Annotation of imach/src/imach.c, revision 1.258
1.258 ! brouard 1: /* $Id: imach.c,v 1.257 2017/03/29 16:53:30 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.258 ! brouard 4: Revision 1.257 2017/03/29 16:53:30 brouard
! 5: Summary: Temp
! 6:
1.257 brouard 7: Revision 1.256 2017/03/27 05:50:23 brouard
8: Summary: Temporary
9:
1.256 brouard 10: Revision 1.255 2017/03/08 16:02:28 brouard
11: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
12:
1.255 brouard 13: Revision 1.254 2017/03/08 07:13:00 brouard
14: Summary: Fixing data parameter line
15:
1.254 brouard 16: Revision 1.253 2016/12/15 11:59:41 brouard
17: Summary: 0.99 in progress
18:
1.253 brouard 19: Revision 1.252 2016/09/15 21:15:37 brouard
20: *** empty log message ***
21:
1.252 brouard 22: Revision 1.251 2016/09/15 15:01:13 brouard
23: Summary: not working
24:
1.251 brouard 25: Revision 1.250 2016/09/08 16:07:27 brouard
26: Summary: continue
27:
1.250 brouard 28: Revision 1.249 2016/09/07 17:14:18 brouard
29: Summary: Starting values from frequencies
30:
1.249 brouard 31: Revision 1.248 2016/09/07 14:10:18 brouard
32: *** empty log message ***
33:
1.248 brouard 34: Revision 1.247 2016/09/02 11:11:21 brouard
35: *** empty log message ***
36:
1.247 brouard 37: Revision 1.246 2016/09/02 08:49:22 brouard
38: *** empty log message ***
39:
1.246 brouard 40: Revision 1.245 2016/09/02 07:25:01 brouard
41: *** empty log message ***
42:
1.245 brouard 43: Revision 1.244 2016/09/02 07:17:34 brouard
44: *** empty log message ***
45:
1.244 brouard 46: Revision 1.243 2016/09/02 06:45:35 brouard
47: *** empty log message ***
48:
1.243 brouard 49: Revision 1.242 2016/08/30 15:01:20 brouard
50: Summary: Fixing a lots
51:
1.242 brouard 52: Revision 1.241 2016/08/29 17:17:25 brouard
53: Summary: gnuplot problem in Back projection to fix
54:
1.241 brouard 55: Revision 1.240 2016/08/29 07:53:18 brouard
56: Summary: Better
57:
1.240 brouard 58: Revision 1.239 2016/08/26 15:51:03 brouard
59: Summary: Improvement in Powell output in order to copy and paste
60:
61: Author:
62:
1.239 brouard 63: Revision 1.238 2016/08/26 14:23:35 brouard
64: Summary: Starting tests of 0.99
65:
1.238 brouard 66: Revision 1.237 2016/08/26 09:20:19 brouard
67: Summary: to valgrind
68:
1.237 brouard 69: Revision 1.236 2016/08/25 10:50:18 brouard
70: *** empty log message ***
71:
1.236 brouard 72: Revision 1.235 2016/08/25 06:59:23 brouard
73: *** empty log message ***
74:
1.235 brouard 75: Revision 1.234 2016/08/23 16:51:20 brouard
76: *** empty log message ***
77:
1.234 brouard 78: Revision 1.233 2016/08/23 07:40:50 brouard
79: Summary: not working
80:
1.233 brouard 81: Revision 1.232 2016/08/22 14:20:21 brouard
82: Summary: not working
83:
1.232 brouard 84: Revision 1.231 2016/08/22 07:17:15 brouard
85: Summary: not working
86:
1.231 brouard 87: Revision 1.230 2016/08/22 06:55:53 brouard
88: Summary: Not working
89:
1.230 brouard 90: Revision 1.229 2016/07/23 09:45:53 brouard
91: Summary: Completing for func too
92:
1.229 brouard 93: Revision 1.228 2016/07/22 17:45:30 brouard
94: Summary: Fixing some arrays, still debugging
95:
1.227 brouard 96: Revision 1.226 2016/07/12 18:42:34 brouard
97: Summary: temp
98:
1.226 brouard 99: Revision 1.225 2016/07/12 08:40:03 brouard
100: Summary: saving but not running
101:
1.225 brouard 102: Revision 1.224 2016/07/01 13:16:01 brouard
103: Summary: Fixes
104:
1.224 brouard 105: Revision 1.223 2016/02/19 09:23:35 brouard
106: Summary: temporary
107:
1.223 brouard 108: Revision 1.222 2016/02/17 08:14:50 brouard
109: Summary: Probably last 0.98 stable version 0.98r6
110:
1.222 brouard 111: Revision 1.221 2016/02/15 23:35:36 brouard
112: Summary: minor bug
113:
1.220 brouard 114: Revision 1.219 2016/02/15 00:48:12 brouard
115: *** empty log message ***
116:
1.219 brouard 117: Revision 1.218 2016/02/12 11:29:23 brouard
118: Summary: 0.99 Back projections
119:
1.218 brouard 120: Revision 1.217 2015/12/23 17:18:31 brouard
121: Summary: Experimental backcast
122:
1.217 brouard 123: Revision 1.216 2015/12/18 17:32:11 brouard
124: Summary: 0.98r4 Warning and status=-2
125:
126: Version 0.98r4 is now:
127: - displaying an error when status is -1, date of interview unknown and date of death known;
128: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
129: Older changes concerning s=-2, dating from 2005 have been supersed.
130:
1.216 brouard 131: Revision 1.215 2015/12/16 08:52:24 brouard
132: Summary: 0.98r4 working
133:
1.215 brouard 134: Revision 1.214 2015/12/16 06:57:54 brouard
135: Summary: temporary not working
136:
1.214 brouard 137: Revision 1.213 2015/12/11 18:22:17 brouard
138: Summary: 0.98r4
139:
1.213 brouard 140: Revision 1.212 2015/11/21 12:47:24 brouard
141: Summary: minor typo
142:
1.212 brouard 143: Revision 1.211 2015/11/21 12:41:11 brouard
144: Summary: 0.98r3 with some graph of projected cross-sectional
145:
146: Author: Nicolas Brouard
147:
1.211 brouard 148: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 149: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 150: Summary: Adding ftolpl parameter
151: Author: N Brouard
152:
153: We had difficulties to get smoothed confidence intervals. It was due
154: to the period prevalence which wasn't computed accurately. The inner
155: parameter ftolpl is now an outer parameter of the .imach parameter
156: file after estepm. If ftolpl is small 1.e-4 and estepm too,
157: computation are long.
158:
1.209 brouard 159: Revision 1.208 2015/11/17 14:31:57 brouard
160: Summary: temporary
161:
1.208 brouard 162: Revision 1.207 2015/10/27 17:36:57 brouard
163: *** empty log message ***
164:
1.207 brouard 165: Revision 1.206 2015/10/24 07:14:11 brouard
166: *** empty log message ***
167:
1.206 brouard 168: Revision 1.205 2015/10/23 15:50:53 brouard
169: Summary: 0.98r3 some clarification for graphs on likelihood contributions
170:
1.205 brouard 171: Revision 1.204 2015/10/01 16:20:26 brouard
172: Summary: Some new graphs of contribution to likelihood
173:
1.204 brouard 174: Revision 1.203 2015/09/30 17:45:14 brouard
175: Summary: looking at better estimation of the hessian
176:
177: Also a better criteria for convergence to the period prevalence And
178: therefore adding the number of years needed to converge. (The
179: prevalence in any alive state shold sum to one
180:
1.203 brouard 181: Revision 1.202 2015/09/22 19:45:16 brouard
182: Summary: Adding some overall graph on contribution to likelihood. Might change
183:
1.202 brouard 184: Revision 1.201 2015/09/15 17:34:58 brouard
185: Summary: 0.98r0
186:
187: - Some new graphs like suvival functions
188: - Some bugs fixed like model=1+age+V2.
189:
1.201 brouard 190: Revision 1.200 2015/09/09 16:53:55 brouard
191: Summary: Big bug thanks to Flavia
192:
193: Even model=1+age+V2. did not work anymore
194:
1.200 brouard 195: Revision 1.199 2015/09/07 14:09:23 brouard
196: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
197:
1.199 brouard 198: Revision 1.198 2015/09/03 07:14:39 brouard
199: Summary: 0.98q5 Flavia
200:
1.198 brouard 201: Revision 1.197 2015/09/01 18:24:39 brouard
202: *** empty log message ***
203:
1.197 brouard 204: Revision 1.196 2015/08/18 23:17:52 brouard
205: Summary: 0.98q5
206:
1.196 brouard 207: Revision 1.195 2015/08/18 16:28:39 brouard
208: Summary: Adding a hack for testing purpose
209:
210: After reading the title, ftol and model lines, if the comment line has
211: a q, starting with #q, the answer at the end of the run is quit. It
212: permits to run test files in batch with ctest. The former workaround was
213: $ echo q | imach foo.imach
214:
1.195 brouard 215: Revision 1.194 2015/08/18 13:32:00 brouard
216: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
217:
1.194 brouard 218: Revision 1.193 2015/08/04 07:17:42 brouard
219: Summary: 0.98q4
220:
1.193 brouard 221: Revision 1.192 2015/07/16 16:49:02 brouard
222: Summary: Fixing some outputs
223:
1.192 brouard 224: Revision 1.191 2015/07/14 10:00:33 brouard
225: Summary: Some fixes
226:
1.191 brouard 227: Revision 1.190 2015/05/05 08:51:13 brouard
228: Summary: Adding digits in output parameters (7 digits instead of 6)
229:
230: Fix 1+age+.
231:
1.190 brouard 232: Revision 1.189 2015/04/30 14:45:16 brouard
233: Summary: 0.98q2
234:
1.189 brouard 235: Revision 1.188 2015/04/30 08:27:53 brouard
236: *** empty log message ***
237:
1.188 brouard 238: Revision 1.187 2015/04/29 09:11:15 brouard
239: *** empty log message ***
240:
1.187 brouard 241: Revision 1.186 2015/04/23 12:01:52 brouard
242: Summary: V1*age is working now, version 0.98q1
243:
244: Some codes had been disabled in order to simplify and Vn*age was
245: working in the optimization phase, ie, giving correct MLE parameters,
246: but, as usual, outputs were not correct and program core dumped.
247:
1.186 brouard 248: Revision 1.185 2015/03/11 13:26:42 brouard
249: Summary: Inclusion of compile and links command line for Intel Compiler
250:
1.185 brouard 251: Revision 1.184 2015/03/11 11:52:39 brouard
252: Summary: Back from Windows 8. Intel Compiler
253:
1.184 brouard 254: Revision 1.183 2015/03/10 20:34:32 brouard
255: Summary: 0.98q0, trying with directest, mnbrak fixed
256:
257: We use directest instead of original Powell test; probably no
258: incidence on the results, but better justifications;
259: We fixed Numerical Recipes mnbrak routine which was wrong and gave
260: wrong results.
261:
1.183 brouard 262: Revision 1.182 2015/02/12 08:19:57 brouard
263: Summary: Trying to keep directest which seems simpler and more general
264: Author: Nicolas Brouard
265:
1.182 brouard 266: Revision 1.181 2015/02/11 23:22:24 brouard
267: Summary: Comments on Powell added
268:
269: Author:
270:
1.181 brouard 271: Revision 1.180 2015/02/11 17:33:45 brouard
272: Summary: Finishing move from main to function (hpijx and prevalence_limit)
273:
1.180 brouard 274: Revision 1.179 2015/01/04 09:57:06 brouard
275: Summary: back to OS/X
276:
1.179 brouard 277: Revision 1.178 2015/01/04 09:35:48 brouard
278: *** empty log message ***
279:
1.178 brouard 280: Revision 1.177 2015/01/03 18:40:56 brouard
281: Summary: Still testing ilc32 on OSX
282:
1.177 brouard 283: Revision 1.176 2015/01/03 16:45:04 brouard
284: *** empty log message ***
285:
1.176 brouard 286: Revision 1.175 2015/01/03 16:33:42 brouard
287: *** empty log message ***
288:
1.175 brouard 289: Revision 1.174 2015/01/03 16:15:49 brouard
290: Summary: Still in cross-compilation
291:
1.174 brouard 292: Revision 1.173 2015/01/03 12:06:26 brouard
293: Summary: trying to detect cross-compilation
294:
1.173 brouard 295: Revision 1.172 2014/12/27 12:07:47 brouard
296: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
297:
1.172 brouard 298: Revision 1.171 2014/12/23 13:26:59 brouard
299: Summary: Back from Visual C
300:
301: Still problem with utsname.h on Windows
302:
1.171 brouard 303: Revision 1.170 2014/12/23 11:17:12 brouard
304: Summary: Cleaning some \%% back to %%
305:
306: The escape was mandatory for a specific compiler (which one?), but too many warnings.
307:
1.170 brouard 308: Revision 1.169 2014/12/22 23:08:31 brouard
309: Summary: 0.98p
310:
311: Outputs some informations on compiler used, OS etc. Testing on different platforms.
312:
1.169 brouard 313: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 314: Summary: update
1.169 brouard 315:
1.168 brouard 316: Revision 1.167 2014/12/22 13:50:56 brouard
317: Summary: Testing uname and compiler version and if compiled 32 or 64
318:
319: Testing on Linux 64
320:
1.167 brouard 321: Revision 1.166 2014/12/22 11:40:47 brouard
322: *** empty log message ***
323:
1.166 brouard 324: Revision 1.165 2014/12/16 11:20:36 brouard
325: Summary: After compiling on Visual C
326:
327: * imach.c (Module): Merging 1.61 to 1.162
328:
1.165 brouard 329: Revision 1.164 2014/12/16 10:52:11 brouard
330: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
331:
332: * imach.c (Module): Merging 1.61 to 1.162
333:
1.164 brouard 334: Revision 1.163 2014/12/16 10:30:11 brouard
335: * imach.c (Module): Merging 1.61 to 1.162
336:
1.163 brouard 337: Revision 1.162 2014/09/25 11:43:39 brouard
338: Summary: temporary backup 0.99!
339:
1.162 brouard 340: Revision 1.1 2014/09/16 11:06:58 brouard
341: Summary: With some code (wrong) for nlopt
342:
343: Author:
344:
345: Revision 1.161 2014/09/15 20:41:41 brouard
346: Summary: Problem with macro SQR on Intel compiler
347:
1.161 brouard 348: Revision 1.160 2014/09/02 09:24:05 brouard
349: *** empty log message ***
350:
1.160 brouard 351: Revision 1.159 2014/09/01 10:34:10 brouard
352: Summary: WIN32
353: Author: Brouard
354:
1.159 brouard 355: Revision 1.158 2014/08/27 17:11:51 brouard
356: *** empty log message ***
357:
1.158 brouard 358: Revision 1.157 2014/08/27 16:26:55 brouard
359: Summary: Preparing windows Visual studio version
360: Author: Brouard
361:
362: In order to compile on Visual studio, time.h is now correct and time_t
363: and tm struct should be used. difftime should be used but sometimes I
364: just make the differences in raw time format (time(&now).
365: Trying to suppress #ifdef LINUX
366: Add xdg-open for __linux in order to open default browser.
367:
1.157 brouard 368: Revision 1.156 2014/08/25 20:10:10 brouard
369: *** empty log message ***
370:
1.156 brouard 371: Revision 1.155 2014/08/25 18:32:34 brouard
372: Summary: New compile, minor changes
373: Author: Brouard
374:
1.155 brouard 375: Revision 1.154 2014/06/20 17:32:08 brouard
376: Summary: Outputs now all graphs of convergence to period prevalence
377:
1.154 brouard 378: Revision 1.153 2014/06/20 16:45:46 brouard
379: Summary: If 3 live state, convergence to period prevalence on same graph
380: Author: Brouard
381:
1.153 brouard 382: Revision 1.152 2014/06/18 17:54:09 brouard
383: Summary: open browser, use gnuplot on same dir than imach if not found in the path
384:
1.152 brouard 385: Revision 1.151 2014/06/18 16:43:30 brouard
386: *** empty log message ***
387:
1.151 brouard 388: Revision 1.150 2014/06/18 16:42:35 brouard
389: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
390: Author: brouard
391:
1.150 brouard 392: Revision 1.149 2014/06/18 15:51:14 brouard
393: Summary: Some fixes in parameter files errors
394: Author: Nicolas Brouard
395:
1.149 brouard 396: Revision 1.148 2014/06/17 17:38:48 brouard
397: Summary: Nothing new
398: Author: Brouard
399:
400: Just a new packaging for OS/X version 0.98nS
401:
1.148 brouard 402: Revision 1.147 2014/06/16 10:33:11 brouard
403: *** empty log message ***
404:
1.147 brouard 405: Revision 1.146 2014/06/16 10:20:28 brouard
406: Summary: Merge
407: Author: Brouard
408:
409: Merge, before building revised version.
410:
1.146 brouard 411: Revision 1.145 2014/06/10 21:23:15 brouard
412: Summary: Debugging with valgrind
413: Author: Nicolas Brouard
414:
415: Lot of changes in order to output the results with some covariates
416: After the Edimburgh REVES conference 2014, it seems mandatory to
417: improve the code.
418: No more memory valgrind error but a lot has to be done in order to
419: continue the work of splitting the code into subroutines.
420: Also, decodemodel has been improved. Tricode is still not
421: optimal. nbcode should be improved. Documentation has been added in
422: the source code.
423:
1.144 brouard 424: Revision 1.143 2014/01/26 09:45:38 brouard
425: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
426:
427: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
428: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
429:
1.143 brouard 430: Revision 1.142 2014/01/26 03:57:36 brouard
431: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
432:
433: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
434:
1.142 brouard 435: Revision 1.141 2014/01/26 02:42:01 brouard
436: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
437:
1.141 brouard 438: Revision 1.140 2011/09/02 10:37:54 brouard
439: Summary: times.h is ok with mingw32 now.
440:
1.140 brouard 441: Revision 1.139 2010/06/14 07:50:17 brouard
442: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
443: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
444:
1.139 brouard 445: Revision 1.138 2010/04/30 18:19:40 brouard
446: *** empty log message ***
447:
1.138 brouard 448: Revision 1.137 2010/04/29 18:11:38 brouard
449: (Module): Checking covariates for more complex models
450: than V1+V2. A lot of change to be done. Unstable.
451:
1.137 brouard 452: Revision 1.136 2010/04/26 20:30:53 brouard
453: (Module): merging some libgsl code. Fixing computation
454: of likelione (using inter/intrapolation if mle = 0) in order to
455: get same likelihood as if mle=1.
456: Some cleaning of code and comments added.
457:
1.136 brouard 458: Revision 1.135 2009/10/29 15:33:14 brouard
459: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
460:
1.135 brouard 461: Revision 1.134 2009/10/29 13:18:53 brouard
462: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
463:
1.134 brouard 464: Revision 1.133 2009/07/06 10:21:25 brouard
465: just nforces
466:
1.133 brouard 467: Revision 1.132 2009/07/06 08:22:05 brouard
468: Many tings
469:
1.132 brouard 470: Revision 1.131 2009/06/20 16:22:47 brouard
471: Some dimensions resccaled
472:
1.131 brouard 473: Revision 1.130 2009/05/26 06:44:34 brouard
474: (Module): Max Covariate is now set to 20 instead of 8. A
475: lot of cleaning with variables initialized to 0. Trying to make
476: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
477:
1.130 brouard 478: Revision 1.129 2007/08/31 13:49:27 lievre
479: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
480:
1.129 lievre 481: Revision 1.128 2006/06/30 13:02:05 brouard
482: (Module): Clarifications on computing e.j
483:
1.128 brouard 484: Revision 1.127 2006/04/28 18:11:50 brouard
485: (Module): Yes the sum of survivors was wrong since
486: imach-114 because nhstepm was no more computed in the age
487: loop. Now we define nhstepma in the age loop.
488: (Module): In order to speed up (in case of numerous covariates) we
489: compute health expectancies (without variances) in a first step
490: and then all the health expectancies with variances or standard
491: deviation (needs data from the Hessian matrices) which slows the
492: computation.
493: In the future we should be able to stop the program is only health
494: expectancies and graph are needed without standard deviations.
495:
1.127 brouard 496: Revision 1.126 2006/04/28 17:23:28 brouard
497: (Module): Yes the sum of survivors was wrong since
498: imach-114 because nhstepm was no more computed in the age
499: loop. Now we define nhstepma in the age loop.
500: Version 0.98h
501:
1.126 brouard 502: Revision 1.125 2006/04/04 15:20:31 lievre
503: Errors in calculation of health expectancies. Age was not initialized.
504: Forecasting file added.
505:
506: Revision 1.124 2006/03/22 17:13:53 lievre
507: Parameters are printed with %lf instead of %f (more numbers after the comma).
508: The log-likelihood is printed in the log file
509:
510: Revision 1.123 2006/03/20 10:52:43 brouard
511: * imach.c (Module): <title> changed, corresponds to .htm file
512: name. <head> headers where missing.
513:
514: * imach.c (Module): Weights can have a decimal point as for
515: English (a comma might work with a correct LC_NUMERIC environment,
516: otherwise the weight is truncated).
517: Modification of warning when the covariates values are not 0 or
518: 1.
519: Version 0.98g
520:
521: Revision 1.122 2006/03/20 09:45:41 brouard
522: (Module): Weights can have a decimal point as for
523: English (a comma might work with a correct LC_NUMERIC environment,
524: otherwise the weight is truncated).
525: Modification of warning when the covariates values are not 0 or
526: 1.
527: Version 0.98g
528:
529: Revision 1.121 2006/03/16 17:45:01 lievre
530: * imach.c (Module): Comments concerning covariates added
531:
532: * imach.c (Module): refinements in the computation of lli if
533: status=-2 in order to have more reliable computation if stepm is
534: not 1 month. Version 0.98f
535:
536: Revision 1.120 2006/03/16 15:10:38 lievre
537: (Module): refinements in the computation of lli if
538: status=-2 in order to have more reliable computation if stepm is
539: not 1 month. Version 0.98f
540:
541: Revision 1.119 2006/03/15 17:42:26 brouard
542: (Module): Bug if status = -2, the loglikelihood was
543: computed as likelihood omitting the logarithm. Version O.98e
544:
545: Revision 1.118 2006/03/14 18:20:07 brouard
546: (Module): varevsij Comments added explaining the second
547: table of variances if popbased=1 .
548: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
549: (Module): Function pstamp added
550: (Module): Version 0.98d
551:
552: Revision 1.117 2006/03/14 17:16:22 brouard
553: (Module): varevsij Comments added explaining the second
554: table of variances if popbased=1 .
555: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
556: (Module): Function pstamp added
557: (Module): Version 0.98d
558:
559: Revision 1.116 2006/03/06 10:29:27 brouard
560: (Module): Variance-covariance wrong links and
561: varian-covariance of ej. is needed (Saito).
562:
563: Revision 1.115 2006/02/27 12:17:45 brouard
564: (Module): One freematrix added in mlikeli! 0.98c
565:
566: Revision 1.114 2006/02/26 12:57:58 brouard
567: (Module): Some improvements in processing parameter
568: filename with strsep.
569:
570: Revision 1.113 2006/02/24 14:20:24 brouard
571: (Module): Memory leaks checks with valgrind and:
572: datafile was not closed, some imatrix were not freed and on matrix
573: allocation too.
574:
575: Revision 1.112 2006/01/30 09:55:26 brouard
576: (Module): Back to gnuplot.exe instead of wgnuplot.exe
577:
578: Revision 1.111 2006/01/25 20:38:18 brouard
579: (Module): Lots of cleaning and bugs added (Gompertz)
580: (Module): Comments can be added in data file. Missing date values
581: can be a simple dot '.'.
582:
583: Revision 1.110 2006/01/25 00:51:50 brouard
584: (Module): Lots of cleaning and bugs added (Gompertz)
585:
586: Revision 1.109 2006/01/24 19:37:15 brouard
587: (Module): Comments (lines starting with a #) are allowed in data.
588:
589: Revision 1.108 2006/01/19 18:05:42 lievre
590: Gnuplot problem appeared...
591: To be fixed
592:
593: Revision 1.107 2006/01/19 16:20:37 brouard
594: Test existence of gnuplot in imach path
595:
596: Revision 1.106 2006/01/19 13:24:36 brouard
597: Some cleaning and links added in html output
598:
599: Revision 1.105 2006/01/05 20:23:19 lievre
600: *** empty log message ***
601:
602: Revision 1.104 2005/09/30 16:11:43 lievre
603: (Module): sump fixed, loop imx fixed, and simplifications.
604: (Module): If the status is missing at the last wave but we know
605: that the person is alive, then we can code his/her status as -2
606: (instead of missing=-1 in earlier versions) and his/her
607: contributions to the likelihood is 1 - Prob of dying from last
608: health status (= 1-p13= p11+p12 in the easiest case of somebody in
609: the healthy state at last known wave). Version is 0.98
610:
611: Revision 1.103 2005/09/30 15:54:49 lievre
612: (Module): sump fixed, loop imx fixed, and simplifications.
613:
614: Revision 1.102 2004/09/15 17:31:30 brouard
615: Add the possibility to read data file including tab characters.
616:
617: Revision 1.101 2004/09/15 10:38:38 brouard
618: Fix on curr_time
619:
620: Revision 1.100 2004/07/12 18:29:06 brouard
621: Add version for Mac OS X. Just define UNIX in Makefile
622:
623: Revision 1.99 2004/06/05 08:57:40 brouard
624: *** empty log message ***
625:
626: Revision 1.98 2004/05/16 15:05:56 brouard
627: New version 0.97 . First attempt to estimate force of mortality
628: directly from the data i.e. without the need of knowing the health
629: state at each age, but using a Gompertz model: log u =a + b*age .
630: This is the basic analysis of mortality and should be done before any
631: other analysis, in order to test if the mortality estimated from the
632: cross-longitudinal survey is different from the mortality estimated
633: from other sources like vital statistic data.
634:
635: The same imach parameter file can be used but the option for mle should be -3.
636:
1.133 brouard 637: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 638: former routines in order to include the new code within the former code.
639:
640: The output is very simple: only an estimate of the intercept and of
641: the slope with 95% confident intervals.
642:
643: Current limitations:
644: A) Even if you enter covariates, i.e. with the
645: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
646: B) There is no computation of Life Expectancy nor Life Table.
647:
648: Revision 1.97 2004/02/20 13:25:42 lievre
649: Version 0.96d. Population forecasting command line is (temporarily)
650: suppressed.
651:
652: Revision 1.96 2003/07/15 15:38:55 brouard
653: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
654: rewritten within the same printf. Workaround: many printfs.
655:
656: Revision 1.95 2003/07/08 07:54:34 brouard
657: * imach.c (Repository):
658: (Repository): Using imachwizard code to output a more meaningful covariance
659: matrix (cov(a12,c31) instead of numbers.
660:
661: Revision 1.94 2003/06/27 13:00:02 brouard
662: Just cleaning
663:
664: Revision 1.93 2003/06/25 16:33:55 brouard
665: (Module): On windows (cygwin) function asctime_r doesn't
666: exist so I changed back to asctime which exists.
667: (Module): Version 0.96b
668:
669: Revision 1.92 2003/06/25 16:30:45 brouard
670: (Module): On windows (cygwin) function asctime_r doesn't
671: exist so I changed back to asctime which exists.
672:
673: Revision 1.91 2003/06/25 15:30:29 brouard
674: * imach.c (Repository): Duplicated warning errors corrected.
675: (Repository): Elapsed time after each iteration is now output. It
676: helps to forecast when convergence will be reached. Elapsed time
677: is stamped in powell. We created a new html file for the graphs
678: concerning matrix of covariance. It has extension -cov.htm.
679:
680: Revision 1.90 2003/06/24 12:34:15 brouard
681: (Module): Some bugs corrected for windows. Also, when
682: mle=-1 a template is output in file "or"mypar.txt with the design
683: of the covariance matrix to be input.
684:
685: Revision 1.89 2003/06/24 12:30:52 brouard
686: (Module): Some bugs corrected for windows. Also, when
687: mle=-1 a template is output in file "or"mypar.txt with the design
688: of the covariance matrix to be input.
689:
690: Revision 1.88 2003/06/23 17:54:56 brouard
691: * 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.
692:
693: Revision 1.87 2003/06/18 12:26:01 brouard
694: Version 0.96
695:
696: Revision 1.86 2003/06/17 20:04:08 brouard
697: (Module): Change position of html and gnuplot routines and added
698: routine fileappend.
699:
700: Revision 1.85 2003/06/17 13:12:43 brouard
701: * imach.c (Repository): Check when date of death was earlier that
702: current date of interview. It may happen when the death was just
703: prior to the death. In this case, dh was negative and likelihood
704: was wrong (infinity). We still send an "Error" but patch by
705: assuming that the date of death was just one stepm after the
706: interview.
707: (Repository): Because some people have very long ID (first column)
708: we changed int to long in num[] and we added a new lvector for
709: memory allocation. But we also truncated to 8 characters (left
710: truncation)
711: (Repository): No more line truncation errors.
712:
713: Revision 1.84 2003/06/13 21:44:43 brouard
714: * imach.c (Repository): Replace "freqsummary" at a correct
715: place. It differs from routine "prevalence" which may be called
716: many times. Probs is memory consuming and must be used with
717: parcimony.
718: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
719:
720: Revision 1.83 2003/06/10 13:39:11 lievre
721: *** empty log message ***
722:
723: Revision 1.82 2003/06/05 15:57:20 brouard
724: Add log in imach.c and fullversion number is now printed.
725:
726: */
727: /*
728: Interpolated Markov Chain
729:
730: Short summary of the programme:
731:
1.227 brouard 732: This program computes Healthy Life Expectancies or State-specific
733: (if states aren't health statuses) Expectancies from
734: cross-longitudinal data. Cross-longitudinal data consist in:
735:
736: -1- a first survey ("cross") where individuals from different ages
737: are interviewed on their health status or degree of disability (in
738: the case of a health survey which is our main interest)
739:
740: -2- at least a second wave of interviews ("longitudinal") which
741: measure each change (if any) in individual health status. Health
742: expectancies are computed from the time spent in each health state
743: according to a model. More health states you consider, more time is
744: necessary to reach the Maximum Likelihood of the parameters involved
745: in the model. The simplest model is the multinomial logistic model
746: where pij is the probability to be observed in state j at the second
747: wave conditional to be observed in state i at the first
748: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
749: etc , where 'age' is age and 'sex' is a covariate. If you want to
750: have a more complex model than "constant and age", you should modify
751: the program where the markup *Covariates have to be included here
752: again* invites you to do it. More covariates you add, slower the
1.126 brouard 753: convergence.
754:
755: The advantage of this computer programme, compared to a simple
756: multinomial logistic model, is clear when the delay between waves is not
757: identical for each individual. Also, if a individual missed an
758: intermediate interview, the information is lost, but taken into
759: account using an interpolation or extrapolation.
760:
761: hPijx is the probability to be observed in state i at age x+h
762: conditional to the observed state i at age x. The delay 'h' can be
763: split into an exact number (nh*stepm) of unobserved intermediate
764: states. This elementary transition (by month, quarter,
765: semester or year) is modelled as a multinomial logistic. The hPx
766: matrix is simply the matrix product of nh*stepm elementary matrices
767: and the contribution of each individual to the likelihood is simply
768: hPijx.
769:
770: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 771: of the life expectancies. It also computes the period (stable) prevalence.
772:
773: Back prevalence and projections:
1.227 brouard 774:
775: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
776: double agemaxpar, double ftolpl, int *ncvyearp, double
777: dateprev1,double dateprev2, int firstpass, int lastpass, int
778: mobilavproj)
779:
780: Computes the back prevalence limit for any combination of
781: covariate values k at any age between ageminpar and agemaxpar and
782: returns it in **bprlim. In the loops,
783:
784: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
785: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
786:
787: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 788: Computes for any combination of covariates k and any age between bage and fage
789: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
790: oldm=oldms;savm=savms;
1.227 brouard 791:
792: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 793: Computes the transition matrix starting at age 'age' over
794: 'nhstepm*hstepm*stepm' months (i.e. until
795: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 796: nhstepm*hstepm matrices.
