8072:
8073: return -2*L*num/sump;
8074: }
8075:
1.136 brouard 8076: #ifdef GSL
8077: /******************* Gompertz_f Likelihood ******************************/
8078: double gompertz_f(const gsl_vector *v, void *params)
8079: {
8080: double A,B,LL=0.0,sump=0.,num=0.;
8081: double *x= (double *) v->data;
8082: int i,n=0; /* n is the size of the sample */
8083:
8084: for (i=0;i<=imx-1 ; i++) {
8085: sump=sump+weight[i];
8086: /* sump=sump+1;*/
8087: num=num+1;
8088: }
8089:
8090:
8091: /* for (i=0; i<=imx; i++)
8092: 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]);*/
8093: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8094: for (i=1;i<=imx ; i++)
8095: {
8096: if (cens[i] == 1 && wav[i]>1)
8097: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8098:
8099: if (cens[i] == 0 && wav[i]>1)
8100: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8101: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8102:
8103: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8104: if (wav[i] > 1 ) { /* ??? */
8105: LL=LL+A*weight[i];
8106: /* 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]);*/
8107: }
8108: }
8109:
8110: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8111: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8112:
8113: return -2*LL*num/sump;
8114: }
8115: #endif
8116:
1.126 brouard 8117: /******************* Printing html file ***********/
1.201 brouard 8118: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8119: int lastpass, int stepm, int weightopt, char model[],\
8120: int imx, double p[],double **matcov,double agemortsup){
8121: int i,k;
8122:
8123: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8124: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8125: for (i=1;i<=2;i++)
8126: 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 8127: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8128: fprintf(fichtm,"</ul>");
8129:
8130: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8131:
8132: 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>");
8133:
8134: for (k=agegomp;k<(agemortsup-2);k++)
8135: 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]);
8136:
8137:
8138: fflush(fichtm);
8139: }
8140:
8141: /******************* Gnuplot file **************/
1.201 brouard 8142: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8143:
8144: char dirfileres[132],optfileres[132];
1.164 brouard 8145:
1.126 brouard 8146: int ng;
8147:
8148:
8149: /*#ifdef windows */
8150: fprintf(ficgp,"cd \"%s\" \n",pathc);
8151: /*#endif */
8152:
8153:
8154: strcpy(dirfileres,optionfilefiname);
8155: strcpy(optfileres,"vpl");
1.199 brouard 8156: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8157: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8158: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8159: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8160: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8161:
8162: }
8163:
1.136 brouard 8164: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8165: {
1.126 brouard 8166:
1.136 brouard 8167: /*-------- data file ----------*/
8168: FILE *fic;
8169: char dummy[]=" ";
1.240 brouard 8170: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8171: int lstra;
1.136 brouard 8172: int linei, month, year,iout;
8173: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8174: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8175: char *stratrunc;
1.223 brouard 8176:
1.240 brouard 8177: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8178: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8179:
1.240 brouard 8180: for(v=1; v <=ncovcol;v++){
8181: DummyV[v]=0;
8182: FixedV[v]=0;
8183: }
8184: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8185: DummyV[v]=1;
8186: FixedV[v]=0;
8187: }
8188: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8189: DummyV[v]=0;
8190: FixedV[v]=1;
8191: }
8192: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8193: DummyV[v]=1;
8194: FixedV[v]=1;
8195: }
8196: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8197: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8198: 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]);
8199: }
1.126 brouard 8200:
1.136 brouard 8201: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8202: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8203: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8204: }
1.126 brouard 8205:
1.136 brouard 8206: i=1;
8207: linei=0;
8208: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8209: linei=linei+1;
8210: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8211: if(line[j] == '\t')
8212: line[j] = ' ';
8213: }
8214: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8215: ;
8216: };
8217: line[j+1]=0; /* Trims blanks at end of line */
8218: if(line[0]=='#'){
8219: fprintf(ficlog,"Comment line\n%s\n",line);
8220: printf("Comment line\n%s\n",line);
8221: continue;
8222: }
8223: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8224: strcpy(line, linetmp);
1.223 brouard 8225:
8226: /* Loops on waves */
8227: for (j=maxwav;j>=1;j--){
8228: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8229: cutv(stra, strb, line, ' ');
8230: if(strb[0]=='.') { /* Missing value */
8231: lval=-1;
8232: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8233: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8234: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8235: 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);
8236: 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);
8237: return 1;
8238: }
8239: }else{
8240: errno=0;
8241: /* what_kind_of_number(strb); */
8242: dval=strtod(strb,&endptr);
8243: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8244: /* if(strb != endptr && *endptr == '\0') */
8245: /* dval=dlval; */
8246: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8247: if( strb[0]=='\0' || (*endptr != '\0')){
8248: 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);
8249: 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);
8250: return 1;
8251: }
8252: cotqvar[j][iv][i]=dval;
8253: cotvar[j][ntv+iv][i]=dval;
8254: }
8255: strcpy(line,stra);
1.223 brouard 8256: }/* end loop ntqv */
1.225 brouard 8257:
1.223 brouard 8258: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8259: cutv(stra, strb, line, ' ');
8260: if(strb[0]=='.') { /* Missing value */
8261: lval=-1;
8262: }else{
8263: errno=0;
8264: lval=strtol(strb,&endptr,10);
8265: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8266: if( strb[0]=='\0' || (*endptr != '\0')){
8267: 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);
8268: 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);
8269: return 1;
8270: }
8271: }
8272: if(lval <-1 || lval >1){
8273: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8274: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8275: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8276: For example, for multinomial values like 1, 2 and 3,\n \
8277: build V1=0 V2=0 for the reference value (1),\n \
8278: V1=1 V2=0 for (2) \n \
1.223 brouard 8279: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8280: output of IMaCh is often meaningless.\n \
1.223 brouard 8281: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8282: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8283: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8284: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8285: For example, for multinomial values like 1, 2 and 3,\n \
8286: build V1=0 V2=0 for the reference value (1),\n \
8287: V1=1 V2=0 for (2) \n \
1.223 brouard 8288: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8289: output of IMaCh is often meaningless.\n \
1.223 brouard 8290: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8291: return 1;
8292: }
8293: cotvar[j][iv][i]=(double)(lval);
8294: strcpy(line,stra);
1.223 brouard 8295: }/* end loop ntv */
1.225 brouard 8296:
1.223 brouard 8297: /* Statuses at wave */
1.137 brouard 8298: cutv(stra, strb, line, ' ');
1.223 brouard 8299: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8300: lval=-1;
1.136 brouard 8301: }else{
1.238 brouard 8302: errno=0;
8303: lval=strtol(strb,&endptr,10);
8304: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8305: if( strb[0]=='\0' || (*endptr != '\0')){
8306: 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);
8307: 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);
8308: return 1;
8309: }
1.136 brouard 8310: }
1.225 brouard 8311:
1.136 brouard 8312: s[j][i]=lval;
1.225 brouard 8313:
1.223 brouard 8314: /* Date of Interview */
1.136 brouard 8315: strcpy(line,stra);
8316: cutv(stra, strb,line,' ');
1.169 brouard 8317: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8318: }
1.169 brouard 8319: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8320: month=99;
8321: year=9999;
1.136 brouard 8322: }else{
1.225 brouard 8323: 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);
8324: 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);
8325: return 1;
1.136 brouard 8326: }
8327: anint[j][i]= (double) year;
8328: mint[j][i]= (double)month;
8329: strcpy(line,stra);
1.223 brouard 8330: } /* End loop on waves */
1.225 brouard 8331:
1.223 brouard 8332: /* Date of death */
1.136 brouard 8333: cutv(stra, strb,line,' ');
1.169 brouard 8334: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8335: }
1.169 brouard 8336: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8337: month=99;
8338: year=9999;
8339: }else{
1.141 brouard 8340: 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 8341: 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);
8342: return 1;
1.136 brouard 8343: }
8344: andc[i]=(double) year;
8345: moisdc[i]=(double) month;
8346: strcpy(line,stra);
8347:
1.223 brouard 8348: /* Date of birth */
1.136 brouard 8349: cutv(stra, strb,line,' ');
1.169 brouard 8350: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8351: }
1.169 brouard 8352: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8353: month=99;
8354: year=9999;
8355: }else{
1.141 brouard 8356: 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);
8357: 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 8358: return 1;
1.136 brouard 8359: }
8360: if (year==9999) {
1.141 brouard 8361: 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);
8362: 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 8363: return 1;
8364:
1.136 brouard 8365: }
8366: annais[i]=(double)(year);
8367: moisnais[i]=(double)(month);
8368: strcpy(line,stra);
1.225 brouard 8369:
1.223 brouard 8370: /* Sample weight */
1.136 brouard 8371: cutv(stra, strb,line,' ');
8372: errno=0;
8373: dval=strtod(strb,&endptr);
8374: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8375: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8376: 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 8377: fflush(ficlog);
8378: return 1;
8379: }
8380: weight[i]=dval;
8381: strcpy(line,stra);
1.225 brouard 8382:
1.223 brouard 8383: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8384: cutv(stra, strb, line, ' ');
8385: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8386: lval=-1;
1.223 brouard 8387: }else{
1.225 brouard 8388: errno=0;
8389: /* what_kind_of_number(strb); */
8390: dval=strtod(strb,&endptr);
8391: /* if(strb != endptr && *endptr == '\0') */
8392: /* dval=dlval; */
8393: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8394: if( strb[0]=='\0' || (*endptr != '\0')){
8395: 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);
8396: 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);
8397: return 1;
8398: }
8399: coqvar[iv][i]=dval;
1.226 brouard 8400: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8401: }
8402: strcpy(line,stra);
8403: }/* end loop nqv */
1.136 brouard 8404:
1.223 brouard 8405: /* Covariate values */
1.136 brouard 8406: for (j=ncovcol;j>=1;j--){
8407: cutv(stra, strb,line,' ');
1.223 brouard 8408: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8409: lval=-1;
1.136 brouard 8410: }else{
1.225 brouard 8411: errno=0;
8412: lval=strtol(strb,&endptr,10);
8413: if( strb[0]=='\0' || (*endptr != '\0')){
8414: 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);
8415: 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);
8416: return 1;
8417: }
1.136 brouard 8418: }
8419: if(lval <-1 || lval >1){
1.225 brouard 8420: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8421: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8422: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8423: For example, for multinomial values like 1, 2 and 3,\n \
8424: build V1=0 V2=0 for the reference value (1),\n \
8425: V1=1 V2=0 for (2) \n \
1.136 brouard 8426: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8427: output of IMaCh is often meaningless.\n \
1.136 brouard 8428: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8429: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8430: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8431: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8432: For example, for multinomial values like 1, 2 and 3,\n \
8433: build V1=0 V2=0 for the reference value (1),\n \
8434: V1=1 V2=0 for (2) \n \
1.136 brouard 8435: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8436: output of IMaCh is often meaningless.\n \
1.136 brouard 8437: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8438: return 1;
1.136 brouard 8439: }
8440: covar[j][i]=(double)(lval);
8441: strcpy(line,stra);
8442: }
8443: lstra=strlen(stra);
1.225 brouard 8444:
1.136 brouard 8445: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8446: stratrunc = &(stra[lstra-9]);
8447: num[i]=atol(stratrunc);
8448: }
8449: else
8450: num[i]=atol(stra);
8451: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8452: 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;}*/
8453:
8454: i=i+1;
8455: } /* End loop reading data */
1.225 brouard 8456:
1.136 brouard 8457: *imax=i-1; /* Number of individuals */
8458: fclose(fic);
1.225 brouard 8459:
1.136 brouard 8460: return (0);
1.164 brouard 8461: /* endread: */
1.225 brouard 8462: printf("Exiting readdata: ");
8463: fclose(fic);
8464: return (1);
1.223 brouard 8465: }
1.126 brouard 8466:
1.234 brouard 8467: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8468: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8469: while (*p2 == ' ')
1.234 brouard 8470: p2++;
8471: /* while ((*p1++ = *p2++) !=0) */
8472: /* ; */
8473: /* do */
8474: /* while (*p2 == ' ') */
8475: /* p2++; */
8476: /* while (*p1++ == *p2++); */
8477: *stri=p2;
1.145 brouard 8478: }
8479:
1.235 brouard 8480: int decoderesult ( char resultline[], int nres)
1.230 brouard 8481: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8482: {
1.235 brouard 8483: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8484: char resultsav[MAXLINE];
1.234 brouard 8485: int resultmodel[MAXLINE];
8486: int modelresult[MAXLINE];
1.230 brouard 8487: char stra[80], strb[80], strc[80], strd[80],stre[80];
8488:
1.234 brouard 8489: removefirstspace(&resultline);
1.233 brouard 8490: printf("decoderesult:%s\n",resultline);
1.230 brouard 8491:
8492: if (strstr(resultline,"v") !=0){
8493: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8494: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8495: return 1;
8496: }
8497: trimbb(resultsav, resultline);
8498: if (strlen(resultsav) >1){
8499: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8500: }
1.253 brouard 8501: if(j == 0){ /* Resultline but no = */
8502: TKresult[nres]=0; /* Combination for the nresult and the model */
8503: return (0);
8504: }
8505:
1.234 brouard 8506: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8507: 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);
8508: 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);
8509: }
8510: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8511: if(nbocc(resultsav,'=') >1){
8512: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8513: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8514: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8515: }else
8516: cutl(strc,strd,resultsav,'=');
1.230 brouard 8517: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8518:
1.230 brouard 8519: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8520: Tvarsel[k]=atoi(strc);
8521: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8522: /* cptcovsel++; */
8523: if (nbocc(stra,'=') >0)
8524: strcpy(resultsav,stra); /* and analyzes it */
8525: }
1.235 brouard 8526: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8527: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8528: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8529: match=0;
1.236 brouard 8530: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8531: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8532: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8533: match=1;
8534: break;
8535: }
8536: }
8537: if(match == 0){
8538: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8539: }
8540: }
8541: }
1.235 brouard 8542: /* Checking for missing or useless values in comparison of current model needs */
8543: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8544: match=0;
1.235 brouard 8545: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8546: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8547: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8548: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8549: ++match;
8550: }
8551: }
8552: }
8553: if(match == 0){
8554: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8555: }else if(match > 1){
8556: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8557: }
8558: }
1.235 brouard 8559:
1.234 brouard 8560: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8561: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8562: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8563: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8564: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8565: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8566: /* 1 0 0 0 */
8567: /* 2 1 0 0 */
8568: /* 3 0 1 0 */
8569: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8570: /* 5 0 0 1 */
8571: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8572: /* 7 0 1 1 */
8573: /* 8 1 1 1 */
1.237 brouard 8574: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8575: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8576: /* V5*age V5 known which value for nres? */
8577: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8578: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8579: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8580: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8581: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8582: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8583: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8584: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8585: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8586: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8587: k4++;;
8588: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8589: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8590: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8591: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8592: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8593: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8594: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8595: k4q++;;
8596: }
8597: }
1.234 brouard 8598:
1.235 brouard 8599: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8600: return (0);
8601: }
1.235 brouard 8602:
1.230 brouard 8603: int decodemodel( char model[], int lastobs)
8604: /**< This routine decodes the model and returns:
1.224 brouard 8605: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8606: * - nagesqr = 1 if age*age in the model, otherwise 0.
8607: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8608: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8609: * - cptcovage number of covariates with age*products =2
8610: * - cptcovs number of simple covariates
8611: * - 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
8612: * which is a new column after the 9 (ncovcol) variables.
8613: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8614: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8615: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8616: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8617: */
1.136 brouard 8618: {
1.238 brouard 8619: int i, j, k, ks, v;
1.227 brouard 8620: int j1, k1, k2, k3, k4;
1.136 brouard 8621: char modelsav[80];
1.145 brouard 8622: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8623: char *strpt;
1.136 brouard 8624:
1.145 brouard 8625: /*removespace(model);*/
1.136 brouard 8626: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8627: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8628: if (strstr(model,"AGE") !=0){
1.192 brouard 8629: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8630: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8631: return 1;
8632: }
1.141 brouard 8633: if (strstr(model,"v") !=0){
8634: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8635: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8636: return 1;
8637: }
1.187 brouard 8638: strcpy(modelsav,model);
8639: if ((strpt=strstr(model,"age*age")) !=0){
8640: printf(" strpt=%s, model=%s\n",strpt, model);
8641: if(strpt != model){
1.234 brouard 8642: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8643: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8644: corresponding column of parameters.\n",model);
1.234 brouard 8645: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8646: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8647: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8648: return 1;
1.225 brouard 8649: }
1.187 brouard 8650: nagesqr=1;
8651: if (strstr(model,"+age*age") !=0)
1.234 brouard 8652: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8653: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8654: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8655: else
1.234 brouard 8656: substrchaine(modelsav, model, "age*age");
1.187 brouard 8657: }else
8658: nagesqr=0;
8659: if (strlen(modelsav) >1){
8660: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8661: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8662: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8663: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8664: * cst, age and age*age
8665: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8666: /* including age products which are counted in cptcovage.
8667: * but the covariates which are products must be treated
8668: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8669: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8670: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8671:
8672:
1.187 brouard 8673: /* Design
8674: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8675: * < ncovcol=8 >
8676: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8677: * k= 1 2 3 4 5 6 7 8
8678: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8679: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8680: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8681: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8682: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8683: * Tage[++cptcovage]=k
8684: * if products, new covar are created after ncovcol with k1
8685: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8686: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8687: * 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
8688: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8689: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8690: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8691: * < ncovcol=8 >
8692: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8693: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8694: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8695: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8696: * p Tprod[1]@2={ 6, 5}
8697: *p Tvard[1][1]@4= {7, 8, 5, 6}
8698: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8699: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8700: *How to reorganize?
8701: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8702: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8703: * {2, 1, 4, 8, 5, 6, 3, 7}
8704: * Struct []
8705: */
1.225 brouard 8706:
1.187 brouard 8707: /* This loop fills the array Tvar from the string 'model'.*/
8708: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8709: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8710: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8711: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8712: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8713: /* k=1 Tvar[1]=2 (from V2) */
8714: /* k=5 Tvar[5] */
8715: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8716: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8717: /* } */
1.198 brouard 8718: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8719: /*
8720: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8721: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8722: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8723: }
1.187 brouard 8724: cptcovage=0;
8725: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8726: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8727: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8728: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8729: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8730: /*scanf("%d",i);*/
8731: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8732: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8733: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8734: /* covar is not filled and then is empty */
8735: cptcovprod--;
8736: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8737: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8738: Typevar[k]=1; /* 1 for age product */
8739: cptcovage++; /* Sums the number of covariates which include age as a product */
8740: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8741: /*printf("stre=%s ", stre);*/
8742: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8743: cptcovprod--;
8744: cutl(stre,strb,strc,'V');
8745: Tvar[k]=atoi(stre);
8746: Typevar[k]=1; /* 1 for age product */
8747: cptcovage++;
8748: Tage[cptcovage]=k;
8749: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8750: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8751: cptcovn++;
8752: cptcovprodnoage++;k1++;
8753: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8754: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8755: because this model-covariate is a construction we invent a new column
8756: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8757: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8758: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8759: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8760: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8761: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8762: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8763: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8764: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8765: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8766: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8767: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8768: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8769: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8770: for (i=1; i<=lastobs;i++){
8771: /* Computes the new covariate which is a product of
8772: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8773: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8774: }
8775: } /* End age is not in the model */
8776: } /* End if model includes a product */
8777: else { /* no more sum */
8778: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8779: /* scanf("%d",i);*/
8780: cutl(strd,strc,strb,'V');
8781: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8782: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8783: Tvar[k]=atoi(strd);
8784: Typevar[k]=0; /* 0 for simple covariates */
8785: }
8786: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8787: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8788: scanf("%d",i);*/
1.187 brouard 8789: } /* end of loop + on total covariates */
8790: } /* end if strlen(modelsave == 0) age*age might exist */
8791: } /* end if strlen(model == 0) */
1.136 brouard 8792:
8793: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8794: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8795:
1.136 brouard 8796: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8797: printf("cptcovprod=%d ", cptcovprod);
8798: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8799: scanf("%d ",i);*/
8800:
8801:
1.230 brouard 8802: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8803: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8804: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8805: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8806: k = 1 2 3 4 5 6 7 8 9
8807: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8808: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8809: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8810: Dummy[k] 1 0 0 0 3 1 1 2 3
8811: Tmodelind[combination of covar]=k;
1.225 brouard 8812: */
8813: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8814: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8815: /* 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 8816: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8817: printf("Model=%s\n\
8818: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8819: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8820: 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);
8821: fprintf(ficlog,"Model=%s\n\
8822: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8823: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8824: 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 8825: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8826: 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 */
8827: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8828: Fixed[k]= 0;
8829: Dummy[k]= 0;
1.225 brouard 8830: ncoveff++;
1.232 brouard 8831: ncovf++;
1.234 brouard 8832: nsd++;
8833: modell[k].maintype= FTYPE;
8834: TvarsD[nsd]=Tvar[k];
8835: TvarsDind[nsd]=k;
8836: TvarF[ncovf]=Tvar[k];
8837: TvarFind[ncovf]=k;
8838: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8839: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8840: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8841: Fixed[k]= 0;
8842: Dummy[k]= 0;
8843: ncoveff++;
8844: ncovf++;
8845: modell[k].maintype= FTYPE;
8846: TvarF[ncovf]=Tvar[k];
8847: TvarFind[ncovf]=k;
1.230 brouard 8848: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8849: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8850: }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 8851: Fixed[k]= 0;
8852: Dummy[k]= 1;
1.230 brouard 8853: nqfveff++;
1.234 brouard 8854: modell[k].maintype= FTYPE;
8855: modell[k].subtype= FQ;
8856: nsq++;
8857: TvarsQ[nsq]=Tvar[k];
8858: TvarsQind[nsq]=k;
1.232 brouard 8859: ncovf++;
1.234 brouard 8860: TvarF[ncovf]=Tvar[k];
8861: TvarFind[ncovf]=k;
1.231 brouard 8862: 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 8863: 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 8864: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8865: Fixed[k]= 1;
8866: Dummy[k]= 0;
1.225 brouard 8867: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8868: modell[k].maintype= VTYPE;
8869: modell[k].subtype= VD;
8870: nsd++;
8871: TvarsD[nsd]=Tvar[k];
8872: TvarsDind[nsd]=k;
8873: ncovv++; /* Only simple time varying variables */
8874: TvarV[ncovv]=Tvar[k];
1.242 brouard 8875: 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 8876: 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 */
8877: 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 8878: 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);
8879: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8880: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8881: Fixed[k]= 1;
8882: Dummy[k]= 1;
8883: nqtveff++;
8884: modell[k].maintype= VTYPE;
8885: modell[k].subtype= VQ;
8886: ncovv++; /* Only simple time varying variables */
8887: nsq++;
8888: TvarsQ[nsq]=Tvar[k];
8889: TvarsQind[nsq]=k;
8890: TvarV[ncovv]=Tvar[k];
1.242 brouard 8891: 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 8892: 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 */
8893: 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 8894: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8895: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8896: 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 8897: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8898: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8899: ncova++;
8900: TvarA[ncova]=Tvar[k];
8901: TvarAind[ncova]=k;
1.231 brouard 8902: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8903: Fixed[k]= 2;
8904: Dummy[k]= 2;
8905: modell[k].maintype= ATYPE;
8906: modell[k].subtype= APFD;
8907: /* ncoveff++; */
1.227 brouard 8908: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8909: Fixed[k]= 2;
8910: Dummy[k]= 3;
8911: modell[k].maintype= ATYPE;
8912: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8913: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8914: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8915: Fixed[k]= 3;
8916: Dummy[k]= 2;
8917: modell[k].maintype= ATYPE;
8918: modell[k].subtype= APVD; /* Product age * varying dummy */
8919: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8920: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8921: Fixed[k]= 3;
8922: Dummy[k]= 3;
8923: modell[k].maintype= ATYPE;
8924: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8925: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8926: }
8927: }else if (Typevar[k] == 2) { /* product without age */
8928: k1=Tposprod[k];
8929: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8930: if(Tvard[k1][2] <=ncovcol){
8931: Fixed[k]= 1;
8932: Dummy[k]= 0;
8933: modell[k].maintype= FTYPE;
8934: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8935: ncovf++; /* Fixed variables without age */
8936: TvarF[ncovf]=Tvar[k];
8937: TvarFind[ncovf]=k;
8938: }else if(Tvard[k1][2] <=ncovcol+nqv){
8939: Fixed[k]= 0; /* or 2 ?*/
8940: Dummy[k]= 1;
8941: modell[k].maintype= FTYPE;
8942: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8943: ncovf++; /* Varying variables without age */
8944: TvarF[ncovf]=Tvar[k];
8945: TvarFind[ncovf]=k;
8946: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8947: Fixed[k]= 1;
8948: Dummy[k]= 0;
8949: modell[k].maintype= VTYPE;
8950: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8951: ncovv++; /* Varying variables without age */
8952: TvarV[ncovv]=Tvar[k];
8953: TvarVind[ncovv]=k;
8954: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8955: Fixed[k]= 1;
8956: Dummy[k]= 1;
8957: modell[k].maintype= VTYPE;
8958: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8959: ncovv++; /* Varying variables without age */
8960: TvarV[ncovv]=Tvar[k];
8961: TvarVind[ncovv]=k;
8962: }
1.227 brouard 8963: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8964: if(Tvard[k1][2] <=ncovcol){
8965: Fixed[k]= 0; /* or 2 ?*/
8966: Dummy[k]= 1;
8967: modell[k].maintype= FTYPE;
8968: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8969: ncovf++; /* Fixed variables without age */
8970: TvarF[ncovf]=Tvar[k];
8971: TvarFind[ncovf]=k;
8972: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8973: Fixed[k]= 1;
8974: Dummy[k]= 1;
8975: modell[k].maintype= VTYPE;
8976: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8977: ncovv++; /* Varying variables without age */
8978: TvarV[ncovv]=Tvar[k];
8979: TvarVind[ncovv]=k;
8980: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8981: Fixed[k]= 1;
8982: Dummy[k]= 1;
8983: modell[k].maintype= VTYPE;
8984: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8985: ncovv++; /* Varying variables without age */
8986: TvarV[ncovv]=Tvar[k];
8987: TvarVind[ncovv]=k;
8988: ncovv++; /* Varying variables without age */
8989: TvarV[ncovv]=Tvar[k];
8990: TvarVind[ncovv]=k;
8991: }
1.227 brouard 8992: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8993: if(Tvard[k1][2] <=ncovcol){
8994: Fixed[k]= 1;
8995: Dummy[k]= 1;
8996: modell[k].maintype= VTYPE;
8997: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8998: ncovv++; /* Varying variables without age */
8999: TvarV[ncovv]=Tvar[k];
9000: TvarVind[ncovv]=k;
9001: }else if(Tvard[k1][2] <=ncovcol+nqv){
9002: Fixed[k]= 1;
9003: Dummy[k]= 1;
9004: modell[k].maintype= VTYPE;
9005: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9006: ncovv++; /* Varying variables without age */
9007: TvarV[ncovv]=Tvar[k];
9008: TvarVind[ncovv]=k;
9009: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9010: Fixed[k]= 1;
9011: Dummy[k]= 0;
9012: modell[k].maintype= VTYPE;
9013: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9014: ncovv++; /* Varying variables without age */
9015: TvarV[ncovv]=Tvar[k];
9016: TvarVind[ncovv]=k;
9017: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9018: Fixed[k]= 1;
9019: Dummy[k]= 1;
9020: modell[k].maintype= VTYPE;
9021: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9022: ncovv++; /* Varying variables without age */
9023: TvarV[ncovv]=Tvar[k];
9024: TvarVind[ncovv]=k;
9025: }
1.227 brouard 9026: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9027: if(Tvard[k1][2] <=ncovcol){
9028: Fixed[k]= 1;
9029: Dummy[k]= 1;
9030: modell[k].maintype= VTYPE;
9031: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9032: ncovv++; /* Varying variables without age */
9033: TvarV[ncovv]=Tvar[k];
9034: TvarVind[ncovv]=k;
9035: }else if(Tvard[k1][2] <=ncovcol+nqv){
9036: Fixed[k]= 1;
9037: Dummy[k]= 1;
9038: modell[k].maintype= VTYPE;
9039: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9040: ncovv++; /* Varying variables without age */
9041: TvarV[ncovv]=Tvar[k];
9042: TvarVind[ncovv]=k;
9043: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9044: Fixed[k]= 1;