797:
798: Returns p3mat[i][j][h] after calling
799: p3mat[i][j][h]=matprod2(newm,
800: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
801: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
802: oldm);
1.226 brouard 803:
804: Important routines
805:
806: - func (or funcone), computes logit (pij) distinguishing
807: o fixed variables (single or product dummies or quantitative);
808: o varying variables by:
809: (1) wave (single, product dummies, quantitative),
810: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
811: % fixed dummy (treated) or quantitative (not done because time-consuming);
812: % varying dummy (not done) or quantitative (not done);
813: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
814: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
815: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
816: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
817: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 818:
1.226 brouard 819:
820:
1.133 brouard 821: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
822: Institut national d'études démographiques, Paris.
1.126 brouard 823: This software have been partly granted by Euro-REVES, a concerted action
824: from the European Union.
825: It is copyrighted identically to a GNU software product, ie programme and
826: software can be distributed freely for non commercial use. Latest version
827: can be accessed at http://euroreves.ined.fr/imach .
828:
829: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
830: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
831:
832: **********************************************************************/
833: /*
834: main
835: read parameterfile
836: read datafile
837: concatwav
838: freqsummary
839: if (mle >= 1)
840: mlikeli
841: print results files
842: if mle==1
843: computes hessian
844: read end of parameter file: agemin, agemax, bage, fage, estepm
845: begin-prev-date,...
846: open gnuplot file
847: open html file
1.145 brouard 848: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
849: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
850: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
851: freexexit2 possible for memory heap.
852:
853: h Pij x | pij_nom ficrestpij
854: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
855: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
856: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
857:
858: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
859: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
860: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
861: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
862: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
863:
1.126 brouard 864: forecasting if prevfcast==1 prevforecast call prevalence()
865: health expectancies
866: Variance-covariance of DFLE
867: prevalence()
868: movingaverage()
869: varevsij()
870: if popbased==1 varevsij(,popbased)
871: total life expectancies
872: Variance of period (stable) prevalence
873: end
874: */
875:
1.187 brouard 876: /* #define DEBUG */
877: /* #define DEBUGBRENT */
1.203 brouard 878: /* #define DEBUGLINMIN */
879: /* #define DEBUGHESS */
880: #define DEBUGHESSIJ
1.224 brouard 881: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 882: #define POWELL /* Instead of NLOPT */
1.224 brouard 883: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 884: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
885: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 886:
887: #include <math.h>
888: #include <stdio.h>
889: #include <stdlib.h>
890: #include <string.h>
1.226 brouard 891: #include <ctype.h>
1.159 brouard 892:
893: #ifdef _WIN32
894: #include <io.h>
1.172 brouard 895: #include <windows.h>
896: #include <tchar.h>
1.159 brouard 897: #else
1.126 brouard 898: #include <unistd.h>
1.159 brouard 899: #endif
1.126 brouard 900:
901: #include <limits.h>
902: #include <sys/types.h>
1.171 brouard 903:
904: #if defined(__GNUC__)
905: #include <sys/utsname.h> /* Doesn't work on Windows */
906: #endif
907:
1.126 brouard 908: #include <sys/stat.h>
909: #include <errno.h>
1.159 brouard 910: /* extern int errno; */
1.126 brouard 911:
1.157 brouard 912: /* #ifdef LINUX */
913: /* #include <time.h> */
914: /* #include "timeval.h" */
915: /* #else */
916: /* #include <sys/time.h> */
917: /* #endif */
918:
1.126 brouard 919: #include <time.h>
920:
1.136 brouard 921: #ifdef GSL
922: #include <gsl/gsl_errno.h>
923: #include <gsl/gsl_multimin.h>
924: #endif
925:
1.167 brouard 926:
1.162 brouard 927: #ifdef NLOPT
928: #include <nlopt.h>
929: typedef struct {
930: double (* function)(double [] );
931: } myfunc_data ;
932: #endif
933:
1.126 brouard 934: /* #include <libintl.h> */
935: /* #define _(String) gettext (String) */
936:
1.251 brouard 937: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 938:
939: #define GNUPLOTPROGRAM "gnuplot"
940: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
941: #define FILENAMELENGTH 132
942:
943: #define GLOCK_ERROR_NOPATH -1 /* empty path */
944: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
945:
1.144 brouard 946: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
947: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 948:
949: #define NINTERVMAX 8
1.144 brouard 950: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
951: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
952: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 953: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 954: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
955: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 956: #define MAXN 20000
1.144 brouard 957: #define YEARM 12. /**< Number of months per year */
1.218 brouard 958: /* #define AGESUP 130 */
959: #define AGESUP 150
960: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 961: #define AGEBASE 40
1.194 brouard 962: #define AGEOVERFLOW 1.e20
1.164 brouard 963: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 964: #ifdef _WIN32
965: #define DIRSEPARATOR '\\'
966: #define CHARSEPARATOR "\\"
967: #define ODIRSEPARATOR '/'
968: #else
1.126 brouard 969: #define DIRSEPARATOR '/'
970: #define CHARSEPARATOR "/"
971: #define ODIRSEPARATOR '\\'
972: #endif
973:
1.258 ! brouard 974: /* $Id: imach.c,v 1.257 2017/03/29 16:53:30 brouard Exp $ */
1.126 brouard 975: /* $State: Exp $ */
1.196 brouard 976: #include "version.h"
977: char version[]=__IMACH_VERSION__;
1.224 brouard 978: 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";
1.258 ! brouard 979: char fullversion[]="$Revision: 1.257 $ $Date: 2017/03/29 16:53:30 $";
1.126 brouard 980: char strstart[80];
981: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 982: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 983: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 984: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
985: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
986: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 987: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
988: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 989: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
990: int cptcovprodnoage=0; /**< Number of covariate products without age */
991: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 992: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
993: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 994: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 995: int nsd=0; /**< Total number of single dummy variables (output) */
996: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 997: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 998: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 999: int ntveff=0; /**< ntveff number of effective time varying variables */
1000: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1001: int cptcov=0; /* Working variable */
1.218 brouard 1002: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1003: int npar=NPARMAX;
1004: int nlstate=2; /* Number of live states */
1005: int ndeath=1; /* Number of dead states */
1.130 brouard 1006: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1007: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1008: int popbased=0;
1009:
1010: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1011: int maxwav=0; /* Maxim number of waves */
1012: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1013: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1014: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1015: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1016: int mle=1, weightopt=0;
1.126 brouard 1017: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1018: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1019: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1020: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1021: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1022: int selected(int kvar); /* Is covariate kvar selected for printing results */
1023:
1.130 brouard 1024: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1025: double **matprod2(); /* test */
1.126 brouard 1026: double **oldm, **newm, **savm; /* Working pointers to matrices */
1027: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1028: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1029:
1.136 brouard 1030: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1031: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1032: FILE *ficlog, *ficrespow;
1.130 brouard 1033: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1034: double fretone; /* Only one call to likelihood */
1.130 brouard 1035: long ipmx=0; /* Number of contributions */
1.126 brouard 1036: double sw; /* Sum of weights */
1037: char filerespow[FILENAMELENGTH];
1038: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1039: FILE *ficresilk;
1040: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1041: FILE *ficresprobmorprev;
1042: FILE *fichtm, *fichtmcov; /* Html File */
1043: FILE *ficreseij;
1044: char filerese[FILENAMELENGTH];
1045: FILE *ficresstdeij;
1046: char fileresstde[FILENAMELENGTH];
1047: FILE *ficrescveij;
1048: char filerescve[FILENAMELENGTH];
1049: FILE *ficresvij;
1050: char fileresv[FILENAMELENGTH];
1051: FILE *ficresvpl;
1052: char fileresvpl[FILENAMELENGTH];
1053: char title[MAXLINE];
1.234 brouard 1054: char model[MAXLINE]; /**< The model line */
1.217 brouard 1055: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1056: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1057: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1058: char command[FILENAMELENGTH];
1059: int outcmd=0;
1060:
1.217 brouard 1061: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1062: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1063: char filelog[FILENAMELENGTH]; /* Log file */
1064: char filerest[FILENAMELENGTH];
1065: char fileregp[FILENAMELENGTH];
1066: char popfile[FILENAMELENGTH];
1067:
1068: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1069:
1.157 brouard 1070: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1071: /* struct timezone tzp; */
1072: /* extern int gettimeofday(); */
1073: struct tm tml, *gmtime(), *localtime();
1074:
1075: extern time_t time();
1076:
1077: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1078: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1079: struct tm tm;
1080:
1.126 brouard 1081: char strcurr[80], strfor[80];
1082:
1083: char *endptr;
1084: long lval;
1085: double dval;
1086:
1087: #define NR_END 1
1088: #define FREE_ARG char*
1089: #define FTOL 1.0e-10
1090:
1091: #define NRANSI
1.240 brouard 1092: #define ITMAX 200
1093: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1094:
1095: #define TOL 2.0e-4
1096:
1097: #define CGOLD 0.3819660
1098: #define ZEPS 1.0e-10
1099: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1100:
1101: #define GOLD 1.618034
1102: #define GLIMIT 100.0
1103: #define TINY 1.0e-20
1104:
1105: static double maxarg1,maxarg2;
1106: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1107: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1108:
1109: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1110: #define rint(a) floor(a+0.5)
1.166 brouard 1111: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1112: #define mytinydouble 1.0e-16
1.166 brouard 1113: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1114: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1115: /* static double dsqrarg; */
1116: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1117: static double sqrarg;
1118: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1119: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1120: int agegomp= AGEGOMP;
1121:
1122: int imx;
1123: int stepm=1;
1124: /* Stepm, step in month: minimum step interpolation*/
1125:
1126: int estepm;
1127: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1128:
1129: int m,nb;
1130: long *num;
1.197 brouard 1131: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1132: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1133: covariate for which somebody answered excluding
1134: undefined. Usually 2: 0 and 1. */
1135: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1136: covariate for which somebody answered including
1137: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1138: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1139: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1140: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1141: double *ageexmed,*agecens;
1142: double dateintmean=0;
1143:
1144: double *weight;
1145: int **s; /* Status */
1.141 brouard 1146: double *agedc;
1.145 brouard 1147: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1148: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1149: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1150: double **coqvar; /* Fixed quantitative covariate iqv */
1151: double ***cotvar; /* Time varying covariate itv */
1152: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1153: double idx;
1154: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1155: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1156: /*k 1 2 3 4 5 6 7 8 9 */
1157: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1158: /* Tndvar[k] 1 2 3 4 5 */
1159: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1160: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1161: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1162: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1163: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1164: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1165: /* Tprod[i]=k 4 7 */
1166: /* Tage[i]=k 5 8 */
1167: /* */
1168: /* Type */
1169: /* V 1 2 3 4 5 */
1170: /* F F V V V */
1171: /* D Q D D Q */
1172: /* */
1173: int *TvarsD;
1174: int *TvarsDind;
1175: int *TvarsQ;
1176: int *TvarsQind;
1177:
1.235 brouard 1178: #define MAXRESULTLINES 10
1179: int nresult=0;
1.258 ! brouard 1180: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1181: int TKresult[MAXRESULTLINES];
1.237 brouard 1182: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1183: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1184: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1185: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1186: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1187: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1188:
1.234 brouard 1189: /* 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 1190: 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 */
1191: 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 */
1192: 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 */
1193: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1194: 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 */
1195: 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 1196: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1197: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1198: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1199: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1200: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1201: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1202: 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 */
1203: 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 */
1204:
1.230 brouard 1205: int *Tvarsel; /**< Selected covariates for output */
1206: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1207: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1208: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1209: 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 1210: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1211: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1212: int *Tage;
1.227 brouard 1213: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1214: 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 1215: 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*/
1216: 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 1217: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1218: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1219: int **Tvard;
1220: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1221: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1222: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1223: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1224: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1225: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1226: double *lsurv, *lpop, *tpop;
1227:
1.231 brouard 1228: #define FD 1; /* Fixed dummy covariate */
1229: #define FQ 2; /* Fixed quantitative covariate */
1230: #define FP 3; /* Fixed product covariate */
1231: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1232: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1233: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1234: #define VD 10; /* Varying dummy covariate */
1235: #define VQ 11; /* Varying quantitative covariate */
1236: #define VP 12; /* Varying product covariate */
1237: #define VPDD 13; /* Varying product dummy*dummy covariate */
1238: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1239: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1240: #define APFD 16; /* Age product * fixed dummy covariate */
1241: #define APFQ 17; /* Age product * fixed quantitative covariate */
1242: #define APVD 18; /* Age product * varying dummy covariate */
1243: #define APVQ 19; /* Age product * varying quantitative covariate */
1244:
1245: #define FTYPE 1; /* Fixed covariate */
1246: #define VTYPE 2; /* Varying covariate (loop in wave) */
1247: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1248:
1249: struct kmodel{
1250: int maintype; /* main type */
1251: int subtype; /* subtype */
1252: };
1253: struct kmodel modell[NCOVMAX];
1254:
1.143 brouard 1255: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1256: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1257:
1258: /**************** split *************************/
1259: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1260: {
1261: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1262: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1263: */
1264: char *ss; /* pointer */
1.186 brouard 1265: int l1=0, l2=0; /* length counters */
1.126 brouard 1266:
1267: l1 = strlen(path ); /* length of path */
1268: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1269: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1270: if ( ss == NULL ) { /* no directory, so determine current directory */
1271: strcpy( name, path ); /* we got the fullname name because no directory */
1272: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1273: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1274: /* get current working directory */
1275: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1276: #ifdef WIN32
1277: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1278: #else
1279: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1280: #endif
1.126 brouard 1281: return( GLOCK_ERROR_GETCWD );
1282: }
1283: /* got dirc from getcwd*/
1284: printf(" DIRC = %s \n",dirc);
1.205 brouard 1285: } else { /* strip directory from path */
1.126 brouard 1286: ss++; /* after this, the filename */
1287: l2 = strlen( ss ); /* length of filename */
1288: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1289: strcpy( name, ss ); /* save file name */
1290: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1291: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1292: printf(" DIRC2 = %s \n",dirc);
1293: }
1294: /* We add a separator at the end of dirc if not exists */
1295: l1 = strlen( dirc ); /* length of directory */
1296: if( dirc[l1-1] != DIRSEPARATOR ){
1297: dirc[l1] = DIRSEPARATOR;
1298: dirc[l1+1] = 0;
1299: printf(" DIRC3 = %s \n",dirc);
1300: }
1301: ss = strrchr( name, '.' ); /* find last / */
1302: if (ss >0){
1303: ss++;
1304: strcpy(ext,ss); /* save extension */
1305: l1= strlen( name);
1306: l2= strlen(ss)+1;
1307: strncpy( finame, name, l1-l2);
1308: finame[l1-l2]= 0;
1309: }
1310:
1311: return( 0 ); /* we're done */
1312: }
1313:
1314:
1315: /******************************************/
1316:
1317: void replace_back_to_slash(char *s, char*t)
1318: {
1319: int i;
1320: int lg=0;
1321: i=0;
1322: lg=strlen(t);
1323: for(i=0; i<= lg; i++) {
1324: (s[i] = t[i]);
1325: if (t[i]== '\\') s[i]='/';
1326: }
1327: }
1328:
1.132 brouard 1329: char *trimbb(char *out, char *in)
1.137 brouard 1330: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1331: char *s;
1332: s=out;
1333: while (*in != '\0'){
1.137 brouard 1334: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1335: in++;
1336: }
1337: *out++ = *in++;
1338: }
1339: *out='\0';
1340: return s;
1341: }
1342:
1.187 brouard 1343: /* char *substrchaine(char *out, char *in, char *chain) */
1344: /* { */
1345: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1346: /* char *s, *t; */
1347: /* t=in;s=out; */
1348: /* while ((*in != *chain) && (*in != '\0')){ */
1349: /* *out++ = *in++; */
1350: /* } */
1351:
1352: /* /\* *in matches *chain *\/ */
1353: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1354: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1355: /* } */
1356: /* in--; chain--; */
1357: /* while ( (*in != '\0')){ */
1358: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1359: /* *out++ = *in++; */
1360: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1361: /* } */
1362: /* *out='\0'; */
1363: /* out=s; */
1364: /* return out; */
1365: /* } */
1366: char *substrchaine(char *out, char *in, char *chain)
1367: {
1368: /* Substract chain 'chain' from 'in', return and output 'out' */
1369: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1370:
1371: char *strloc;
1372:
1373: strcpy (out, in);
1374: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1375: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1376: if(strloc != NULL){
1377: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1378: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1379: /* strcpy (strloc, strloc +strlen(chain));*/
1380: }
1381: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1382: return out;
1383: }
1384:
1385:
1.145 brouard 1386: char *cutl(char *blocc, char *alocc, char *in, char occ)
1387: {
1.187 brouard 1388: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1389: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1390: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1391: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1392: */
1.160 brouard 1393: char *s, *t;
1.145 brouard 1394: t=in;s=in;
1395: while ((*in != occ) && (*in != '\0')){
1396: *alocc++ = *in++;
1397: }
1398: if( *in == occ){
1399: *(alocc)='\0';
1400: s=++in;
1401: }
1402:
1403: if (s == t) {/* occ not found */
1404: *(alocc-(in-s))='\0';
1405: in=s;
1406: }
1407: while ( *in != '\0'){
1408: *blocc++ = *in++;
1409: }
1410:
1411: *blocc='\0';
1412: return t;
1413: }
1.137 brouard 1414: char *cutv(char *blocc, char *alocc, char *in, char occ)
1415: {
1.187 brouard 1416: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1417: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1418: gives blocc="abcdef2ghi" and alocc="j".
1419: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1420: */
1421: char *s, *t;
1422: t=in;s=in;
1423: while (*in != '\0'){
1424: while( *in == occ){
1425: *blocc++ = *in++;
1426: s=in;
1427: }
1428: *blocc++ = *in++;
1429: }
1430: if (s == t) /* occ not found */
1431: *(blocc-(in-s))='\0';
1432: else
1433: *(blocc-(in-s)-1)='\0';
1434: in=s;
1435: while ( *in != '\0'){
1436: *alocc++ = *in++;
1437: }
1438:
1439: *alocc='\0';
1440: return s;
1441: }
1442:
1.126 brouard 1443: int nbocc(char *s, char occ)
1444: {
1445: int i,j=0;
1446: int lg=20;
1447: i=0;
1448: lg=strlen(s);
1449: for(i=0; i<= lg; i++) {
1.234 brouard 1450: if (s[i] == occ ) j++;
1.126 brouard 1451: }
1452: return j;
1453: }
1454:
1.137 brouard 1455: /* void cutv(char *u,char *v, char*t, char occ) */
1456: /* { */
1457: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1458: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1459: /* gives u="abcdef2ghi" and v="j" *\/ */
1460: /* int i,lg,j,p=0; */
1461: /* i=0; */
1462: /* lg=strlen(t); */
1463: /* for(j=0; j<=lg-1; j++) { */
1464: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1465: /* } */
1.126 brouard 1466:
1.137 brouard 1467: /* for(j=0; j<p; j++) { */
1468: /* (u[j] = t[j]); */
1469: /* } */
1470: /* u[p]='\0'; */
1.126 brouard 1471:
1.137 brouard 1472: /* for(j=0; j<= lg; j++) { */
1473: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1474: /* } */
1475: /* } */
1.126 brouard 1476:
1.160 brouard 1477: #ifdef _WIN32
1478: char * strsep(char **pp, const char *delim)
1479: {
1480: char *p, *q;
1481:
1482: if ((p = *pp) == NULL)
1483: return 0;
1484: if ((q = strpbrk (p, delim)) != NULL)
1485: {
1486: *pp = q + 1;
1487: *q = '\0';
1488: }
1489: else
1490: *pp = 0;
1491: return p;
1492: }
1493: #endif
1494:
1.126 brouard 1495: /********************** nrerror ********************/
1496:
1497: void nrerror(char error_text[])
1498: {
1499: fprintf(stderr,"ERREUR ...\n");
1500: fprintf(stderr,"%s\n",error_text);
1501: exit(EXIT_FAILURE);
1502: }
1503: /*********************** vector *******************/
1504: double *vector(int nl, int nh)
1505: {
1506: double *v;
1507: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1508: if (!v) nrerror("allocation failure in vector");
1509: return v-nl+NR_END;
1510: }
1511:
1512: /************************ free vector ******************/
1513: void free_vector(double*v, int nl, int nh)
1514: {
1515: free((FREE_ARG)(v+nl-NR_END));
1516: }
1517:
1518: /************************ivector *******************************/
1519: int *ivector(long nl,long nh)
1520: {
1521: int *v;
1522: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1523: if (!v) nrerror("allocation failure in ivector");
1524: return v-nl+NR_END;
1525: }
1526:
1527: /******************free ivector **************************/
1528: void free_ivector(int *v, long nl, long nh)
1529: {
1530: free((FREE_ARG)(v+nl-NR_END));
1531: }
1532:
1533: /************************lvector *******************************/
1534: long *lvector(long nl,long nh)
1535: {
1536: long *v;
1537: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1538: if (!v) nrerror("allocation failure in ivector");
1539: return v-nl+NR_END;
1540: }
1541:
1542: /******************free lvector **************************/
1543: void free_lvector(long *v, long nl, long nh)
1544: {
1545: free((FREE_ARG)(v+nl-NR_END));
1546: }
1547:
1548: /******************* imatrix *******************************/
1549: int **imatrix(long nrl, long nrh, long ncl, long nch)
1550: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1551: {
1552: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1553: int **m;
1554:
1555: /* allocate pointers to rows */
1556: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1557: if (!m) nrerror("allocation failure 1 in matrix()");
1558: m += NR_END;
1559: m -= nrl;
1560:
1561:
1562: /* allocate rows and set pointers to them */
1563: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1564: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1565: m[nrl] += NR_END;
1566: m[nrl] -= ncl;
1567:
1568: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1569:
1570: /* return pointer to array of pointers to rows */
1571: return m;
1572: }
1573:
1574: /****************** free_imatrix *************************/
1575: void free_imatrix(m,nrl,nrh,ncl,nch)
1576: int **m;
1577: long nch,ncl,nrh,nrl;
1578: /* free an int matrix allocated by imatrix() */
1579: {
1580: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1581: free((FREE_ARG) (m+nrl-NR_END));
1582: }
1583:
1584: /******************* matrix *******************************/
1585: double **matrix(long nrl, long nrh, long ncl, long nch)
1586: {
1587: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1588: double **m;
1589:
1590: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1591: if (!m) nrerror("allocation failure 1 in matrix()");
1592: m += NR_END;
1593: m -= nrl;
1594:
1595: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1596: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1597: m[nrl] += NR_END;
1598: m[nrl] -= ncl;
1599:
1600: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1601: return m;
1.145 brouard 1602: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1603: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1604: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1605: */
1606: }
1607:
1608: /*************************free matrix ************************/
1609: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1610: {
1611: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1612: free((FREE_ARG)(m+nrl-NR_END));
1613: }
1614:
1615: /******************* ma3x *******************************/
1616: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1617: {
1618: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1619: double ***m;
1620:
1621: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1622: if (!m) nrerror("allocation failure 1 in matrix()");
1623: m += NR_END;
1624: m -= nrl;
1625:
1626: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1627: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1628: m[nrl] += NR_END;
1629: m[nrl] -= ncl;
1630:
1631: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1632:
1633: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1634: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1635: m[nrl][ncl] += NR_END;
1636: m[nrl][ncl] -= nll;
1637: for (j=ncl+1; j<=nch; j++)
1638: m[nrl][j]=m[nrl][j-1]+nlay;
1639:
1640: for (i=nrl+1; i<=nrh; i++) {
1641: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1642: for (j=ncl+1; j<=nch; j++)
1643: m[i][j]=m[i][j-1]+nlay;
1644: }
1645: return m;
1646: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1647: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1648: */
1649: }
1650:
1651: /*************************free ma3x ************************/
1652: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1653: {
1654: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1655: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1656: free((FREE_ARG)(m+nrl-NR_END));
1657: }
1658:
1659: /*************** function subdirf ***********/
1660: char *subdirf(char fileres[])
1661: {
1662: /* Caution optionfilefiname is hidden */
1663: strcpy(tmpout,optionfilefiname);
1664: strcat(tmpout,"/"); /* Add to the right */
1665: strcat(tmpout,fileres);
1666: return tmpout;
1667: }
1668:
1669: /*************** function subdirf2 ***********/
1670: char *subdirf2(char fileres[], char *preop)
1671: {
1672:
1673: /* Caution optionfilefiname is hidden */
1674: strcpy(tmpout,optionfilefiname);
1675: strcat(tmpout,"/");
1676: strcat(tmpout,preop);
1677: strcat(tmpout,fileres);
1678: return tmpout;
1679: }
1680:
1681: /*************** function subdirf3 ***********/
1682: char *subdirf3(char fileres[], char *preop, char *preop2)
1683: {
1684:
1685: /* Caution optionfilefiname is hidden */
1686: strcpy(tmpout,optionfilefiname);
1687: strcat(tmpout,"/");
1688: strcat(tmpout,preop);
1689: strcat(tmpout,preop2);
1690: strcat(tmpout,fileres);
1691: return tmpout;
1692: }
1.213 brouard 1693:
1694: /*************** function subdirfext ***********/
1695: char *subdirfext(char fileres[], char *preop, char *postop)
1696: {
1697:
1698: strcpy(tmpout,preop);
1699: strcat(tmpout,fileres);
1700: strcat(tmpout,postop);
1701: return tmpout;
1702: }
1.126 brouard 1703:
1.213 brouard 1704: /*************** function subdirfext3 ***********/
1705: char *subdirfext3(char fileres[], char *preop, char *postop)
1706: {
1707:
1708: /* Caution optionfilefiname is hidden */
1709: strcpy(tmpout,optionfilefiname);
1710: strcat(tmpout,"/");
1711: strcat(tmpout,preop);
1712: strcat(tmpout,fileres);
1713: strcat(tmpout,postop);
1714: return tmpout;
1715: }
1716:
1.162 brouard 1717: char *asc_diff_time(long time_sec, char ascdiff[])
1718: {
1719: long sec_left, days, hours, minutes;
1720: days = (time_sec) / (60*60*24);
1721: sec_left = (time_sec) % (60*60*24);
1722: hours = (sec_left) / (60*60) ;
1723: sec_left = (sec_left) %(60*60);
1724: minutes = (sec_left) /60;
1725: sec_left = (sec_left) % (60);
1726: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1727: return ascdiff;
1728: }
1729:
1.126 brouard 1730: /***************** f1dim *************************/
1731: extern int ncom;
1732: extern double *pcom,*xicom;
1733: extern double (*nrfunc)(double []);
1734:
1735: double f1dim(double x)
1736: {
1737: int j;
1738: double f;
1739: double *xt;
1740:
1741: xt=vector(1,ncom);
1742: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1743: f=(*nrfunc)(xt);
1744: free_vector(xt,1,ncom);
1745: return f;
1746: }
1747:
1748: /*****************brent *************************/
1749: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1750: {
1751: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1752: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1753: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1754: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1755: * returned function value.