9045: Dummy[k]= 1;
9046: modell[k].maintype= VTYPE;
9047: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9048: ncovv++; /* Varying variables without age */
9049: TvarV[ncovv]=Tvar[k];
9050: TvarVind[ncovv]=k;
9051: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9052: Fixed[k]= 1;
9053: Dummy[k]= 1;
9054: modell[k].maintype= VTYPE;
9055: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9056: ncovv++; /* Varying variables without age */
9057: TvarV[ncovv]=Tvar[k];
9058: TvarVind[ncovv]=k;
9059: }
1.227 brouard 9060: }else{
1.240 brouard 9061: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9062: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9063: } /*end k1*/
1.225 brouard 9064: }else{
1.226 brouard 9065: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9066: 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 9067: }
1.227 brouard 9068: 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 9069: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9070: 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]);
9071: }
9072: /* Searching for doublons in the model */
9073: for(k1=1; k1<= cptcovt;k1++){
9074: for(k2=1; k2 <k1;k2++){
9075: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9076: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9077: if(Tvar[k1]==Tvar[k2]){
9078: 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]]);
9079: 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);
9080: return(1);
9081: }
9082: }else if (Typevar[k1] ==2){
9083: k3=Tposprod[k1];
9084: k4=Tposprod[k2];
9085: 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])) ){
9086: 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]]);
9087: 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);
9088: return(1);
9089: }
9090: }
1.227 brouard 9091: }
9092: }
1.225 brouard 9093: }
9094: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9095: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9096: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9097: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9098: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9099: /*endread:*/
1.225 brouard 9100: printf("Exiting decodemodel: ");
9101: return (1);
1.136 brouard 9102: }
9103:
1.169 brouard 9104: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9105: {/* Check ages at death */
1.136 brouard 9106: int i, m;
1.218 brouard 9107: int firstone=0;
9108:
1.136 brouard 9109: for (i=1; i<=imx; i++) {
9110: for(m=2; (m<= maxwav); m++) {
9111: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9112: anint[m][i]=9999;
1.216 brouard 9113: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9114: s[m][i]=-1;
1.136 brouard 9115: }
9116: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9117: *nberr = *nberr + 1;
1.218 brouard 9118: if(firstone == 0){
9119: firstone=1;
1.260 brouard 9120: 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 9121: }
1.260 brouard 9122: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
9123: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9124: }
9125: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9126: (*nberr)++;
1.259 brouard 9127: 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);
9128: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
9129: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9130: }
9131: }
9132: }
9133:
9134: for (i=1; i<=imx; i++) {
9135: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9136: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9137: 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 9138: if (s[m][i] >= nlstate+1) {
1.169 brouard 9139: if(agedc[i]>0){
9140: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9141: agev[m][i]=agedc[i];
1.214 brouard 9142: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9143: }else {
1.136 brouard 9144: if ((int)andc[i]!=9999){
9145: nbwarn++;
9146: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9147: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9148: agev[m][i]=-1;
9149: }
9150: }
1.169 brouard 9151: } /* agedc > 0 */
1.214 brouard 9152: } /* end if */
1.136 brouard 9153: else if(s[m][i] !=9){ /* Standard case, age in fractional
9154: years but with the precision of a month */
9155: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9156: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9157: agev[m][i]=1;
9158: else if(agev[m][i] < *agemin){
9159: *agemin=agev[m][i];
9160: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9161: }
9162: else if(agev[m][i] >*agemax){
9163: *agemax=agev[m][i];
1.156 brouard 9164: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9165: }
9166: /*agev[m][i]=anint[m][i]-annais[i];*/
9167: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9168: } /* en if 9*/
1.136 brouard 9169: else { /* =9 */
1.214 brouard 9170: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9171: agev[m][i]=1;
9172: s[m][i]=-1;
9173: }
9174: }
1.214 brouard 9175: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9176: agev[m][i]=1;
1.214 brouard 9177: else{
9178: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9179: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9180: agev[m][i]=0;
9181: }
9182: } /* End for lastpass */
9183: }
1.136 brouard 9184:
9185: for (i=1; i<=imx; i++) {
9186: for(m=firstpass; (m<=lastpass); m++){
9187: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9188: (*nberr)++;
1.136 brouard 9189: 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);
9190: 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);
9191: return 1;
9192: }
9193: }
9194: }
9195:
9196: /*for (i=1; i<=imx; i++){
9197: for (m=firstpass; (m<lastpass); m++){
9198: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9199: }
9200:
9201: }*/
9202:
9203:
1.139 brouard 9204: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9205: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9206:
9207: return (0);
1.164 brouard 9208: /* endread:*/
1.136 brouard 9209: printf("Exiting calandcheckages: ");
9210: return (1);
9211: }
9212:
1.172 brouard 9213: #if defined(_MSC_VER)
9214: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9215: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9216: //#include "stdafx.h"
9217: //#include <stdio.h>
9218: //#include <tchar.h>
9219: //#include <windows.h>
9220: //#include <iostream>
9221: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9222:
9223: LPFN_ISWOW64PROCESS fnIsWow64Process;
9224:
9225: BOOL IsWow64()
9226: {
9227: BOOL bIsWow64 = FALSE;
9228:
9229: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9230: // (HANDLE, PBOOL);
9231:
9232: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9233:
9234: HMODULE module = GetModuleHandle(_T("kernel32"));
9235: const char funcName[] = "IsWow64Process";
9236: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9237: GetProcAddress(module, funcName);
9238:
9239: if (NULL != fnIsWow64Process)
9240: {
9241: if (!fnIsWow64Process(GetCurrentProcess(),
9242: &bIsWow64))
9243: //throw std::exception("Unknown error");
9244: printf("Unknown error\n");
9245: }
9246: return bIsWow64 != FALSE;
9247: }
9248: #endif
1.177 brouard 9249:
1.191 brouard 9250: void syscompilerinfo(int logged)
1.167 brouard 9251: {
9252: /* #include "syscompilerinfo.h"*/
1.185 brouard 9253: /* command line Intel compiler 32bit windows, XP compatible:*/
9254: /* /GS /W3 /Gy
9255: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9256: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9257: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9258: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9259: */
9260: /* 64 bits */
1.185 brouard 9261: /*
9262: /GS /W3 /Gy
9263: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9264: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9265: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9266: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9267: /* Optimization are useless and O3 is slower than O2 */
9268: /*
9269: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9270: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9271: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9272: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9273: */
1.186 brouard 9274: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9275: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9276: /PDB:"visual studio
9277: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9278: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9279: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9280: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9281: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9282: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9283: uiAccess='false'"
9284: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9285: /NOLOGO /TLBID:1
9286: */
1.177 brouard 9287: #if defined __INTEL_COMPILER
1.178 brouard 9288: #if defined(__GNUC__)
9289: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9290: #endif
1.177 brouard 9291: #elif defined(__GNUC__)
1.179 brouard 9292: #ifndef __APPLE__
1.174 brouard 9293: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9294: #endif
1.177 brouard 9295: struct utsname sysInfo;
1.178 brouard 9296: int cross = CROSS;
9297: if (cross){
9298: printf("Cross-");
1.191 brouard 9299: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9300: }
1.174 brouard 9301: #endif
9302:
1.171 brouard 9303: #include <stdint.h>
1.178 brouard 9304:
1.191 brouard 9305: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9306: #if defined(__clang__)
1.191 brouard 9307: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9308: #endif
9309: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9310: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9311: #endif
9312: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9313: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9314: #endif
9315: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9316: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9317: #endif
9318: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9319: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9320: #endif
9321: #if defined(_MSC_VER)
1.191 brouard 9322: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9323: #endif
9324: #if defined(__PGI)
1.191 brouard 9325: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9326: #endif
9327: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9328: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9329: #endif
1.191 brouard 9330: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9331:
1.167 brouard 9332: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9333: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9334: // Windows (x64 and x86)
1.191 brouard 9335: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9336: #elif __unix__ // all unices, not all compilers
9337: // Unix
1.191 brouard 9338: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9339: #elif __linux__
9340: // linux
1.191 brouard 9341: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9342: #elif __APPLE__
1.174 brouard 9343: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9344: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9345: #endif
9346:
9347: /* __MINGW32__ */
9348: /* __CYGWIN__ */
9349: /* __MINGW64__ */
9350: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9351: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9352: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9353: /* _WIN64 // Defined for applications for Win64. */
9354: /* _M_X64 // Defined for compilations that target x64 processors. */
9355: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9356:
1.167 brouard 9357: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9358: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9359: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9360: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9361: #else
1.191 brouard 9362: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9363: #endif
9364:
1.169 brouard 9365: #if defined(__GNUC__)
9366: # if defined(__GNUC_PATCHLEVEL__)
9367: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9368: + __GNUC_MINOR__ * 100 \
9369: + __GNUC_PATCHLEVEL__)
9370: # else
9371: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9372: + __GNUC_MINOR__ * 100)
9373: # endif
1.174 brouard 9374: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9375: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9376:
9377: if (uname(&sysInfo) != -1) {
9378: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9379: 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 9380: }
9381: else
9382: perror("uname() error");
1.179 brouard 9383: //#ifndef __INTEL_COMPILER
9384: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9385: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9386: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9387: #endif
1.169 brouard 9388: #endif
1.172 brouard 9389:
9390: // void main()
9391: // {
1.169 brouard 9392: #if defined(_MSC_VER)
1.174 brouard 9393: if (IsWow64()){
1.191 brouard 9394: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9395: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9396: }
9397: else{
1.191 brouard 9398: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9399: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9400: }
1.172 brouard 9401: // printf("\nPress Enter to continue...");
9402: // getchar();
9403: // }
9404:
1.169 brouard 9405: #endif
9406:
1.167 brouard 9407:
1.219 brouard 9408: }
1.136 brouard 9409:
1.219 brouard 9410: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9411: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9412: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9413: /* double ftolpl = 1.e-10; */
1.180 brouard 9414: double age, agebase, agelim;
1.203 brouard 9415: double tot;
1.180 brouard 9416:
1.202 brouard 9417: strcpy(filerespl,"PL_");
9418: strcat(filerespl,fileresu);
9419: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9420: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9421: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9422: }
1.227 brouard 9423: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9424: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9425: pstamp(ficrespl);
1.203 brouard 9426: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9427: fprintf(ficrespl,"#Age ");
9428: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9429: fprintf(ficrespl,"\n");
1.180 brouard 9430:
1.219 brouard 9431: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9432:
1.219 brouard 9433: agebase=ageminpar;
9434: agelim=agemaxpar;
1.180 brouard 9435:
1.227 brouard 9436: /* i1=pow(2,ncoveff); */
1.234 brouard 9437: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9438: if (cptcovn < 1){i1=1;}
1.180 brouard 9439:
1.238 brouard 9440: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9441: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9442: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9443: continue;
1.235 brouard 9444:
1.238 brouard 9445: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9446: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9447: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9448: /* k=k+1; */
9449: /* to clean */
9450: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9451: fprintf(ficrespl,"#******");
9452: printf("#******");
9453: fprintf(ficlog,"#******");
9454: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9455: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9456: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9457: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9458: }
9459: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9460: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9461: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9462: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9463: }
9464: fprintf(ficrespl,"******\n");
9465: printf("******\n");
9466: fprintf(ficlog,"******\n");
9467: if(invalidvarcomb[k]){
9468: printf("\nCombination (%d) ignored because no case \n",k);
9469: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9470: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9471: continue;
9472: }
1.219 brouard 9473:
1.238 brouard 9474: fprintf(ficrespl,"#Age ");
9475: for(j=1;j<=cptcoveff;j++) {
9476: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9477: }
9478: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9479: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9480:
1.238 brouard 9481: for (age=agebase; age<=agelim; age++){
9482: /* for (age=agebase; age<=agebase; age++){ */
9483: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9484: fprintf(ficrespl,"%.0f ",age );
9485: for(j=1;j<=cptcoveff;j++)
9486: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9487: tot=0.;
9488: for(i=1; i<=nlstate;i++){
9489: tot += prlim[i][i];
9490: fprintf(ficrespl," %.5f", prlim[i][i]);
9491: }
9492: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9493: } /* Age */
9494: /* was end of cptcod */
9495: } /* cptcov */
9496: } /* nres */
1.219 brouard 9497: return 0;
1.180 brouard 9498: }
9499:
1.218 brouard 9500: 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){
9501: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9502:
9503: /* Computes the back prevalence limit for any combination of covariate values
9504: * at any age between ageminpar and agemaxpar
9505: */
1.235 brouard 9506: int i, j, k, i1, nres=0 ;
1.217 brouard 9507: /* double ftolpl = 1.e-10; */
9508: double age, agebase, agelim;
9509: double tot;
1.218 brouard 9510: /* double ***mobaverage; */
9511: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9512:
9513: strcpy(fileresplb,"PLB_");
9514: strcat(fileresplb,fileresu);
9515: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9516: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9517: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9518: }
9519: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9520: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9521: pstamp(ficresplb);
9522: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9523: fprintf(ficresplb,"#Age ");
9524: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9525: fprintf(ficresplb,"\n");
9526:
1.218 brouard 9527:
9528: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9529:
9530: agebase=ageminpar;
9531: agelim=agemaxpar;
9532:
9533:
1.227 brouard 9534: i1=pow(2,cptcoveff);
1.218 brouard 9535: if (cptcovn < 1){i1=1;}
1.227 brouard 9536:
1.238 brouard 9537: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9538: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9539: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9540: continue;
9541: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9542: fprintf(ficresplb,"#******");
9543: printf("#******");
9544: fprintf(ficlog,"#******");
9545: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9546: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9547: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9548: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9549: }
9550: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9551: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9552: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9553: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9554: }
9555: fprintf(ficresplb,"******\n");
9556: printf("******\n");
9557: fprintf(ficlog,"******\n");
9558: if(invalidvarcomb[k]){
9559: printf("\nCombination (%d) ignored because no cases \n",k);
9560: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9561: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9562: continue;
9563: }
1.218 brouard 9564:
1.238 brouard 9565: fprintf(ficresplb,"#Age ");
9566: for(j=1;j<=cptcoveff;j++) {
9567: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9568: }
9569: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9570: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9571:
9572:
1.238 brouard 9573: for (age=agebase; age<=agelim; age++){
9574: /* for (age=agebase; age<=agebase; age++){ */
9575: if(mobilavproj > 0){
9576: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9577: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9578: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9579: }else if (mobilavproj == 0){
9580: 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);
9581: 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);
9582: exit(1);
9583: }else{
9584: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9585: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9586: }
9587: fprintf(ficresplb,"%.0f ",age );
9588: for(j=1;j<=cptcoveff;j++)
9589: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9590: tot=0.;
9591: for(i=1; i<=nlstate;i++){
9592: tot += bprlim[i][i];
9593: fprintf(ficresplb," %.5f", bprlim[i][i]);
9594: }
9595: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9596: } /* Age */
9597: /* was end of cptcod */
1.255 brouard 9598: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9599: } /* end of any combination */
9600: } /* end of nres */
1.218 brouard 9601: /* hBijx(p, bage, fage); */
9602: /* fclose(ficrespijb); */
9603:
9604: return 0;
1.217 brouard 9605: }
1.218 brouard 9606:
1.180 brouard 9607: int hPijx(double *p, int bage, int fage){
9608: /*------------- h Pij x at various ages ------------*/
9609:
9610: int stepsize;
9611: int agelim;
9612: int hstepm;
9613: int nhstepm;
1.235 brouard 9614: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9615:
9616: double agedeb;
9617: double ***p3mat;
9618:
1.201 brouard 9619: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9620: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9621: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9622: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9623: }
9624: printf("Computing pij: result on file '%s' \n", filerespij);
9625: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9626:
9627: stepsize=(int) (stepm+YEARM-1)/YEARM;
9628: /*if (stepm<=24) stepsize=2;*/
9629:
9630: agelim=AGESUP;
9631: hstepm=stepsize*YEARM; /* Every year of age */
9632: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9633:
1.180 brouard 9634: /* hstepm=1; aff par mois*/
9635: pstamp(ficrespij);
9636: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9637: i1= pow(2,cptcoveff);
1.218 brouard 9638: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9639: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9640: /* k=k+1; */
1.235 brouard 9641: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9642: for(k=1; k<=i1;k++){
1.253 brouard 9643: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9644: continue;
1.183 brouard 9645: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9646: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9647: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9648: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9649: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9650: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9651: }
1.183 brouard 9652: fprintf(ficrespij,"******\n");
9653:
9654: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9655: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9656: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9657:
9658: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9659:
1.183 brouard 9660: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9661: oldm=oldms;savm=savms;
1.235 brouard 9662: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9663: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9664: for(i=1; i<=nlstate;i++)
9665: for(j=1; j<=nlstate+ndeath;j++)
9666: fprintf(ficrespij," %1d-%1d",i,j);
9667: fprintf(ficrespij,"\n");
9668: for (h=0; h<=nhstepm; h++){
9669: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9670: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9671: for(i=1; i<=nlstate;i++)
9672: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9673: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9674: fprintf(ficrespij,"\n");
9675: }
1.183 brouard 9676: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9677: fprintf(ficrespij,"\n");
9678: }
1.180 brouard 9679: /*}*/
9680: }
1.218 brouard 9681: return 0;
1.180 brouard 9682: }
1.218 brouard 9683:
9684: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9685: /*------------- h Bij x at various ages ------------*/
9686:
9687: int stepsize;
1.218 brouard 9688: /* int agelim; */
9689: int ageminl;
1.217 brouard 9690: int hstepm;
9691: int nhstepm;
1.238 brouard 9692: int h, i, i1, j, k, nres;
1.218 brouard 9693:
1.217 brouard 9694: double agedeb;
9695: double ***p3mat;
1.218 brouard 9696:
9697: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9698: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9699: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9700: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9701: }
9702: printf("Computing pij back: result on file '%s' \n", filerespijb);
9703: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9704:
9705: stepsize=(int) (stepm+YEARM-1)/YEARM;
9706: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9707:
1.218 brouard 9708: /* agelim=AGESUP; */
9709: ageminl=30;
9710: hstepm=stepsize*YEARM; /* Every year of age */
9711: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9712:
9713: /* hstepm=1; aff par mois*/
9714: pstamp(ficrespijb);
1.255 brouard 9715: 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 9716: i1= pow(2,cptcoveff);
1.218 brouard 9717: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9718: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9719: /* k=k+1; */
1.238 brouard 9720: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9721: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9722: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9723: continue;
9724: fprintf(ficrespijb,"\n#****** ");
9725: for(j=1;j<=cptcoveff;j++)
9726: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9727: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9728: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9729: }
9730: fprintf(ficrespijb,"******\n");
9731: if(invalidvarcomb[k]){
9732: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9733: continue;
9734: }
9735:
9736: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9737: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9738: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9739: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9740: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9741:
9742: /* nhstepm=nhstepm*YEARM; aff par mois*/
9743:
9744: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9745: /* oldm=oldms;savm=savms; */
9746: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9747: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9748: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9749: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9750: for(i=1; i<=nlstate;i++)
9751: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9752: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9753: fprintf(ficrespijb,"\n");
1.238 brouard 9754: for (h=0; h<=nhstepm; h++){
9755: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9756: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9757: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9758: for(i=1; i<=nlstate;i++)
9759: for(j=1; j<=nlstate+ndeath;j++)
9760: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9761: fprintf(ficrespijb,"\n");
9762: }
9763: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9764: fprintf(ficrespijb,"\n");
9765: } /* end age deb */
9766: } /* end combination */
9767: } /* end nres */
1.218 brouard 9768: return 0;
9769: } /* hBijx */
1.217 brouard 9770:
1.180 brouard 9771:
1.136 brouard 9772: /***********************************************/
9773: /**************** Main Program *****************/
9774: /***********************************************/
9775:
9776: int main(int argc, char *argv[])
9777: {
9778: #ifdef GSL
9779: const gsl_multimin_fminimizer_type *T;
9780: size_t iteri = 0, it;
9781: int rval = GSL_CONTINUE;
9782: int status = GSL_SUCCESS;
9783: double ssval;
9784: #endif
9785: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9786: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9787: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9788: int jj, ll, li, lj, lk;
1.136 brouard 9789: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9790: int num_filled;
1.136 brouard 9791: int itimes;
9792: int NDIM=2;
9793: int vpopbased=0;
1.235 brouard 9794: int nres=0;
1.258 brouard 9795: int endishere=0;
1.136 brouard 9796:
1.164 brouard 9797: char ca[32], cb[32];
1.136 brouard 9798: /* FILE *fichtm; *//* Html File */
9799: /* FILE *ficgp;*/ /*Gnuplot File */
9800: struct stat info;
1.191 brouard 9801: double agedeb=0.;
1.194 brouard 9802:
9803: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9804: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9805:
1.165 brouard 9806: double fret;
1.191 brouard 9807: double dum=0.; /* Dummy variable */
1.136 brouard 9808: double ***p3mat;
1.218 brouard 9809: /* double ***mobaverage; */
1.164 brouard 9810:
9811: char line[MAXLINE];
1.197 brouard 9812: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9813:
1.234 brouard 9814: char modeltemp[MAXLINE];
1.230 brouard 9815: char resultline[MAXLINE];
9816:
1.136 brouard 9817: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9818: char *tok, *val; /* pathtot */
1.136 brouard 9819: int firstobs=1, lastobs=10;
1.195 brouard 9820: int c, h , cpt, c2;
1.191 brouard 9821: int jl=0;
9822: int i1, j1, jk, stepsize=0;
1.194 brouard 9823: int count=0;
9824:
1.164 brouard 9825: int *tab;
1.136 brouard 9826: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9827: int backcast=0;
1.136 brouard 9828: int mobilav=0,popforecast=0;
1.191 brouard 9829: int hstepm=0, nhstepm=0;
1.136 brouard 9830: int agemortsup;
9831: float sumlpop=0.;
9832: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9833: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9834:
1.191 brouard 9835: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9836: double ftolpl=FTOL;
9837: double **prlim;
1.217 brouard 9838: double **bprlim;
1.136 brouard 9839: double ***param; /* Matrix of parameters */
1.251 brouard 9840: double ***paramstart; /* Matrix of starting parameter values */
9841: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9842: double **matcov; /* Matrix of covariance */
1.203 brouard 9843: double **hess; /* Hessian matrix */
1.136 brouard 9844: double ***delti3; /* Scale */
9845: double *delti; /* Scale */
9846: double ***eij, ***vareij;
9847: double **varpl; /* Variances of prevalence limits by age */
9848: double *epj, vepp;
1.164 brouard 9849:
1.136 brouard 9850: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9851: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9852:
1.136 brouard 9853: double **ximort;
1.145 brouard 9854: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9855: int *dcwave;
9856:
1.164 brouard 9857: char z[1]="c";
1.136 brouard 9858:
9859: /*char *strt;*/
9860: char strtend[80];
1.126 brouard 9861:
1.164 brouard 9862:
1.126 brouard 9863: /* setlocale (LC_ALL, ""); */
9864: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9865: /* textdomain (PACKAGE); */
9866: /* setlocale (LC_CTYPE, ""); */
9867: /* setlocale (LC_MESSAGES, ""); */
9868:
9869: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9870: rstart_time = time(NULL);
9871: /* (void) gettimeofday(&start_time,&tzp);*/
9872: start_time = *localtime(&rstart_time);
1.126 brouard 9873: curr_time=start_time;
1.157 brouard 9874: /*tml = *localtime(&start_time.tm_sec);*/
9875: /* strcpy(strstart,asctime(&tml)); */
9876: strcpy(strstart,asctime(&start_time));
1.126 brouard 9877:
9878: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9879: /* tp.tm_sec = tp.tm_sec +86400; */
9880: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9881: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9882: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9883: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9884: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9885: /* strt=asctime(&tmg); */
9886: /* printf("Time(after) =%s",strstart); */
9887: /* (void) time (&time_value);
9888: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9889: * tm = *localtime(&time_value);
9890: * strstart=asctime(&tm);
9891: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9892: */
9893:
9894: nberr=0; /* Number of errors and warnings */
9895: nbwarn=0;
1.184 brouard 9896: #ifdef WIN32
9897: _getcwd(pathcd, size);
9898: #else
1.126 brouard 9899: getcwd(pathcd, size);
1.184 brouard 9900: #endif
1.191 brouard 9901: syscompilerinfo(0);
1.196 brouard 9902: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9903: if(argc <=1){
9904: printf("\nEnter the parameter file name: ");
1.205 brouard 9905: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9906: printf("ERROR Empty parameter file name\n");
9907: goto end;
9908: }
1.126 brouard 9909: i=strlen(pathr);
9910: if(pathr[i-1]=='\n')
9911: pathr[i-1]='\0';
1.156 brouard 9912: i=strlen(pathr);
1.205 brouard 9913: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9914: pathr[i-1]='\0';
1.205 brouard 9915: }
9916: i=strlen(pathr);
9917: if( i==0 ){
9918: printf("ERROR Empty parameter file name\n");
9919: goto end;
9920: }
9921: for (tok = pathr; tok != NULL; ){
1.126 brouard 9922: printf("Pathr |%s|\n",pathr);
9923: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9924: printf("val= |%s| pathr=%s\n",val,pathr);
9925: strcpy (pathtot, val);
9926: if(pathr[0] == '\0') break; /* Dirty */
9927: }
9928: }
9929: else{
9930: strcpy(pathtot,argv[1]);
9931: }
9932: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9933: /*cygwin_split_path(pathtot,path,optionfile);
9934: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9935: /* cutv(path,optionfile,pathtot,'\\');*/
9936:
9937: /* Split argv[0], imach program to get pathimach */
9938: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9939: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9940: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9941: /* strcpy(pathimach,argv[0]); */
9942: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9943: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9944: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9945: #ifdef WIN32
9946: _chdir(path); /* Can be a relative path */
9947: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9948: #else
1.126 brouard 9949: chdir(path); /* Can be a relative path */
1.184 brouard 9950: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9951: #endif
9952: printf("Current directory %s!\n",pathcd);
1.126 brouard 9953: strcpy(command,"mkdir ");
9954: strcat(command,optionfilefiname);
9955: if((outcmd=system(command)) != 0){
1.169 brouard 9956: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9957: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9958: /* fclose(ficlog); */
9959: /* exit(1); */
9960: }
9961: /* if((imk=mkdir(optionfilefiname))<0){ */
9962: /* perror("mkdir"); */
9963: /* } */
9964:
9965: /*-------- arguments in the command line --------*/
9966:
1.186 brouard 9967: /* Main Log file */
1.126 brouard 9968: strcat(filelog, optionfilefiname);
9969: strcat(filelog,".log"); /* */
9970: if((ficlog=fopen(filelog,"w"))==NULL) {
9971: printf("Problem with logfile %s\n",filelog);
9972: goto end;
9973: }
9974: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9975: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9976: fprintf(ficlog,"\nEnter the parameter file name: \n");
9977: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9978: path=%s \n\
9979: optionfile=%s\n\
9980: optionfilext=%s\n\
1.156 brouard 9981: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9982:
1.197 brouard 9983: syscompilerinfo(1);
1.167 brouard 9984:
1.126 brouard 9985: printf("Local time (at start):%s",strstart);
9986: fprintf(ficlog,"Local time (at start): %s",strstart);
9987: fflush(ficlog);
9988: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9989: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9990:
9991: /* */
9992: strcpy(fileres,"r");
9993: strcat(fileres, optionfilefiname);
1.201 brouard 9994: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9995: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9996: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9997:
1.186 brouard 9998: /* Main ---------arguments file --------*/
1.126 brouard 9999:
10000: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10001: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10002: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10003: fflush(ficlog);
1.149 brouard 10004: /* goto end; */
10005: exit(70);
1.126 brouard 10006: }
10007:
10008:
10009:
10010: strcpy(filereso,"o");
1.201 brouard 10011: strcat(filereso,fileresu);
1.126 brouard 10012: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10013: printf("Problem with Output resultfile: %s\n", filereso);
10014: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10015: fflush(ficlog);
10016: goto end;
10017: }
10018:
10019: /* Reads comments: lines beginning with '#' */
10020: numlinepar=0;
1.197 brouard 10021:
10022: /* First parameter line */
10023: while(fgets(line, MAXLINE, ficpar)) {
10024: /* If line starts with a # it is a comment */
10025: if (line[0] == '#') {
10026: numlinepar++;
10027: fputs(line,stdout);
10028: fputs(line,ficparo);
10029: fputs(line,ficlog);
10030: continue;
10031: }else
10032: break;
10033: }
10034: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10035: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10036: if (num_filled != 5) {
10037: printf("Should be 5 parameters\n");
10038: }
1.126 brouard 10039: numlinepar++;
1.197 brouard 10040: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10041: }
10042: /* Second parameter line */
10043: while(fgets(line, MAXLINE, ficpar)) {
10044: /* If line starts with a # it is a comment */
10045: if (line[0] == '#') {
10046: numlinepar++;
10047: fputs(line,stdout);
10048: fputs(line,ficparo);
10049: fputs(line,ficlog);
10050: continue;
10051: }else
10052: break;
10053: }
1.223 brouard 10054: 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", \
10055: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10056: if (num_filled != 11) {
10057: 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 10058: printf("but line=%s\n",line);
1.197 brouard 10059: }
1.223 brouard 10060: 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 10061: }
1.203 brouard 10062: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10063: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10064: /* Third parameter line */
10065: while(fgets(line, MAXLINE, ficpar)) {
10066: /* If line starts with a # it is a comment */
10067: if (line[0] == '#') {
10068: numlinepar++;
10069: fputs(line,stdout);
10070: fputs(line,ficparo);
10071: fputs(line,ficlog);
10072: continue;
10073: }else
10074: break;
10075: }
1.201 brouard 10076: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10077: if (num_filled == 0)
10078: model[0]='\0';
10079: else if (num_filled != 1){
1.197 brouard 10080: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10081: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10082: model[0]='\0';
10083: goto end;
10084: }
10085: else{
10086: if (model[0]=='+'){
10087: for(i=1; i<=strlen(model);i++)
10088: modeltemp[i-1]=model[i];
1.201 brouard 10089: strcpy(model,modeltemp);
1.197 brouard 10090: }
10091: }
1.199 brouard 10092: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10093: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10094: }
10095: /* 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); */
10096: /* numlinepar=numlinepar+3; /\* In general *\/ */
10097: /* 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 10098: 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);
10099: 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 10100: fflush(ficlog);
1.190 brouard 10101: /* if(model[0]=='#'|| model[0]== '\0'){ */
10102: if(model[0]=='#'){
1.187 brouard 10103: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10104: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10105: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10106: if(mle != -1){
10107: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10108: exit(1);
10109: }
10110: }
1.126 brouard 10111: while((c=getc(ficpar))=='#' && c!= EOF){
10112: ungetc(c,ficpar);
10113: fgets(line, MAXLINE, ficpar);
10114: numlinepar++;
1.195 brouard 10115: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10116: z[0]=line[1];
10117: }
10118: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10119: fputs(line, stdout);
10120: //puts(line);
1.126 brouard 10121: fputs(line,ficparo);
10122: fputs(line,ficlog);
10123: }
10124: ungetc(c,ficpar);
10125:
10126:
1.145 brouard 10127: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10128: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10129: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10130: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10131: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10132: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10133: v1+v2*age+v2*v3 makes cptcovn = 3
10134: */
10135: if (strlen(model)>1)
1.187 brouard 10136: 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 10137: else
1.187 brouard 10138: ncovmodel=2; /* Constant and age */
1.133 brouard 10139: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10140: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10141: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10142: 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);
10143: 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);
10144: fflush(stdout);
10145: fclose (ficlog);
10146: goto end;
10147: }
1.126 brouard 10148: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10149: delti=delti3[1][1];
10150: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10151: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10152: /* We could also provide initial parameters values giving by simple logistic regression
10153: * only one way, that is without matrix product. We will have nlstate maximizations */
10154: /* for(i=1;i<nlstate;i++){ */
10155: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10156: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10157: /* } */
1.126 brouard 10158: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10159: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10160: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10161: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10162: fclose (ficparo);
10163: fclose (ficlog);
10164: goto end;
10165: exit(0);
1.220 brouard 10166: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10167: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10168: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10169: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10170: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10171: matcov=matrix(1,npar,1,npar);
1.203 brouard 10172: hess=matrix(1,npar,1,npar);
1.220 brouard 10173: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10174: /* Read guessed parameters */
1.126 brouard 10175: /* Reads comments: lines beginning with '#' */
10176: while((c=getc(ficpar))=='#' && c!= EOF){
10177: ungetc(c,ficpar);
10178: fgets(line, MAXLINE, ficpar);
10179: numlinepar++;
1.141 brouard 10180: fputs(line,stdout);
1.126 brouard 10181: fputs(line,ficparo);
10182: fputs(line,ficlog);
10183: }
10184: ungetc(c,ficpar);
10185:
10186: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10187: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10188: for(i=1; i <=nlstate; i++){
1.234 brouard 10189: j=0;
1.126 brouard 10190: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10191: if(jj==i) continue;
10192: j++;
10193: fscanf(ficpar,"%1d%1d",&i1,&j1);
10194: if ((i1 != i) || (j1 != jj)){
10195: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10196: It might be a problem of design; if ncovcol and the model are correct\n \
10197: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10198: exit(1);
10199: }
10200: fprintf(ficparo,"%1d%1d",i1,j1);
10201: if(mle==1)
10202: printf("%1d%1d",i,jj);
10203: fprintf(ficlog,"%1d%1d",i,jj);
10204: for(k=1; k<=ncovmodel;k++){
10205: fscanf(ficpar," %lf",¶m[i][j][k]);
10206: if(mle==1){
10207: printf(" %lf",param[i][j][k]);
10208: fprintf(ficlog," %lf",param[i][j][k]);
10209: }
10210: else
10211: fprintf(ficlog," %lf",param[i][j][k]);
10212: fprintf(ficparo," %lf",param[i][j][k]);
10213: }
10214: fscanf(ficpar,"\n");
10215: numlinepar++;
10216: if(mle==1)
10217: printf("\n");
10218: fprintf(ficlog,"\n");
10219: fprintf(ficparo,"\n");
1.126 brouard 10220: }
10221: }
10222: fflush(ficlog);
1.234 brouard 10223:
1.251 brouard 10224: /* Reads parameters values */
1.126 brouard 10225: p=param[1][1];
1.251 brouard 10226: pstart=paramstart[1][1];
1.126 brouard 10227:
10228: /* Reads comments: lines beginning with '#' */
10229: while((c=getc(ficpar))=='#' && c!= EOF){
10230: ungetc(c,ficpar);
10231: fgets(line, MAXLINE, ficpar);
10232: numlinepar++;
1.141 brouard 10233: fputs(line,stdout);
1.126 brouard 10234: fputs(line,ficparo);
10235: fputs(line,ficlog);
10236: }
10237: ungetc(c,ficpar);
10238:
10239: for(i=1; i <=nlstate; i++){
10240: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10241: fscanf(ficpar,"%1d%1d",&i1,&j1);
10242: if ( (i1-i) * (j1-j) != 0){
10243: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10244: exit(1);
10245: }
10246: printf("%1d%1d",i,j);
10247: fprintf(ficparo,"%1d%1d",i1,j1);
10248: fprintf(ficlog,"%1d%1d",i1,j1);
10249: for(k=1; k<=ncovmodel;k++){
10250: fscanf(ficpar,"%le",&delti3[i][j][k]);
10251: printf(" %le",delti3[i][j][k]);
10252: fprintf(ficparo," %le",delti3[i][j][k]);
10253: fprintf(ficlog," %le",delti3[i][j][k]);
10254: }
10255: fscanf(ficpar,"\n");
10256: numlinepar++;
10257: printf("\n");
10258: fprintf(ficparo,"\n");
10259: fprintf(ficlog,"\n");
1.126 brouard 10260: }
10261: }
10262: fflush(ficlog);
1.234 brouard 10263:
1.145 brouard 10264: /* Reads covariance matrix */
1.126 brouard 10265: delti=delti3[1][1];
1.220 brouard 10266:
10267:
1.126 brouard 10268: /* 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 10269:
1.126 brouard 10270: /* Reads comments: lines beginning with '#' */
10271: while((c=getc(ficpar))=='#' && c!= EOF){
10272: ungetc(c,ficpar);
10273: fgets(line, MAXLINE, ficpar);
10274: numlinepar++;
1.141 brouard 10275: fputs(line,stdout);
1.126 brouard 10276: fputs(line,ficparo);
10277: fputs(line,ficlog);
10278: }
10279: ungetc(c,ficpar);
1.220 brouard 10280:
1.126 brouard 10281: matcov=matrix(1,npar,1,npar);
1.203 brouard 10282: hess=matrix(1,npar,1,npar);
1.131 brouard 10283: for(i=1; i <=npar; i++)
10284: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10285:
1.194 brouard 10286: /* Scans npar lines */
1.126 brouard 10287: for(i=1; i <=npar; i++){
1.226 brouard 10288: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10289: if(count != 3){
1.226 brouard 10290: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10291: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10292: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10293: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10294: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10295: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10296: exit(1);
1.220 brouard 10297: }else{
1.226 brouard 10298: if(mle==1)
10299: printf("%1d%1d%d",i1,j1,jk);
10300: }
10301: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10302: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10303: for(j=1; j <=i; j++){
1.226 brouard 10304: fscanf(ficpar," %le",&matcov[i][j]);
10305: if(mle==1){
10306: printf(" %.5le",matcov[i][j]);
10307: }
10308: fprintf(ficlog," %.5le",matcov[i][j]);
10309: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10310: }
10311: fscanf(ficpar,"\n");
10312: numlinepar++;
10313: if(mle==1)
1.220 brouard 10314: printf("\n");
1.126 brouard 10315: fprintf(ficlog,"\n");
10316: fprintf(ficparo,"\n");
10317: }
1.194 brouard 10318: /* End of read covariance matrix npar lines */
1.126 brouard 10319: for(i=1; i <=npar; i++)
10320: for(j=i+1;j<=npar;j++)
1.226 brouard 10321: matcov[i][j]=matcov[j][i];
1.126 brouard 10322:
10323: if(mle==1)
10324: printf("\n");
10325: fprintf(ficlog,"\n");
10326:
10327: fflush(ficlog);
10328:
10329: /*-------- Rewriting parameter file ----------*/
10330: strcpy(rfileres,"r"); /* "Rparameterfile */
10331: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10332: strcat(rfileres,"."); /* */
10333: strcat(rfileres,optionfilext); /* Other files have txt extension */
10334: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10335: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10336: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10337: }
10338: fprintf(ficres,"#%s\n",version);
10339: } /* End of mle != -3 */
1.218 brouard 10340:
1.186 brouard 10341: /* Main data
10342: */
1.126 brouard 10343: n= lastobs;
10344: num=lvector(1,n);
10345: moisnais=vector(1,n);
10346: annais=vector(1,n);
10347: moisdc=vector(1,n);
10348: andc=vector(1,n);
1.220 brouard 10349: weight=vector(1,n);
1.126 brouard 10350: agedc=vector(1,n);
10351: cod=ivector(1,n);
1.220 brouard 10352: for(i=1;i<=n;i++){
1.234 brouard 10353: num[i]=0;
10354: moisnais[i]=0;
10355: annais[i]=0;
10356: moisdc[i]=0;
10357: andc[i]=0;
10358: agedc[i]=0;
10359: cod[i]=0;
10360: weight[i]=1.0; /* Equal weights, 1 by default */
10361: }
1.126 brouard 10362: mint=matrix(1,maxwav,1,n);
10363: anint=matrix(1,maxwav,1,n);
1.131 brouard 10364: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10365: tab=ivector(1,NCOVMAX);
1.144 brouard 10366: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10367: 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 10368:
1.136 brouard 10369: /* Reads data from file datafile */
10370: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10371: goto end;
10372:
10373: /* Calculation of the number of parameters from char model */
1.234 brouard 10374: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10375: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10376: k=3 V4 Tvar[k=3]= 4 (from V4)
10377: k=2 V1 Tvar[k=2]= 1 (from V1)
10378: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10379: */
10380:
10381: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10382: TvarsDind=ivector(1,NCOVMAX); /* */
10383: TvarsD=ivector(1,NCOVMAX); /* */
10384: TvarsQind=ivector(1,NCOVMAX); /* */
10385: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10386: TvarF=ivector(1,NCOVMAX); /* */
10387: TvarFind=ivector(1,NCOVMAX); /* */
10388: TvarV=ivector(1,NCOVMAX); /* */
10389: TvarVind=ivector(1,NCOVMAX); /* */
10390: TvarA=ivector(1,NCOVMAX); /* */
10391: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10392: TvarFD=ivector(1,NCOVMAX); /* */
10393: TvarFDind=ivector(1,NCOVMAX); /* */
10394: TvarFQ=ivector(1,NCOVMAX); /* */
10395: TvarFQind=ivector(1,NCOVMAX); /* */
10396: TvarVD=ivector(1,NCOVMAX); /* */
10397: TvarVDind=ivector(1,NCOVMAX); /* */
10398: TvarVQ=ivector(1,NCOVMAX); /* */
10399: TvarVQind=ivector(1,NCOVMAX); /* */
10400:
1.230 brouard 10401: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10402: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10403: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10404: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10405: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10406: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10407: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10408: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10409: */
10410: /* For model-covariate k tells which data-covariate to use but
10411: because this model-covariate is a construction we invent a new column
10412: ncovcol + k1
10413: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10414: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10415: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10416: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10417: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10418: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10419: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10420: */
1.145 brouard 10421: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10422: 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 10423: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10424: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10425: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10426: 4 covariates (3 plus signs)
10427: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10428: */
1.230 brouard 10429: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10430: * individual dummy, fixed or varying:
10431: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10432: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10433: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10434: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10435: * Tmodelind[1]@9={9,0,3,2,}*/
10436: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10437: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10438: * individual quantitative, fixed or varying:
10439: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10440: * 3, 1, 0, 0, 0, 0, 0, 0},
10441: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10442: /* Main decodemodel */
10443:
1.187 brouard 10444:
1.223 brouard 10445: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10446: goto end;
10447:
1.137 brouard 10448: if((double)(lastobs-imx)/(double)imx > 1.10){
10449: nbwarn++;
10450: 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);
10451: 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);
10452: }
1.136 brouard 10453: /* if(mle==1){*/
1.137 brouard 10454: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10455: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10456: }
10457:
10458: /*-calculation of age at interview from date of interview and age at death -*/
10459: agev=matrix(1,maxwav,1,imx);
10460:
10461: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10462: goto end;
10463:
1.126 brouard 10464:
1.136 brouard 10465: agegomp=(int)agemin;
10466: free_vector(moisnais,1,n);
10467: free_vector(annais,1,n);
1.126 brouard 10468: /* free_matrix(mint,1,maxwav,1,n);
10469: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10470: /* free_vector(moisdc,1,n); */
10471: /* free_vector(andc,1,n); */
1.145 brouard 10472: /* */
10473:
1.126 brouard 10474: wav=ivector(1,imx);
1.214 brouard 10475: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10476: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10477: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10478: 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.*/
10479: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10480: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10481:
10482: /* Concatenates waves */
1.214 brouard 10483: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10484: Death is a valid wave (if date is known).