1756: */
1.126 brouard 1757: int iter;
1758: double a,b,d,etemp;
1.159 brouard 1759: double fu=0,fv,fw,fx;
1.164 brouard 1760: double ftemp=0.;
1.126 brouard 1761: double p,q,r,tol1,tol2,u,v,w,x,xm;
1762: double e=0.0;
1763:
1764: a=(ax < cx ? ax : cx);
1765: b=(ax > cx ? ax : cx);
1766: x=w=v=bx;
1767: fw=fv=fx=(*f)(x);
1768: for (iter=1;iter<=ITMAX;iter++) {
1769: xm=0.5*(a+b);
1770: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1771: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1772: printf(".");fflush(stdout);
1773: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1774: #ifdef DEBUGBRENT
1.126 brouard 1775: 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);
1776: 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);
1777: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1778: #endif
1779: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1780: *xmin=x;
1781: return fx;
1782: }
1783: ftemp=fu;
1784: if (fabs(e) > tol1) {
1785: r=(x-w)*(fx-fv);
1786: q=(x-v)*(fx-fw);
1787: p=(x-v)*q-(x-w)*r;
1788: q=2.0*(q-r);
1789: if (q > 0.0) p = -p;
1790: q=fabs(q);
1791: etemp=e;
1792: e=d;
1793: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1794: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1795: else {
1.224 brouard 1796: d=p/q;
1797: u=x+d;
1798: if (u-a < tol2 || b-u < tol2)
1799: d=SIGN(tol1,xm-x);
1.126 brouard 1800: }
1801: } else {
1802: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1803: }
1804: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1805: fu=(*f)(u);
1806: if (fu <= fx) {
1807: if (u >= x) a=x; else b=x;
1808: SHFT(v,w,x,u)
1.183 brouard 1809: SHFT(fv,fw,fx,fu)
1810: } else {
1811: if (u < x) a=u; else b=u;
1812: if (fu <= fw || w == x) {
1.224 brouard 1813: v=w;
1814: w=u;
1815: fv=fw;
1816: fw=fu;
1.183 brouard 1817: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1818: v=u;
1819: fv=fu;
1.183 brouard 1820: }
1821: }
1.126 brouard 1822: }
1823: nrerror("Too many iterations in brent");
1824: *xmin=x;
1825: return fx;
1826: }
1827:
1828: /****************** mnbrak ***********************/
1829:
1830: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1831: double (*func)(double))
1.183 brouard 1832: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1833: the downhill direction (defined by the function as evaluated at the initial points) and returns
1834: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1835: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1836: */
1.126 brouard 1837: double ulim,u,r,q, dum;
1838: double fu;
1.187 brouard 1839:
1840: double scale=10.;
1841: int iterscale=0;
1842:
1843: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1844: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1845:
1846:
1847: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1848: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1849: /* *bx = *ax - (*ax - *bx)/scale; */
1850: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1851: /* } */
1852:
1.126 brouard 1853: if (*fb > *fa) {
1854: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1855: SHFT(dum,*fb,*fa,dum)
1856: }
1.126 brouard 1857: *cx=(*bx)+GOLD*(*bx-*ax);
1858: *fc=(*func)(*cx);
1.183 brouard 1859: #ifdef DEBUG
1.224 brouard 1860: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1861: 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 1862: #endif
1.224 brouard 1863: 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 1864: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1865: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1866: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1867: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1868: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1869: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1870: fu=(*func)(u);
1.163 brouard 1871: #ifdef DEBUG
1872: /* f(x)=A(x-u)**2+f(u) */
1873: double A, fparabu;
1874: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1875: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1876: 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);
1877: 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 1878: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1879: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1880: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1881: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1882: #endif
1.184 brouard 1883: #ifdef MNBRAKORIGINAL
1.183 brouard 1884: #else
1.191 brouard 1885: /* if (fu > *fc) { */
1886: /* #ifdef DEBUG */
1887: /* printf("mnbrak4 fu > fc \n"); */
1888: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1889: /* #endif */
1890: /* /\* 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 *\\/ *\/ */
1891: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1892: /* dum=u; /\* Shifting c and u *\/ */
1893: /* u = *cx; */
1894: /* *cx = dum; */
1895: /* dum = fu; */
1896: /* fu = *fc; */
1897: /* *fc =dum; */
1898: /* } else { /\* end *\/ */
1899: /* #ifdef DEBUG */
1900: /* printf("mnbrak3 fu < fc \n"); */
1901: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1902: /* #endif */
1903: /* dum=u; /\* Shifting c and u *\/ */
1904: /* u = *cx; */
1905: /* *cx = dum; */
1906: /* dum = fu; */
1907: /* fu = *fc; */
1908: /* *fc =dum; */
1909: /* } */
1.224 brouard 1910: #ifdef DEBUGMNBRAK
1911: double A, fparabu;
1912: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1913: fparabu= *fa - A*(*ax-u)*(*ax-u);
1914: 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);
1915: 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 1916: #endif
1.191 brouard 1917: dum=u; /* Shifting c and u */
1918: u = *cx;
1919: *cx = dum;
1920: dum = fu;
1921: fu = *fc;
1922: *fc =dum;
1.183 brouard 1923: #endif
1.162 brouard 1924: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1925: #ifdef DEBUG
1.224 brouard 1926: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1927: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1928: #endif
1.126 brouard 1929: fu=(*func)(u);
1930: if (fu < *fc) {
1.183 brouard 1931: #ifdef DEBUG
1.224 brouard 1932: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1933: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1934: #endif
1935: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1936: SHFT(*fb,*fc,fu,(*func)(u))
1937: #ifdef DEBUG
1938: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1939: #endif
1940: }
1.162 brouard 1941: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1942: #ifdef DEBUG
1.224 brouard 1943: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1944: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1945: #endif
1.126 brouard 1946: u=ulim;
1947: fu=(*func)(u);
1.183 brouard 1948: } else { /* u could be left to b (if r > q parabola has a maximum) */
1949: #ifdef DEBUG
1.224 brouard 1950: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1951: 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 1952: #endif
1.126 brouard 1953: u=(*cx)+GOLD*(*cx-*bx);
1954: fu=(*func)(u);
1.224 brouard 1955: #ifdef DEBUG
1956: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1957: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1958: #endif
1.183 brouard 1959: } /* end tests */
1.126 brouard 1960: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1961: SHFT(*fa,*fb,*fc,fu)
1962: #ifdef DEBUG
1.224 brouard 1963: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1964: 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 1965: #endif
1966: } /* 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 1967: }
1968:
1969: /*************** linmin ************************/
1.162 brouard 1970: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1971: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1972: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1973: the value of func at the returned location p . This is actually all accomplished by calling the
1974: routines mnbrak and brent .*/
1.126 brouard 1975: int ncom;
1976: double *pcom,*xicom;
1977: double (*nrfunc)(double []);
1978:
1.224 brouard 1979: #ifdef LINMINORIGINAL
1.126 brouard 1980: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1981: #else
1982: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1983: #endif
1.126 brouard 1984: {
1985: double brent(double ax, double bx, double cx,
1986: double (*f)(double), double tol, double *xmin);
1987: double f1dim(double x);
1988: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1989: double *fc, double (*func)(double));
1990: int j;
1991: double xx,xmin,bx,ax;
1992: double fx,fb,fa;
1.187 brouard 1993:
1.203 brouard 1994: #ifdef LINMINORIGINAL
1995: #else
1996: double scale=10., axs, xxs; /* Scale added for infinity */
1997: #endif
1998:
1.126 brouard 1999: ncom=n;
2000: pcom=vector(1,n);
2001: xicom=vector(1,n);
2002: nrfunc=func;
2003: for (j=1;j<=n;j++) {
2004: pcom[j]=p[j];
1.202 brouard 2005: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2006: }
1.187 brouard 2007:
1.203 brouard 2008: #ifdef LINMINORIGINAL
2009: xx=1.;
2010: #else
2011: axs=0.0;
2012: xxs=1.;
2013: do{
2014: xx= xxs;
2015: #endif
1.187 brouard 2016: ax=0.;
2017: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2018: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2019: /* 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)) */
2020: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2021: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2022: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2023: /* 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 2024: #ifdef LINMINORIGINAL
2025: #else
2026: if (fx != fx){
1.224 brouard 2027: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2028: printf("|");
2029: fprintf(ficlog,"|");
1.203 brouard 2030: #ifdef DEBUGLINMIN
1.224 brouard 2031: 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 2032: #endif
2033: }
1.224 brouard 2034: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2035: #endif
2036:
1.191 brouard 2037: #ifdef DEBUGLINMIN
2038: 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 2039: 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 2040: #endif
1.224 brouard 2041: #ifdef LINMINORIGINAL
2042: #else
2043: if(fb == fx){ /* Flat function in the direction */
2044: xmin=xx;
2045: *flat=1;
2046: }else{
2047: *flat=0;
2048: #endif
2049: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2050: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2051: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2052: /* fmin = f(p[j] + xmin * xi[j]) */
2053: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2054: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2055: #ifdef DEBUG
1.224 brouard 2056: 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);
2057: 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);
2058: #endif
2059: #ifdef LINMINORIGINAL
2060: #else
2061: }
1.126 brouard 2062: #endif
1.191 brouard 2063: #ifdef DEBUGLINMIN
2064: printf("linmin end ");
1.202 brouard 2065: fprintf(ficlog,"linmin end ");
1.191 brouard 2066: #endif
1.126 brouard 2067: for (j=1;j<=n;j++) {
1.203 brouard 2068: #ifdef LINMINORIGINAL
2069: xi[j] *= xmin;
2070: #else
2071: #ifdef DEBUGLINMIN
2072: if(xxs <1.0)
2073: printf(" before xi[%d]=%12.8f", j,xi[j]);
2074: #endif
2075: 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) */
2076: #ifdef DEBUGLINMIN
2077: if(xxs <1.0)
2078: 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 );
2079: #endif
2080: #endif
1.187 brouard 2081: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2082: }
1.191 brouard 2083: #ifdef DEBUGLINMIN
1.203 brouard 2084: printf("\n");
1.191 brouard 2085: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2086: 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 2087: for (j=1;j<=n;j++) {
1.202 brouard 2088: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2089: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2090: if(j % ncovmodel == 0){
1.191 brouard 2091: printf("\n");
1.202 brouard 2092: fprintf(ficlog,"\n");
2093: }
1.191 brouard 2094: }
1.203 brouard 2095: #else
1.191 brouard 2096: #endif
1.126 brouard 2097: free_vector(xicom,1,n);
2098: free_vector(pcom,1,n);
2099: }
2100:
2101:
2102: /*************** powell ************************/
1.162 brouard 2103: /*
2104: Minimization of a function func of n variables. Input consists of an initial starting point
2105: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2106: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2107: such that failure to decrease by more than this amount on one iteration signals doneness. On
2108: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2109: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2110: */
1.224 brouard 2111: #ifdef LINMINORIGINAL
2112: #else
2113: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2114: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2115: #endif
1.126 brouard 2116: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2117: double (*func)(double []))
2118: {
1.224 brouard 2119: #ifdef LINMINORIGINAL
2120: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2121: double (*func)(double []));
1.224 brouard 2122: #else
1.241 brouard 2123: void linmin(double p[], double xi[], int n, double *fret,
2124: double (*func)(double []),int *flat);
1.224 brouard 2125: #endif
1.239 brouard 2126: int i,ibig,j,jk,k;
1.126 brouard 2127: double del,t,*pt,*ptt,*xit;
1.181 brouard 2128: double directest;
1.126 brouard 2129: double fp,fptt;
2130: double *xits;
2131: int niterf, itmp;
1.224 brouard 2132: #ifdef LINMINORIGINAL
2133: #else
2134:
2135: flatdir=ivector(1,n);
2136: for (j=1;j<=n;j++) flatdir[j]=0;
2137: #endif
1.126 brouard 2138:
2139: pt=vector(1,n);
2140: ptt=vector(1,n);
2141: xit=vector(1,n);
2142: xits=vector(1,n);
2143: *fret=(*func)(p);
2144: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2145: rcurr_time = time(NULL);
1.126 brouard 2146: for (*iter=1;;++(*iter)) {
1.187 brouard 2147: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2148: ibig=0;
2149: del=0.0;
1.157 brouard 2150: rlast_time=rcurr_time;
2151: /* (void) gettimeofday(&curr_time,&tzp); */
2152: rcurr_time = time(NULL);
2153: curr_time = *localtime(&rcurr_time);
2154: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2155: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2156: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2157: for (i=1;i<=n;i++) {
1.126 brouard 2158: fprintf(ficrespow," %.12lf", p[i]);
2159: }
1.239 brouard 2160: fprintf(ficrespow,"\n");fflush(ficrespow);
2161: printf("\n#model= 1 + age ");
2162: fprintf(ficlog,"\n#model= 1 + age ");
2163: if(nagesqr==1){
1.241 brouard 2164: printf(" + age*age ");
2165: fprintf(ficlog," + age*age ");
1.239 brouard 2166: }
2167: for(j=1;j <=ncovmodel-2;j++){
2168: if(Typevar[j]==0) {
2169: printf(" + V%d ",Tvar[j]);
2170: fprintf(ficlog," + V%d ",Tvar[j]);
2171: }else if(Typevar[j]==1) {
2172: printf(" + V%d*age ",Tvar[j]);
2173: fprintf(ficlog," + V%d*age ",Tvar[j]);
2174: }else if(Typevar[j]==2) {
2175: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2176: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2177: }
2178: }
1.126 brouard 2179: printf("\n");
1.239 brouard 2180: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2181: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2182: fprintf(ficlog,"\n");
1.239 brouard 2183: for(i=1,jk=1; i <=nlstate; i++){
2184: for(k=1; k <=(nlstate+ndeath); k++){
2185: if (k != i) {
2186: printf("%d%d ",i,k);
2187: fprintf(ficlog,"%d%d ",i,k);
2188: for(j=1; j <=ncovmodel; j++){
2189: printf("%12.7f ",p[jk]);
2190: fprintf(ficlog,"%12.7f ",p[jk]);
2191: jk++;
2192: }
2193: printf("\n");
2194: fprintf(ficlog,"\n");
2195: }
2196: }
2197: }
1.241 brouard 2198: if(*iter <=3 && *iter >1){
1.157 brouard 2199: tml = *localtime(&rcurr_time);
2200: strcpy(strcurr,asctime(&tml));
2201: rforecast_time=rcurr_time;
1.126 brouard 2202: itmp = strlen(strcurr);
2203: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2204: strcurr[itmp-1]='\0';
1.162 brouard 2205: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2206: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2207: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2208: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2209: forecast_time = *localtime(&rforecast_time);
2210: strcpy(strfor,asctime(&forecast_time));
2211: itmp = strlen(strfor);
2212: if(strfor[itmp-1]=='\n')
2213: strfor[itmp-1]='\0';
2214: 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);
2215: 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 2216: }
2217: }
1.187 brouard 2218: for (i=1;i<=n;i++) { /* For each direction i */
2219: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2220: fptt=(*fret);
2221: #ifdef DEBUG
1.203 brouard 2222: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2223: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2224: #endif
1.203 brouard 2225: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2226: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2227: #ifdef LINMINORIGINAL
1.188 brouard 2228: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2229: #else
2230: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2231: flatdir[i]=flat; /* Function is vanishing in that direction i */
2232: #endif
2233: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2234: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2235: /* because that direction will be replaced unless the gain del is small */
2236: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2237: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2238: /* with the new direction. */
2239: del=fabs(fptt-(*fret));
2240: ibig=i;
1.126 brouard 2241: }
2242: #ifdef DEBUG
2243: printf("%d %.12e",i,(*fret));
2244: fprintf(ficlog,"%d %.12e",i,(*fret));
2245: for (j=1;j<=n;j++) {
1.224 brouard 2246: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2247: printf(" x(%d)=%.12e",j,xit[j]);
2248: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2249: }
2250: for(j=1;j<=n;j++) {
1.225 brouard 2251: printf(" p(%d)=%.12e",j,p[j]);
2252: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2253: }
2254: printf("\n");
2255: fprintf(ficlog,"\n");
2256: #endif
1.187 brouard 2257: } /* end loop on each direction i */
2258: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2259: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2260: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2261: for(j=1;j<=n;j++) {
1.225 brouard 2262: if(flatdir[j] >0){
2263: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2264: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2265: }
2266: /* printf("\n"); */
2267: /* fprintf(ficlog,"\n"); */
2268: }
1.243 brouard 2269: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2270: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2271: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2272: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2273: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2274: /* decreased of more than 3.84 */
2275: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2276: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2277: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2278:
1.188 brouard 2279: /* Starting the program with initial values given by a former maximization will simply change */
2280: /* the scales of the directions and the directions, because the are reset to canonical directions */
2281: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2282: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2283: #ifdef DEBUG
2284: int k[2],l;
2285: k[0]=1;
2286: k[1]=-1;
2287: printf("Max: %.12e",(*func)(p));
2288: fprintf(ficlog,"Max: %.12e",(*func)(p));
2289: for (j=1;j<=n;j++) {
2290: printf(" %.12e",p[j]);
2291: fprintf(ficlog," %.12e",p[j]);
2292: }
2293: printf("\n");
2294: fprintf(ficlog,"\n");
2295: for(l=0;l<=1;l++) {
2296: for (j=1;j<=n;j++) {
2297: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2298: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2299: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2300: }
2301: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2302: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2303: }
2304: #endif
2305:
1.224 brouard 2306: #ifdef LINMINORIGINAL
2307: #else
2308: free_ivector(flatdir,1,n);
2309: #endif
1.126 brouard 2310: free_vector(xit,1,n);
2311: free_vector(xits,1,n);
2312: free_vector(ptt,1,n);
2313: free_vector(pt,1,n);
2314: return;
1.192 brouard 2315: } /* enough precision */
1.240 brouard 2316: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2317: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2318: ptt[j]=2.0*p[j]-pt[j];
2319: xit[j]=p[j]-pt[j];
2320: pt[j]=p[j];
2321: }
1.181 brouard 2322: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2323: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2324: if (*iter <=4) {
1.225 brouard 2325: #else
2326: #endif
1.224 brouard 2327: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2328: #else
1.161 brouard 2329: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2330: #endif
1.162 brouard 2331: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2332: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2333: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2334: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2335: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2336: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2337: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2338: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2339: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2340: /* Even if f3 <f1, directest can be negative and t >0 */
2341: /* mu² and del² are equal when f3=f1 */
2342: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2343: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2344: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2345: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2346: #ifdef NRCORIGINAL
2347: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2348: #else
2349: 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 2350: t= t- del*SQR(fp-fptt);
1.183 brouard 2351: #endif
1.202 brouard 2352: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2353: #ifdef DEBUG
1.181 brouard 2354: 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);
2355: 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 2356: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2357: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2358: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2359: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2360: 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);
2361: 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);
2362: #endif
1.183 brouard 2363: #ifdef POWELLORIGINAL
2364: if (t < 0.0) { /* Then we use it for new direction */
2365: #else
1.182 brouard 2366: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2367: 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 2368: 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 2369: 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 2370: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2371: }
1.181 brouard 2372: if (directest < 0.0) { /* Then we use it for new direction */
2373: #endif
1.191 brouard 2374: #ifdef DEBUGLINMIN
1.234 brouard 2375: printf("Before linmin in direction P%d-P0\n",n);
2376: for (j=1;j<=n;j++) {
2377: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2378: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2379: if(j % ncovmodel == 0){
2380: printf("\n");
2381: fprintf(ficlog,"\n");
2382: }
2383: }
1.224 brouard 2384: #endif
2385: #ifdef LINMINORIGINAL
1.234 brouard 2386: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2387: #else
1.234 brouard 2388: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2389: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2390: #endif
1.234 brouard 2391:
1.191 brouard 2392: #ifdef DEBUGLINMIN
1.234 brouard 2393: for (j=1;j<=n;j++) {
2394: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2395: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2396: if(j % ncovmodel == 0){
2397: printf("\n");
2398: fprintf(ficlog,"\n");
2399: }
2400: }
1.224 brouard 2401: #endif
1.234 brouard 2402: for (j=1;j<=n;j++) {
2403: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2404: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2405: }
1.224 brouard 2406: #ifdef LINMINORIGINAL
2407: #else
1.234 brouard 2408: for (j=1, flatd=0;j<=n;j++) {
2409: if(flatdir[j]>0)
2410: flatd++;
2411: }
2412: if(flatd >0){
1.255 brouard 2413: printf("%d flat directions: ",flatd);
2414: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2415: for (j=1;j<=n;j++) {
2416: if(flatdir[j]>0){
2417: printf("%d ",j);
2418: fprintf(ficlog,"%d ",j);
2419: }
2420: }
2421: printf("\n");
2422: fprintf(ficlog,"\n");
2423: }
1.191 brouard 2424: #endif
1.234 brouard 2425: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2426: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2427:
1.126 brouard 2428: #ifdef DEBUG
1.234 brouard 2429: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2430: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2431: for(j=1;j<=n;j++){
2432: printf(" %lf",xit[j]);
2433: fprintf(ficlog," %lf",xit[j]);
2434: }
2435: printf("\n");
2436: fprintf(ficlog,"\n");
1.126 brouard 2437: #endif
1.192 brouard 2438: } /* end of t or directest negative */
1.224 brouard 2439: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2440: #else
1.234 brouard 2441: } /* end if (fptt < fp) */
1.192 brouard 2442: #endif
1.225 brouard 2443: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2444: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2445: #else
1.224 brouard 2446: #endif
1.234 brouard 2447: } /* loop iteration */
1.126 brouard 2448: }
1.234 brouard 2449:
1.126 brouard 2450: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2451:
1.235 brouard 2452: 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 2453: {
1.235 brouard 2454: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2455: (and selected quantitative values in nres)
2456: by left multiplying the unit
1.234 brouard 2457: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2458: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2459: /* Wx is row vector: population in state 1, population in state 2, population dead */
2460: /* or prevalence in state 1, prevalence in state 2, 0 */
2461: /* newm is the matrix after multiplications, its rows are identical at a factor */
2462: /* Initial matrix pimij */
2463: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2464: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2465: /* 0, 0 , 1} */
2466: /*
2467: * and after some iteration: */
2468: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2469: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2470: /* 0, 0 , 1} */
2471: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2472: /* {0.51571254859325999, 0.4842874514067399, */
2473: /* 0.51326036147820708, 0.48673963852179264} */
2474: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2475:
1.126 brouard 2476: int i, ii,j,k;
1.209 brouard 2477: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2478: /* double **matprod2(); */ /* test */
1.218 brouard 2479: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2480: double **newm;
1.209 brouard 2481: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2482: int ncvloop=0;
1.169 brouard 2483:
1.209 brouard 2484: min=vector(1,nlstate);
2485: max=vector(1,nlstate);
2486: meandiff=vector(1,nlstate);
2487:
1.218 brouard 2488: /* Starting with matrix unity */
1.126 brouard 2489: for (ii=1;ii<=nlstate+ndeath;ii++)
2490: for (j=1;j<=nlstate+ndeath;j++){
2491: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2492: }
1.169 brouard 2493:
2494: cov[1]=1.;
2495:
2496: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2497: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2498: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2499: ncvloop++;
1.126 brouard 2500: newm=savm;
2501: /* Covariates have to be included here again */
1.138 brouard 2502: cov[2]=agefin;
1.187 brouard 2503: if(nagesqr==1)
2504: cov[3]= agefin*agefin;;
1.234 brouard 2505: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2506: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2507: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2508: /* 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 2509: }
2510: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2511: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2512: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2513: /* 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 2514: }
1.237 brouard 2515: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2516: if(Dummy[Tvar[Tage[k]]]){
2517: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2518: } else{
1.235 brouard 2519: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2520: }
1.235 brouard 2521: /* 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 2522: }
1.237 brouard 2523: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2524: /* 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 2525: if(Dummy[Tvard[k][1]==0]){
2526: if(Dummy[Tvard[k][2]==0]){
2527: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2528: }else{
2529: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2530: }
2531: }else{
2532: if(Dummy[Tvard[k][2]==0]){
2533: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2534: }else{
2535: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2536: }
2537: }
1.234 brouard 2538: }
1.138 brouard 2539: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2540: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2541: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2542: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2543: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2544: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2545: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2546:
1.126 brouard 2547: savm=oldm;
2548: oldm=newm;
1.209 brouard 2549:
2550: for(j=1; j<=nlstate; j++){
2551: max[j]=0.;
2552: min[j]=1.;
2553: }
2554: for(i=1;i<=nlstate;i++){
2555: sumnew=0;
2556: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2557: for(j=1; j<=nlstate; j++){
2558: prlim[i][j]= newm[i][j]/(1-sumnew);
2559: max[j]=FMAX(max[j],prlim[i][j]);
2560: min[j]=FMIN(min[j],prlim[i][j]);
2561: }
2562: }
2563:
1.126 brouard 2564: maxmax=0.;
1.209 brouard 2565: for(j=1; j<=nlstate; j++){
2566: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2567: maxmax=FMAX(maxmax,meandiff[j]);
2568: /* 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 2569: } /* j loop */
1.203 brouard 2570: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2571: /* 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 2572: if(maxmax < ftolpl){
1.209 brouard 2573: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2574: free_vector(min,1,nlstate);
2575: free_vector(max,1,nlstate);
2576: free_vector(meandiff,1,nlstate);
1.126 brouard 2577: return prlim;
2578: }
1.169 brouard 2579: } /* age loop */
1.208 brouard 2580: /* After some age loop it doesn't converge */
1.209 brouard 2581: 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 2582: 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 2583: /* 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); */
2584: free_vector(min,1,nlstate);
2585: free_vector(max,1,nlstate);
2586: free_vector(meandiff,1,nlstate);
1.208 brouard 2587:
1.169 brouard 2588: return prlim; /* should not reach here */
1.126 brouard 2589: }
2590:
1.217 brouard 2591:
2592: /**** Back Prevalence limit (stable or period prevalence) ****************/
2593:
1.218 brouard 2594: /* 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) */
2595: /* 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 2596: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2597: {
1.218 brouard 2598: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2599: matrix by transitions matrix until convergence is reached with precision ftolpl */
2600: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2601: /* Wx is row vector: population in state 1, population in state 2, population dead */
2602: /* or prevalence in state 1, prevalence in state 2, 0 */
2603: /* newm is the matrix after multiplications, its rows are identical at a factor */
2604: /* Initial matrix pimij */
2605: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2606: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2607: /* 0, 0 , 1} */
2608: /*
2609: * and after some iteration: */
2610: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2611: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2612: /* 0, 0 , 1} */
2613: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2614: /* {0.51571254859325999, 0.4842874514067399, */
2615: /* 0.51326036147820708, 0.48673963852179264} */
2616: /* If we start from prlim again, prlim tends to a constant matrix */
2617:
2618: int i, ii,j,k;
1.247 brouard 2619: int first=0;
1.217 brouard 2620: double *min, *max, *meandiff, maxmax,sumnew=0.;
2621: /* double **matprod2(); */ /* test */
2622: double **out, cov[NCOVMAX+1], **bmij();
2623: double **newm;
1.218 brouard 2624: double **dnewm, **doldm, **dsavm; /* for use */
2625: double **oldm, **savm; /* for use */
2626:
1.217 brouard 2627: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2628: int ncvloop=0;
2629:
2630: min=vector(1,nlstate);
2631: max=vector(1,nlstate);
2632: meandiff=vector(1,nlstate);
2633:
1.218 brouard 2634: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2635: oldm=oldms; savm=savms;
2636:
2637: /* Starting with matrix unity */
2638: for (ii=1;ii<=nlstate+ndeath;ii++)
2639: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2640: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2641: }
2642:
2643: cov[1]=1.;
2644:
2645: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2646: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2647: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2648: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2649: ncvloop++;
1.218 brouard 2650: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2651: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2652: /* Covariates have to be included here again */
2653: cov[2]=agefin;
2654: if(nagesqr==1)
2655: cov[3]= agefin*agefin;;
1.242 brouard 2656: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2657: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2658: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2659: /* 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)); */
2660: }
2661: /* for (k=1; k<=cptcovn;k++) { */
2662: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2663: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2664: /* /\* 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])]); *\/ */
2665: /* } */
2666: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2667: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2668: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2669: /* 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]); */
2670: }
2671: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2672: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2673: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2674: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2675: for (k=1; k<=cptcovage;k++){ /* For product with age */
2676: if(Dummy[Tvar[Tage[k]]]){
2677: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2678: } else{
2679: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2680: }
2681: /* 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]); */
2682: }
2683: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2684: /* 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]); */
2685: if(Dummy[Tvard[k][1]==0]){
2686: if(Dummy[Tvard[k][2]==0]){
2687: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2688: }else{
2689: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2690: }
2691: }else{
2692: if(Dummy[Tvard[k][2]==0]){
2693: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2694: }else{
2695: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2696: }
2697: }
1.217 brouard 2698: }
2699:
2700: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2701: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2702: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2703: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2704: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2705: /* ij should be linked to the correct index of cov */
2706: /* age and covariate values ij are in 'cov', but we need to pass
2707: * ij for the observed prevalence at age and status and covariate
2708: * number: prevacurrent[(int)agefin][ii][ij]
2709: */
2710: /* 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 *\/ */
2711: /* 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 *\/ */
2712: 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 2713: savm=oldm;
2714: oldm=newm;
2715: for(j=1; j<=nlstate; j++){
2716: max[j]=0.;
2717: min[j]=1.;
2718: }
2719: for(j=1; j<=nlstate; j++){
2720: for(i=1;i<=nlstate;i++){
1.234 brouard 2721: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2722: bprlim[i][j]= newm[i][j];
2723: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2724: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2725: }
2726: }
1.218 brouard 2727:
1.217 brouard 2728: maxmax=0.;
2729: for(i=1; i<=nlstate; i++){
2730: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2731: maxmax=FMAX(maxmax,meandiff[i]);
2732: /* 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); */
2733: } /* j loop */
2734: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2735: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2736: if(maxmax < ftolpl){
1.220 brouard 2737: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2738: free_vector(min,1,nlstate);
2739: free_vector(max,1,nlstate);
2740: free_vector(meandiff,1,nlstate);
2741: return bprlim;
2742: }
2743: } /* age loop */
2744: /* After some age loop it doesn't converge */
1.247 brouard 2745: if(first){
2746: first=1;
2747: 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\
2748: 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);
2749: }
2750: 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 2751: 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);
2752: /* 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); */
2753: free_vector(min,1,nlstate);
2754: free_vector(max,1,nlstate);
2755: free_vector(meandiff,1,nlstate);
2756:
2757: return bprlim; /* should not reach here */
2758: }
2759:
1.126 brouard 2760: /*************** transition probabilities ***************/
2761:
2762: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2763: {
1.138 brouard 2764: /* According to parameters values stored in x and the covariate's values stored in cov,
2765: computes the probability to be observed in state j being in state i by appying the
2766: model to the ncovmodel covariates (including constant and age).
2767: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2768: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2769: ncth covariate in the global vector x is given by the formula:
2770: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2771: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2772: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2773: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2774: Outputs ps[i][j] the probability to be observed in j being in j according to
2775: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2776: */
2777: double s1, lnpijopii;
1.126 brouard 2778: /*double t34;*/
1.164 brouard 2779: int i,j, nc, ii, jj;
1.126 brouard 2780:
1.223 brouard 2781: for(i=1; i<= nlstate; i++){
2782: for(j=1; j<i;j++){
2783: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2784: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2785: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2786: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2787: }
2788: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2789: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2790: }
2791: for(j=i+1; j<=nlstate+ndeath;j++){
2792: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2793: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2794: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2795: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2796: }
2797: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2798: }
2799: }
1.218 brouard 2800:
1.223 brouard 2801: for(i=1; i<= nlstate; i++){
2802: s1=0;
2803: for(j=1; j<i; j++){
2804: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2805: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2806: }
2807: for(j=i+1; j<=nlstate+ndeath; j++){
2808: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2809: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2810: }
2811: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2812: ps[i][i]=1./(s1+1.);
2813: /* Computing other pijs */
2814: for(j=1; j<i; j++)
2815: ps[i][j]= exp(ps[i][j])*ps[i][i];
2816: for(j=i+1; j<=nlstate+ndeath; j++)
2817: ps[i][j]= exp(ps[i][j])*ps[i][i];
2818: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2819: } /* end i */
1.218 brouard 2820:
1.223 brouard 2821: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2822: for(jj=1; jj<= nlstate+ndeath; jj++){
2823: ps[ii][jj]=0;
2824: ps[ii][ii]=1;
2825: }
2826: }
1.218 brouard 2827:
2828:
1.223 brouard 2829: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2830: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2831: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2832: /* } */
2833: /* printf("\n "); */
2834: /* } */
2835: /* printf("\n ");printf("%lf ",cov[2]);*/
2836: /*
2837: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2838: goto end;*/
1.223 brouard 2839: return ps;
1.126 brouard 2840: }
2841:
1.218 brouard 2842: /*************** backward transition probabilities ***************/
2843:
2844: /* 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 ) */
2845: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2846: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2847: {
1.222 brouard 2848: /* Computes the backward probability at age agefin and covariate ij
2849: * and returns in **ps as well as **bmij.
2850: */
1.218 brouard 2851: int i, ii, j,k;
1.222 brouard 2852:
2853: double **out, **pmij();
2854: double sumnew=0.;
1.218 brouard 2855: double agefin;
1.222 brouard 2856:
2857: double **dnewm, **dsavm, **doldm;
2858: double **bbmij;
2859:
1.218 brouard 2860: doldm=ddoldms; /* global pointers */
1.222 brouard 2861: dnewm=ddnewms;
2862: dsavm=ddsavms;
2863:
2864: agefin=cov[2];
2865: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2866: the observed prevalence (with this covariate ij) */
2867: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2868: /* We do have the matrix Px in savm and we need pij */
2869: for (j=1;j<=nlstate+ndeath;j++){
2870: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2871: for (ii=1;ii<=nlstate;ii++){
2872: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2873: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2874: for (ii=1;ii<=nlstate+ndeath;ii++){
2875: if(sumnew >= 1.e-10){
2876: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2877: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2878: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2879: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2880: /* }else */
2881: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2882: }else{
1.242 brouard 2883: ;
2884: /* 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 2885: }
2886: } /*End ii */
2887: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2888: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2889: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2890: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2891: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2892: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2893: /* left Product of this matrix by diag matrix of prevalences (savm) */
2894: for (j=1;j<=nlstate+ndeath;j++){
2895: for (ii=1;ii<=nlstate+ndeath;ii++){
2896: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2897: }
2898: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2899: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2900: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2901: /* end bmij */
2902: return ps;
1.218 brouard 2903: }
1.217 brouard 2904: /*************** transition probabilities ***************/
2905:
1.218 brouard 2906: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2907: {
2908: /* According to parameters values stored in x and the covariate's values stored in cov,
2909: computes the probability to be observed in state j being in state i by appying the
2910: model to the ncovmodel covariates (including constant and age).
2911: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2912: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2913: ncth covariate in the global vector x is given by the formula:
2914: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2915: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2916: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2917: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2918: Outputs ps[i][j] the probability to be observed in j being in j according to
2919: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2920: */
2921: double s1, lnpijopii;
2922: /*double t34;*/
2923: int i,j, nc, ii, jj;
2924:
1.234 brouard 2925: for(i=1; i<= nlstate; i++){
2926: for(j=1; j<i;j++){
2927: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2928: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2929: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2930: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2931: }
2932: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2933: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2934: }
2935: for(j=i+1; j<=nlstate+ndeath;j++){
2936: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2937: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2938: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2939: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2940: }
2941: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2942: }
2943: }
2944:
2945: for(i=1; i<= nlstate; i++){
2946: s1=0;
2947: for(j=1; j<i; j++){
2948: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2949: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2950: }
2951: for(j=i+1; j<=nlstate+ndeath; j++){
2952: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2953: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2954: }
2955: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2956: ps[i][i]=1./(s1+1.);
2957: /* Computing other pijs */
2958: for(j=1; j<i; j++)
2959: ps[i][j]= exp(ps[i][j])*ps[i][i];
2960: for(j=i+1; j<=nlstate+ndeath; j++)
2961: ps[i][j]= exp(ps[i][j])*ps[i][i];
2962: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2963: } /* end i */
2964:
2965: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2966: for(jj=1; jj<= nlstate+ndeath; jj++){
2967: ps[ii][jj]=0;
2968: ps[ii][ii]=1;
2969: }
2970: }
2971: /* Added for backcast */ /* Transposed matrix too */
2972: for(jj=1; jj<= nlstate+ndeath; jj++){
2973: s1=0.;
2974: for(ii=1; ii<= nlstate+ndeath; ii++){
2975: s1+=ps[ii][jj];
2976: }
2977: for(ii=1; ii<= nlstate; ii++){
2978: ps[ii][jj]=ps[ii][jj]/s1;
2979: }
2980: }
2981: /* Transposition */
2982: for(jj=1; jj<= nlstate+ndeath; jj++){
2983: for(ii=jj; ii<= nlstate+ndeath; ii++){
2984: s1=ps[ii][jj];
2985: ps[ii][jj]=ps[jj][ii];
2986: ps[jj][ii]=s1;
2987: }
2988: }
2989: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2990: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2991: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2992: /* } */
2993: /* printf("\n "); */
2994: /* } */
2995: /* printf("\n ");printf("%lf ",cov[2]);*/
2996: /*
2997: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2998: goto end;*/
2999: return ps;
1.217 brouard 3000: }
3001:
3002:
1.126 brouard 3003: /**************** Product of 2 matrices ******************/
3004:
1.145 brouard 3005: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3006: {
3007: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3008: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3009: /* in, b, out are matrice of pointers which should have been initialized
3010: before: only the contents of out is modified. The function returns
3011: a pointer to pointers identical to out */
1.145 brouard 3012: int i, j, k;
1.126 brouard 3013: for(i=nrl; i<= nrh; i++)
1.145 brouard 3014: for(k=ncolol; k<=ncoloh; k++){
3015: out[i][k]=0.;
3016: for(j=ncl; j<=nch; j++)
3017: out[i][k] +=in[i][j]*b[j][k];
3018: }
1.126 brouard 3019: return out;
3020: }
3021:
3022:
3023: /************* Higher Matrix Product ***************/
3024:
1.235 brouard 3025: 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 3026: {
1.218 brouard 3027: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3028: 'nhstepm*hstepm*stepm' months (i.e. until
3029: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3030: nhstepm*hstepm matrices.