10485: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10486: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10487: and mw[mi+1][i]. dh depends on stepm.
10488: */
10489:
1.126 brouard 10490: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10491: /* Concatenates waves */
1.145 brouard 10492:
1.215 brouard 10493: free_vector(moisdc,1,n);
10494: free_vector(andc,1,n);
10495:
1.126 brouard 10496: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10497: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10498: ncodemax[1]=1;
1.145 brouard 10499: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10500: cptcoveff=0;
1.220 brouard 10501: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10502: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10503: }
10504:
10505: ncovcombmax=pow(2,cptcoveff);
10506: invalidvarcomb=ivector(1, ncovcombmax);
10507: for(i=1;i<ncovcombmax;i++)
10508: invalidvarcomb[i]=0;
10509:
1.211 brouard 10510: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10511: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10512: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10513:
1.200 brouard 10514: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10515: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10516: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10517: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10518: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10519: * (currently 0 or 1) in the data.
10520: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10521: * corresponding modality (h,j).
10522: */
10523:
1.145 brouard 10524: h=0;
10525: /*if (cptcovn > 0) */
1.126 brouard 10526: m=pow(2,cptcoveff);
10527:
1.144 brouard 10528: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10529: * For k=4 covariates, h goes from 1 to m=2**k
10530: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10531: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10532: * h\k 1 2 3 4
1.143 brouard 10533: *______________________________
10534: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10535: * 2 2 1 1 1
10536: * 3 i=2 1 2 1 1
10537: * 4 2 2 1 1
10538: * 5 i=3 1 i=2 1 2 1
10539: * 6 2 1 2 1
10540: * 7 i=4 1 2 2 1
10541: * 8 2 2 2 1
1.197 brouard 10542: * 9 i=5 1 i=3 1 i=2 1 2
10543: * 10 2 1 1 2
10544: * 11 i=6 1 2 1 2
10545: * 12 2 2 1 2
10546: * 13 i=7 1 i=4 1 2 2
10547: * 14 2 1 2 2
10548: * 15 i=8 1 2 2 2
10549: * 16 2 2 2 2
1.143 brouard 10550: */
1.212 brouard 10551: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10552: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10553: * and the value of each covariate?
10554: * V1=1, V2=1, V3=2, V4=1 ?
10555: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10556: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10557: * In order to get the real value in the data, we use nbcode
10558: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10559: * We are keeping this crazy system in order to be able (in the future?)
10560: * to have more than 2 values (0 or 1) for a covariate.
10561: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10562: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10563: * bbbbbbbb
10564: * 76543210
10565: * h-1 00000101 (6-1=5)
1.219 brouard 10566: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10567: * &
10568: * 1 00000001 (1)
1.219 brouard 10569: * 00000000 = 1 & ((h-1) >> (k-1))
10570: * +1= 00000001 =1
1.211 brouard 10571: *
10572: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10573: * h' 1101 =2^3+2^2+0x2^1+2^0
10574: * >>k' 11
10575: * & 00000001
10576: * = 00000001
10577: * +1 = 00000010=2 = codtabm(14,3)
10578: * Reverse h=6 and m=16?
10579: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10580: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10581: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10582: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10583: * V3=decodtabm(14,3,2**4)=2
10584: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10585: *(h-1) >> (j-1) 0011 =13 >> 2
10586: * &1 000000001
10587: * = 000000001
10588: * +1= 000000010 =2
10589: * 2211
10590: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10591: * V3=2
1.220 brouard 10592: * codtabm and decodtabm are identical
1.211 brouard 10593: */
10594:
1.145 brouard 10595:
10596: free_ivector(Ndum,-1,NCOVMAX);
10597:
10598:
1.126 brouard 10599:
1.186 brouard 10600: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10601: strcpy(optionfilegnuplot,optionfilefiname);
10602: if(mle==-3)
1.201 brouard 10603: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10604: strcat(optionfilegnuplot,".gp");
10605:
10606: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10607: printf("Problem with file %s",optionfilegnuplot);
10608: }
10609: else{
1.204 brouard 10610: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10611: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10612: //fprintf(ficgp,"set missing 'NaNq'\n");
10613: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10614: }
10615: /* fclose(ficgp);*/
1.186 brouard 10616:
10617:
10618: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10619:
10620: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10621: if(mle==-3)
1.201 brouard 10622: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10623: strcat(optionfilehtm,".htm");
10624: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10625: printf("Problem with %s \n",optionfilehtm);
10626: exit(0);
1.126 brouard 10627: }
10628:
10629: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10630: strcat(optionfilehtmcov,"-cov.htm");
10631: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10632: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10633: }
10634: else{
10635: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10636: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10637: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10638: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10639: }
10640:
1.213 brouard 10641: 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 10642: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10643: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10644: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10645: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10646: \n\
10647: <hr size=\"2\" color=\"#EC5E5E\">\
10648: <ul><li><h4>Parameter files</h4>\n\
10649: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10650: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10651: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10652: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10653: - Date and time at start: %s</ul>\n",\
10654: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10655: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10656: fileres,fileres,\
10657: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10658: fflush(fichtm);
10659:
10660: strcpy(pathr,path);
10661: strcat(pathr,optionfilefiname);
1.184 brouard 10662: #ifdef WIN32
10663: _chdir(optionfilefiname); /* Move to directory named optionfile */
10664: #else
1.126 brouard 10665: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10666: #endif
10667:
1.126 brouard 10668:
1.220 brouard 10669: /* Calculates basic frequencies. Computes observed prevalence at single age
10670: and for any valid combination of covariates
1.126 brouard 10671: and prints on file fileres'p'. */
1.251 brouard 10672: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10673: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10674:
10675: fprintf(fichtm,"\n");
10676: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10677: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10678: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10679: imx,agemin,agemax,jmin,jmax,jmean);
10680: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10681: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10682: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10683: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10684: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10685:
1.126 brouard 10686: /* For Powell, parameters are in a vector p[] starting at p[1]
10687: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10688: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10689:
10690: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10691: /* For mortality only */
1.126 brouard 10692: if (mle==-3){
1.136 brouard 10693: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10694: for(i=1;i<=NDIM;i++)
10695: for(j=1;j<=NDIM;j++)
10696: ximort[i][j]=0.;
1.186 brouard 10697: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10698: cens=ivector(1,n);
10699: ageexmed=vector(1,n);
10700: agecens=vector(1,n);
10701: dcwave=ivector(1,n);
1.223 brouard 10702:
1.126 brouard 10703: for (i=1; i<=imx; i++){
10704: dcwave[i]=-1;
10705: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10706: if (s[m][i]>nlstate) {
10707: dcwave[i]=m;
10708: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10709: break;
10710: }
1.126 brouard 10711: }
1.226 brouard 10712:
1.126 brouard 10713: for (i=1; i<=imx; i++) {
10714: if (wav[i]>0){
1.226 brouard 10715: ageexmed[i]=agev[mw[1][i]][i];
10716: j=wav[i];
10717: agecens[i]=1.;
10718:
10719: if (ageexmed[i]> 1 && wav[i] > 0){
10720: agecens[i]=agev[mw[j][i]][i];
10721: cens[i]= 1;
10722: }else if (ageexmed[i]< 1)
10723: cens[i]= -1;
10724: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10725: cens[i]=0 ;
1.126 brouard 10726: }
10727: else cens[i]=-1;
10728: }
10729:
10730: for (i=1;i<=NDIM;i++) {
10731: for (j=1;j<=NDIM;j++)
1.226 brouard 10732: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10733: }
10734:
1.145 brouard 10735: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10736: /*printf("%lf %lf", p[1], p[2]);*/
10737:
10738:
1.136 brouard 10739: #ifdef GSL
10740: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10741: #else
1.126 brouard 10742: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10743: #endif
1.201 brouard 10744: strcpy(filerespow,"POW-MORT_");
10745: strcat(filerespow,fileresu);
1.126 brouard 10746: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10747: printf("Problem with resultfile: %s\n", filerespow);
10748: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10749: }
1.136 brouard 10750: #ifdef GSL
10751: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10752: #else
1.126 brouard 10753: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10754: #endif
1.126 brouard 10755: /* for (i=1;i<=nlstate;i++)
10756: for(j=1;j<=nlstate+ndeath;j++)
10757: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10758: */
10759: fprintf(ficrespow,"\n");
1.136 brouard 10760: #ifdef GSL
10761: /* gsl starts here */
10762: T = gsl_multimin_fminimizer_nmsimplex;
10763: gsl_multimin_fminimizer *sfm = NULL;
10764: gsl_vector *ss, *x;
10765: gsl_multimin_function minex_func;
10766:
10767: /* Initial vertex size vector */
10768: ss = gsl_vector_alloc (NDIM);
10769:
10770: if (ss == NULL){
10771: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10772: }
10773: /* Set all step sizes to 1 */
10774: gsl_vector_set_all (ss, 0.001);
10775:
10776: /* Starting point */
1.126 brouard 10777:
1.136 brouard 10778: x = gsl_vector_alloc (NDIM);
10779:
10780: if (x == NULL){
10781: gsl_vector_free(ss);
10782: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10783: }
10784:
10785: /* Initialize method and iterate */
10786: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10787: /* gsl_vector_set(x, 0, 0.0268); */
10788: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10789: gsl_vector_set(x, 0, p[1]);
10790: gsl_vector_set(x, 1, p[2]);
10791:
10792: minex_func.f = &gompertz_f;
10793: minex_func.n = NDIM;
10794: minex_func.params = (void *)&p; /* ??? */
10795:
10796: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10797: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10798:
10799: printf("Iterations beginning .....\n\n");
10800: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10801:
10802: iteri=0;
10803: while (rval == GSL_CONTINUE){
10804: iteri++;
10805: status = gsl_multimin_fminimizer_iterate(sfm);
10806:
10807: if (status) printf("error: %s\n", gsl_strerror (status));
10808: fflush(0);
10809:
10810: if (status)
10811: break;
10812:
10813: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10814: ssval = gsl_multimin_fminimizer_size (sfm);
10815:
10816: if (rval == GSL_SUCCESS)
10817: printf ("converged to a local maximum at\n");
10818:
10819: printf("%5d ", iteri);
10820: for (it = 0; it < NDIM; it++){
10821: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10822: }
10823: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10824: }
10825:
10826: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10827:
10828: gsl_vector_free(x); /* initial values */
10829: gsl_vector_free(ss); /* inital step size */
10830: for (it=0; it<NDIM; it++){
10831: p[it+1]=gsl_vector_get(sfm->x,it);
10832: fprintf(ficrespow," %.12lf", p[it]);
10833: }
10834: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10835: #endif
10836: #ifdef POWELL
10837: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10838: #endif
1.126 brouard 10839: fclose(ficrespow);
10840:
1.203 brouard 10841: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10842:
10843: for(i=1; i <=NDIM; i++)
10844: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10845: matcov[i][j]=matcov[j][i];
1.126 brouard 10846:
10847: printf("\nCovariance matrix\n ");
1.203 brouard 10848: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10849: for(i=1; i <=NDIM; i++) {
10850: for(j=1;j<=NDIM;j++){
1.220 brouard 10851: printf("%f ",matcov[i][j]);
10852: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10853: }
1.203 brouard 10854: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10855: }
10856:
10857: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10858: for (i=1;i<=NDIM;i++) {
1.126 brouard 10859: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10860: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10861: }
1.126 brouard 10862: lsurv=vector(1,AGESUP);
10863: lpop=vector(1,AGESUP);
10864: tpop=vector(1,AGESUP);
10865: lsurv[agegomp]=100000;
10866:
10867: for (k=agegomp;k<=AGESUP;k++) {
10868: agemortsup=k;
10869: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10870: }
10871:
10872: for (k=agegomp;k<agemortsup;k++)
10873: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10874:
10875: for (k=agegomp;k<agemortsup;k++){
10876: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10877: sumlpop=sumlpop+lpop[k];
10878: }
10879:
10880: tpop[agegomp]=sumlpop;
10881: for (k=agegomp;k<(agemortsup-3);k++){
10882: /* tpop[k+1]=2;*/
10883: tpop[k+1]=tpop[k]-lpop[k];
10884: }
10885:
10886:
10887: printf("\nAge lx qx dx Lx Tx e(x)\n");
10888: for (k=agegomp;k<(agemortsup-2);k++)
10889: 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]);
10890:
10891:
10892: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10893: ageminpar=50;
10894: agemaxpar=100;
1.194 brouard 10895: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10896: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10897: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10898: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10899: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10900: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10901: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10902: }else{
10903: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10904: 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 10905: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10906: }
1.201 brouard 10907: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10908: stepm, weightopt,\
10909: model,imx,p,matcov,agemortsup);
10910:
10911: free_vector(lsurv,1,AGESUP);
10912: free_vector(lpop,1,AGESUP);
10913: free_vector(tpop,1,AGESUP);
1.220 brouard 10914: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10915: free_ivector(cens,1,n);
10916: free_vector(agecens,1,n);
10917: free_ivector(dcwave,1,n);
1.220 brouard 10918: #ifdef GSL
1.136 brouard 10919: #endif
1.186 brouard 10920: } /* Endof if mle==-3 mortality only */
1.205 brouard 10921: /* Standard */
10922: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10923: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10924: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10925: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10926: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10927: for (k=1; k<=npar;k++)
10928: printf(" %d %8.5f",k,p[k]);
10929: printf("\n");
1.205 brouard 10930: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10931: /* mlikeli uses func not funcone */
1.247 brouard 10932: /* for(i=1;i<nlstate;i++){ */
10933: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10934: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10935: /* } */
1.205 brouard 10936: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10937: }
10938: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10939: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10940: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10941: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10942: }
10943: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10944: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10945: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10946: for (k=1; k<=npar;k++)
10947: printf(" %d %8.5f",k,p[k]);
10948: printf("\n");
10949:
10950: /*--------- results files --------------*/
1.224 brouard 10951: 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 10952:
10953:
10954: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10955: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10956: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10957: for(i=1,jk=1; i <=nlstate; i++){