3031: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3032: (typically every 2 years instead of every month which is too big
3033: for the memory).
3034: Model is determined by parameters x and covariates have to be
3035: included manually here.
3036:
3037: */
3038:
3039: int i, j, d, h, k;
1.131 brouard 3040: double **out, cov[NCOVMAX+1];
1.126 brouard 3041: double **newm;
1.187 brouard 3042: double agexact;
1.214 brouard 3043: double agebegin, ageend;
1.126 brouard 3044:
3045: /* Hstepm could be zero and should return the unit matrix */
3046: for (i=1;i<=nlstate+ndeath;i++)
3047: for (j=1;j<=nlstate+ndeath;j++){
3048: oldm[i][j]=(i==j ? 1.0 : 0.0);
3049: po[i][j][0]=(i==j ? 1.0 : 0.0);
3050: }
3051: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3052: for(h=1; h <=nhstepm; h++){
3053: for(d=1; d <=hstepm; d++){
3054: newm=savm;
3055: /* Covariates have to be included here again */
3056: cov[1]=1.;
1.214 brouard 3057: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3058: cov[2]=agexact;
3059: if(nagesqr==1)
1.227 brouard 3060: cov[3]= agexact*agexact;
1.235 brouard 3061: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3062: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3063: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3064: /* 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)); */
3065: }
3066: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3067: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3068: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3069: /* 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]); */
3070: }
3071: for (k=1; k<=cptcovage;k++){
3072: if(Dummy[Tvar[Tage[k]]]){
3073: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3074: } else{
3075: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3076: }
3077: /* 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]); */
3078: }
3079: for (k=1; k<=cptcovprod;k++){ /* */
3080: /* 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]); */
3081: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3082: }
3083: /* for (k=1; k<=cptcovn;k++) */
3084: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3085: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3086: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3087: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3088: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3089:
3090:
1.126 brouard 3091: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3092: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3093: /* right multiplication of oldm by the current matrix */
1.126 brouard 3094: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3095: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3096: /* if((int)age == 70){ */
3097: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3098: /* for(i=1; i<=nlstate+ndeath; i++) { */
3099: /* printf("%d pmmij ",i); */
3100: /* for(j=1;j<=nlstate+ndeath;j++) { */
3101: /* printf("%f ",pmmij[i][j]); */
3102: /* } */
3103: /* printf(" oldm "); */
3104: /* for(j=1;j<=nlstate+ndeath;j++) { */
3105: /* printf("%f ",oldm[i][j]); */
3106: /* } */
3107: /* printf("\n"); */
3108: /* } */
3109: /* } */
1.126 brouard 3110: savm=oldm;
3111: oldm=newm;
3112: }
3113: for(i=1; i<=nlstate+ndeath; i++)
3114: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3115: po[i][j][h]=newm[i][j];
3116: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3117: }
1.128 brouard 3118: /*printf("h=%d ",h);*/
1.126 brouard 3119: } /* end h */
1.218 brouard 3120: /* printf("\n H=%d \n",h); */
1.126 brouard 3121: return po;
3122: }
3123:
1.217 brouard 3124: /************* Higher Back Matrix Product ***************/
1.218 brouard 3125: /* 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 3126: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3127: {
1.218 brouard 3128: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3129: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3130: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3131: nhstepm*hstepm matrices.
3132: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3133: (typically every 2 years instead of every month which is too big
1.217 brouard 3134: for the memory).
1.218 brouard 3135: Model is determined by parameters x and covariates have to be
3136: included manually here.
1.217 brouard 3137:
1.222 brouard 3138: */
1.217 brouard 3139:
3140: int i, j, d, h, k;
3141: double **out, cov[NCOVMAX+1];
3142: double **newm;
3143: double agexact;
3144: double agebegin, ageend;
1.222 brouard 3145: double **oldm, **savm;
1.217 brouard 3146:
1.222 brouard 3147: oldm=oldms;savm=savms;
1.217 brouard 3148: /* Hstepm could be zero and should return the unit matrix */
3149: for (i=1;i<=nlstate+ndeath;i++)
3150: for (j=1;j<=nlstate+ndeath;j++){
3151: oldm[i][j]=(i==j ? 1.0 : 0.0);
3152: po[i][j][0]=(i==j ? 1.0 : 0.0);
3153: }
3154: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3155: for(h=1; h <=nhstepm; h++){
3156: for(d=1; d <=hstepm; d++){
3157: newm=savm;
3158: /* Covariates have to be included here again */
3159: cov[1]=1.;
3160: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3161: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3162: cov[2]=agexact;
3163: if(nagesqr==1)
1.222 brouard 3164: cov[3]= agexact*agexact;
1.218 brouard 3165: for (k=1; k<=cptcovn;k++)
1.222 brouard 3166: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3167: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3168: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3169: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3170: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3171: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3172: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3173: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3174: /* 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 3175:
3176:
1.217 brouard 3177: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3178: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3179: /* Careful transposed matrix */
1.222 brouard 3180: /* age is in cov[2] */
1.218 brouard 3181: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3182: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3183: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3184: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3185: /* if((int)age == 70){ */
3186: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3187: /* for(i=1; i<=nlstate+ndeath; i++) { */
3188: /* printf("%d pmmij ",i); */
3189: /* for(j=1;j<=nlstate+ndeath;j++) { */
3190: /* printf("%f ",pmmij[i][j]); */
3191: /* } */
3192: /* printf(" oldm "); */
3193: /* for(j=1;j<=nlstate+ndeath;j++) { */
3194: /* printf("%f ",oldm[i][j]); */
3195: /* } */
3196: /* printf("\n"); */
3197: /* } */
3198: /* } */
3199: savm=oldm;
3200: oldm=newm;
3201: }
3202: for(i=1; i<=nlstate+ndeath; i++)
3203: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3204: po[i][j][h]=newm[i][j];
3205: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3206: }
3207: /*printf("h=%d ",h);*/
3208: } /* end h */
1.222 brouard 3209: /* printf("\n H=%d \n",h); */
1.217 brouard 3210: return po;
3211: }
3212:
3213:
1.162 brouard 3214: #ifdef NLOPT
3215: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3216: double fret;
3217: double *xt;
3218: int j;
3219: myfunc_data *d2 = (myfunc_data *) pd;
3220: /* xt = (p1-1); */
3221: xt=vector(1,n);
3222: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3223:
3224: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3225: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3226: printf("Function = %.12lf ",fret);
3227: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3228: printf("\n");
3229: free_vector(xt,1,n);
3230: return fret;
3231: }
3232: #endif
1.126 brouard 3233:
3234: /*************** log-likelihood *************/
3235: double func( double *x)
3236: {
1.226 brouard 3237: int i, ii, j, k, mi, d, kk;
3238: int ioffset=0;
3239: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3240: double **out;
3241: double lli; /* Individual log likelihood */
3242: int s1, s2;
1.228 brouard 3243: 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 3244: double bbh, survp;
3245: long ipmx;
3246: double agexact;
3247: /*extern weight */
3248: /* We are differentiating ll according to initial status */
3249: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3250: /*for(i=1;i<imx;i++)
3251: printf(" %d\n",s[4][i]);
3252: */
1.162 brouard 3253:
1.226 brouard 3254: ++countcallfunc;
1.162 brouard 3255:
1.226 brouard 3256: cov[1]=1.;
1.126 brouard 3257:
1.226 brouard 3258: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3259: ioffset=0;
1.226 brouard 3260: if(mle==1){
3261: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3262: /* Computes the values of the ncovmodel covariates of the model
3263: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3264: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3265: to be observed in j being in i according to the model.
3266: */
1.243 brouard 3267: ioffset=2+nagesqr ;
1.233 brouard 3268: /* Fixed */
1.234 brouard 3269: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3270: 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)*/
3271: }
1.226 brouard 3272: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3273: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3274: has been calculated etc */
3275: /* For an individual i, wav[i] gives the number of effective waves */
3276: /* We compute the contribution to Likelihood of each effective transition
3277: mw[mi][i] is real wave of the mi th effectve wave */
3278: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3279: s2=s[mw[mi+1][i]][i];
3280: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3281: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3282: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3283: */
3284: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3285: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3286: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3287: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3288: }
3289: for (ii=1;ii<=nlstate+ndeath;ii++)
3290: for (j=1;j<=nlstate+ndeath;j++){
3291: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3292: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3293: }
3294: for(d=0; d<dh[mi][i]; d++){
3295: newm=savm;
3296: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3297: cov[2]=agexact;
3298: if(nagesqr==1)
3299: cov[3]= agexact*agexact; /* Should be changed here */
3300: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3301: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3302: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3303: else
3304: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3305: }
3306: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3307: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3308: savm=oldm;
3309: oldm=newm;
3310: } /* end mult */
3311:
3312: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3313: /* But now since version 0.9 we anticipate for bias at large stepm.
3314: * If stepm is larger than one month (smallest stepm) and if the exact delay
3315: * (in months) between two waves is not a multiple of stepm, we rounded to
3316: * the nearest (and in case of equal distance, to the lowest) interval but now
3317: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3318: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3319: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3320: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3321: * -stepm/2 to stepm/2 .
3322: * For stepm=1 the results are the same as for previous versions of Imach.
3323: * For stepm > 1 the results are less biased than in previous versions.
3324: */
1.234 brouard 3325: s1=s[mw[mi][i]][i];
3326: s2=s[mw[mi+1][i]][i];
3327: bbh=(double)bh[mi][i]/(double)stepm;
3328: /* bias bh is positive if real duration
3329: * is higher than the multiple of stepm and negative otherwise.
3330: */
3331: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3332: if( s2 > nlstate){
3333: /* i.e. if s2 is a death state and if the date of death is known
3334: then the contribution to the likelihood is the probability to
3335: die between last step unit time and current step unit time,
3336: which is also equal to probability to die before dh
3337: minus probability to die before dh-stepm .
3338: In version up to 0.92 likelihood was computed
3339: as if date of death was unknown. Death was treated as any other
3340: health state: the date of the interview describes the actual state
3341: and not the date of a change in health state. The former idea was
3342: to consider that at each interview the state was recorded
3343: (healthy, disable or death) and IMaCh was corrected; but when we
3344: introduced the exact date of death then we should have modified
3345: the contribution of an exact death to the likelihood. This new
3346: contribution is smaller and very dependent of the step unit
3347: stepm. It is no more the probability to die between last interview
3348: and month of death but the probability to survive from last
3349: interview up to one month before death multiplied by the
3350: probability to die within a month. Thanks to Chris
3351: Jackson for correcting this bug. Former versions increased
3352: mortality artificially. The bad side is that we add another loop
3353: which slows down the processing. The difference can be up to 10%
3354: lower mortality.
3355: */
3356: /* If, at the beginning of the maximization mostly, the
3357: cumulative probability or probability to be dead is
3358: constant (ie = 1) over time d, the difference is equal to
3359: 0. out[s1][3] = savm[s1][3]: probability, being at state
3360: s1 at precedent wave, to be dead a month before current
3361: wave is equal to probability, being at state s1 at
3362: precedent wave, to be dead at mont of the current
3363: wave. Then the observed probability (that this person died)
3364: is null according to current estimated parameter. In fact,
3365: it should be very low but not zero otherwise the log go to
3366: infinity.
3367: */
1.183 brouard 3368: /* #ifdef INFINITYORIGINAL */
3369: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3370: /* #else */
3371: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3372: /* lli=log(mytinydouble); */
3373: /* else */
3374: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3375: /* #endif */
1.226 brouard 3376: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3377:
1.226 brouard 3378: } else if ( s2==-1 ) { /* alive */
3379: for (j=1,survp=0. ; j<=nlstate; j++)
3380: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3381: /*survp += out[s1][j]; */
3382: lli= log(survp);
3383: }
3384: else if (s2==-4) {
3385: for (j=3,survp=0. ; j<=nlstate; j++)
3386: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3387: lli= log(survp);
3388: }
3389: else if (s2==-5) {
3390: for (j=1,survp=0. ; j<=2; j++)
3391: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3392: lli= log(survp);
3393: }
3394: else{
3395: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3396: /* 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 */
3397: }
3398: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3399: /*if(lli ==000.0)*/
3400: /*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); */
3401: ipmx +=1;
3402: sw += weight[i];
3403: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3404: /* if (lli < log(mytinydouble)){ */
3405: /* 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); */
3406: /* 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]); */
3407: /* } */
3408: } /* end of wave */
3409: } /* end of individual */
3410: } else if(mle==2){
3411: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3412: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3413: for(mi=1; mi<= wav[i]-1; mi++){
3414: for (ii=1;ii<=nlstate+ndeath;ii++)
3415: for (j=1;j<=nlstate+ndeath;j++){
3416: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3417: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3418: }
3419: for(d=0; d<=dh[mi][i]; d++){
3420: newm=savm;
3421: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3422: cov[2]=agexact;
3423: if(nagesqr==1)
3424: cov[3]= agexact*agexact;
3425: for (kk=1; kk<=cptcovage;kk++) {
3426: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3427: }
3428: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3429: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3430: savm=oldm;
3431: oldm=newm;
3432: } /* end mult */
3433:
3434: s1=s[mw[mi][i]][i];
3435: s2=s[mw[mi+1][i]][i];
3436: bbh=(double)bh[mi][i]/(double)stepm;
3437: 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 */
3438: ipmx +=1;
3439: sw += weight[i];
3440: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3441: } /* end of wave */
3442: } /* end of individual */
3443: } else if(mle==3){ /* exponential inter-extrapolation */
3444: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3445: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3446: for(mi=1; mi<= wav[i]-1; mi++){
3447: for (ii=1;ii<=nlstate+ndeath;ii++)
3448: for (j=1;j<=nlstate+ndeath;j++){
3449: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3450: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3451: }
3452: for(d=0; d<dh[mi][i]; d++){
3453: newm=savm;
3454: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3455: cov[2]=agexact;
3456: if(nagesqr==1)
3457: cov[3]= agexact*agexact;
3458: for (kk=1; kk<=cptcovage;kk++) {
3459: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3460: }
3461: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3462: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3463: savm=oldm;
3464: oldm=newm;
3465: } /* end mult */
3466:
3467: s1=s[mw[mi][i]][i];
3468: s2=s[mw[mi+1][i]][i];
3469: bbh=(double)bh[mi][i]/(double)stepm;
3470: 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 */
3471: ipmx +=1;
3472: sw += weight[i];
3473: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3474: } /* end of wave */
3475: } /* end of individual */
3476: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3477: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3478: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3479: for(mi=1; mi<= wav[i]-1; mi++){
3480: for (ii=1;ii<=nlstate+ndeath;ii++)
3481: for (j=1;j<=nlstate+ndeath;j++){
3482: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3483: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3484: }
3485: for(d=0; d<dh[mi][i]; d++){
3486: newm=savm;
3487: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3488: cov[2]=agexact;
3489: if(nagesqr==1)
3490: cov[3]= agexact*agexact;
3491: for (kk=1; kk<=cptcovage;kk++) {
3492: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3493: }
1.126 brouard 3494:
1.226 brouard 3495: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3496: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3497: savm=oldm;
3498: oldm=newm;
3499: } /* end mult */
3500:
3501: s1=s[mw[mi][i]][i];
3502: s2=s[mw[mi+1][i]][i];
3503: if( s2 > nlstate){
3504: lli=log(out[s1][s2] - savm[s1][s2]);
3505: } else if ( s2==-1 ) { /* alive */
3506: for (j=1,survp=0. ; j<=nlstate; j++)
3507: survp += out[s1][j];
3508: lli= log(survp);
3509: }else{
3510: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3511: }
3512: ipmx +=1;
3513: sw += weight[i];
3514: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3515: /* 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 3516: } /* end of wave */
3517: } /* end of individual */
3518: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3519: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3520: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3521: for(mi=1; mi<= wav[i]-1; mi++){
3522: for (ii=1;ii<=nlstate+ndeath;ii++)
3523: for (j=1;j<=nlstate+ndeath;j++){
3524: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3525: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3526: }
3527: for(d=0; d<dh[mi][i]; d++){
3528: newm=savm;
3529: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3530: cov[2]=agexact;
3531: if(nagesqr==1)
3532: cov[3]= agexact*agexact;
3533: for (kk=1; kk<=cptcovage;kk++) {
3534: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3535: }
1.126 brouard 3536:
1.226 brouard 3537: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3538: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3539: savm=oldm;
3540: oldm=newm;
3541: } /* end mult */
3542:
3543: s1=s[mw[mi][i]][i];
3544: s2=s[mw[mi+1][i]][i];
3545: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3546: ipmx +=1;
3547: sw += weight[i];
3548: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3549: /*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]);*/
3550: } /* end of wave */
3551: } /* end of individual */
3552: } /* End of if */
3553: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3554: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3555: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3556: return -l;
1.126 brouard 3557: }
3558:
3559: /*************** log-likelihood *************/
3560: double funcone( double *x)
3561: {
1.228 brouard 3562: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3563: int i, ii, j, k, mi, d, kk;
1.228 brouard 3564: int ioffset=0;
1.131 brouard 3565: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3566: double **out;
3567: double lli; /* Individual log likelihood */
3568: double llt;
3569: int s1, s2;
1.228 brouard 3570: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3571:
1.126 brouard 3572: double bbh, survp;
1.187 brouard 3573: double agexact;
1.214 brouard 3574: double agebegin, ageend;
1.126 brouard 3575: /*extern weight */
3576: /* We are differentiating ll according to initial status */
3577: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3578: /*for(i=1;i<imx;i++)
3579: printf(" %d\n",s[4][i]);
3580: */
3581: cov[1]=1.;
3582:
3583: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3584: ioffset=0;
3585: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3586: /* ioffset=2+nagesqr+cptcovage; */
3587: ioffset=2+nagesqr;
1.232 brouard 3588: /* Fixed */
1.224 brouard 3589: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3590: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3591: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3592: 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)*/
3593: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3594: /* cov[2+6]=covar[Tvar[6]][i]; */
3595: /* cov[2+6]=covar[2][i]; V2 */
3596: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3597: /* cov[2+7]=covar[Tvar[7]][i]; */
3598: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3599: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3600: /* cov[2+9]=covar[Tvar[9]][i]; */
3601: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3602: }
1.232 brouard 3603: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3604: /* 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?)*\/ */
3605: /* } */
1.231 brouard 3606: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3607: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3608: /* } */
1.225 brouard 3609:
1.233 brouard 3610:
3611: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3612: /* Wave varying (but not age varying) */
3613: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3614: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3615: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3616: }
1.232 brouard 3617: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3618: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3619: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3620: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3621: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3622: /* 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 3623: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3624: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3625: /* /\* 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]); *\/ */
3626: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3627: /* } */
1.126 brouard 3628: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3629: for (j=1;j<=nlstate+ndeath;j++){
3630: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3631: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3632: }
1.214 brouard 3633:
3634: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3635: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3636: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3637: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3638: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3639: and mw[mi+1][i]. dh depends on stepm.*/
3640: newm=savm;
1.247 brouard 3641: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3642: cov[2]=agexact;
3643: if(nagesqr==1)
3644: cov[3]= agexact*agexact;
3645: for (kk=1; kk<=cptcovage;kk++) {
3646: if(!FixedV[Tvar[Tage[kk]]])
3647: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3648: else
3649: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3650: }
3651: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3652: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3653: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3654: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3655: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3656: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3657: savm=oldm;
3658: oldm=newm;
1.126 brouard 3659: } /* end mult */
3660:
3661: s1=s[mw[mi][i]][i];
3662: s2=s[mw[mi+1][i]][i];
1.217 brouard 3663: /* if(s2==-1){ */
3664: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3665: /* /\* exit(1); *\/ */
3666: /* } */
1.126 brouard 3667: bbh=(double)bh[mi][i]/(double)stepm;
3668: /* bias is positive if real duration
3669: * is higher than the multiple of stepm and negative otherwise.
3670: */
3671: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3672: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3673: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3674: for (j=1,survp=0. ; j<=nlstate; j++)
3675: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3676: lli= log(survp);
1.126 brouard 3677: }else if (mle==1){
1.242 brouard 3678: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3679: } else if(mle==2){
1.242 brouard 3680: 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 3681: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3682: 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 3683: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3684: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3685: } else{ /* mle=0 back to 1 */
1.242 brouard 3686: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3687: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3688: } /* End of if */
3689: ipmx +=1;
3690: sw += weight[i];
3691: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3692: /*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 3693: if(globpr){
1.246 brouard 3694: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3695: %11.6f %11.6f %11.6f ", \
1.242 brouard 3696: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3697: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3698: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3699: llt +=ll[k]*gipmx/gsw;
3700: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3701: }
3702: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3703: }
1.232 brouard 3704: } /* end of wave */
3705: } /* end of individual */
3706: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3707: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3708: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3709: if(globpr==0){ /* First time we count the contributions and weights */
3710: gipmx=ipmx;
3711: gsw=sw;
3712: }
3713: return -l;
1.126 brouard 3714: }
3715:
3716:
3717: /*************** function likelione ***********/
3718: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3719: {
3720: /* This routine should help understanding what is done with
3721: the selection of individuals/waves and
3722: to check the exact contribution to the likelihood.
3723: Plotting could be done.
3724: */
3725: int k;
3726:
3727: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3728: strcpy(fileresilk,"ILK_");
1.202 brouard 3729: strcat(fileresilk,fileresu);
1.126 brouard 3730: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3731: printf("Problem with resultfile: %s\n", fileresilk);
3732: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3733: }
1.214 brouard 3734: 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");
3735: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3736: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3737: for(k=1; k<=nlstate; k++)
3738: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3739: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3740: }
3741:
3742: *fretone=(*funcone)(p);
3743: if(*globpri !=0){
3744: fclose(ficresilk);
1.205 brouard 3745: if (mle ==0)
3746: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3747: else if(mle >=1)
3748: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3749: 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 3750:
1.208 brouard 3751:
3752: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3753: 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 3754: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3755: }
1.207 brouard 3756: 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 3757: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3758: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3759: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3760: fflush(fichtm);
1.205 brouard 3761: }
1.126 brouard 3762: return;
3763: }
3764:
3765:
3766: /*********** Maximum Likelihood Estimation ***************/
3767:
3768: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3769: {
1.165 brouard 3770: int i,j, iter=0;
1.126 brouard 3771: double **xi;
3772: double fret;
3773: double fretone; /* Only one call to likelihood */
3774: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3775:
3776: #ifdef NLOPT
3777: int creturn;
3778: nlopt_opt opt;
3779: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3780: double *lb;
3781: double minf; /* the minimum objective value, upon return */
3782: double * p1; /* Shifted parameters from 0 instead of 1 */
3783: myfunc_data dinst, *d = &dinst;
3784: #endif
3785:
3786:
1.126 brouard 3787: xi=matrix(1,npar,1,npar);
3788: for (i=1;i<=npar;i++)
3789: for (j=1;j<=npar;j++)
3790: xi[i][j]=(i==j ? 1.0 : 0.0);
3791: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3792: strcpy(filerespow,"POW_");
1.126 brouard 3793: strcat(filerespow,fileres);
3794: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3795: printf("Problem with resultfile: %s\n", filerespow);
3796: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3797: }
3798: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3799: for (i=1;i<=nlstate;i++)
3800: for(j=1;j<=nlstate+ndeath;j++)
3801: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3802: fprintf(ficrespow,"\n");
1.162 brouard 3803: #ifdef POWELL
1.126 brouard 3804: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3805: #endif
1.126 brouard 3806:
1.162 brouard 3807: #ifdef NLOPT
3808: #ifdef NEWUOA
3809: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3810: #else
3811: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3812: #endif
3813: lb=vector(0,npar-1);
3814: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3815: nlopt_set_lower_bounds(opt, lb);
3816: nlopt_set_initial_step1(opt, 0.1);
3817:
3818: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3819: d->function = func;
3820: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3821: nlopt_set_min_objective(opt, myfunc, d);
3822: nlopt_set_xtol_rel(opt, ftol);
3823: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3824: printf("nlopt failed! %d\n",creturn);
3825: }
3826: else {
3827: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3828: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3829: iter=1; /* not equal */
3830: }
3831: nlopt_destroy(opt);
3832: #endif
1.126 brouard 3833: free_matrix(xi,1,npar,1,npar);
3834: fclose(ficrespow);
1.203 brouard 3835: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3836: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3837: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3838:
3839: }
3840:
3841: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3842: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3843: {
3844: double **a,**y,*x,pd;
1.203 brouard 3845: /* double **hess; */
1.164 brouard 3846: int i, j;
1.126 brouard 3847: int *indx;
3848:
3849: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3850: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3851: void lubksb(double **a, int npar, int *indx, double b[]) ;
3852: void ludcmp(double **a, int npar, int *indx, double *d) ;
3853: double gompertz(double p[]);
1.203 brouard 3854: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3855:
3856: printf("\nCalculation of the hessian matrix. Wait...\n");
3857: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3858: for (i=1;i<=npar;i++){
1.203 brouard 3859: printf("%d-",i);fflush(stdout);
3860: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3861:
3862: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3863:
3864: /* printf(" %f ",p[i]);
3865: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3866: }
3867:
3868: for (i=1;i<=npar;i++) {
3869: for (j=1;j<=npar;j++) {
3870: if (j>i) {
1.203 brouard 3871: printf(".%d-%d",i,j);fflush(stdout);
3872: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3873: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3874:
3875: hess[j][i]=hess[i][j];
3876: /*printf(" %lf ",hess[i][j]);*/
3877: }
3878: }
3879: }
3880: printf("\n");
3881: fprintf(ficlog,"\n");
3882:
3883: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3884: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3885:
3886: a=matrix(1,npar,1,npar);
3887: y=matrix(1,npar,1,npar);
3888: x=vector(1,npar);
3889: indx=ivector(1,npar);
3890: for (i=1;i<=npar;i++)
3891: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3892: ludcmp(a,npar,indx,&pd);
3893:
3894: for (j=1;j<=npar;j++) {
3895: for (i=1;i<=npar;i++) x[i]=0;
3896: x[j]=1;
3897: lubksb(a,npar,indx,x);
3898: for (i=1;i<=npar;i++){
3899: matcov[i][j]=x[i];
3900: }
3901: }
3902:
3903: printf("\n#Hessian matrix#\n");
3904: fprintf(ficlog,"\n#Hessian matrix#\n");
3905: for (i=1;i<=npar;i++) {
3906: for (j=1;j<=npar;j++) {
1.203 brouard 3907: printf("%.6e ",hess[i][j]);
3908: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3909: }
3910: printf("\n");
3911: fprintf(ficlog,"\n");
3912: }
3913:
1.203 brouard 3914: /* printf("\n#Covariance matrix#\n"); */
3915: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3916: /* for (i=1;i<=npar;i++) { */
3917: /* for (j=1;j<=npar;j++) { */
3918: /* printf("%.6e ",matcov[i][j]); */
3919: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3920: /* } */
3921: /* printf("\n"); */
3922: /* fprintf(ficlog,"\n"); */
3923: /* } */
3924:
1.126 brouard 3925: /* Recompute Inverse */
1.203 brouard 3926: /* for (i=1;i<=npar;i++) */
3927: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3928: /* ludcmp(a,npar,indx,&pd); */
3929:
3930: /* printf("\n#Hessian matrix recomputed#\n"); */
3931:
3932: /* for (j=1;j<=npar;j++) { */
3933: /* for (i=1;i<=npar;i++) x[i]=0; */
3934: /* x[j]=1; */
3935: /* lubksb(a,npar,indx,x); */
3936: /* for (i=1;i<=npar;i++){ */
3937: /* y[i][j]=x[i]; */
3938: /* printf("%.3e ",y[i][j]); */
3939: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3940: /* } */
3941: /* printf("\n"); */
3942: /* fprintf(ficlog,"\n"); */
3943: /* } */
3944:
3945: /* Verifying the inverse matrix */
3946: #ifdef DEBUGHESS
3947: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3948:
1.203 brouard 3949: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3950: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3951:
3952: for (j=1;j<=npar;j++) {
3953: for (i=1;i<=npar;i++){
1.203 brouard 3954: printf("%.2f ",y[i][j]);
3955: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3956: }
3957: printf("\n");
3958: fprintf(ficlog,"\n");
3959: }
1.203 brouard 3960: #endif
1.126 brouard 3961:
3962: free_matrix(a,1,npar,1,npar);
3963: free_matrix(y,1,npar,1,npar);
3964: free_vector(x,1,npar);
3965: free_ivector(indx,1,npar);
1.203 brouard 3966: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3967:
3968:
3969: }
3970:
3971: /*************** hessian matrix ****************/
3972: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3973: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3974: int i;
3975: int l=1, lmax=20;
1.203 brouard 3976: double k1,k2, res, fx;
1.132 brouard 3977: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3978: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3979: int k=0,kmax=10;
3980: double l1;
3981:
3982: fx=func(x);
3983: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3984: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3985: l1=pow(10,l);
3986: delts=delt;
3987: for(k=1 ; k <kmax; k=k+1){
3988: delt = delta*(l1*k);
3989: p2[theta]=x[theta] +delt;
1.145 brouard 3990: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3991: p2[theta]=x[theta]-delt;
3992: k2=func(p2)-fx;
3993: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3994: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3995:
1.203 brouard 3996: #ifdef DEBUGHESSII
1.126 brouard 3997: 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);
3998: 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);
3999: #endif
4000: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4001: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4002: k=kmax;
4003: }
4004: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4005: k=kmax; l=lmax*10;
1.126 brouard 4006: }
4007: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4008: delts=delt;
4009: }
1.203 brouard 4010: } /* End loop k */
1.126 brouard 4011: }
4012: delti[theta]=delts;
4013: return res;
4014:
4015: }
4016:
1.203 brouard 4017: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4018: {
4019: int i;
1.164 brouard 4020: int l=1, lmax=20;
1.126 brouard 4021: double k1,k2,k3,k4,res,fx;
1.132 brouard 4022: double p2[MAXPARM+1];
1.203 brouard 4023: int k, kmax=1;
4024: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4025:
4026: int firstime=0;
1.203 brouard 4027:
1.126 brouard 4028: fx=func(x);
1.203 brouard 4029: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4030: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4031: p2[thetai]=x[thetai]+delti[thetai]*k;
4032: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4033: k1=func(p2)-fx;
4034:
1.203 brouard 4035: p2[thetai]=x[thetai]+delti[thetai]*k;
4036: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4037: k2=func(p2)-fx;
4038:
1.203 brouard 4039: p2[thetai]=x[thetai]-delti[thetai]*k;
4040: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4041: k3=func(p2)-fx;
4042:
1.203 brouard 4043: p2[thetai]=x[thetai]-delti[thetai]*k;
4044: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4045: k4=func(p2)-fx;
1.203 brouard 4046: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4047: if(k1*k2*k3*k4 <0.){
1.208 brouard 4048: firstime=1;
1.203 brouard 4049: kmax=kmax+10;
1.208 brouard 4050: }
4051: if(kmax >=10 || firstime ==1){
1.246 brouard 4052: 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);
4053: 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 4054: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4055: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
4056: }
4057: #ifdef DEBUGHESSIJ
4058: v1=hess[thetai][thetai];
4059: v2=hess[thetaj][thetaj];
4060: cv12=res;
4061: /* Computing eigen value of Hessian matrix */
4062: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4063: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4064: if ((lc2 <0) || (lc1 <0) ){
4065: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4066: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4067: 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);
4068: 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);
4069: }
1.126 brouard 4070: #endif
4071: }
4072: return res;
4073: }
4074:
1.203 brouard 4075: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4076: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4077: /* { */
4078: /* int i; */
4079: /* int l=1, lmax=20; */
4080: /* double k1,k2,k3,k4,res,fx; */
4081: /* double p2[MAXPARM+1]; */
4082: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4083: /* int k=0,kmax=10; */
4084: /* double l1; */
4085:
4086: /* fx=func(x); */
4087: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4088: /* l1=pow(10,l); */
4089: /* delts=delt; */
4090: /* for(k=1 ; k <kmax; k=k+1){ */
4091: /* delt = delti*(l1*k); */
4092: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4093: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4094: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4095: /* k1=func(p2)-fx; */
4096:
4097: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4098: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4099: /* k2=func(p2)-fx; */
4100:
4101: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4102: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4103: /* k3=func(p2)-fx; */
4104:
4105: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4106: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4107: /* k4=func(p2)-fx; */
4108: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4109: /* #ifdef DEBUGHESSIJ */
4110: /* 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); */
4111: /* 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); */
4112: /* #endif */
4113: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4114: /* k=kmax; */
4115: /* } */
4116: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4117: /* k=kmax; l=lmax*10; */
4118: /* } */
4119: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4120: /* delts=delt; */
4121: /* } */
4122: /* } /\* End loop k *\/ */
4123: /* } */
4124: /* delti[theta]=delts; */
4125: /* return res; */
4126: /* } */
4127:
4128:
1.126 brouard 4129: /************** Inverse of matrix **************/
4130: void ludcmp(double **a, int n, int *indx, double *d)
4131: {
4132: int i,imax,j,k;
4133: double big,dum,sum,temp;
4134: double *vv;
4135:
4136: vv=vector(1,n);
4137: *d=1.0;
4138: for (i=1;i<=n;i++) {
4139: big=0.0;
4140: for (j=1;j<=n;j++)
4141: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4142: if (big == 0.0){
4143: printf(" Singular Hessian matrix at row %d:\n",i);
4144: for (j=1;j<=n;j++) {
4145: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4146: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4147: }
4148: fflush(ficlog);
4149: fclose(ficlog);
4150: nrerror("Singular matrix in routine ludcmp");
4151: }
1.126 brouard 4152: vv[i]=1.0/big;
4153: }
4154: for (j=1;j<=n;j++) {
4155: for (i=1;i<j;i++) {
4156: sum=a[i][j];
4157: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4158: a[i][j]=sum;
4159: }
4160: big=0.0;
4161: for (i=j;i<=n;i++) {
4162: sum=a[i][j];
4163: for (k=1;k<j;k++)
4164: sum -= a[i][k]*a[k][j];
4165: a[i][j]=sum;
4166: if ( (dum=vv[i]*fabs(sum)) >= big) {
4167: big=dum;
4168: imax=i;
4169: }
4170: }
4171: if (j != imax) {
4172: for (k=1;k<=n;k++) {
4173: dum=a[imax][k];
4174: a[imax][k]=a[j][k];
4175: a[j][k]=dum;
4176: }
4177: *d = -(*d);
4178: vv[imax]=vv[j];
4179: }
4180: indx[j]=imax;
4181: if (a[j][j] == 0.0) a[j][j]=TINY;
4182: if (j != n) {
4183: dum=1.0/(a[j][j]);
4184: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4185: }
4186: }
4187: free_vector(vv,1,n); /* Doesn't work */
4188: ;
4189: }
4190:
4191: void lubksb(double **a, int n, int *indx, double b[])
4192: {
4193: int i,ii=0,ip,j;
4194: double sum;
4195:
4196: for (i=1;i<=n;i++) {
4197: ip=indx[i];
4198: sum=b[ip];
4199: b[ip]=b[i];
4200: if (ii)
4201: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4202: else if (sum) ii=i;
4203: b[i]=sum;
4204: }
4205: for (i=n;i>=1;i--) {
4206: sum=b[i];
4207: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4208: b[i]=sum/a[i][i];
4209: }
4210: }
4211:
4212: void pstamp(FILE *fichier)
4213: {
1.196 brouard 4214: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4215: }
4216:
1.253 brouard 4217: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4218:
4219: /* y=a+bx regression */
4220: double sumx = 0.0; /* sum of x */
4221: double sumx2 = 0.0; /* sum of x**2 */
4222: double sumxy = 0.0; /* sum of x * y */
4223: double sumy = 0.0; /* sum of y */
4224: double sumy2 = 0.0; /* sum of y**2 */
4225: double sume2; /* sum of square or residuals */
4226: double yhat;
4227:
4228: double denom=0;
4229: int i;
4230: int ne=*no;
4231:
4232: for ( i=ifi, ne=0;i<=ila;i++) {
4233: if(!isfinite(x[i]) || !isfinite(y[i])){
4234: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4235: continue;
4236: }
4237: ne=ne+1;
4238: sumx += x[i];
4239: sumx2 += x[i]*x[i];
4240: sumxy += x[i] * y[i];
4241: sumy += y[i];
4242: sumy2 += y[i]*y[i];
4243: denom = (ne * sumx2 - sumx*sumx);
4244: /* 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); */
4245: }
4246:
4247: denom = (ne * sumx2 - sumx*sumx);
4248: if (denom == 0) {
4249: // vertical, slope m is infinity
4250: *b = INFINITY;
4251: *a = 0;
4252: if (r) *r = 0;
4253: return 1;
4254: }
4255:
4256: *b = (ne * sumxy - sumx * sumy) / denom;
4257: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4258: if (r!=NULL) {
4259: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4260: sqrt((sumx2 - sumx*sumx/ne) *
4261: (sumy2 - sumy*sumy/ne));
4262: }
4263: *no=ne;
4264: for ( i=ifi, ne=0;i<=ila;i++) {
4265: if(!isfinite(x[i]) || !isfinite(y[i])){
4266: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4267: continue;
4268: }
4269: ne=ne+1;
4270: yhat = y[i] - *a -*b* x[i];
4271: sume2 += yhat * yhat ;
4272:
4273: denom = (ne * sumx2 - sumx*sumx);
4274: /* 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); */
4275: }
4276: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4277: *sa= *sb * sqrt(sumx2/ne);
4278:
4279: return 0;
4280: }
4281:
1.126 brouard 4282: /************ Frequencies ********************/
1.251 brouard 4283: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4284: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4285: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4286: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4287:
1.253 brouard 4288: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4289: int iind=0, iage=0;
4290: int mi; /* Effective wave */
4291: int first;
4292: double ***freq; /* Frequencies */
1.253 brouard 4293: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4294: int no;
1.226 brouard 4295: double *meanq;
4296: double **meanqt;
4297: double *pp, **prop, *posprop, *pospropt;
4298: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4299: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4300: double agebegin, ageend;
4301:
4302: pp=vector(1,nlstate);
1.251 brouard 4303: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4304: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4305: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4306: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4307: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4308: meanqt=matrix(1,lastpass,1,nqtveff);
4309: strcpy(fileresp,"P_");
4310: strcat(fileresp,fileresu);
4311: /*strcat(fileresphtm,fileresu);*/
4312: if((ficresp=fopen(fileresp,"w"))==NULL) {
4313: printf("Problem with prevalence resultfile: %s\n", fileresp);
4314: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4315: exit(0);
4316: }
1.240 brouard 4317:
1.226 brouard 4318: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4319: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4320: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4321: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4322: fflush(ficlog);
4323: exit(70);
4324: }
4325: else{
4326: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4327: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4328: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4329: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4330: }
1.237 brouard 4331: 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 4332:
1.226 brouard 4333: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4334: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4335: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4336: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4337: fflush(ficlog);
4338: exit(70);
1.240 brouard 4339: } else{
1.226 brouard 4340: 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 4341: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4342: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4343: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4344: }
1.240 brouard 4345: 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);
4346:
1.253 brouard 4347: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4348: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4349: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4350: j1=0;
1.126 brouard 4351:
1.227 brouard 4352: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4353: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4354: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4355:
4356:
1.226 brouard 4357: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4358: reference=low_education V1=0,V2=0
4359: med_educ V1=1 V2=0,
4360: high_educ V1=0 V2=1
4361: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4362: */
1.249 brouard 4363: dateintsum=0;
4364: k2cpt=0;
4365:
1.253 brouard 4366: if(cptcoveff == 0 )
4367: nl=1; /* Constant model only */
4368: else
4369: nl=2;
4370: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4371: if(nj==1)
4372: j=0; /* First pass for the constant */
4373: else
4374: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4375: first=1;
4376: 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 */
4377: posproptt=0.;
4378: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4379: scanf("%d", i);*/
4380: for (i=-5; i<=nlstate+ndeath; i++)
4381: for (jk=-5; jk<=nlstate+ndeath; jk++)
4382: for(m=iagemin; m <= iagemax+3; m++)
4383: freq[i][jk][m]=0;
4384:
4385: for (i=1; i<=nlstate; i++) {
1.240 brouard 4386: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4387: prop[i][m]=0;
4388: posprop[i]=0;
4389: pospropt[i]=0;
4390: }
4391: /* for (z1=1; z1<= nqfveff; z1++) { */
4392: /* meanq[z1]+=0.; */
4393: /* for(m=1;m<=lastpass;m++){ */
4394: /* meanqt[m][z1]=0.; */
4395: /* } */
4396: /* } */
4397:
4398: /* dateintsum=0; */
4399: /* k2cpt=0; */
4400:
4401: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4402: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4403: bool=1;
4404: if(j !=0){
4405: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4406: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4407: /* for (z1=1; z1<= nqfveff; z1++) { */
4408: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4409: /* } */
4410: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4411: /* if(Tvaraff[z1] ==-20){ */
4412: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4413: /* }else if(Tvaraff[z1] ==-10){ */
4414: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4415: /* }else */
4416: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4417: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4418: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4419: /* 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",
4420: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4421: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4422: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4423: } /* Onlyf fixed */
4424: } /* end z1 */
4425: } /* cptcovn > 0 */
4426: } /* end any */
4427: }/* end j==0 */
4428: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4429: /* for(m=firstpass; m<=lastpass; m++){ */
4430: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4431: m=mw[mi][iind];
4432: if(j!=0){
4433: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4434: for (z1=1; z1<=cptcoveff; z1++) {
4435: if( Fixed[Tmodelind[z1]]==1){
4436: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4437: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4438: value is -1, we don't select. It differs from the
4439: constant and age model which counts them. */
4440: bool=0; /* not selected */
4441: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4442: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4443: bool=0;
4444: }
4445: }
4446: }
4447: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4448: } /* end j==0 */
4449: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4450: if(bool==1){
4451: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4452: and mw[mi+1][iind]. dh depends on stepm. */
4453: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4454: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4455: if(m >=firstpass && m <=lastpass){
4456: k2=anint[m][iind]+(mint[m][iind]/12.);
4457: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4458: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4459: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4460: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4461: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4462: if (m<lastpass) {
4463: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4464: /* 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]); */
4465: if(s[m][iind]==-1)
4466: 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.));
4467: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4468: /* if((int)agev[m][iind] == 55) */
4469: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4470: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4471: 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 4472: }
1.251 brouard 4473: } /* end if between passes */
4474: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4475: dateintsum=dateintsum+k2; /* on all covariates ?*/
4476: k2cpt++;
4477: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4478: }
1.251 brouard 4479: }else{
4480: bool=1;
4481: }/* end bool 2 */
4482: } /* end m */
4483: } /* end bool */
4484: } /* end iind = 1 to imx */
4485: /* prop[s][age] is feeded for any initial and valid live state as well as
4486: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4487:
4488:
4489: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4490: pstamp(ficresp);
4491: if (cptcoveff>0 && j!=0){
4492: printf( "\n#********** Variable ");
4493: fprintf(ficresp, "\n#********** Variable ");
4494: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4495: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4496: fprintf(ficlog, "\n#********** Variable ");
4497: for (z1=1; z1<=cptcoveff; z1++){
4498: if(!FixedV[Tvaraff[z1]]){
4499: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4500: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4501: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4502: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4503: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4504: }else{
1.251 brouard 4505: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4506: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4507: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4508: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4509: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4510: }
4511: }
4512: printf( "**********\n#");
4513: fprintf(ficresp, "**********\n#");
4514: fprintf(ficresphtm, "**********</h3>\n");
4515: fprintf(ficresphtmfr, "**********</h3>\n");
4516: fprintf(ficlog, "**********\n");
4517: }
4518: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4519: for(i=1; i<=nlstate;i++) {
4520: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4521: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4522: }
4523: fprintf(ficresp, "\n");
4524: fprintf(ficresphtm, "\n");
4525:
4526: /* Header of frequency table by age */
4527: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4528: fprintf(ficresphtmfr,"<th>Age</th> ");
4529: for(jk=-1; jk <=nlstate+ndeath; jk++){
4530: for(m=-1; m <=nlstate+ndeath; m++){
4531: if(jk!=0 && m!=0)
4532: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4533: }
1.226 brouard 4534: }
1.251 brouard 4535: fprintf(ficresphtmfr, "\n");
4536:
4537: /* For each age */
4538: for(iage=iagemin; iage <= iagemax+3; iage++){
4539: fprintf(ficresphtm,"<tr>");
4540: if(iage==iagemax+1){
4541: fprintf(ficlog,"1");
4542: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4543: }else if(iage==iagemax+2){
4544: fprintf(ficlog,"0");
4545: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4546: }else if(iage==iagemax+3){
4547: fprintf(ficlog,"Total");
4548: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4549: }else{
1.240 brouard 4550: if(first==1){
1.251 brouard 4551: first=0;
4552: printf("See log file for details...\n");
4553: }
4554: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4555: fprintf(ficlog,"Age %d", iage);
4556: }
4557: for(jk=1; jk <=nlstate ; jk++){
4558: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4559: pp[jk] += freq[jk][m][iage];
4560: }
4561: for(jk=1; jk <=nlstate ; jk++){
4562: for(m=-1, pos=0; m <=0 ; m++)
4563: pos += freq[jk][m][iage];
4564: if(pp[jk]>=1.e-10){
4565: if(first==1){
4566: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4567: }
4568: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4569: }else{
4570: if(first==1)
4571: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4572: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4573: }
4574: }
4575:
1.251 brouard 4576: for(jk=1; jk <=nlstate ; jk++){
4577: /* posprop[jk]=0; */
4578: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4579: pp[jk] += freq[jk][m][iage];
4580: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4581:
4582: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4583: pos += pp[jk]; /* pos is the total number of transitions until this age */
4584: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4585: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4586: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4587: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4588: }
1.251 brouard 4589: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4590: if(pos>=1.e-5){
1.251 brouard 4591: if(first==1)
4592: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4593: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4594: }else{
4595: if(first==1)
4596: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4597: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4598: }
4599: if( iage <= iagemax){
4600: if(pos>=1.e-5){
4601: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4602: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4603: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4604: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4605: }
4606: else{
4607: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4608: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4609: }
1.240 brouard 4610: }
1.251 brouard 4611: pospropt[jk] +=posprop[jk];
4612: } /* end loop jk */
4613: /* pospropt=0.; */
4614: for(jk=-1; jk <=nlstate+ndeath; jk++){
4615: for(m=-1; m <=nlstate+ndeath; m++){
4616: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4617: if(first==1){
4618: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4619: }
1.253 brouard 4620: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4621: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4622: }
4623: if(jk!=0 && m!=0)
4624: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4625: }
1.251 brouard 4626: } /* end loop jk */
4627: posproptt=0.;
4628: for(jk=1; jk <=nlstate; jk++){
4629: posproptt += pospropt[jk];
4630: }
4631: fprintf(ficresphtmfr,"</tr>\n ");
4632: if(iage <= iagemax){
4633: fprintf(ficresp,"\n");
4634: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4635: }
1.251 brouard 4636: if(first==1)
4637: printf("Others in log...\n");
4638: fprintf(ficlog,"\n");
4639: } /* end loop age iage */
4640: fprintf(ficresphtm,"<tr><th>Tot</th>");
4641: for(jk=1; jk <=nlstate ; jk++){
4642: if(posproptt < 1.e-5){
4643: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4644: }else{
4645: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4646: }
1.226 brouard 4647: }
1.251 brouard 4648: fprintf(ficresphtm,"</tr>\n");
4649: fprintf(ficresphtm,"</table>\n");
4650: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4651: if(posproptt < 1.e-5){
1.251 brouard 4652: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4653: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4654: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4655: invalidvarcomb[j1]=1;
1.226 brouard 4656: }else{
1.251 brouard 4657: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4658: invalidvarcomb[j1]=0;
1.226 brouard 4659: }
1.251 brouard 4660: fprintf(ficresphtmfr,"</table>\n");
4661: fprintf(ficlog,"\n");
4662: if(j!=0){
4663: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4664: for(i=1,jk=1; i <=nlstate; i++){
4665: for(k=1; k <=(nlstate+ndeath); k++){
4666: if (k != i) {
4667: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 brouard 4668: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4669: if(j1==1){ /* All dummy covariates to zero */
4670: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4671: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4672: printf("%d%d ",i,k);
4673: fprintf(ficlog,"%d%d ",i,k);
4674: 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]));
4675: 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]));
4676: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4677: }
1.253 brouard 4678: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4679: for(iage=iagemin; iage <= iagemax+3; iage++){
4680: x[iage]= (double)iage;
4681: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
4682: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
4683: }
4684: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
4685: pstart[jk]=b;
4686: pstart[jk-1]=a;
1.252 brouard 4687: }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 */
4688: 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]);
4689: 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 4690: 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 4691: printf("%d%d ",i,k);
4692: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4693: 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]));
4694: }else{ /* Other cases, like quantitative fixed or varying covariates */
4695: ;
4696: }
4697: /* printf("%12.7f )", param[i][jj][k]); */
4698: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4699: jk++;
4700: } /* end jj */
4701: } /* end k!= i */
4702: } /* end k */
4703: } /* end i, jk */
4704: } /* end j !=0 */
4705: } /* end selected combination of covariate j1 */
4706: if(j==0){ /* We can estimate starting values from the occurences in each case */
4707: printf("#Freqsummary: Starting values for the constants:\n");
4708: fprintf(ficlog,"\n");
4709: for(i=1,jk=1; i <=nlstate; i++){
4710: for(k=1; k <=(nlstate+ndeath); k++){
4711: if (k != i) {
4712: printf("%d%d ",i,k);
4713: fprintf(ficlog,"%d%d ",i,k);
4714: for(jj=1; jj <=ncovmodel; jj++){
1.253 brouard 4715: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
4716: if(jj==1){ /* Age has to be done */
4717: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4718: 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]));
4719: 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]));
4720: }
4721: /* printf("%12.7f )", param[i][jj][k]); */
4722: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4723: jk++;
1.250 brouard 4724: }
1.251 brouard 4725: printf("\n");
4726: fprintf(ficlog,"\n");
1.250 brouard 4727: }
4728: }
4729: }
1.251 brouard 4730: printf("#Freqsummary\n");
4731: fprintf(ficlog,"\n");
4732: for(jk=-1; jk <=nlstate+ndeath; jk++){
4733: for(m=-1; m <=nlstate+ndeath; m++){
4734: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4735: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4736: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4737: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4738: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4739: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4740: /* } */
4741: }
4742: } /* end loop jk */
4743:
4744: printf("\n");
4745: fprintf(ficlog,"\n");
4746: } /* end j=0 */
1.249 brouard 4747: } /* end j */
1.252 brouard 4748:
1.253 brouard 4749: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4750: for(i=1, jk=1; i <=nlstate; i++){
4751: for(j=1; j <=nlstate+ndeath; j++){
4752: if(j!=i){
4753: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4754: printf("%1d%1d",i,j);
4755: fprintf(ficparo,"%1d%1d",i,j);
4756: for(k=1; k<=ncovmodel;k++){
4757: /* printf(" %lf",param[i][j][k]); */
4758: /* fprintf(ficparo," %lf",param[i][j][k]); */
4759: p[jk]=pstart[jk];
4760: printf(" %f ",pstart[jk]);
4761: fprintf(ficparo," %f ",pstart[jk]);
4762: jk++;
4763: }
4764: printf("\n");
4765: fprintf(ficparo,"\n");
4766: }
4767: }
4768: }
4769: } /* end mle=-2 */
1.226 brouard 4770: dateintmean=dateintsum/k2cpt;
1.240 brouard 4771:
1.226 brouard 4772: fclose(ficresp);
4773: fclose(ficresphtm);
4774: fclose(ficresphtmfr);
4775: free_vector(meanq,1,nqfveff);
4776: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4777: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4778: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4779: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4780: free_vector(pospropt,1,nlstate);
4781: free_vector(posprop,1,nlstate);
1.251 brouard 4782: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4783: free_vector(pp,1,nlstate);
4784: /* End of freqsummary */
4785: }
1.126 brouard 4786:
4787: /************ Prevalence ********************/
1.227 brouard 4788: 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)
4789: {
4790: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4791: in each health status at the date of interview (if between dateprev1 and dateprev2).
4792: We still use firstpass and lastpass as another selection.
4793: */
1.126 brouard 4794:
1.227 brouard 4795: int i, m, jk, j1, bool, z1,j, iv;
4796: int mi; /* Effective wave */
4797: int iage;
4798: double agebegin, ageend;
4799:
4800: double **prop;
4801: double posprop;
4802: double y2; /* in fractional years */
4803: int iagemin, iagemax;
4804: int first; /** to stop verbosity which is redirected to log file */
4805:
4806: iagemin= (int) agemin;
4807: iagemax= (int) agemax;
4808: /*pp=vector(1,nlstate);*/
1.251 brouard 4809: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4810: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4811: j1=0;
1.222 brouard 4812:
1.227 brouard 4813: /*j=cptcoveff;*/
4814: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4815:
1.227 brouard 4816: first=1;
4817: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4818: for (i=1; i<=nlstate; i++)
1.251 brouard 4819: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4820: prop[i][iage]=0.0;
4821: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4822: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4823: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4824:
4825: for (i=1; i<=imx; i++) { /* Each individual */
4826: bool=1;
4827: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4828: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4829: m=mw[mi][i];
4830: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4831: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4832: for (z1=1; z1<=cptcoveff; z1++){
4833: if( Fixed[Tmodelind[z1]]==1){
4834: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4835: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4836: bool=0;
4837: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4838: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4839: bool=0;
4840: }
4841: }
4842: if(bool==1){ /* Otherwise we skip that wave/person */
4843: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4844: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4845: if(m >=firstpass && m <=lastpass){
4846: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4847: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4848: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4849: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4850: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4851: 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);
4852: exit(1);
4853: }
4854: if (s[m][i]>0 && s[m][i]<=nlstate) {
4855: /*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]]);*/
4856: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4857: prop[s[m][i]][iagemax+3] += weight[i];
4858: } /* end valid statuses */
4859: } /* end selection of dates */
4860: } /* end selection of waves */
4861: } /* end bool */
4862: } /* end wave */
4863: } /* end individual */
4864: for(i=iagemin; i <= iagemax+3; i++){
4865: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4866: posprop += prop[jk][i];
4867: }
4868:
4869: for(jk=1; jk <=nlstate ; jk++){
4870: if( i <= iagemax){
4871: if(posprop>=1.e-5){
4872: probs[i][jk][j1]= prop[jk][i]/posprop;
4873: } else{
4874: if(first==1){
4875: first=0;
4876: 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]);
4877: }
4878: }
4879: }
4880: }/* end jk */
4881: }/* end i */
1.222 brouard 4882: /*} *//* end i1 */
1.227 brouard 4883: } /* end j1 */
1.222 brouard 4884:
1.227 brouard 4885: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4886: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4887: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4888: } /* End of prevalence */
1.126 brouard 4889:
4890: /************* Waves Concatenation ***************/
4891:
4892: 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)
4893: {
4894: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4895: Death is a valid wave (if date is known).
4896: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4897: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4898: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4899: */
1.126 brouard 4900:
1.224 brouard 4901: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4902: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4903: double sum=0., jmean=0.;*/
1.224 brouard 4904: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4905: int j, k=0,jk, ju, jl;
4906: double sum=0.;
4907: first=0;
1.214 brouard 4908: firstwo=0;
1.217 brouard 4909: firsthree=0;
1.218 brouard 4910: firstfour=0;
1.164 brouard 4911: jmin=100000;
1.126 brouard 4912: jmax=-1;
4913: jmean=0.;
1.224 brouard 4914:
4915: /* Treating live states */
1.214 brouard 4916: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4917: mi=0; /* First valid wave */
1.227 brouard 4918: mli=0; /* Last valid wave */
1.126 brouard 4919: m=firstpass;
1.214 brouard 4920: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4921: 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 */
4922: mli=m-1;/* mw[++mi][i]=m-1; */
4923: }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 */
4924: mw[++mi][i]=m;
4925: mli=m;
1.224 brouard 4926: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4927: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4928: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4929: }
1.227 brouard 4930: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4931: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4932: break;
1.224 brouard 4933: #else
1.227 brouard 4934: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4935: if(firsthree == 0){
4936: 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);
4937: firsthree=1;
4938: }
4939: 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);
4940: mw[++mi][i]=m;
4941: mli=m;
4942: }
4943: if(s[m][i]==-2){ /* Vital status is really unknown */
4944: nbwarn++;
4945: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4946: 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);
4947: 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);
4948: }
4949: break;
4950: }
4951: break;
1.224 brouard 4952: #endif
1.227 brouard 4953: }/* End m >= lastpass */
1.126 brouard 4954: }/* end while */
1.224 brouard 4955:
1.227 brouard 4956: /* 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 4957: /* After last pass */
1.224 brouard 4958: /* Treating death states */
1.214 brouard 4959: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4960: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4961: /* } */
1.126 brouard 4962: mi++; /* Death is another wave */
4963: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4964: /* Only death is a correct wave */
1.126 brouard 4965: mw[mi][i]=m;
1.257 brouard 4966: } /* else not in a death state */
1.224 brouard 4967: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 4968: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 4969: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4970: 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 */
4971: nbwarn++;
4972: if(firstfiv==0){
4973: 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 );
4974: firstfiv=1;
4975: }else{
4976: 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 );
4977: }
4978: }else{ /* Death occured afer last wave potential bias */
4979: nberr++;
4980: if(firstwo==0){
1.257 brouard 4981: 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 4982: firstwo=1;
4983: }
1.257 brouard 4984: 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 4985: }
1.257 brouard 4986: }else{ /* if date of interview is unknown */
1.227 brouard 4987: /* death is known but not confirmed by death status at any wave */
4988: if(firstfour==0){
4989: 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 );
4990: firstfour=1;
4991: }
4992: 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 4993: }
1.224 brouard 4994: } /* end if date of death is known */
4995: #endif
4996: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4997: /* wav[i]=mw[mi][i]; */
1.126 brouard 4998: if(mi==0){
4999: nbwarn++;
5000: if(first==0){
1.227 brouard 5001: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5002: first=1;
1.126 brouard 5003: }
5004: if(first==1){
1.227 brouard 5005: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5006: }
5007: } /* end mi==0 */
5008: } /* End individuals */
1.214 brouard 5009: /* wav and mw are no more changed */
1.223 brouard 5010:
1.214 brouard 5011:
1.126 brouard 5012: for(i=1; i<=imx; i++){
5013: for(mi=1; mi<wav[i];mi++){
5014: if (stepm <=0)
1.227 brouard 5015: dh[mi][i]=1;
1.126 brouard 5016: else{
1.227 brouard 5017: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
5018: if (agedc[i] < 2*AGESUP) {
5019: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5020: if(j==0) j=1; /* Survives at least one month after exam */
5021: else if(j<0){
5022: nberr++;
5023: 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]);
5024: j=1; /* Temporary Dangerous patch */
5025: 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);
5026: 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]);
5027: 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);
5028: }
5029: k=k+1;
5030: if (j >= jmax){
5031: jmax=j;
5032: ijmax=i;
5033: }
5034: if (j <= jmin){
5035: jmin=j;
5036: ijmin=i;
5037: }
5038: sum=sum+j;
5039: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5040: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5041: }
5042: }
5043: else{
5044: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5045: /* 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 5046:
1.227 brouard 5047: k=k+1;
5048: if (j >= jmax) {
5049: jmax=j;
5050: ijmax=i;
5051: }
5052: else if (j <= jmin){
5053: jmin=j;
5054: ijmin=i;
5055: }
5056: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5057: /*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]);*/
5058: if(j<0){
5059: nberr++;
5060: 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]);
5061: 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]);
5062: }
5063: sum=sum+j;
5064: }
5065: jk= j/stepm;
5066: jl= j -jk*stepm;
5067: ju= j -(jk+1)*stepm;
5068: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5069: if(jl==0){
5070: dh[mi][i]=jk;
5071: bh[mi][i]=0;
5072: }else{ /* We want a negative bias in order to only have interpolation ie
5073: * to avoid the price of an extra matrix product in likelihood */
5074: dh[mi][i]=jk+1;
5075: bh[mi][i]=ju;
5076: }
5077: }else{
5078: if(jl <= -ju){
5079: dh[mi][i]=jk;
5080: bh[mi][i]=jl; /* bias is positive if real duration
5081: * is higher than the multiple of stepm and negative otherwise.
5082: */
5083: }
5084: else{
5085: dh[mi][i]=jk+1;
5086: bh[mi][i]=ju;
5087: }
5088: if(dh[mi][i]==0){
5089: dh[mi][i]=1; /* At least one step */
5090: bh[mi][i]=ju; /* At least one step */
5091: /* 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);*/
5092: }
5093: } /* end if mle */
1.126 brouard 5094: }
5095: } /* end wave */
5096: }
5097: jmean=sum/k;
5098: 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 5099: 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 5100: }
1.126 brouard 5101:
5102: /*********** Tricode ****************************/
1.220 brouard 5103: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5104: {
5105: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5106: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5107: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5108: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5109: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5110: */
1.130 brouard 5111:
1.242 brouard 5112: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5113: int modmaxcovj=0; /* Modality max of covariates j */
5114: int cptcode=0; /* Modality max of covariates j */
5115: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5116:
5117:
1.242 brouard 5118: /* cptcoveff=0; */
5119: /* *cptcov=0; */
1.126 brouard 5120:
1.242 brouard 5121: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5122:
1.242 brouard 5123: /* Loop on covariates without age and products and no quantitative variable */
5124: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5125: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5126: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5127: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5128: switch(Fixed[k]) {
5129: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5130: 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*/
5131: ij=(int)(covar[Tvar[k]][i]);
5132: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5133: * If product of Vn*Vm, still boolean *:
5134: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5135: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5136: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5137: modality of the nth covariate of individual i. */
5138: if (ij > modmaxcovj)
5139: modmaxcovj=ij;
5140: else if (ij < modmincovj)
5141: modmincovj=ij;
5142: if ((ij < -1) && (ij > NCOVMAX)){
5143: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5144: exit(1);
5145: }else
5146: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5147: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5148: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5149: /* getting the maximum value of the modality of the covariate
5150: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5151: female ies 1, then modmaxcovj=1.