10958: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10959: if (k != i) {
10960: printf("%d%d ",i,k);
10961: fprintf(ficlog,"%d%d ",i,k);
10962: fprintf(ficres,"%1d%1d ",i,k);
10963: for(j=1; j <=ncovmodel; j++){
10964: printf("%12.7f ",p[jk]);
10965: fprintf(ficlog,"%12.7f ",p[jk]);
10966: fprintf(ficres,"%12.7f ",p[jk]);
10967: jk++;
10968: }
10969: printf("\n");
10970: fprintf(ficlog,"\n");
10971: fprintf(ficres,"\n");
10972: }
1.126 brouard 10973: }
10974: }
1.203 brouard 10975: if(mle != 0){
10976: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10977: ftolhess=ftol; /* Usually correct */
1.203 brouard 10978: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10979: 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");
10980: 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");
10981: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10982: for(k=1; k <=(nlstate+ndeath); k++){
10983: if (k != i) {
10984: printf("%d%d ",i,k);
10985: fprintf(ficlog,"%d%d ",i,k);
10986: for(j=1; j <=ncovmodel; j++){
10987: 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]));
10988: 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]));
10989: jk++;
10990: }
10991: printf("\n");
10992: fprintf(ficlog,"\n");
10993: }
10994: }
1.193 brouard 10995: }
1.203 brouard 10996: } /* end of hesscov and Wald tests */
1.225 brouard 10997:
1.203 brouard 10998: /* */
1.126 brouard 10999: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11000: printf("# Scales (for hessian or gradient estimation)\n");
11001: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11002: for(i=1,jk=1; i <=nlstate; i++){
11003: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11004: if (j!=i) {
11005: fprintf(ficres,"%1d%1d",i,j);
11006: printf("%1d%1d",i,j);
11007: fprintf(ficlog,"%1d%1d",i,j);
11008: for(k=1; k<=ncovmodel;k++){
11009: printf(" %.5e",delti[jk]);
11010: fprintf(ficlog," %.5e",delti[jk]);
11011: fprintf(ficres," %.5e",delti[jk]);
11012: jk++;
11013: }
11014: printf("\n");
11015: fprintf(ficlog,"\n");
11016: fprintf(ficres,"\n");
11017: }
1.126 brouard 11018: }
11019: }
11020:
11021: 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 11022: if(mle >= 1) /* To big for the screen */
1.126 brouard 11023: 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");
11024: 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");
11025: /* # 121 Var(a12)\n\ */
11026: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11027: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11028: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11029: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11030: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11031: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11032: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11033:
11034:
11035: /* Just to have a covariance matrix which will be more understandable
11036: even is we still don't want to manage dictionary of variables
11037: */
11038: for(itimes=1;itimes<=2;itimes++){
11039: jj=0;
11040: for(i=1; i <=nlstate; i++){
1.225 brouard 11041: for(j=1; j <=nlstate+ndeath; j++){
11042: if(j==i) continue;
11043: for(k=1; k<=ncovmodel;k++){
11044: jj++;
11045: ca[0]= k+'a'-1;ca[1]='\0';
11046: if(itimes==1){
11047: if(mle>=1)
11048: printf("#%1d%1d%d",i,j,k);
11049: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11050: fprintf(ficres,"#%1d%1d%d",i,j,k);
11051: }else{
11052: if(mle>=1)
11053: printf("%1d%1d%d",i,j,k);
11054: fprintf(ficlog,"%1d%1d%d",i,j,k);
11055: fprintf(ficres,"%1d%1d%d",i,j,k);
11056: }
11057: ll=0;
11058: for(li=1;li <=nlstate; li++){
11059: for(lj=1;lj <=nlstate+ndeath; lj++){
11060: if(lj==li) continue;
11061: for(lk=1;lk<=ncovmodel;lk++){
11062: ll++;
11063: if(ll<=jj){
11064: cb[0]= lk +'a'-1;cb[1]='\0';
11065: if(ll<jj){
11066: if(itimes==1){
11067: if(mle>=1)
11068: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11069: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11070: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11071: }else{
11072: if(mle>=1)
11073: printf(" %.5e",matcov[jj][ll]);
11074: fprintf(ficlog," %.5e",matcov[jj][ll]);
11075: fprintf(ficres," %.5e",matcov[jj][ll]);
11076: }
11077: }else{
11078: if(itimes==1){
11079: if(mle>=1)
11080: printf(" Var(%s%1d%1d)",ca,i,j);
11081: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11082: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11083: }else{
11084: if(mle>=1)
11085: printf(" %.7e",matcov[jj][ll]);
11086: fprintf(ficlog," %.7e",matcov[jj][ll]);
11087: fprintf(ficres," %.7e",matcov[jj][ll]);
11088: }
11089: }
11090: }
11091: } /* end lk */
11092: } /* end lj */
11093: } /* end li */
11094: if(mle>=1)
11095: printf("\n");
11096: fprintf(ficlog,"\n");
11097: fprintf(ficres,"\n");
11098: numlinepar++;
11099: } /* end k*/
11100: } /*end j */
1.126 brouard 11101: } /* end i */
11102: } /* end itimes */
11103:
11104: fflush(ficlog);
11105: fflush(ficres);
1.225 brouard 11106: while(fgets(line, MAXLINE, ficpar)) {
11107: /* If line starts with a # it is a comment */
11108: if (line[0] == '#') {
11109: numlinepar++;
11110: fputs(line,stdout);
11111: fputs(line,ficparo);
11112: fputs(line,ficlog);
11113: continue;
11114: }else
11115: break;
11116: }
11117:
1.209 brouard 11118: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11119: /* ungetc(c,ficpar); */
11120: /* fgets(line, MAXLINE, ficpar); */
11121: /* fputs(line,stdout); */
11122: /* fputs(line,ficparo); */
11123: /* } */
11124: /* ungetc(c,ficpar); */
1.126 brouard 11125:
11126: estepm=0;
1.209 brouard 11127: 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 11128:
11129: if (num_filled != 6) {
11130: 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);
11131: 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);
11132: goto end;
11133: }
11134: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11135: }
11136: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11137: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11138:
1.209 brouard 11139: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11140: if (estepm==0 || estepm < stepm) estepm=stepm;
11141: if (fage <= 2) {
11142: bage = ageminpar;
11143: fage = agemaxpar;
11144: }
11145:
11146: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11147: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11148: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11149:
1.186 brouard 11150: /* Other stuffs, more or less useful */
1.254 brouard 11151: while(fgets(line, MAXLINE, ficpar)) {
11152: /* If line starts with a # it is a comment */
11153: if (line[0] == '#') {
11154: numlinepar++;
11155: fputs(line,stdout);
11156: fputs(line,ficparo);
11157: fputs(line,ficlog);
11158: continue;
11159: }else
11160: break;
11161: }
11162:
11163: 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){
11164:
11165: if (num_filled != 7) {
11166: 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);
11167: 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);
11168: goto end;
11169: }
11170: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11171: 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);
11172: 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);
11173: 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 11174: }
1.254 brouard 11175:
11176: while(fgets(line, MAXLINE, ficpar)) {
11177: /* If line starts with a # it is a comment */
11178: if (line[0] == '#') {
11179: numlinepar++;
11180: fputs(line,stdout);
11181: fputs(line,ficparo);
11182: fputs(line,ficlog);
11183: continue;
11184: }else
11185: break;
1.126 brouard 11186: }
11187:
11188:
11189: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11190: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11191:
1.254 brouard 11192: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11193: if (num_filled != 1) {
11194: 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);
11195: 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);
11196: goto end;
11197: }
11198: printf("pop_based=%d\n",popbased);
11199: fprintf(ficlog,"pop_based=%d\n",popbased);
11200: fprintf(ficparo,"pop_based=%d\n",popbased);
11201: fprintf(ficres,"pop_based=%d\n",popbased);
11202: }
11203:
1.258 brouard 11204: /* Results */
11205: nresult=0;
11206: do{
11207: if(!fgets(line, MAXLINE, ficpar)){
11208: endishere=1;
11209: parameterline=14;
11210: }else if (line[0] == '#') {
11211: /* If line starts with a # it is a comment */
1.254 brouard 11212: numlinepar++;
11213: fputs(line,stdout);
11214: fputs(line,ficparo);
11215: fputs(line,ficlog);
11216: continue;
1.258 brouard 11217: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11218: parameterline=11;
11219: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11220: parameterline=12;
11221: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11222: parameterline=13;
11223: else{
11224: parameterline=14;
1.254 brouard 11225: }
1.258 brouard 11226: switch (parameterline){
11227: case 11:
11228: 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){
11229: if (num_filled != 8) {
11230: 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);
11231: 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);
11232: goto end;
11233: }
11234: 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);
11235: 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);
11236: 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);
11237: 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);
11238: /* day and month of proj2 are not used but only year anproj2.*/
11239: }
1.254 brouard 11240: break;
1.258 brouard 11241: case 12:
11242: /*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);*/
11243: 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){
11244: if (num_filled != 8) {
11245: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11246: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 finloal-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11247: goto end;
11248: }
11249: 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);
11250: 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);
11251: 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);
11252: 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);
11253: /* day and month of proj2 are not used but only year anproj2.*/
11254: }
1.230 brouard 11255: break;
1.258 brouard 11256: case 13:
11257: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11258: if (num_filled == 0){
11259: resultline[0]='\0';
11260: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11261: 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);
11262: break;
11263: } else if (num_filled != 1){
11264: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11265: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11266: }
11267: nresult++; /* Sum of resultlines */
11268: printf("Result %d: result=%s\n",nresult, resultline);
11269: if(nresult > MAXRESULTLINES){
11270: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11271: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11272: goto end;
11273: }
11274: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11275: fprintf(ficparo,"result: %s\n",resultline);
11276: fprintf(ficres,"result: %s\n",resultline);
11277: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11278: break;
1.258 brouard 11279: case 14:
1.259 brouard 11280: if(ncovmodel >2 && nresult==0 ){
11281: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11282: goto end;
11283: }
1.259 brouard 11284: break;
1.258 brouard 11285: default:
11286: nresult=1;
11287: decoderesult(".",nresult ); /* No covariate */
11288: }
11289: } /* End switch parameterline */
11290: }while(endishere==0); /* End do */
1.126 brouard 11291:
1.230 brouard 11292: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11293: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11294:
11295: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11296: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11297: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11298: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11299: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11300: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11301: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11302: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11303: }else{
1.218 brouard 11304: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11305: }
11306: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11307: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11308: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11309:
1.225 brouard 11310: /*------------ free_vector -------------*/
11311: /* chdir(path); */
1.220 brouard 11312:
1.215 brouard 11313: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11314: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11315: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11316: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11317: free_lvector(num,1,n);
11318: free_vector(agedc,1,n);
11319: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11320: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11321: fclose(ficparo);
11322: fclose(ficres);
1.220 brouard 11323:
11324:
1.186 brouard 11325: /* Other results (useful)*/
1.220 brouard 11326:
11327:
1.126 brouard 11328: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11329: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11330: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11331: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11332: fclose(ficrespl);
11333:
11334: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11335: /*#include "hpijx.h"*/
11336: hPijx(p, bage, fage);
1.145 brouard 11337: fclose(ficrespij);
1.227 brouard 11338:
1.220 brouard 11339: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11340: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11341: k=1;
1.126 brouard 11342: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11343:
1.219 brouard 11344: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11345: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11346: for(i=1;i<=AGESUP;i++)
1.219 brouard 11347: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11348: for(k=1;k<=ncovcombmax;k++)
11349: probs[i][j][k]=0.;
1.219 brouard 11350: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11351: if (mobilav!=0 ||mobilavproj !=0 ) {
11352: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11353: for(i=1;i<=AGESUP;i++)
11354: for(j=1;j<=nlstate;j++)
11355: for(k=1;k<=ncovcombmax;k++)
11356: mobaverages[i][j][k]=0.;
1.219 brouard 11357: mobaverage=mobaverages;
11358: if (mobilav!=0) {
1.235 brouard 11359: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11360: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11361: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11362: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11363: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11364: }
1.219 brouard 11365: }
11366: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11367: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11368: else if (mobilavproj !=0) {
1.235 brouard 11369: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11370: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11371: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11372: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11373: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11374: }
1.219 brouard 11375: }
11376: }/* end if moving average */
1.227 brouard 11377:
1.126 brouard 11378: /*---------- Forecasting ------------------*/
11379: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11380: if(prevfcast==1){
11381: /* if(stepm ==1){*/
1.225 brouard 11382: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11383: }
1.217 brouard 11384: if(backcast==1){
1.219 brouard 11385: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11386: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11387: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11388:
11389: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11390:
11391: bprlim=matrix(1,nlstate,1,nlstate);
11392: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11393: fclose(ficresplb);
11394:
1.222 brouard 11395: hBijx(p, bage, fage, mobaverage);
11396: fclose(ficrespijb);
1.219 brouard 11397: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11398:
11399: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11400: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11401: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11402: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11403: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11404: }
1.217 brouard 11405:
1.186 brouard 11406:
11407: /* ------ Other prevalence ratios------------ */
1.126 brouard 11408:
1.215 brouard 11409: free_ivector(wav,1,imx);
11410: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11411: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11412: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11413:
11414:
1.127 brouard 11415: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11416:
1.201 brouard 11417: strcpy(filerese,"E_");
11418: strcat(filerese,fileresu);
1.126 brouard 11419: if((ficreseij=fopen(filerese,"w"))==NULL) {
11420: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11421: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11422: }
1.208 brouard 11423: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11424: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11425:
11426: pstamp(ficreseij);
1.219 brouard 11427:
1.235 brouard 11428: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11429: if (cptcovn < 1){i1=1;}
11430:
11431: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11432: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11433: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11434: continue;