5152: */
5153: } /* end for loop on individuals i */
5154: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5155: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5156: cptcode=modmaxcovj;
5157: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5158: /*for (i=0; i<=cptcode; i++) {*/
5159: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5160: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5161: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5162: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5163: if( j != -1){
5164: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5165: covariate for which somebody answered excluding
5166: undefined. Usually 2: 0 and 1. */
5167: }
5168: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5169: covariate for which somebody answered including
5170: undefined. Usually 3: -1, 0 and 1. */
5171: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5172: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5173: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5174:
1.242 brouard 5175: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5176: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5177: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5178: /* modmincovj=3; modmaxcovj = 7; */
5179: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5180: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5181: /* defining two dummy variables: variables V1_1 and V1_2.*/
5182: /* nbcode[Tvar[j]][ij]=k; */
5183: /* nbcode[Tvar[j]][1]=0; */
5184: /* nbcode[Tvar[j]][2]=1; */
5185: /* nbcode[Tvar[j]][3]=2; */
5186: /* To be continued (not working yet). */
5187: ij=0; /* ij is similar to i but can jump over null modalities */
5188: 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*/
5189: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5190: break;
5191: }
5192: ij++;
5193: 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*/
5194: cptcode = ij; /* New max modality for covar j */
5195: } /* end of loop on modality i=-1 to 1 or more */
5196: break;
5197: case 1: /* Testing on varying covariate, could be simple and
5198: * should look at waves or product of fixed *
5199: * varying. No time to test -1, assuming 0 and 1 only */
5200: ij=0;
5201: for(i=0; i<=1;i++){
5202: nbcode[Tvar[k]][++ij]=i;
5203: }
5204: break;
5205: default:
5206: break;
5207: } /* end switch */
5208: } /* end dummy test */
5209:
5210: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5211: /* /\*recode from 0 *\/ */
5212: /* k is a modality. If we have model=V1+V1*sex */
5213: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5214: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5215: /* } */
5216: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5217: /* if (ij > ncodemax[j]) { */
5218: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5219: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5220: /* break; */
5221: /* } */
5222: /* } /\* end of loop on modality k *\/ */
5223: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5224:
5225: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5226: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5227: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5228: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5229: 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 */
5230: 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 */
5231: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5232: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5233:
5234: ij=0;
5235: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5236: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5237: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5238: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5239: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5240: /* If product not in single variable we don't print results */
5241: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5242: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5243: 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*/
5244: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5245: 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 */
5246: if(Fixed[k]!=0)
5247: anyvaryingduminmodel=1;
5248: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5249: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5250: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5251: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5252: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5253: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5254: }
5255: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5256: /* ij--; */
5257: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5258: *cptcov=ij; /*Number of total real effective covariates: effective
5259: * because they can be excluded from the model and real
5260: * if in the model but excluded because missing values, but how to get k from ij?*/
5261: for(j=ij+1; j<= cptcovt; j++){
5262: Tvaraff[j]=0;
5263: Tmodelind[j]=0;
5264: }
5265: for(j=ntveff+1; j<= cptcovt; j++){
5266: TmodelInvind[j]=0;
5267: }
5268: /* To be sorted */
5269: ;
5270: }
1.126 brouard 5271:
1.145 brouard 5272:
1.126 brouard 5273: /*********** Health Expectancies ****************/
5274:
1.235 brouard 5275: 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 5276:
5277: {
5278: /* Health expectancies, no variances */
1.164 brouard 5279: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5280: int nhstepma, nstepma; /* Decreasing with age */
5281: double age, agelim, hf;
5282: double ***p3mat;
5283: double eip;
5284:
1.238 brouard 5285: /* pstamp(ficreseij); */
1.126 brouard 5286: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5287: fprintf(ficreseij,"# Age");
5288: for(i=1; i<=nlstate;i++){
5289: for(j=1; j<=nlstate;j++){
5290: fprintf(ficreseij," e%1d%1d ",i,j);
5291: }
5292: fprintf(ficreseij," e%1d. ",i);
5293: }
5294: fprintf(ficreseij,"\n");
5295:
5296:
5297: if(estepm < stepm){
5298: printf ("Problem %d lower than %d\n",estepm, stepm);
5299: }
5300: else hstepm=estepm;
5301: /* We compute the life expectancy from trapezoids spaced every estepm months
5302: * This is mainly to measure the difference between two models: for example
5303: * if stepm=24 months pijx are given only every 2 years and by summing them
5304: * we are calculating an estimate of the Life Expectancy assuming a linear
5305: * progression in between and thus overestimating or underestimating according
5306: * to the curvature of the survival function. If, for the same date, we
5307: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5308: * to compare the new estimate of Life expectancy with the same linear
5309: * hypothesis. A more precise result, taking into account a more precise
5310: * curvature will be obtained if estepm is as small as stepm. */
5311:
5312: /* For example we decided to compute the life expectancy with the smallest unit */
5313: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5314: nhstepm is the number of hstepm from age to agelim
5315: nstepm is the number of stepm from age to agelin.
5316: Look at hpijx to understand the reason of that which relies in memory size
5317: and note for a fixed period like estepm months */
5318: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5319: survival function given by stepm (the optimization length). Unfortunately it
5320: means that if the survival funtion is printed only each two years of age and if
5321: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5322: results. So we changed our mind and took the option of the best precision.
5323: */
5324: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5325:
5326: agelim=AGESUP;
5327: /* If stepm=6 months */
5328: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5329: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5330:
5331: /* nhstepm age range expressed in number of stepm */
5332: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5333: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5334: /* if (stepm >= YEARM) hstepm=1;*/
5335: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5336: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5337:
5338: for (age=bage; age<=fage; age ++){
5339: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5340: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5341: /* if (stepm >= YEARM) hstepm=1;*/
5342: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5343:
5344: /* If stepm=6 months */
5345: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5346: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5347:
1.235 brouard 5348: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5349:
5350: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5351:
5352: printf("%d|",(int)age);fflush(stdout);
5353: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5354:
5355: /* Computing expectancies */
5356: for(i=1; i<=nlstate;i++)
5357: for(j=1; j<=nlstate;j++)
5358: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5359: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5360:
5361: /* 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]);*/
5362:
5363: }
5364:
5365: fprintf(ficreseij,"%3.0f",age );
5366: for(i=1; i<=nlstate;i++){
5367: eip=0;
5368: for(j=1; j<=nlstate;j++){
5369: eip +=eij[i][j][(int)age];
5370: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5371: }
5372: fprintf(ficreseij,"%9.4f", eip );
5373: }
5374: fprintf(ficreseij,"\n");
5375:
5376: }
5377: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5378: printf("\n");
5379: fprintf(ficlog,"\n");
5380:
5381: }
5382:
1.235 brouard 5383: 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 5384:
5385: {
5386: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5387: to initial status i, ei. .
1.126 brouard 5388: */
5389: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5390: int nhstepma, nstepma; /* Decreasing with age */
5391: double age, agelim, hf;
5392: double ***p3matp, ***p3matm, ***varhe;
5393: double **dnewm,**doldm;
5394: double *xp, *xm;
5395: double **gp, **gm;
5396: double ***gradg, ***trgradg;
5397: int theta;
5398:
5399: double eip, vip;
5400:
5401: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5402: xp=vector(1,npar);
5403: xm=vector(1,npar);
5404: dnewm=matrix(1,nlstate*nlstate,1,npar);
5405: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5406:
5407: pstamp(ficresstdeij);
5408: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5409: fprintf(ficresstdeij,"# Age");
5410: for(i=1; i<=nlstate;i++){
5411: for(j=1; j<=nlstate;j++)
5412: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5413: fprintf(ficresstdeij," e%1d. ",i);
5414: }
5415: fprintf(ficresstdeij,"\n");
5416:
5417: pstamp(ficrescveij);
5418: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5419: fprintf(ficrescveij,"# Age");
5420: for(i=1; i<=nlstate;i++)
5421: for(j=1; j<=nlstate;j++){
5422: cptj= (j-1)*nlstate+i;
5423: for(i2=1; i2<=nlstate;i2++)
5424: for(j2=1; j2<=nlstate;j2++){
5425: cptj2= (j2-1)*nlstate+i2;
5426: if(cptj2 <= cptj)
5427: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5428: }
5429: }
5430: fprintf(ficrescveij,"\n");
5431:
5432: if(estepm < stepm){
5433: printf ("Problem %d lower than %d\n",estepm, stepm);
5434: }
5435: else hstepm=estepm;
5436: /* We compute the life expectancy from trapezoids spaced every estepm months
5437: * This is mainly to measure the difference between two models: for example
5438: * if stepm=24 months pijx are given only every 2 years and by summing them
5439: * we are calculating an estimate of the Life Expectancy assuming a linear
5440: * progression in between and thus overestimating or underestimating according
5441: * to the curvature of the survival function. If, for the same date, we
5442: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5443: * to compare the new estimate of Life expectancy with the same linear
5444: * hypothesis. A more precise result, taking into account a more precise
5445: * curvature will be obtained if estepm is as small as stepm. */
5446:
5447: /* For example we decided to compute the life expectancy with the smallest unit */
5448: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5449: nhstepm is the number of hstepm from age to agelim
5450: nstepm is the number of stepm from age to agelin.
5451: Look at hpijx to understand the reason of that which relies in memory size
5452: and note for a fixed period like estepm months */
5453: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5454: survival function given by stepm (the optimization length). Unfortunately it
5455: means that if the survival funtion is printed only each two years of age and if
5456: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5457: results. So we changed our mind and took the option of the best precision.
5458: */
5459: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5460:
5461: /* If stepm=6 months */
5462: /* nhstepm age range expressed in number of stepm */
5463: agelim=AGESUP;
5464: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5465: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5466: /* if (stepm >= YEARM) hstepm=1;*/
5467: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5468:
5469: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5470: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5471: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5472: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5473: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5474: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5475:
5476: for (age=bage; age<=fage; age ++){
5477: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5478: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5479: /* if (stepm >= YEARM) hstepm=1;*/
5480: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5481:
1.126 brouard 5482: /* If stepm=6 months */
5483: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5484: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5485:
5486: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5487:
1.126 brouard 5488: /* Computing Variances of health expectancies */
5489: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5490: decrease memory allocation */
5491: for(theta=1; theta <=npar; theta++){
5492: for(i=1; i<=npar; i++){
1.222 brouard 5493: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5494: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5495: }
1.235 brouard 5496: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5497: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5498:
1.126 brouard 5499: for(j=1; j<= nlstate; j++){
1.222 brouard 5500: for(i=1; i<=nlstate; i++){
5501: for(h=0; h<=nhstepm-1; h++){
5502: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5503: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5504: }
5505: }
1.126 brouard 5506: }
1.218 brouard 5507:
1.126 brouard 5508: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5509: for(h=0; h<=nhstepm-1; h++){
5510: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5511: }
1.126 brouard 5512: }/* End theta */
5513:
5514:
5515: for(h=0; h<=nhstepm-1; h++)
5516: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5517: for(theta=1; theta <=npar; theta++)
5518: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5519:
1.218 brouard 5520:
1.222 brouard 5521: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5522: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5523: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5524:
1.222 brouard 5525: printf("%d|",(int)age);fflush(stdout);
5526: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5527: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5528: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5529: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5530: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5531: for(ij=1;ij<=nlstate*nlstate;ij++)
5532: for(ji=1;ji<=nlstate*nlstate;ji++)
5533: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5534: }
5535: }
1.218 brouard 5536:
1.126 brouard 5537: /* Computing expectancies */
1.235 brouard 5538: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5539: for(i=1; i<=nlstate;i++)
5540: for(j=1; j<=nlstate;j++)
1.222 brouard 5541: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5542: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5543:
1.222 brouard 5544: /* 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 5545:
1.222 brouard 5546: }
1.218 brouard 5547:
1.126 brouard 5548: fprintf(ficresstdeij,"%3.0f",age );
5549: for(i=1; i<=nlstate;i++){
5550: eip=0.;
5551: vip=0.;
5552: for(j=1; j<=nlstate;j++){
1.222 brouard 5553: eip += eij[i][j][(int)age];
5554: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5555: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5556: 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 5557: }
5558: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5559: }
5560: fprintf(ficresstdeij,"\n");
1.218 brouard 5561:
1.126 brouard 5562: fprintf(ficrescveij,"%3.0f",age );
5563: for(i=1; i<=nlstate;i++)
5564: for(j=1; j<=nlstate;j++){
1.222 brouard 5565: cptj= (j-1)*nlstate+i;
5566: for(i2=1; i2<=nlstate;i2++)
5567: for(j2=1; j2<=nlstate;j2++){
5568: cptj2= (j2-1)*nlstate+i2;
5569: if(cptj2 <= cptj)
5570: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5571: }
1.126 brouard 5572: }
5573: fprintf(ficrescveij,"\n");
1.218 brouard 5574:
1.126 brouard 5575: }
5576: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5577: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5578: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5579: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5580: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5581: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5582: printf("\n");
5583: fprintf(ficlog,"\n");
1.218 brouard 5584:
1.126 brouard 5585: free_vector(xm,1,npar);
5586: free_vector(xp,1,npar);
5587: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5588: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5589: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5590: }
1.218 brouard 5591:
1.126 brouard 5592: /************ Variance ******************/
1.235 brouard 5593: 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 5594: {
5595: /* Variance of health expectancies */
5596: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5597: /* double **newm;*/
5598: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5599:
5600: /* int movingaverage(); */
5601: double **dnewm,**doldm;
5602: double **dnewmp,**doldmp;
5603: int i, j, nhstepm, hstepm, h, nstepm ;
5604: int k;
5605: double *xp;
5606: double **gp, **gm; /* for var eij */
5607: double ***gradg, ***trgradg; /*for var eij */
5608: double **gradgp, **trgradgp; /* for var p point j */
5609: double *gpp, *gmp; /* for var p point j */
5610: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5611: double ***p3mat;
5612: double age,agelim, hf;
5613: /* double ***mobaverage; */
5614: int theta;
5615: char digit[4];
5616: char digitp[25];
5617:
5618: char fileresprobmorprev[FILENAMELENGTH];
5619:
5620: if(popbased==1){
5621: if(mobilav!=0)
5622: strcpy(digitp,"-POPULBASED-MOBILAV_");
5623: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5624: }
5625: else
5626: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5627:
1.218 brouard 5628: /* if (mobilav!=0) { */
5629: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5630: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5631: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5632: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5633: /* } */
5634: /* } */
5635:
5636: strcpy(fileresprobmorprev,"PRMORPREV-");
5637: sprintf(digit,"%-d",ij);
5638: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5639: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5640: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5641: strcat(fileresprobmorprev,fileresu);
5642: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5643: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5644: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5645: }
5646: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5647: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5648: pstamp(ficresprobmorprev);
5649: 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 5650: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5651: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5652: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5653: }
5654: for(j=1;j<=cptcoveff;j++)
5655: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5656: fprintf(ficresprobmorprev,"\n");
5657:
1.218 brouard 5658: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5659: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5660: fprintf(ficresprobmorprev," p.%-d SE",j);
5661: for(i=1; i<=nlstate;i++)
5662: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5663: }
5664: fprintf(ficresprobmorprev,"\n");
5665:
5666: fprintf(ficgp,"\n# Routine varevsij");
5667: fprintf(ficgp,"\nunset title \n");
5668: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5669: 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");
5670: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5671: /* } */
5672: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5673: pstamp(ficresvij);
5674: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5675: if(popbased==1)
5676: 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);
5677: else
5678: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5679: fprintf(ficresvij,"# Age");
5680: for(i=1; i<=nlstate;i++)
5681: for(j=1; j<=nlstate;j++)
5682: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5683: fprintf(ficresvij,"\n");
5684:
5685: xp=vector(1,npar);
5686: dnewm=matrix(1,nlstate,1,npar);
5687: doldm=matrix(1,nlstate,1,nlstate);
5688: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5689: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5690:
5691: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5692: gpp=vector(nlstate+1,nlstate+ndeath);
5693: gmp=vector(nlstate+1,nlstate+ndeath);
5694: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5695:
1.218 brouard 5696: if(estepm < stepm){
5697: printf ("Problem %d lower than %d\n",estepm, stepm);
5698: }
5699: else hstepm=estepm;
5700: /* For example we decided to compute the life expectancy with the smallest unit */
5701: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5702: nhstepm is the number of hstepm from age to agelim
5703: nstepm is the number of stepm from age to agelim.
5704: Look at function hpijx to understand why because of memory size limitations,
5705: we decided (b) to get a life expectancy respecting the most precise curvature of the
5706: survival function given by stepm (the optimization length). Unfortunately it
5707: means that if the survival funtion is printed every two years of age and if
5708: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5709: results. So we changed our mind and took the option of the best precision.
5710: */
5711: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5712: agelim = AGESUP;
5713: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5714: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5715: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5716: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5717: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5718: gp=matrix(0,nhstepm,1,nlstate);
5719: gm=matrix(0,nhstepm,1,nlstate);
5720:
5721:
5722: for(theta=1; theta <=npar; theta++){
5723: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5724: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5725: }
5726:
1.242 brouard 5727: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5728:
5729: if (popbased==1) {
5730: if(mobilav ==0){
5731: for(i=1; i<=nlstate;i++)
5732: prlim[i][i]=probs[(int)age][i][ij];
5733: }else{ /* mobilav */
5734: for(i=1; i<=nlstate;i++)
5735: prlim[i][i]=mobaverage[(int)age][i][ij];
5736: }
5737: }
5738:
1.235 brouard 5739: 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 5740: for(j=1; j<= nlstate; j++){
5741: for(h=0; h<=nhstepm; h++){
5742: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5743: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5744: }
5745: }
5746: /* Next for computing probability of death (h=1 means
5747: computed over hstepm matrices product = hstepm*stepm months)
5748: as a weighted average of prlim.
5749: */
5750: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5751: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5752: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5753: }
5754: /* end probability of death */
5755:
5756: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5757: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5758:
1.242 brouard 5759: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5760:
5761: if (popbased==1) {
5762: if(mobilav ==0){
5763: for(i=1; i<=nlstate;i++)
5764: prlim[i][i]=probs[(int)age][i][ij];
5765: }else{ /* mobilav */
5766: for(i=1; i<=nlstate;i++)
5767: prlim[i][i]=mobaverage[(int)age][i][ij];
5768: }
5769: }
5770:
1.235 brouard 5771: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5772:
5773: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5774: for(h=0; h<=nhstepm; h++){
5775: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5776: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5777: }
5778: }
5779: /* This for computing probability of death (h=1 means
5780: computed over hstepm matrices product = hstepm*stepm months)
5781: as a weighted average of prlim.
5782: */
5783: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5784: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5785: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5786: }
5787: /* end probability of death */
5788:
5789: for(j=1; j<= nlstate; j++) /* vareij */
5790: for(h=0; h<=nhstepm; h++){
5791: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5792: }
5793:
5794: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5795: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5796: }
5797:
5798: } /* End theta */
5799:
5800: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5801:
5802: for(h=0; h<=nhstepm; h++) /* veij */
5803: for(j=1; j<=nlstate;j++)
5804: for(theta=1; theta <=npar; theta++)
5805: trgradg[h][j][theta]=gradg[h][theta][j];
5806:
5807: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5808: for(theta=1; theta <=npar; theta++)
5809: trgradgp[j][theta]=gradgp[theta][j];
5810:
5811:
5812: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5813: for(i=1;i<=nlstate;i++)
5814: for(j=1;j<=nlstate;j++)
5815: vareij[i][j][(int)age] =0.;
5816:
5817: for(h=0;h<=nhstepm;h++){
5818: for(k=0;k<=nhstepm;k++){
5819: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5820: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5821: for(i=1;i<=nlstate;i++)
5822: for(j=1;j<=nlstate;j++)
5823: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5824: }
5825: }
5826:
5827: /* pptj */
5828: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5829: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5830: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5831: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5832: varppt[j][i]=doldmp[j][i];
5833: /* end ppptj */
5834: /* x centered again */
5835:
1.242 brouard 5836: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5837:
5838: if (popbased==1) {
5839: if(mobilav ==0){
5840: for(i=1; i<=nlstate;i++)
5841: prlim[i][i]=probs[(int)age][i][ij];
5842: }else{ /* mobilav */
5843: for(i=1; i<=nlstate;i++)
5844: prlim[i][i]=mobaverage[(int)age][i][ij];
5845: }
5846: }
5847:
5848: /* This for computing probability of death (h=1 means
5849: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5850: as a weighted average of prlim.
5851: */
1.235 brouard 5852: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5853: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5854: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5855: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5856: }
5857: /* end probability of death */
5858:
5859: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5860: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5861: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5862: for(i=1; i<=nlstate;i++){
5863: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5864: }
5865: }
5866: fprintf(ficresprobmorprev,"\n");
5867:
5868: fprintf(ficresvij,"%.0f ",age );
5869: for(i=1; i<=nlstate;i++)
5870: for(j=1; j<=nlstate;j++){
5871: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5872: }
5873: fprintf(ficresvij,"\n");
5874: free_matrix(gp,0,nhstepm,1,nlstate);
5875: free_matrix(gm,0,nhstepm,1,nlstate);
5876: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5877: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5878: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5879: } /* End age */
5880: free_vector(gpp,nlstate+1,nlstate+ndeath);
5881: free_vector(gmp,nlstate+1,nlstate+ndeath);
5882: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5883: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5884: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5885: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5886: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5887: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5888: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5889: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5890: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5891: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5892: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5893: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5894: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5895: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5896: 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);
5897: /* 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 5898: */
1.218 brouard 5899: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5900: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5901:
1.218 brouard 5902: free_vector(xp,1,npar);
5903: free_matrix(doldm,1,nlstate,1,nlstate);
5904: free_matrix(dnewm,1,nlstate,1,npar);
5905: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5906: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5907: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5908: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5909: fclose(ficresprobmorprev);
5910: fflush(ficgp);
5911: fflush(fichtm);
5912: } /* end varevsij */
1.126 brouard 5913:
5914: /************ Variance of prevlim ******************/
1.235 brouard 5915: 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 5916: {
1.205 brouard 5917: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5918: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5919:
1.126 brouard 5920: double **dnewm,**doldm;
5921: int i, j, nhstepm, hstepm;
5922: double *xp;
5923: double *gp, *gm;
5924: double **gradg, **trgradg;
1.208 brouard 5925: double **mgm, **mgp;
1.126 brouard 5926: double age,agelim;
5927: int theta;
5928:
5929: pstamp(ficresvpl);
5930: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5931: fprintf(ficresvpl,"# Age ");
5932: if(nresult >=1)
5933: fprintf(ficresvpl," Result# ");
1.126 brouard 5934: for(i=1; i<=nlstate;i++)
5935: fprintf(ficresvpl," %1d-%1d",i,i);
5936: fprintf(ficresvpl,"\n");
5937:
5938: xp=vector(1,npar);
5939: dnewm=matrix(1,nlstate,1,npar);
5940: doldm=matrix(1,nlstate,1,nlstate);
5941:
5942: hstepm=1*YEARM; /* Every year of age */
5943: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5944: agelim = AGESUP;
5945: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5946: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5947: if (stepm >= YEARM) hstepm=1;
5948: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5949: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5950: mgp=matrix(1,npar,1,nlstate);
5951: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5952: gp=vector(1,nlstate);
5953: gm=vector(1,nlstate);
5954:
5955: for(theta=1; theta <=npar; theta++){
5956: for(i=1; i<=npar; i++){ /* Computes gradient */
5957: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5958: }
1.209 brouard 5959: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5960: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5961: else
1.235 brouard 5962: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5963: for(i=1;i<=nlstate;i++){
1.126 brouard 5964: gp[i] = prlim[i][i];
1.208 brouard 5965: mgp[theta][i] = prlim[i][i];
5966: }
1.126 brouard 5967: for(i=1; i<=npar; i++) /* Computes gradient */
5968: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5969: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5970: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5971: else
1.235 brouard 5972: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5973: for(i=1;i<=nlstate;i++){
1.126 brouard 5974: gm[i] = prlim[i][i];
1.208 brouard 5975: mgm[theta][i] = prlim[i][i];
5976: }
1.126 brouard 5977: for(i=1;i<=nlstate;i++)
5978: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5979: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5980: } /* End theta */
5981:
5982: trgradg =matrix(1,nlstate,1,npar);
5983:
5984: for(j=1; j<=nlstate;j++)
5985: for(theta=1; theta <=npar; theta++)
5986: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5987: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5988: /* printf("\nmgm mgp %d ",(int)age); */
5989: /* for(j=1; j<=nlstate;j++){ */
5990: /* printf(" %d ",j); */
5991: /* for(theta=1; theta <=npar; theta++) */
5992: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5993: /* printf("\n "); */
5994: /* } */
5995: /* } */
5996: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5997: /* printf("\n gradg %d ",(int)age); */
5998: /* for(j=1; j<=nlstate;j++){ */
5999: /* printf("%d ",j); */
6000: /* for(theta=1; theta <=npar; theta++) */
6001: /* printf("%d %lf ",theta,gradg[theta][j]); */
6002: /* printf("\n "); */
6003: /* } */
6004: /* } */
1.126 brouard 6005:
6006: for(i=1;i<=nlstate;i++)
6007: varpl[i][(int)age] =0.;
1.209 brouard 6008: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6009: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6010: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6011: }else{
1.126 brouard 6012: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6013: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6014: }
1.126 brouard 6015: for(i=1;i<=nlstate;i++)
6016: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6017:
6018: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6019: if(nresult >=1)
6020: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6021: for(i=1; i<=nlstate;i++)
6022: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6023: fprintf(ficresvpl,"\n");
6024: free_vector(gp,1,nlstate);
6025: free_vector(gm,1,nlstate);
1.208 brouard 6026: free_matrix(mgm,1,npar,1,nlstate);
6027: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6028: free_matrix(gradg,1,npar,1,nlstate);
6029: free_matrix(trgradg,1,nlstate,1,npar);
6030: } /* End age */
6031:
6032: free_vector(xp,1,npar);
6033: free_matrix(doldm,1,nlstate,1,npar);
6034: free_matrix(dnewm,1,nlstate,1,nlstate);
6035:
6036: }
6037:
6038: /************ Variance of one-step probabilities ******************/
6039: 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 6040: {
6041: int i, j=0, k1, l1, tj;
6042: int k2, l2, j1, z1;
6043: int k=0, l;
6044: int first=1, first1, first2;
6045: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6046: double **dnewm,**doldm;
6047: double *xp;
6048: double *gp, *gm;
6049: double **gradg, **trgradg;
6050: double **mu;
6051: double age, cov[NCOVMAX+1];
6052: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6053: int theta;
6054: char fileresprob[FILENAMELENGTH];
6055: char fileresprobcov[FILENAMELENGTH];
6056: char fileresprobcor[FILENAMELENGTH];
6057: double ***varpij;
6058:
6059: strcpy(fileresprob,"PROB_");
6060: strcat(fileresprob,fileres);
6061: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6062: printf("Problem with resultfile: %s\n", fileresprob);
6063: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6064: }
6065: strcpy(fileresprobcov,"PROBCOV_");
6066: strcat(fileresprobcov,fileresu);
6067: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6068: printf("Problem with resultfile: %s\n", fileresprobcov);
6069: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6070: }
6071: strcpy(fileresprobcor,"PROBCOR_");
6072: strcat(fileresprobcor,fileresu);
6073: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6074: printf("Problem with resultfile: %s\n", fileresprobcor);
6075: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6076: }
6077: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6078: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6079: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6080: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6081: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6082: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6083: pstamp(ficresprob);
6084: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6085: fprintf(ficresprob,"# Age");
6086: pstamp(ficresprobcov);
6087: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6088: fprintf(ficresprobcov,"# Age");
6089: pstamp(ficresprobcor);
6090: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6091: fprintf(ficresprobcor,"# Age");
1.126 brouard 6092:
6093:
1.222 brouard 6094: for(i=1; i<=nlstate;i++)
6095: for(j=1; j<=(nlstate+ndeath);j++){
6096: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6097: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6098: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6099: }
6100: /* fprintf(ficresprob,"\n");
6101: fprintf(ficresprobcov,"\n");
6102: fprintf(ficresprobcor,"\n");
6103: */
6104: xp=vector(1,npar);
6105: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6106: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6107: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6108: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6109: first=1;
6110: fprintf(ficgp,"\n# Routine varprob");
6111: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6112: fprintf(fichtm,"\n");
6113:
6114: 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);
6115: 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);
6116: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6117: and drawn. It helps understanding how is the covariance between two incidences.\
6118: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6119: 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 6120: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6121: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6122: standard deviations wide on each axis. <br>\
6123: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6124: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6125: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6126:
1.222 brouard 6127: cov[1]=1;
6128: /* tj=cptcoveff; */
1.225 brouard 6129: tj = (int) pow(2,cptcoveff);
1.222 brouard 6130: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6131: j1=0;
1.224 brouard 6132: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6133: if (cptcovn>0) {
6134: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6135: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6136: fprintf(ficresprob, "**********\n#\n");
6137: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6138: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6139: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6140:
1.222 brouard 6141: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6142: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6143: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6144:
6145:
1.222 brouard 6146: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6147: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6148: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6149:
1.222 brouard 6150: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6151: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6152: fprintf(ficresprobcor, "**********\n#");
6153: if(invalidvarcomb[j1]){
6154: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6155: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6156: continue;
6157: }
6158: }
6159: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6160: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6161: gp=vector(1,(nlstate)*(nlstate+ndeath));
6162: gm=vector(1,(nlstate)*(nlstate+ndeath));
6163: for (age=bage; age<=fage; age ++){
6164: cov[2]=age;
6165: if(nagesqr==1)
6166: cov[3]= age*age;
6167: for (k=1; k<=cptcovn;k++) {
6168: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6169: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6170: * 1 1 1 1 1
6171: * 2 2 1 1 1
6172: * 3 1 2 1 1
6173: */
6174: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6175: }
6176: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6177: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6178: for (k=1; k<=cptcovprod;k++)
6179: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6180:
6181:
1.222 brouard 6182: for(theta=1; theta <=npar; theta++){
6183: for(i=1; i<=npar; i++)
6184: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6185:
1.222 brouard 6186: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6187:
1.222 brouard 6188: k=0;
6189: for(i=1; i<= (nlstate); i++){
6190: for(j=1; j<=(nlstate+ndeath);j++){
6191: k=k+1;
6192: gp[k]=pmmij[i][j];
6193: }
6194: }
1.220 brouard 6195:
1.222 brouard 6196: for(i=1; i<=npar; i++)
6197: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6198:
1.222 brouard 6199: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6200: k=0;
6201: for(i=1; i<=(nlstate); i++){
6202: for(j=1; j<=(nlstate+ndeath);j++){
6203: k=k+1;
6204: gm[k]=pmmij[i][j];
6205: }
6206: }
1.220 brouard 6207:
1.222 brouard 6208: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6209: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6210: }
1.126 brouard 6211:
1.222 brouard 6212: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6213: for(theta=1; theta <=npar; theta++)
6214: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6215:
1.222 brouard 6216: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6217: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6218:
1.222 brouard 6219: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6220:
1.222 brouard 6221: k=0;
6222: for(i=1; i<=(nlstate); i++){
6223: for(j=1; j<=(nlstate+ndeath);j++){
6224: k=k+1;
6225: mu[k][(int) age]=pmmij[i][j];
6226: }
6227: }
6228: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6229: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6230: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6231:
1.222 brouard 6232: /*printf("\n%d ",(int)age);
6233: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6234: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6235: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6236: }*/
1.220 brouard 6237:
1.222 brouard 6238: fprintf(ficresprob,"\n%d ",(int)age);
6239: fprintf(ficresprobcov,"\n%d ",(int)age);
6240: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6241:
1.222 brouard 6242: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6243: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6244: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6245: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6246: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6247: }
6248: i=0;
6249: for (k=1; k<=(nlstate);k++){
6250: for (l=1; l<=(nlstate+ndeath);l++){
6251: i++;
6252: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6253: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6254: for (j=1; j<=i;j++){
6255: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6256: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6257: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6258: }
6259: }
6260: }/* end of loop for state */
6261: } /* end of loop for age */
6262: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6263: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6264: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6265: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6266:
6267: /* Confidence intervalle of pij */
6268: /*
6269: fprintf(ficgp,"\nunset parametric;unset label");
6270: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6271: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6272: 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);
6273: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6274: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6275: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6276: */
6277:
6278: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6279: first1=1;first2=2;
6280: for (k2=1; k2<=(nlstate);k2++){
6281: for (l2=1; l2<=(nlstate+ndeath);l2++){
6282: if(l2==k2) continue;
6283: j=(k2-1)*(nlstate+ndeath)+l2;
6284: for (k1=1; k1<=(nlstate);k1++){
6285: for (l1=1; l1<=(nlstate+ndeath);l1++){
6286: if(l1==k1) continue;
6287: i=(k1-1)*(nlstate+ndeath)+l1;
6288: if(i<=j) continue;
6289: for (age=bage; age<=fage; age ++){
6290: if ((int)age %5==0){
6291: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6292: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6293: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6294: mu1=mu[i][(int) age]/stepm*YEARM ;
6295: mu2=mu[j][(int) age]/stepm*YEARM;
6296: c12=cv12/sqrt(v1*v2);
6297: /* Computing eigen value of matrix of covariance */
6298: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6299: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6300: if ((lc2 <0) || (lc1 <0) ){
6301: if(first2==1){
6302: first1=0;
6303: 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);
6304: }
6305: 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);
6306: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6307: /* lc2=fabs(lc2); */
6308: }
1.220 brouard 6309:
1.222 brouard 6310: /* Eigen vectors */
6311: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6312: /*v21=sqrt(1.-v11*v11); *//* error */
6313: v21=(lc1-v1)/cv12*v11;
6314: v12=-v21;
6315: v22=v11;
6316: tnalp=v21/v11;
6317: if(first1==1){
6318: first1=0;
6319: 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);
6320: }
6321: 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);
6322: /*printf(fignu*/
6323: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6324: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6325: if(first==1){
6326: first=0;
6327: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6328: fprintf(ficgp,"\nset parametric;unset label");
6329: 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);
6330: fprintf(ficgp,"\nset ter svg size 640, 480");
6331: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6332: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6333: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6334: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6335: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6336: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6337: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6338: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6339: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6340: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6341: 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", \
6342: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6343: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6344: }else{
6345: first=0;
6346: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6347: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6348: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6349: 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", \
6350: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6351: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6352: }/* if first */
6353: } /* age mod 5 */
6354: } /* end loop age */
6355: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6356: first=1;
6357: } /*l12 */
6358: } /* k12 */
6359: } /*l1 */
6360: }/* k1 */
6361: } /* loop on combination of covariates j1 */
6362: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6363: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6364: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6365: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6366: free_vector(xp,1,npar);
6367: fclose(ficresprob);
6368: fclose(ficresprobcov);
6369: fclose(ficresprobcor);
6370: fflush(ficgp);
6371: fflush(fichtmcov);
6372: }
1.126 brouard 6373:
6374:
6375: /******************* Printing html file ***********/
1.201 brouard 6376: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6377: int lastpass, int stepm, int weightopt, char model[],\
6378: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 ! brouard 6379: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6380: double jprev1, double mprev1,double anprev1, double dateprev1, \
6381: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6382: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6383:
6384: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6385: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6386: </ul>");
1.237 brouard 6387: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6388: </ul>", model);
1.214 brouard 6389: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6390: 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",
6391: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6392: 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 6393: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6394: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6395: fprintf(fichtm,"\
6396: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6397: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6398: fprintf(fichtm,"\
1.217 brouard 6399: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6400: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6401: fprintf(fichtm,"\
1.126 brouard 6402: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6403: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6404: fprintf(fichtm,"\
1.217 brouard 6405: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6406: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6407: fprintf(fichtm,"\
1.211 brouard 6408: - (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 6409: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6410: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6411: if(prevfcast==1){
6412: fprintf(fichtm,"\
6413: - Prevalence projections by age and states: \
1.201 brouard 6414: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6415: }
1.126 brouard 6416:
1.222 brouard 6417: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6418:
1.225 brouard 6419: m=pow(2,cptcoveff);
1.222 brouard 6420: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6421:
1.222 brouard 6422: jj1=0;
1.237 brouard 6423:
6424: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6425: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6426: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6427: continue;
1.220 brouard 6428:
1.222 brouard 6429: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6430: jj1++;
6431: if (cptcovn > 0) {
6432: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6433: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6434: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6435: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6436: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6437: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6438: }
1.237 brouard 6439: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6440: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6441: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6442: }
6443:
1.230 brouard 6444: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6445: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6446: if(invalidvarcomb[k1]){
6447: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6448: printf("\nCombination (%d) ignored because no cases \n",k1);
6449: continue;
6450: }
6451: }
6452: /* aij, bij */
1.241 brouard 6453: fprintf(fichtm,"<br>- Logit model (yours is: 1+age+%s), for example: logit(pij)=log(pij/pii)= aij+ bij age + V1 age + etc. as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
6454: <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 6455: /* Pij */
1.241 brouard 6456: 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> \
6457: <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 6458: /* Quasi-incidences */
6459: 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 6460: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6461: 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 6462: 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> \
6463: <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 6464: /* Survival functions (period) in state j */
6465: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6466: 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> \
6467: <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 6468: }
6469: /* State specific survival functions (period) */
6470: for(cpt=1; cpt<=nlstate;cpt++){
6471: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6472: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6473: <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 6474: }
6475: /* Period (stable) prevalence in each health state */
6476: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6477: 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 6478: <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 6479: }
6480: if(backcast==1){
6481: /* Period (stable) back prevalence in each health state */
6482: for(cpt=1; cpt<=nlstate;cpt++){
1.255 brouard 6483: 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 6484: <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 6485: }
1.217 brouard 6486: }
1.222 brouard 6487: if(prevfcast==1){
6488: /* Projection of prevalence up to period (stable) prevalence in each health state */
6489: for(cpt=1; cpt<=nlstate;cpt++){
1.258 ! brouard 6490: 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> \
! 6491: <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 6492: }
6493: }
1.220 brouard 6494:
1.222 brouard 6495: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6496: 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> \
6497: <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 6498: }
6499: /* } /\* end i1 *\/ */
6500: }/* End k1 */
6501: fprintf(fichtm,"</ul>");
1.126 brouard 6502:
1.222 brouard 6503: fprintf(fichtm,"\
1.126 brouard 6504: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6505: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6506: - 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 6507: But because parameters are usually highly correlated (a higher incidence of disability \
6508: and a higher incidence of recovery can give very close observed transition) it might \
6509: be very useful to look not only at linear confidence intervals estimated from the \
6510: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6511: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6512: covariance matrix of the one-step probabilities. \
6513: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6514:
1.222 brouard 6515: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6516: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6517: fprintf(fichtm,"\
1.126 brouard 6518: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6519: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6520:
1.222 brouard 6521: fprintf(fichtm,"\
1.126 brouard 6522: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6523: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6524: fprintf(fichtm,"\
1.126 brouard 6525: - 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): \
6526: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6527: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6528: fprintf(fichtm,"\
1.126 brouard 6529: - (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): \
6530: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6531: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6532: fprintf(fichtm,"\
1.128 brouard 6533: - 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 6534: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6535: fprintf(fichtm,"\
1.128 brouard 6536: - 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 6537: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6538: fprintf(fichtm,"\
1.126 brouard 6539: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6540: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6541:
6542: /* if(popforecast==1) fprintf(fichtm,"\n */
6543: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6544: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6545: /* <br>",fileres,fileres,fileres,fileres); */
6546: /* else */
6547: /* 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 6548: fflush(fichtm);
6549: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6550:
1.225 brouard 6551: m=pow(2,cptcoveff);
1.222 brouard 6552: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6553:
1.222 brouard 6554: jj1=0;
1.237 brouard 6555:
1.241 brouard 6556: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6557: for(k1=1; k1<=m;k1++){
1.253 brouard 6558: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6559: continue;
1.222 brouard 6560: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6561: jj1++;
1.126 brouard 6562: if (cptcovn > 0) {
6563: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6564: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6565: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6566: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6567: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6568: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6569: }
6570:
1.126 brouard 6571: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6572:
1.222 brouard 6573: if(invalidvarcomb[k1]){
6574: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6575: continue;
6576: }
1.126 brouard 6577: }
6578: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 ! brouard 6579: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6580: 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 6581: <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 6582: }
6583: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6584: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6585: true period expectancies (those weighted with period prevalences are also\
6586: drawn in addition to the population based expectancies computed using\
1.241 brouard 6587: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6588: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6589: /* } /\* end i1 *\/ */
6590: }/* End k1 */
1.241 brouard 6591: }/* End nres */
1.222 brouard 6592: fprintf(fichtm,"</ul>");
6593: fflush(fichtm);
1.126 brouard 6594: }
6595:
6596: /******************* Gnuplot file **************/
1.223 brouard 6597: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6598:
6599: char dirfileres[132],optfileres[132];
1.223 brouard 6600: char gplotcondition[132];
1.237 brouard 6601: 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 6602: int lv=0, vlv=0, kl=0;
1.130 brouard 6603: int ng=0;
1.201 brouard 6604: int vpopbased;
1.223 brouard 6605: int ioffset; /* variable offset for columns */
1.235 brouard 6606: int nres=0; /* Index of resultline */
1.219 brouard 6607:
1.126 brouard 6608: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6609: /* printf("Problem with file %s",optionfilegnuplot); */
6610: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6611: /* } */
6612:
6613: /*#ifdef windows */
6614: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6615: /*#endif */
1.225 brouard 6616: m=pow(2,cptcoveff);
1.126 brouard 6617:
1.202 brouard 6618: /* Contribution to likelihood */
6619: /* Plot the probability implied in the likelihood */
1.223 brouard 6620: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6621: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6622: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6623: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6624: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6625: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6626: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6627: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6628: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6629: 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));
6630: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6631: 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));
6632: for (i=1; i<= nlstate ; i ++) {
6633: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6634: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6635: 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);
6636: for (j=2; j<= nlstate+ndeath ; j ++) {
6637: 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);
6638: }
6639: fprintf(ficgp,";\nset out; unset ylabel;\n");
6640: }
6641: /* 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 */
6642: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6643: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6644: fprintf(ficgp,"\nset out;unset log\n");
6645: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6646:
1.126 brouard 6647: strcpy(dirfileres,optionfilefiname);
6648: strcpy(optfileres,"vpl");
1.223 brouard 6649: /* 1eme*/
1.238 brouard 6650: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6651: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6652: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6653: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6654: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6655: continue;
6656: /* We are interested in selected combination by the resultline */
1.246 brouard 6657: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6658: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6659: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6660: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6661: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6662: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6663: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6664: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6665: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6666: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6667: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6668: }
6669: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6670: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6671: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6672: }
1.246 brouard 6673: /* printf("\n#\n"); */
1.238 brouard 6674: fprintf(ficgp,"\n#\n");
6675: if(invalidvarcomb[k1]){
6676: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6677: continue;
6678: }
1.235 brouard 6679:
1.241 brouard 6680: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6681: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6682: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.235 brouard 6683:
1.238 brouard 6684: for (i=1; i<= nlstate ; i ++) {
6685: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6686: else fprintf(ficgp," %%*lf (%%*lf)");
6687: }
1.242 brouard 6688: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6689: for (i=1; i<= nlstate ; i ++) {
6690: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6691: else fprintf(ficgp," %%*lf (%%*lf)");
6692: }
1.242 brouard 6693: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6694: for (i=1; i<= nlstate ; i ++) {
6695: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6696: else fprintf(ficgp," %%*lf (%%*lf)");
6697: }
6698: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6699: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6700: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6701: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6702: if(cptcoveff ==0){
1.245 brouard 6703: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6704: }else{
6705: kl=0;
6706: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6707: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6708: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6709: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6710: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6711: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6712: kl++;
1.238 brouard 6713: /* 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 *\/ */
6714: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6715: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6716: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6717: if(k==cptcoveff){
1.245 brouard 6718: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6719: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6720: }else{
6721: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6722: kl++;
6723: }
6724: } /* end covariate */
6725: } /* end if no covariate */
6726: } /* end if backcast */
6727: fprintf(ficgp,"\nset out \n");
6728: } /* nres */
1.201 brouard 6729: } /* k1 */
6730: } /* cpt */
1.235 brouard 6731:
6732:
1.126 brouard 6733: /*2 eme*/
1.238 brouard 6734: for (k1=1; k1<= m ; k1 ++){
6735: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6736: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6737: continue;
6738: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6739: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6740: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6741: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6742: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6743: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6744: vlv= nbcode[Tvaraff[k]][lv];
6745: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6746: }
1.237 brouard 6747: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6748: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6749: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6750: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6751: }
1.211 brouard 6752: fprintf(ficgp,"\n#\n");
1.223 brouard 6753: if(invalidvarcomb[k1]){
6754: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6755: continue;
6756: }
1.219 brouard 6757:
1.241 brouard 6758: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6759: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6760: if(vpopbased==0)
6761: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6762: else
6763: fprintf(ficgp,"\nreplot ");
6764: for (i=1; i<= nlstate+1 ; i ++) {
6765: k=2*i;
6766: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1, vpopbased);
6767: for (j=1; j<= nlstate+1 ; j ++) {
6768: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6769: else fprintf(ficgp," %%*lf (%%*lf)");
6770: }
6771: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6772: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6773: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6774: for (j=1; j<= nlstate+1 ; j ++) {
6775: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6776: else fprintf(ficgp," %%*lf (%%*lf)");
6777: }
6778: fprintf(ficgp,"\" t\"\" w l lt 0,");
6779: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),k1-1,k1-1,vpopbased);
6780: for (j=1; j<= nlstate+1 ; j ++) {
6781: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6782: else fprintf(ficgp," %%*lf (%%*lf)");
6783: }
6784: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6785: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6786: } /* state */
6787: } /* vpopbased */
1.244 brouard 6788: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6789: } /* end nres */
6790: } /* k1 end 2 eme*/
6791:
6792:
6793: /*3eme*/
6794: for (k1=1; k1<= m ; k1 ++){
6795: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6796: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6797: continue;
6798:
6799: for (cpt=1; cpt<= nlstate ; cpt ++) {
6800: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6801: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6802: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6803: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6804: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6805: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6806: vlv= nbcode[Tvaraff[k]][lv];
6807: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6808: }
6809: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6810: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6811: }
6812: fprintf(ficgp,"\n#\n");
6813: if(invalidvarcomb[k1]){
6814: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6815: continue;
6816: }
6817:
6818: /* k=2+nlstate*(2*cpt-2); */
6819: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6820: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6821: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6822: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),k1-1,k1-1,k,cpt);
1.238 brouard 6823: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6824: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6825: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6826: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6827: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6828: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6829:
1.238 brouard 6830: */
6831: for (i=1; i< nlstate ; i ++) {
6832: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+i,cpt,i+1);
6833: /* 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 6834:
1.238 brouard 6835: }
6836: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6837: }
6838: } /* end nres */
6839: } /* end kl 3eme */
1.126 brouard 6840:
1.223 brouard 6841: /* 4eme */
1.201 brouard 6842: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6843: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6844: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6845: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6846: continue;
1.238 brouard 6847: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6848: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6849: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6850: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6851: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6852: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6853: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6854: vlv= nbcode[Tvaraff[k]][lv];
6855: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6856: }
6857: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6858: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6859: }
6860: fprintf(ficgp,"\n#\n");
6861: if(invalidvarcomb[k1]){
6862: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6863: continue;
1.223 brouard 6864: }
1.238 brouard 6865:
1.241 brouard 6866: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6867: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6868: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6869: k=3;
6870: for (i=1; i<= nlstate ; i ++){
6871: if(i==1){
6872: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6873: }else{
6874: fprintf(ficgp,", '' ");
6875: }
6876: l=(nlstate+ndeath)*(i-1)+1;
6877: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6878: for (j=2; j<= nlstate+ndeath ; j ++)
6879: fprintf(ficgp,"+$%d",k+l+j-1);
6880: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6881: } /* nlstate */
6882: fprintf(ficgp,"\nset out\n");
6883: } /* end cpt state*/
6884: } /* end nres */
6885: } /* end covariate k1 */
6886:
1.220 brouard 6887: /* 5eme */
1.201 brouard 6888: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6889: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6890: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6891: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6892: continue;
1.238 brouard 6893: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6894: 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);
6895: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6896: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6897: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6898: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6899: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6900: vlv= nbcode[Tvaraff[k]][lv];
6901: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6902: }
6903: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6904: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6905: }
6906: fprintf(ficgp,"\n#\n");
6907: if(invalidvarcomb[k1]){
6908: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6909: continue;
6910: }
1.227 brouard 6911:
1.241 brouard 6912: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6913: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6914: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6915: k=3;
6916: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6917: if(j==1)
6918: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6919: else
6920: fprintf(ficgp,", '' ");
6921: l=(nlstate+ndeath)*(cpt-1) +j;
6922: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6923: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6924: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6925: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6926: } /* nlstate */
6927: fprintf(ficgp,", '' ");
6928: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6929: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6930: l=(nlstate+ndeath)*(cpt-1) +j;
6931: if(j < nlstate)
6932: fprintf(ficgp,"$%d +",k+l);
6933: else
6934: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6935: }
6936: fprintf(ficgp,"\nset out\n");
6937: } /* end cpt state*/
6938: } /* end covariate */
6939: } /* end nres */
1.227 brouard 6940:
1.220 brouard 6941: /* 6eme */
1.202 brouard 6942: /* CV preval stable (period) for each covariate */
1.237 brouard 6943: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6944: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6945: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6946: continue;
1.255 brouard 6947: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.227 brouard 6948:
1.211 brouard 6949: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6950: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6951: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6952: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6953: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6954: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6955: vlv= nbcode[Tvaraff[k]][lv];
6956: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6957: }
1.237 brouard 6958: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6959: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6960: }
1.211 brouard 6961: fprintf(ficgp,"\n#\n");
1.223 brouard 6962: if(invalidvarcomb[k1]){
1.227 brouard 6963: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6964: continue;
1.223 brouard 6965: }
1.227 brouard 6966:
1.241 brouard 6967: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6968: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6969: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6970: k=3; /* Offset */
1.255 brouard 6971: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 6972: if(i==1)
6973: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6974: else
6975: fprintf(ficgp,", '' ");
1.255 brouard 6976: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 6977: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6978: for (j=2; j<= nlstate ; j ++)
6979: fprintf(ficgp,"+$%d",k+l+j-1);
6980: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6981: } /* nlstate */
1.201 brouard 6982: fprintf(ficgp,"\nset out\n");
1.153 brouard 6983: } /* end cpt state*/
6984: } /* end covariate */
1.227 brouard 6985:
6986:
1.220 brouard 6987: /* 7eme */
1.218 brouard 6988: if(backcast == 1){
1.217 brouard 6989: /* CV back preval stable (period) for each covariate */
1.237 brouard 6990: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6991: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6992: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6993: continue;
1.255 brouard 6994: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
6995: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 6996: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6997: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6998: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6999: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7000: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7001: vlv= nbcode[Tvaraff[k]][lv];
7002: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7003: }
1.237 brouard 7004: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7005: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7006: }
1.227 brouard 7007: fprintf(ficgp,"\n#\n");
7008: if(invalidvarcomb[k1]){
7009: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7010: continue;
7011: }
7012:
1.241 brouard 7013: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 7014: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7015: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7016: k=3; /* Offset */
1.255 brouard 7017: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7018: if(i==1)
7019: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7020: else
7021: fprintf(ficgp,", '' ");
7022: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7023: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7024: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7025: /* 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 7026: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7027: /* for (j=2; j<= nlstate ; j ++) */
7028: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7029: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7030: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7031: } /* nlstate */
7032: fprintf(ficgp,"\nset out\n");
1.218 brouard 7033: } /* end cpt state*/
7034: } /* end covariate */
7035: } /* End if backcast */
7036:
1.223 brouard 7037: /* 8eme */
1.218 brouard 7038: if(prevfcast==1){
7039: /* Projection from cross-sectional to stable (period) for each covariate */
7040:
1.237 brouard 7041: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7042: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7043: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7044: continue;
1.211 brouard 7045: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7046: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7047: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7048: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7049: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7050: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7051: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7052: vlv= nbcode[Tvaraff[k]][lv];
7053: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7054: }
1.237 brouard 7055: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7056: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7057: }
1.227 brouard 7058: fprintf(ficgp,"\n#\n");
7059: if(invalidvarcomb[k1]){
7060: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7061: continue;
7062: }
7063:
7064: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7065: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7066: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7067: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7068: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7069: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7070: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7071: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7072: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7073: if(i==1){
7074: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7075: }else{
7076: fprintf(ficgp,",\\\n '' ");
7077: }
7078: if(cptcoveff ==0){ /* No covariate */
7079: ioffset=2; /* Age is in 2 */
7080: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7081: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7082: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7083: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7084: fprintf(ficgp," u %d:(", ioffset);
7085: if(i==nlstate+1)
7086: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7087: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7088: else
7089: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7090: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7091: }else{ /* more than 2 covariates */
7092: if(cptcoveff ==1){
7093: ioffset=4; /* Age is in 4 */
7094: }else{
7095: ioffset=6; /* Age is in 6 */
7096: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7097: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7098: }
7099: fprintf(ficgp," u %d:(",ioffset);
7100: kl=0;
7101: strcpy(gplotcondition,"(");
7102: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7103: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7104: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7105: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7106: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7107: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7108: kl++;
7109: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7110: kl++;
7111: if(k <cptcoveff && cptcoveff>1)
7112: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7113: }
7114: strcpy(gplotcondition+strlen(gplotcondition),")");
7115: /* 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 *\/ */
7116: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7117: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7118: /* '' 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*/
7119: if(i==nlstate+1){
7120: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7121: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7122: }else{
7123: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7124: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7125: }
7126: } /* end if covariate */
7127: } /* nlstate */
7128: fprintf(ficgp,"\nset out\n");
1.223 brouard 7129: } /* end cpt state*/
7130: } /* end covariate */
7131: } /* End if prevfcast */
1.227 brouard 7132:
7133:
1.238 brouard 7134: /* 9eme writing MLE parameters */
7135: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7136: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7137: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7138: for(k=1; k <=(nlstate+ndeath); k++){
7139: if (k != i) {
1.227 brouard 7140: fprintf(ficgp,"# current state %d\n",k);
7141: for(j=1; j <=ncovmodel; j++){
7142: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7143: jk++;
7144: }
7145: fprintf(ficgp,"\n");
1.126 brouard 7146: }
7147: }
1.223 brouard 7148: }
1.187 brouard 7149: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7150:
1.145 brouard 7151: /*goto avoid;*/
1.238 brouard 7152: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7153: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7154: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7155: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7156: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7157: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7158: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7159: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7160: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7161: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7162: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7163: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7164: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7165: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7166: fprintf(ficgp,"#\n");
1.223 brouard 7167: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7168: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7169: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7170: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7171: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7172: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7173: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7174: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7175: continue;
7176: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7177: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7178: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7179: }
7180: fprintf(ficgp,"\n#\n");
1.241 brouard 7181: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7182: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7183: if (ng==1){
7184: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7185: fprintf(ficgp,"\nunset log y");
7186: }else if (ng==2){
7187: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7188: fprintf(ficgp,"\nset log y");
7189: }else if (ng==3){
7190: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7191: fprintf(ficgp,"\nset log y");
7192: }else
7193: fprintf(ficgp,"\nunset title ");
7194: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7195: i=1;
7196: for(k2=1; k2<=nlstate; k2++) {
7197: k3=i;
7198: for(k=1; k<=(nlstate+ndeath); k++) {
7199: if (k != k2){
7200: switch( ng) {
7201: case 1:
7202: if(nagesqr==0)
7203: fprintf(ficgp," p%d+p%d*x",i,i+1);
7204: else /* nagesqr =1 */
7205: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7206: break;
7207: case 2: /* ng=2 */
7208: if(nagesqr==0)
7209: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7210: else /* nagesqr =1 */
7211: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7212: break;
7213: case 3:
7214: if(nagesqr==0)
7215: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7216: else /* nagesqr =1 */
7217: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7218: break;
7219: }
7220: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7221: ijp=1; /* product no age */
7222: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7223: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7224: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7225: if(j==Tage[ij]) { /* Product by age */
7226: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7227: if(DummyV[j]==0){
1.237 brouard 7228: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7229: }else{ /* quantitative */
7230: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7231: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7232: }
7233: ij++;
7234: }
7235: }else if(j==Tprod[ijp]) { /* */
7236: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7237: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7238: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7239: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7240: /* 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)]); */
7241: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7242: }else{ /* Vn is dummy and Vm is quanti */
7243: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7244: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7245: }
7246: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7247: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7248: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7249: }else{ /* Both quanti */
7250: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7251: }
7252: }
1.238 brouard 7253: ijp++;
1.237 brouard 7254: }
7255: } else{ /* simple covariate */
7256: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7257: if(Dummy[j]==0){
7258: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7259: }else{ /* quantitative */
7260: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7261: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7262: }
1.237 brouard 7263: } /* end simple */
7264: } /* end j */
1.223 brouard 7265: }else{
7266: i=i-ncovmodel;
7267: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7268: fprintf(ficgp," (1.");
7269: }
1.227 brouard 7270:
1.223 brouard 7271: if(ng != 1){
7272: fprintf(ficgp,")/(1");
1.227 brouard 7273:
1.223 brouard 7274: for(k1=1; k1 <=nlstate; k1++){
7275: if(nagesqr==0)
7276: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7277: else /* nagesqr =1 */
7278: 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 7279:
1.223 brouard 7280: ij=1;
7281: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7282: if((j-2)==Tage[ij]) { /* Bug valgrind */
7283: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7284: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7285: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7286: ij++;
7287: }
7288: }
7289: else
1.225 brouard 7290: 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 7291: }
7292: fprintf(ficgp,")");
7293: }
7294: fprintf(ficgp,")");
7295: if(ng ==2)
7296: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7297: else /* ng= 3 */
7298: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7299: }else{ /* end ng <> 1 */
7300: if( k !=k2) /* logit p11 is hard to draw */
7301: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7302: }
7303: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7304: fprintf(ficgp,",");
7305: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7306: fprintf(ficgp,",");
7307: i=i+ncovmodel;
7308: } /* end k */
7309: } /* end k2 */
7310: fprintf(ficgp,"\n set out\n");
7311: } /* end jk */
7312: } /* end ng */
7313: /* avoid: */
7314: fflush(ficgp);
1.126 brouard 7315: } /* end gnuplot */
7316:
7317:
7318: /*************** Moving average **************/
1.219 brouard 7319: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7320: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7321:
1.222 brouard 7322: int i, cpt, cptcod;
7323: int modcovmax =1;
7324: int mobilavrange, mob;
7325: int iage=0;
7326:
7327: double sum=0.;
7328: double age;
7329: double *sumnewp, *sumnewm;
7330: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7331:
7332:
1.225 brouard 7333: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7334: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7335:
7336: sumnewp = vector(1,ncovcombmax);
7337: sumnewm = vector(1,ncovcombmax);
7338: agemingood = vector(1,ncovcombmax);
7339: agemaxgood = vector(1,ncovcombmax);
7340:
7341: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7342: sumnewm[cptcod]=0.;
7343: sumnewp[cptcod]=0.;
7344: agemingood[cptcod]=0;
7345: agemaxgood[cptcod]=0;
7346: }
7347: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7348:
7349: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7350: if(mobilav==1) mobilavrange=5; /* default */
7351: else mobilavrange=mobilav;
7352: for (age=bage; age<=fage; age++)
7353: for (i=1; i<=nlstate;i++)
7354: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7355: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7356: /* We keep the original values on the extreme ages bage, fage and for
7357: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7358: we use a 5 terms etc. until the borders are no more concerned.
7359: */
7360: for (mob=3;mob <=mobilavrange;mob=mob+2){
7361: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7362: for (i=1; i<=nlstate;i++){
7363: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7364: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7365: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7366: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7367: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7368: }
7369: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7370: }
7371: }
7372: }/* end age */
7373: }/* end mob */
7374: }else
7375: return -1;
7376: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7377: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7378: if(invalidvarcomb[cptcod]){
7379: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7380: continue;
7381: }
1.219 brouard 7382:
1.222 brouard 7383: agemingood[cptcod]=fage-(mob-1)/2;
7384: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7385: sumnewm[cptcod]=0.;
7386: for (i=1; i<=nlstate;i++){
7387: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7388: }
7389: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7390: agemingood[cptcod]=age;
7391: }else{ /* bad */
7392: for (i=1; i<=nlstate;i++){
7393: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7394: } /* i */
7395: } /* end bad */
7396: }/* age */
7397: sum=0.;
7398: for (i=1; i<=nlstate;i++){
7399: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7400: }
7401: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7402: 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);
7403: /* for (i=1; i<=nlstate;i++){ */
7404: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7405: /* } /\* i *\/ */
7406: } /* end bad */
7407: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7408: /* From youngest, finding the oldest wrong */
7409: agemaxgood[cptcod]=bage+(mob-1)/2;
7410: for (age=bage+(mob-1)/2; age<=fage; age++){
7411: sumnewm[cptcod]=0.;
7412: for (i=1; i<=nlstate;i++){
7413: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7414: }
7415: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7416: agemaxgood[cptcod]=age;
7417: }else{ /* bad */
7418: for (i=1; i<=nlstate;i++){
7419: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7420: } /* i */
7421: } /* end bad */
7422: }/* age */
7423: sum=0.;
7424: for (i=1; i<=nlstate;i++){
7425: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7426: }
7427: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7428: 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);
7429: /* for (i=1; i<=nlstate;i++){ */
7430: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7431: /* } /\* i *\/ */
7432: } /* end bad */
7433:
7434: for (age=bage; age<=fage; age++){
1.235 brouard 7435: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7436: sumnewp[cptcod]=0.;
7437: sumnewm[cptcod]=0.;
7438: for (i=1; i<=nlstate;i++){
7439: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7440: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7441: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7442: }
7443: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7444: }
7445: /* printf("\n"); */
7446: /* } */
7447: /* brutal averaging */
7448: for (i=1; i<=nlstate;i++){
7449: for (age=1; age<=bage; age++){
7450: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7451: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7452: }
7453: for (age=fage; age<=AGESUP; age++){
7454: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7455: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7456: }
7457: } /* end i status */
7458: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7459: for (age=1; age<=AGESUP; age++){
7460: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7461: mobaverage[(int)age][i][cptcod]=0.;
7462: }
7463: }
7464: }/* end cptcod */
7465: free_vector(sumnewm,1, ncovcombmax);
7466: free_vector(sumnewp,1, ncovcombmax);
7467: free_vector(agemaxgood,1, ncovcombmax);
7468: free_vector(agemingood,1, ncovcombmax);
7469: return 0;
7470: }/* End movingaverage */
1.218 brouard 7471:
1.126 brouard 7472:
7473: /************** Forecasting ******************/
1.235 brouard 7474: 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 7475: /* proj1, year, month, day of starting projection
7476: agemin, agemax range of age
7477: dateprev1 dateprev2 range of dates during which prevalence is computed
7478: anproj2 year of en of projection (same day and month as proj1).
7479: */
1.235 brouard 7480: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7481: double agec; /* generic age */
7482: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7483: double *popeffectif,*popcount;
7484: double ***p3mat;
1.218 brouard 7485: /* double ***mobaverage; */
1.126 brouard 7486: char fileresf[FILENAMELENGTH];
7487:
7488: agelim=AGESUP;
1.211 brouard 7489: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7490: in each health status at the date of interview (if between dateprev1 and dateprev2).
7491: We still use firstpass and lastpass as another selection.