1.219 brouard 11435: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11436: printf("\n#****** ");
1.225 brouard 11437: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11438: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11439: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11440: }
11441: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11442: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11443: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11444: }
11445: fprintf(ficreseij,"******\n");
1.235 brouard 11446: printf("******\n");
1.219 brouard 11447:
11448: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11449: oldm=oldms;savm=savms;
1.235 brouard 11450: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11451:
1.219 brouard 11452: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11453: }
11454: fclose(ficreseij);
1.208 brouard 11455: printf("done evsij\n");fflush(stdout);
11456: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11457:
1.227 brouard 11458: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11459:
11460:
1.201 brouard 11461: strcpy(filerest,"T_");
11462: strcat(filerest,fileresu);
1.127 brouard 11463: if((ficrest=fopen(filerest,"w"))==NULL) {
11464: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11465: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11466: }
1.208 brouard 11467: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11468: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11469:
1.126 brouard 11470:
1.201 brouard 11471: strcpy(fileresstde,"STDE_");
11472: strcat(fileresstde,fileresu);
1.126 brouard 11473: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11474: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11475: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11476: }
1.227 brouard 11477: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11478: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11479:
1.201 brouard 11480: strcpy(filerescve,"CVE_");
11481: strcat(filerescve,fileresu);
1.126 brouard 11482: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11483: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11484: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11485: }
1.227 brouard 11486: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11487: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11488:
1.201 brouard 11489: strcpy(fileresv,"V_");
11490: strcat(fileresv,fileresu);
1.126 brouard 11491: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11492: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11493: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11494: }
1.227 brouard 11495: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11496: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11497:
1.145 brouard 11498: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11499: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11500:
1.235 brouard 11501: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11502: if (cptcovn < 1){i1=1;}
11503:
11504: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11505: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11506: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11507: continue;
1.242 brouard 11508: printf("\n#****** Result for:");
11509: fprintf(ficrest,"\n#****** Result for:");
11510: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11511: for(j=1;j<=cptcoveff;j++){
11512: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11513: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11514: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11515: }
1.235 brouard 11516: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11517: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11518: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11519: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11520: }
1.208 brouard 11521: fprintf(ficrest,"******\n");
1.227 brouard 11522: fprintf(ficlog,"******\n");
11523: printf("******\n");
1.208 brouard 11524:
11525: fprintf(ficresstdeij,"\n#****** ");
11526: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11527: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11528: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11529: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11530: }
1.235 brouard 11531: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11532: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11533: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11534: }
1.208 brouard 11535: fprintf(ficresstdeij,"******\n");
11536: fprintf(ficrescveij,"******\n");
11537:
11538: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11539: /* pstamp(ficresvij); */
1.225 brouard 11540: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11541: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11542: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11543: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11544: }
1.208 brouard 11545: fprintf(ficresvij,"******\n");
11546:
11547: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11548: oldm=oldms;savm=savms;
1.235 brouard 11549: printf(" cvevsij ");
11550: fprintf(ficlog, " cvevsij ");
11551: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11552: printf(" end cvevsij \n ");
11553: fprintf(ficlog, " end cvevsij \n ");
11554:
11555: /*
11556: */
11557: /* goto endfree; */
11558:
11559: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11560: pstamp(ficrest);
11561:
11562:
11563: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11564: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11565: cptcod= 0; /* To be deleted */
11566: printf("varevsij vpopbased=%d \n",vpopbased);
11567: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11568: 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 11569: 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 ");
11570: if(vpopbased==1)
11571: 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);
11572: else
11573: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11574: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11575: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11576: fprintf(ficrest,"\n");
11577: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11578: epj=vector(1,nlstate+1);
11579: printf("Computing age specific period (stable) prevalences in each health state \n");
11580: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11581: for(age=bage; age <=fage ;age++){
1.235 brouard 11582: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11583: if (vpopbased==1) {
11584: if(mobilav ==0){
11585: for(i=1; i<=nlstate;i++)
11586: prlim[i][i]=probs[(int)age][i][k];
11587: }else{ /* mobilav */
11588: for(i=1; i<=nlstate;i++)
11589: prlim[i][i]=mobaverage[(int)age][i][k];
11590: }
11591: }
1.219 brouard 11592:
1.227 brouard 11593: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11594: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11595: /* printf(" age %4.0f ",age); */
11596: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11597: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11598: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11599: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11600: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11601: }
11602: epj[nlstate+1] +=epj[j];
11603: }
11604: /* printf(" age %4.0f \n",age); */
1.219 brouard 11605:
1.227 brouard 11606: for(i=1, vepp=0.;i <=nlstate;i++)
11607: for(j=1;j <=nlstate;j++)
11608: vepp += vareij[i][j][(int)age];
11609: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11610: for(j=1;j <=nlstate;j++){
11611: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11612: }
11613: fprintf(ficrest,"\n");
11614: }
1.208 brouard 11615: } /* End vpopbased */
11616: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11617: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11618: free_vector(epj,1,nlstate+1);
1.235 brouard 11619: printf("done selection\n");fflush(stdout);
11620: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11621:
1.145 brouard 11622: /*}*/
1.235 brouard 11623: } /* End k selection */
1.227 brouard 11624:
11625: printf("done State-specific expectancies\n");fflush(stdout);
11626: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11627:
1.126 brouard 11628: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11629:
1.201 brouard 11630: strcpy(fileresvpl,"VPL_");
11631: strcat(fileresvpl,fileresu);
1.126 brouard 11632: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11633: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11634: exit(0);
11635: }
1.208 brouard 11636: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11637: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11638:
1.145 brouard 11639: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11640: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11641:
1.235 brouard 11642: i1=pow(2,cptcoveff);
11643: if (cptcovn < 1){i1=1;}
11644:
11645: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11646: for(k=1; k<=i1;k++){
1.253 brouard 11647: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11648: continue;
1.227 brouard 11649: fprintf(ficresvpl,"\n#****** ");
11650: printf("\n#****** ");
11651: fprintf(ficlog,"\n#****** ");
11652: for(j=1;j<=cptcoveff;j++) {
11653: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11654: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11655: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11656: }
1.235 brouard 11657: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11658: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11659: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11660: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11661: }
1.227 brouard 11662: fprintf(ficresvpl,"******\n");
11663: printf("******\n");
11664: fprintf(ficlog,"******\n");
11665:
11666: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11667: oldm=oldms;savm=savms;
1.235 brouard 11668: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11669: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11670: /*}*/
1.126 brouard 11671: }
1.227 brouard 11672:
1.126 brouard 11673: fclose(ficresvpl);
1.208 brouard 11674: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11675: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11676:
11677: free_vector(weight,1,n);
11678: free_imatrix(Tvard,1,NCOVMAX,1,2);
11679: free_imatrix(s,1,maxwav+1,1,n);
11680: free_matrix(anint,1,maxwav,1,n);
11681: free_matrix(mint,1,maxwav,1,n);
11682: free_ivector(cod,1,n);
11683: free_ivector(tab,1,NCOVMAX);
11684: fclose(ficresstdeij);
11685: fclose(ficrescveij);
11686: fclose(ficresvij);
11687: fclose(ficrest);
11688: fclose(ficpar);
11689:
11690:
1.126 brouard 11691: /*---------- End : free ----------------*/
1.219 brouard 11692: if (mobilav!=0 ||mobilavproj !=0)
11693: 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 11694: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11695: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11696: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11697: } /* mle==-3 arrives here for freeing */
1.227 brouard 11698: /* endfree:*/
11699: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11700: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11701: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11702: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11703: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11704: free_matrix(coqvar,1,maxwav,1,n);
11705: free_matrix(covar,0,NCOVMAX,1,n);
11706: free_matrix(matcov,1,npar,1,npar);
11707: free_matrix(hess,1,npar,1,npar);
11708: /*free_vector(delti,1,npar);*/
11709: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11710: free_matrix(agev,1,maxwav,1,imx);
11711: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11712:
11713: free_ivector(ncodemax,1,NCOVMAX);
11714: free_ivector(ncodemaxwundef,1,NCOVMAX);
11715: free_ivector(Dummy,-1,NCOVMAX);
11716: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11717: free_ivector(DummyV,1,NCOVMAX);
11718: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11719: free_ivector(Typevar,-1,NCOVMAX);
11720: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11721: free_ivector(TvarsQ,1,NCOVMAX);
11722: free_ivector(TvarsQind,1,NCOVMAX);
11723: free_ivector(TvarsD,1,NCOVMAX);
11724: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11725: free_ivector(TvarFD,1,NCOVMAX);
11726: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11727: free_ivector(TvarF,1,NCOVMAX);
11728: free_ivector(TvarFind,1,NCOVMAX);
11729: free_ivector(TvarV,1,NCOVMAX);
11730: free_ivector(TvarVind,1,NCOVMAX);
11731: free_ivector(TvarA,1,NCOVMAX);
11732: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11733: free_ivector(TvarFQ,1,NCOVMAX);
11734: free_ivector(TvarFQind,1,NCOVMAX);
11735: free_ivector(TvarVD,1,NCOVMAX);
11736: free_ivector(TvarVDind,1,NCOVMAX);
11737: free_ivector(TvarVQ,1,NCOVMAX);
11738: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11739: free_ivector(Tvarsel,1,NCOVMAX);
11740: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11741: free_ivector(Tposprod,1,NCOVMAX);
11742: free_ivector(Tprod,1,NCOVMAX);
11743: free_ivector(Tvaraff,1,NCOVMAX);
11744: free_ivector(invalidvarcomb,1,ncovcombmax);
11745: free_ivector(Tage,1,NCOVMAX);
11746: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11747: free_ivector(TmodelInvind,1,NCOVMAX);
11748: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11749:
11750: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11751: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11752: fflush(fichtm);
11753: fflush(ficgp);
11754:
1.227 brouard 11755:
1.126 brouard 11756: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11757: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11758: 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 11759: }else{
11760: printf("End of Imach\n");
11761: fprintf(ficlog,"End of Imach\n");
11762: }
11763: printf("See log file on %s\n",filelog);
11764: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11765: /*(void) gettimeofday(&end_time,&tzp);*/
11766: rend_time = time(NULL);
11767: end_time = *localtime(&rend_time);
11768: /* tml = *localtime(&end_time.tm_sec); */
11769: strcpy(strtend,asctime(&end_time));
1.126 brouard 11770: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11771: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11772: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11773:
1.157 brouard 11774: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11775: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11776: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11777: /* printf("Total time was %d uSec.\n", total_usecs);*/
11778: /* if(fileappend(fichtm,optionfilehtm)){ */
11779: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11780: fclose(fichtm);
11781: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11782: fclose(fichtmcov);
11783: fclose(ficgp);
11784: fclose(ficlog);
11785: /*------ End -----------*/
1.227 brouard 11786:
11787:
11788: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11789: #ifdef WIN32
1.227 brouard 11790: if (_chdir(pathcd) != 0)
11791: printf("Can't move to directory %s!\n",path);
11792: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11793: #else
1.227 brouard 11794: if(chdir(pathcd) != 0)
11795: printf("Can't move to directory %s!\n", path);
11796: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11797: #endif
1.126 brouard 11798: printf("Current directory %s!\n",pathcd);
11799: /*strcat(plotcmd,CHARSEPARATOR);*/
11800: sprintf(plotcmd,"gnuplot");
1.157 brouard 11801: #ifdef _WIN32
1.126 brouard 11802: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11803: #endif
11804: if(!stat(plotcmd,&info)){
1.158 brouard 11805: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11806: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11807: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11808: }else
11809: strcpy(pplotcmd,plotcmd);
1.157 brouard 11810: #ifdef __unix
1.126 brouard 11811: strcpy(plotcmd,GNUPLOTPROGRAM);
11812: if(!stat(plotcmd,&info)){
1.158 brouard 11813: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11814: }else
11815: strcpy(pplotcmd,plotcmd);
11816: #endif
11817: }else
11818: strcpy(pplotcmd,plotcmd);
11819:
11820: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11821: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11822:
1.126 brouard 11823: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11824: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11825: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11826: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11827: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11828: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11829: }
1.158 brouard 11830: printf(" Successful, please wait...");
1.126 brouard 11831: while (z[0] != 'q') {
11832: /* chdir(path); */
1.154 brouard 11833: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11834: scanf("%s",z);
11835: /* if (z[0] == 'c') system("./imach"); */
11836: if (z[0] == 'e') {
1.158 brouard 11837: #ifdef __APPLE__
1.152 brouard 11838: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11839: #elif __linux
11840: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11841: #else
1.152 brouard 11842: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11843: #endif
11844: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11845: system(pplotcmd);
1.126 brouard 11846: }
11847: else if (z[0] == 'g') system(plotcmd);
11848: else if (z[0] == 'q') exit(0);
11849: }
1.227 brouard 11850: end:
1.126 brouard 11851: while (z[0] != 'q') {
1.195 brouard 11852: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11853: scanf("%s",z);
11854: }
11855: }
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