7492: */
1.214 brouard 7493: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7494: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7495:
1.201 brouard 7496: strcpy(fileresf,"F_");
7497: strcat(fileresf,fileresu);
1.126 brouard 7498: if((ficresf=fopen(fileresf,"w"))==NULL) {
7499: printf("Problem with forecast resultfile: %s\n", fileresf);
7500: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7501: }
1.235 brouard 7502: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7503: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7504:
1.225 brouard 7505: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7506:
7507:
7508: stepsize=(int) (stepm+YEARM-1)/YEARM;
7509: if (stepm<=12) stepsize=1;
7510: if(estepm < stepm){
7511: printf ("Problem %d lower than %d\n",estepm, stepm);
7512: }
7513: else hstepm=estepm;
7514:
7515: hstepm=hstepm/stepm;
7516: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7517: fractional in yp1 */
7518: anprojmean=yp;
7519: yp2=modf((yp1*12),&yp);
7520: mprojmean=yp;
7521: yp1=modf((yp2*30.5),&yp);
7522: jprojmean=yp;
7523: if(jprojmean==0) jprojmean=1;
7524: if(mprojmean==0) jprojmean=1;
7525:
1.227 brouard 7526: i1=pow(2,cptcoveff);
1.126 brouard 7527: if (cptcovn < 1){i1=1;}
7528:
7529: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7530:
7531: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7532:
1.126 brouard 7533: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7534: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7535: for(k=1; k<=i1;k++){
1.253 brouard 7536: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7537: continue;
1.227 brouard 7538: if(invalidvarcomb[k]){
7539: printf("\nCombination (%d) projection ignored because no cases \n",k);
7540: continue;
7541: }
7542: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7543: for(j=1;j<=cptcoveff;j++) {
7544: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7545: }
1.235 brouard 7546: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7547: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7548: }
1.227 brouard 7549: fprintf(ficresf," yearproj age");
7550: for(j=1; j<=nlstate+ndeath;j++){
7551: for(i=1; i<=nlstate;i++)
7552: fprintf(ficresf," p%d%d",i,j);
7553: fprintf(ficresf," wp.%d",j);
7554: }
7555: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7556: fprintf(ficresf,"\n");
7557: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7558: for (agec=fage; agec>=(ageminpar-1); agec--){
7559: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7560: nhstepm = nhstepm/hstepm;
7561: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7562: oldm=oldms;savm=savms;
1.235 brouard 7563: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7564:
7565: for (h=0; h<=nhstepm; h++){
7566: if (h*hstepm/YEARM*stepm ==yearp) {
7567: fprintf(ficresf,"\n");
7568: for(j=1;j<=cptcoveff;j++)
7569: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7570: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7571: }
7572: for(j=1; j<=nlstate+ndeath;j++) {
7573: ppij=0.;
7574: for(i=1; i<=nlstate;i++) {
7575: if (mobilav==1)
7576: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7577: else {
7578: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7579: }
7580: if (h*hstepm/YEARM*stepm== yearp) {
7581: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7582: }
7583: } /* end i */
7584: if (h*hstepm/YEARM*stepm==yearp) {
7585: fprintf(ficresf," %.3f", ppij);
7586: }
7587: }/* end j */
7588: } /* end h */
7589: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7590: } /* end agec */
7591: } /* end yearp */
7592: } /* end k */
1.219 brouard 7593:
1.126 brouard 7594: fclose(ficresf);
1.215 brouard 7595: printf("End of Computing forecasting \n");
7596: fprintf(ficlog,"End of Computing forecasting\n");
7597:
1.126 brouard 7598: }
7599:
1.218 brouard 7600: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7601: /* 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 7602: /* /\* back1, year, month, day of starting backection */
7603: /* agemin, agemax range of age */
7604: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7605: /* anback2 year of en of backection (same day and month as back1). */
7606: /* *\/ */
7607: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7608: /* double agec; /\* generic age *\/ */
7609: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7610: /* double *popeffectif,*popcount; */
7611: /* double ***p3mat; */
7612: /* /\* double ***mobaverage; *\/ */
7613: /* char fileresfb[FILENAMELENGTH]; */
7614:
7615: /* agelim=AGESUP; */
7616: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7617: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7618: /* We still use firstpass and lastpass as another selection. */
7619: /* *\/ */
7620: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7621: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7622: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7623:
7624: /* strcpy(fileresfb,"FB_"); */
7625: /* strcat(fileresfb,fileresu); */
7626: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7627: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7628: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7629: /* } */
7630: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7631: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7632:
1.225 brouard 7633: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7634:
7635: /* /\* if (mobilav!=0) { *\/ */
7636: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7637: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7638: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7639: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7640: /* /\* } *\/ */
7641: /* /\* } *\/ */
7642:
7643: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7644: /* if (stepm<=12) stepsize=1; */
7645: /* if(estepm < stepm){ */
7646: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7647: /* } */
7648: /* else hstepm=estepm; */
7649:
7650: /* hstepm=hstepm/stepm; */
7651: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7652: /* fractional in yp1 *\/ */
7653: /* anprojmean=yp; */
7654: /* yp2=modf((yp1*12),&yp); */
7655: /* mprojmean=yp; */
7656: /* yp1=modf((yp2*30.5),&yp); */
7657: /* jprojmean=yp; */
7658: /* if(jprojmean==0) jprojmean=1; */
7659: /* if(mprojmean==0) jprojmean=1; */
7660:
1.225 brouard 7661: /* i1=cptcoveff; */
1.218 brouard 7662: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7663:
1.218 brouard 7664: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7665:
1.218 brouard 7666: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7667:
7668: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7669: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7670: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7671: /* k=k+1; */
7672: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7673: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7674: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7675: /* } */
7676: /* fprintf(ficresfb," yearbproj age"); */
7677: /* for(j=1; j<=nlstate+ndeath;j++){ */
7678: /* for(i=1; i<=nlstate;i++) */
7679: /* fprintf(ficresfb," p%d%d",i,j); */
7680: /* fprintf(ficresfb," p.%d",j); */
7681: /* } */
7682: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7683: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7684: /* fprintf(ficresfb,"\n"); */
7685: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7686: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7687: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7688: /* nhstepm = nhstepm/hstepm; */
7689: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7690: /* oldm=oldms;savm=savms; */
7691: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7692: /* for (h=0; h<=nhstepm; h++){ */
7693: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7694: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7695: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7696: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7697: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7698: /* } */
7699: /* for(j=1; j<=nlstate+ndeath;j++) { */
7700: /* ppij=0.; */
7701: /* for(i=1; i<=nlstate;i++) { */
7702: /* if (mobilav==1) */
7703: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7704: /* else { */
7705: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7706: /* } */
7707: /* if (h*hstepm/YEARM*stepm== yearp) { */
7708: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7709: /* } */
7710: /* } /\* end i *\/ */
7711: /* if (h*hstepm/YEARM*stepm==yearp) { */
7712: /* fprintf(ficresfb," %.3f", ppij); */
7713: /* } */
7714: /* }/\* end j *\/ */
7715: /* } /\* end h *\/ */
7716: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7717: /* } /\* end agec *\/ */
7718: /* } /\* end yearp *\/ */
7719: /* } /\* end cptcod *\/ */
7720: /* } /\* end cptcov *\/ */
7721:
7722: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7723:
7724: /* fclose(ficresfb); */
7725: /* printf("End of Computing Back forecasting \n"); */
7726: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7727:
1.218 brouard 7728: /* } */
1.217 brouard 7729:
1.126 brouard 7730: /************** Forecasting *****not tested NB*************/
1.227 brouard 7731: /* 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 7732:
1.227 brouard 7733: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7734: /* int *popage; */
7735: /* double calagedatem, agelim, kk1, kk2; */
7736: /* double *popeffectif,*popcount; */
7737: /* double ***p3mat,***tabpop,***tabpopprev; */
7738: /* /\* double ***mobaverage; *\/ */
7739: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7740:
1.227 brouard 7741: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7742: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7743: /* agelim=AGESUP; */
7744: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7745:
1.227 brouard 7746: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7747:
7748:
1.227 brouard 7749: /* strcpy(filerespop,"POP_"); */
7750: /* strcat(filerespop,fileresu); */
7751: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7752: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7753: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7754: /* } */
7755: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7756: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7757:
1.227 brouard 7758: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7759:
1.227 brouard 7760: /* /\* if (mobilav!=0) { *\/ */
7761: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7762: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7763: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7764: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7765: /* /\* } *\/ */
7766: /* /\* } *\/ */
1.126 brouard 7767:
1.227 brouard 7768: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7769: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7770:
1.227 brouard 7771: /* agelim=AGESUP; */
1.126 brouard 7772:
1.227 brouard 7773: /* hstepm=1; */
7774: /* hstepm=hstepm/stepm; */
1.218 brouard 7775:
1.227 brouard 7776: /* if (popforecast==1) { */
7777: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7778: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7779: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7780: /* } */
7781: /* popage=ivector(0,AGESUP); */
7782: /* popeffectif=vector(0,AGESUP); */
7783: /* popcount=vector(0,AGESUP); */
1.126 brouard 7784:
1.227 brouard 7785: /* i=1; */
7786: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7787:
1.227 brouard 7788: /* imx=i; */
7789: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7790: /* } */
1.218 brouard 7791:
1.227 brouard 7792: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7793: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7794: /* k=k+1; */
7795: /* fprintf(ficrespop,"\n#******"); */
7796: /* for(j=1;j<=cptcoveff;j++) { */
7797: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7798: /* } */
7799: /* fprintf(ficrespop,"******\n"); */
7800: /* fprintf(ficrespop,"# Age"); */
7801: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7802: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7803:
1.227 brouard 7804: /* for (cpt=0; cpt<=0;cpt++) { */
7805: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7806:
1.227 brouard 7807: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7808: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7809: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7810:
1.227 brouard 7811: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7812: /* oldm=oldms;savm=savms; */
7813: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7814:
1.227 brouard 7815: /* for (h=0; h<=nhstepm; h++){ */
7816: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7817: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7818: /* } */
7819: /* for(j=1; j<=nlstate+ndeath;j++) { */
7820: /* kk1=0.;kk2=0; */
7821: /* for(i=1; i<=nlstate;i++) { */
7822: /* if (mobilav==1) */
7823: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7824: /* else { */
7825: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7826: /* } */
7827: /* } */
7828: /* if (h==(int)(calagedatem+12*cpt)){ */
7829: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7830: /* /\*fprintf(ficrespop," %.3f", kk1); */
7831: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7832: /* } */
7833: /* } */
7834: /* for(i=1; i<=nlstate;i++){ */
7835: /* kk1=0.; */
7836: /* for(j=1; j<=nlstate;j++){ */
7837: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7838: /* } */
7839: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7840: /* } */
1.218 brouard 7841:
1.227 brouard 7842: /* if (h==(int)(calagedatem+12*cpt)) */
7843: /* for(j=1; j<=nlstate;j++) */
7844: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7845: /* } */
7846: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7847: /* } */
7848: /* } */
1.218 brouard 7849:
1.227 brouard 7850: /* /\******\/ */
1.218 brouard 7851:
1.227 brouard 7852: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7853: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7854: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7855: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7856: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7857:
1.227 brouard 7858: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7859: /* oldm=oldms;savm=savms; */
7860: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7861: /* for (h=0; h<=nhstepm; h++){ */
7862: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7863: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7864: /* } */
7865: /* for(j=1; j<=nlstate+ndeath;j++) { */
7866: /* kk1=0.;kk2=0; */
7867: /* for(i=1; i<=nlstate;i++) { */
7868: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7869: /* } */
7870: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7871: /* } */
7872: /* } */
7873: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7874: /* } */
7875: /* } */
7876: /* } */
7877: /* } */
1.218 brouard 7878:
1.227 brouard 7879: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7880:
1.227 brouard 7881: /* if (popforecast==1) { */
7882: /* free_ivector(popage,0,AGESUP); */
7883: /* free_vector(popeffectif,0,AGESUP); */
7884: /* free_vector(popcount,0,AGESUP); */
7885: /* } */
7886: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7887: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7888: /* fclose(ficrespop); */
7889: /* } /\* End of popforecast *\/ */
1.218 brouard 7890:
1.126 brouard 7891: int fileappend(FILE *fichier, char *optionfich)
7892: {
7893: if((fichier=fopen(optionfich,"a"))==NULL) {
7894: printf("Problem with file: %s\n", optionfich);
7895: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7896: return (0);
7897: }
7898: fflush(fichier);
7899: return (1);
7900: }
7901:
7902:
7903: /**************** function prwizard **********************/
7904: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7905: {
7906:
7907: /* Wizard to print covariance matrix template */
7908:
1.164 brouard 7909: char ca[32], cb[32];
7910: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7911: int numlinepar;
7912:
7913: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7914: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7915: for(i=1; i <=nlstate; i++){
7916: jj=0;
7917: for(j=1; j <=nlstate+ndeath; j++){
7918: if(j==i) continue;
7919: jj++;
7920: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7921: printf("%1d%1d",i,j);
7922: fprintf(ficparo,"%1d%1d",i,j);
7923: for(k=1; k<=ncovmodel;k++){
7924: /* printf(" %lf",param[i][j][k]); */
7925: /* fprintf(ficparo," %lf",param[i][j][k]); */
7926: printf(" 0.");
7927: fprintf(ficparo," 0.");
7928: }
7929: printf("\n");
7930: fprintf(ficparo,"\n");
7931: }
7932: }
7933: printf("# Scales (for hessian or gradient estimation)\n");
7934: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7935: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7936: for(i=1; i <=nlstate; i++){
7937: jj=0;
7938: for(j=1; j <=nlstate+ndeath; j++){
7939: if(j==i) continue;
7940: jj++;
7941: fprintf(ficparo,"%1d%1d",i,j);
7942: printf("%1d%1d",i,j);
7943: fflush(stdout);
7944: for(k=1; k<=ncovmodel;k++){
7945: /* printf(" %le",delti3[i][j][k]); */
7946: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7947: printf(" 0.");
7948: fprintf(ficparo," 0.");
7949: }
7950: numlinepar++;
7951: printf("\n");
7952: fprintf(ficparo,"\n");
7953: }
7954: }
7955: printf("# Covariance matrix\n");
7956: /* # 121 Var(a12)\n\ */
7957: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7958: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7959: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7960: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7961: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7962: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7963: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7964: fflush(stdout);
7965: fprintf(ficparo,"# Covariance matrix\n");
7966: /* # 121 Var(a12)\n\ */
7967: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7968: /* # ...\n\ */
7969: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7970:
7971: for(itimes=1;itimes<=2;itimes++){
7972: jj=0;
7973: for(i=1; i <=nlstate; i++){
7974: for(j=1; j <=nlstate+ndeath; j++){
7975: if(j==i) continue;
7976: for(k=1; k<=ncovmodel;k++){
7977: jj++;
7978: ca[0]= k+'a'-1;ca[1]='\0';
7979: if(itimes==1){
7980: printf("#%1d%1d%d",i,j,k);
7981: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7982: }else{
7983: printf("%1d%1d%d",i,j,k);
7984: fprintf(ficparo,"%1d%1d%d",i,j,k);
7985: /* printf(" %.5le",matcov[i][j]); */
7986: }
7987: ll=0;
7988: for(li=1;li <=nlstate; li++){
7989: for(lj=1;lj <=nlstate+ndeath; lj++){
7990: if(lj==li) continue;
7991: for(lk=1;lk<=ncovmodel;lk++){
7992: ll++;
7993: if(ll<=jj){
7994: cb[0]= lk +'a'-1;cb[1]='\0';
7995: if(ll<jj){
7996: if(itimes==1){
7997: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7998: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7999: }else{
8000: printf(" 0.");
8001: fprintf(ficparo," 0.");
8002: }
8003: }else{
8004: if(itimes==1){
8005: printf(" Var(%s%1d%1d)",ca,i,j);
8006: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8007: }else{
8008: printf(" 0.");
8009: fprintf(ficparo," 0.");
8010: }
8011: }
8012: }
8013: } /* end lk */
8014: } /* end lj */
8015: } /* end li */
8016: printf("\n");
8017: fprintf(ficparo,"\n");
8018: numlinepar++;
8019: } /* end k*/
8020: } /*end j */
8021: } /* end i */
8022: } /* end itimes */
8023:
8024: } /* end of prwizard */
8025: /******************* Gompertz Likelihood ******************************/
8026: double gompertz(double x[])
8027: {
8028: double A,B,L=0.0,sump=0.,num=0.;
8029: int i,n=0; /* n is the size of the sample */
8030:
1.220 brouard 8031: for (i=1;i<=imx ; i++) {
1.126 brouard 8032: sump=sump+weight[i];
8033: /* sump=sump+1;*/
8034: num=num+1;
8035: }
8036:
8037:
8038: /* for (i=0; i<=imx; i++)
8039: 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]);*/
8040:
8041: for (i=1;i<=imx ; i++)
8042: {
8043: if (cens[i] == 1 && wav[i]>1)
8044: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8045:
8046: if (cens[i] == 0 && wav[i]>1)
8047: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8048: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8049:
8050: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8051: if (wav[i] > 1 ) { /* ??? */
8052: L=L+A*weight[i];
8053: /* 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]);*/
8054: }
8055: }
8056:
8057: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8058:
8059: return -2*L*num/sump;
8060: }
8061:
1.136 brouard 8062: #ifdef GSL
8063: /******************* Gompertz_f Likelihood ******************************/
8064: double gompertz_f(const gsl_vector *v, void *params)
8065: {
8066: double A,B,LL=0.0,sump=0.,num=0.;
8067: double *x= (double *) v->data;
8068: int i,n=0; /* n is the size of the sample */
8069:
8070: for (i=0;i<=imx-1 ; i++) {
8071: sump=sump+weight[i];
8072: /* sump=sump+1;*/
8073: num=num+1;
8074: }
8075:
8076:
8077: /* for (i=0; i<=imx; i++)
8078: 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]);*/
8079: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8080: for (i=1;i<=imx ; i++)
8081: {
8082: if (cens[i] == 1 && wav[i]>1)
8083: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8084:
8085: if (cens[i] == 0 && wav[i]>1)
8086: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8087: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8088:
8089: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8090: if (wav[i] > 1 ) { /* ??? */
8091: LL=LL+A*weight[i];
8092: /* 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]);*/
8093: }
8094: }
8095:
8096: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8097: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8098:
8099: return -2*LL*num/sump;
8100: }
8101: #endif
8102:
1.126 brouard 8103: /******************* Printing html file ***********/
1.201 brouard 8104: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8105: int lastpass, int stepm, int weightopt, char model[],\
8106: int imx, double p[],double **matcov,double agemortsup){
8107: int i,k;
8108:
8109: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8110: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8111: for (i=1;i<=2;i++)
8112: 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 8113: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8114: fprintf(fichtm,"</ul>");
8115:
8116: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8117:
8118: 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>");
8119:
8120: for (k=agegomp;k<(agemortsup-2);k++)
8121: 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]);
8122:
8123:
8124: fflush(fichtm);
8125: }
8126:
8127: /******************* Gnuplot file **************/
1.201 brouard 8128: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8129:
8130: char dirfileres[132],optfileres[132];
1.164 brouard 8131:
1.126 brouard 8132: int ng;
8133:
8134:
8135: /*#ifdef windows */
8136: fprintf(ficgp,"cd \"%s\" \n",pathc);
8137: /*#endif */
8138:
8139:
8140: strcpy(dirfileres,optionfilefiname);
8141: strcpy(optfileres,"vpl");
1.199 brouard 8142: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8143: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8144: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8145: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8146: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8147:
8148: }
8149:
1.136 brouard 8150: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8151: {
1.126 brouard 8152:
1.136 brouard 8153: /*-------- data file ----------*/
8154: FILE *fic;
8155: char dummy[]=" ";
1.240 brouard 8156: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8157: int lstra;
1.136 brouard 8158: int linei, month, year,iout;
8159: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8160: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8161: char *stratrunc;
1.223 brouard 8162:
1.240 brouard 8163: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8164: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8165:
1.240 brouard 8166: for(v=1; v <=ncovcol;v++){
8167: DummyV[v]=0;
8168: FixedV[v]=0;
8169: }
8170: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8171: DummyV[v]=1;
8172: FixedV[v]=0;
8173: }
8174: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8175: DummyV[v]=0;
8176: FixedV[v]=1;
8177: }
8178: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8179: DummyV[v]=1;
8180: FixedV[v]=1;
8181: }
8182: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8183: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8184: 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]);
8185: }
1.126 brouard 8186:
1.136 brouard 8187: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8188: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8189: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8190: }
1.126 brouard 8191:
1.136 brouard 8192: i=1;
8193: linei=0;
8194: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8195: linei=linei+1;
8196: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8197: if(line[j] == '\t')
8198: line[j] = ' ';
8199: }
8200: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8201: ;
8202: };
8203: line[j+1]=0; /* Trims blanks at end of line */
8204: if(line[0]=='#'){
8205: fprintf(ficlog,"Comment line\n%s\n",line);
8206: printf("Comment line\n%s\n",line);
8207: continue;
8208: }
8209: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8210: strcpy(line, linetmp);
1.223 brouard 8211:
8212: /* Loops on waves */
8213: for (j=maxwav;j>=1;j--){
8214: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8215: cutv(stra, strb, line, ' ');
8216: if(strb[0]=='.') { /* Missing value */
8217: lval=-1;
8218: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8219: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8220: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8221: 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);
8222: 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);
8223: return 1;
8224: }
8225: }else{
8226: errno=0;
8227: /* what_kind_of_number(strb); */
8228: dval=strtod(strb,&endptr);
8229: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8230: /* if(strb != endptr && *endptr == '\0') */
8231: /* dval=dlval; */
8232: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8233: if( strb[0]=='\0' || (*endptr != '\0')){
8234: 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);
8235: 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);
8236: return 1;
8237: }
8238: cotqvar[j][iv][i]=dval;
8239: cotvar[j][ntv+iv][i]=dval;
8240: }
8241: strcpy(line,stra);
1.223 brouard 8242: }/* end loop ntqv */
1.225 brouard 8243:
1.223 brouard 8244: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8245: cutv(stra, strb, line, ' ');
8246: if(strb[0]=='.') { /* Missing value */
8247: lval=-1;
8248: }else{
8249: errno=0;
8250: lval=strtol(strb,&endptr,10);
8251: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8252: if( strb[0]=='\0' || (*endptr != '\0')){
8253: 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);
8254: 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);
8255: return 1;
8256: }
8257: }
8258: if(lval <-1 || lval >1){
8259: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8260: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8261: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8262: For example, for multinomial values like 1, 2 and 3,\n \
8263: build V1=0 V2=0 for the reference value (1),\n \
8264: V1=1 V2=0 for (2) \n \
1.223 brouard 8265: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8266: output of IMaCh is often meaningless.\n \
1.223 brouard 8267: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8268: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8269: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8270: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8271: For example, for multinomial values like 1, 2 and 3,\n \
8272: build V1=0 V2=0 for the reference value (1),\n \
8273: V1=1 V2=0 for (2) \n \
1.223 brouard 8274: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8275: output of IMaCh is often meaningless.\n \
1.223 brouard 8276: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8277: return 1;
8278: }
8279: cotvar[j][iv][i]=(double)(lval);
8280: strcpy(line,stra);
1.223 brouard 8281: }/* end loop ntv */
1.225 brouard 8282:
1.223 brouard 8283: /* Statuses at wave */
1.137 brouard 8284: cutv(stra, strb, line, ' ');
1.223 brouard 8285: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8286: lval=-1;
1.136 brouard 8287: }else{
1.238 brouard 8288: errno=0;
8289: lval=strtol(strb,&endptr,10);
8290: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8291: if( strb[0]=='\0' || (*endptr != '\0')){
8292: 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);
8293: 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);
8294: return 1;
8295: }
1.136 brouard 8296: }
1.225 brouard 8297:
1.136 brouard 8298: s[j][i]=lval;
1.225 brouard 8299:
1.223 brouard 8300: /* Date of Interview */
1.136 brouard 8301: strcpy(line,stra);
8302: cutv(stra, strb,line,' ');
1.169 brouard 8303: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8304: }
1.169 brouard 8305: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8306: month=99;
8307: year=9999;
1.136 brouard 8308: }else{
1.225 brouard 8309: 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);
8310: 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);
8311: return 1;
1.136 brouard 8312: }
8313: anint[j][i]= (double) year;
8314: mint[j][i]= (double)month;
8315: strcpy(line,stra);
1.223 brouard 8316: } /* End loop on waves */
1.225 brouard 8317:
1.223 brouard 8318: /* Date of death */
1.136 brouard 8319: cutv(stra, strb,line,' ');
1.169 brouard 8320: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8321: }
1.169 brouard 8322: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8323: month=99;
8324: year=9999;
8325: }else{
1.141 brouard 8326: 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 8327: 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);
8328: return 1;
1.136 brouard 8329: }
8330: andc[i]=(double) year;
8331: moisdc[i]=(double) month;
8332: strcpy(line,stra);
8333:
1.223 brouard 8334: /* Date of birth */
1.136 brouard 8335: cutv(stra, strb,line,' ');
1.169 brouard 8336: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8337: }
1.169 brouard 8338: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8339: month=99;
8340: year=9999;
8341: }else{
1.141 brouard 8342: 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);
8343: 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 8344: return 1;
1.136 brouard 8345: }
8346: if (year==9999) {
1.141 brouard 8347: 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);
8348: 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 8349: return 1;
8350:
1.136 brouard 8351: }
8352: annais[i]=(double)(year);
8353: moisnais[i]=(double)(month);
8354: strcpy(line,stra);
1.225 brouard 8355:
1.223 brouard 8356: /* Sample weight */
1.136 brouard 8357: cutv(stra, strb,line,' ');
8358: errno=0;
8359: dval=strtod(strb,&endptr);
8360: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8361: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8362: 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 8363: fflush(ficlog);
8364: return 1;
8365: }
8366: weight[i]=dval;
8367: strcpy(line,stra);
1.225 brouard 8368:
1.223 brouard 8369: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8370: cutv(stra, strb, line, ' ');
8371: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8372: lval=-1;
1.223 brouard 8373: }else{
1.225 brouard 8374: errno=0;
8375: /* what_kind_of_number(strb); */
8376: dval=strtod(strb,&endptr);
8377: /* if(strb != endptr && *endptr == '\0') */
8378: /* dval=dlval; */
8379: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8380: if( strb[0]=='\0' || (*endptr != '\0')){
8381: 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);
8382: 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);
8383: return 1;
8384: }
8385: coqvar[iv][i]=dval;
1.226 brouard 8386: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8387: }
8388: strcpy(line,stra);
8389: }/* end loop nqv */
1.136 brouard 8390:
1.223 brouard 8391: /* Covariate values */
1.136 brouard 8392: for (j=ncovcol;j>=1;j--){
8393: cutv(stra, strb,line,' ');
1.223 brouard 8394: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8395: lval=-1;
1.136 brouard 8396: }else{
1.225 brouard 8397: errno=0;
8398: lval=strtol(strb,&endptr,10);
8399: if( strb[0]=='\0' || (*endptr != '\0')){
8400: 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);
8401: 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);
8402: return 1;
8403: }
1.136 brouard 8404: }
8405: if(lval <-1 || lval >1){
1.225 brouard 8406: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8407: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8408: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8409: For example, for multinomial values like 1, 2 and 3,\n \
8410: build V1=0 V2=0 for the reference value (1),\n \
8411: V1=1 V2=0 for (2) \n \
1.136 brouard 8412: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8413: output of IMaCh is often meaningless.\n \
1.136 brouard 8414: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8415: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8416: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8417: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8418: For example, for multinomial values like 1, 2 and 3,\n \
8419: build V1=0 V2=0 for the reference value (1),\n \
8420: V1=1 V2=0 for (2) \n \
1.136 brouard 8421: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8422: output of IMaCh is often meaningless.\n \
1.136 brouard 8423: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8424: return 1;
1.136 brouard 8425: }
8426: covar[j][i]=(double)(lval);
8427: strcpy(line,stra);
8428: }
8429: lstra=strlen(stra);
1.225 brouard 8430:
1.136 brouard 8431: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8432: stratrunc = &(stra[lstra-9]);
8433: num[i]=atol(stratrunc);
8434: }
8435: else
8436: num[i]=atol(stra);
8437: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8438: 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;}*/
8439:
8440: i=i+1;
8441: } /* End loop reading data */
1.225 brouard 8442:
1.136 brouard 8443: *imax=i-1; /* Number of individuals */
8444: fclose(fic);
1.225 brouard 8445:
1.136 brouard 8446: return (0);
1.164 brouard 8447: /* endread: */
1.225 brouard 8448: printf("Exiting readdata: ");
8449: fclose(fic);
8450: return (1);
1.223 brouard 8451: }
1.126 brouard 8452:
1.234 brouard 8453: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8454: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8455: while (*p2 == ' ')
1.234 brouard 8456: p2++;
8457: /* while ((*p1++ = *p2++) !=0) */
8458: /* ; */
8459: /* do */
8460: /* while (*p2 == ' ') */
8461: /* p2++; */
8462: /* while (*p1++ == *p2++); */
8463: *stri=p2;
1.145 brouard 8464: }
8465:
1.235 brouard 8466: int decoderesult ( char resultline[], int nres)
1.230 brouard 8467: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8468: {
1.235 brouard 8469: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8470: char resultsav[MAXLINE];
1.234 brouard 8471: int resultmodel[MAXLINE];
8472: int modelresult[MAXLINE];
1.230 brouard 8473: char stra[80], strb[80], strc[80], strd[80],stre[80];
8474:
1.234 brouard 8475: removefirstspace(&resultline);
1.233 brouard 8476: printf("decoderesult:%s\n",resultline);
1.230 brouard 8477:
8478: if (strstr(resultline,"v") !=0){
8479: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8480: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8481: return 1;
8482: }
8483: trimbb(resultsav, resultline);
8484: if (strlen(resultsav) >1){
8485: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8486: }
1.253 brouard 8487: if(j == 0){ /* Resultline but no = */
8488: TKresult[nres]=0; /* Combination for the nresult and the model */
8489: return (0);
8490: }
8491:
1.234 brouard 8492: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8493: 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);
8494: 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);
8495: }
8496: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8497: if(nbocc(resultsav,'=') >1){
8498: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8499: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8500: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8501: }else
8502: cutl(strc,strd,resultsav,'=');
1.230 brouard 8503: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8504:
1.230 brouard 8505: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8506: Tvarsel[k]=atoi(strc);
8507: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8508: /* cptcovsel++; */
8509: if (nbocc(stra,'=') >0)
8510: strcpy(resultsav,stra); /* and analyzes it */
8511: }
1.235 brouard 8512: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8513: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8514: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8515: match=0;
1.236 brouard 8516: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8517: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8518: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8519: match=1;
8520: break;
8521: }
8522: }
8523: if(match == 0){
8524: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8525: }
8526: }
8527: }
1.235 brouard 8528: /* Checking for missing or useless values in comparison of current model needs */
8529: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8530: match=0;
1.235 brouard 8531: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8532: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8533: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8534: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8535: ++match;
8536: }
8537: }
8538: }
8539: if(match == 0){
8540: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8541: }else if(match > 1){
8542: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8543: }
8544: }
1.235 brouard 8545:
1.234 brouard 8546: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8547: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8548: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8549: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8550: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8551: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8552: /* 1 0 0 0 */
8553: /* 2 1 0 0 */
8554: /* 3 0 1 0 */
8555: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8556: /* 5 0 0 1 */
8557: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8558: /* 7 0 1 1 */
8559: /* 8 1 1 1 */
1.237 brouard 8560: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8561: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8562: /* V5*age V5 known which value for nres? */
8563: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8564: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8565: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8566: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8567: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8568: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8569: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8570: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8571: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8572: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8573: k4++;;
8574: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8575: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8576: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8577: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8578: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8579: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8580: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8581: k4q++;;
8582: }
8583: }
1.234 brouard 8584:
1.235 brouard 8585: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8586: return (0);
8587: }
1.235 brouard 8588:
1.230 brouard 8589: int decodemodel( char model[], int lastobs)
8590: /**< This routine decodes the model and returns:
1.224 brouard 8591: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8592: * - nagesqr = 1 if age*age in the model, otherwise 0.
8593: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8594: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8595: * - cptcovage number of covariates with age*products =2
8596: * - cptcovs number of simple covariates
8597: * - 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
8598: * which is a new column after the 9 (ncovcol) variables.
8599: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8600: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8601: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8602: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8603: */
1.136 brouard 8604: {
1.238 brouard 8605: int i, j, k, ks, v;
1.227 brouard 8606: int j1, k1, k2, k3, k4;
1.136 brouard 8607: char modelsav[80];
1.145 brouard 8608: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8609: char *strpt;
1.136 brouard 8610:
1.145 brouard 8611: /*removespace(model);*/
1.136 brouard 8612: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8613: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8614: if (strstr(model,"AGE") !=0){
1.192 brouard 8615: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8616: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8617: return 1;
8618: }
1.141 brouard 8619: if (strstr(model,"v") !=0){
8620: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8621: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8622: return 1;
8623: }
1.187 brouard 8624: strcpy(modelsav,model);
8625: if ((strpt=strstr(model,"age*age")) !=0){
8626: printf(" strpt=%s, model=%s\n",strpt, model);
8627: if(strpt != model){
1.234 brouard 8628: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8629: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8630: corresponding column of parameters.\n",model);
1.234 brouard 8631: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8632: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8633: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8634: return 1;
1.225 brouard 8635: }
1.187 brouard 8636: nagesqr=1;
8637: if (strstr(model,"+age*age") !=0)
1.234 brouard 8638: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8639: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8640: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8641: else
1.234 brouard 8642: substrchaine(modelsav, model, "age*age");
1.187 brouard 8643: }else
8644: nagesqr=0;
8645: if (strlen(modelsav) >1){
8646: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8647: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8648: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8649: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8650: * cst, age and age*age
8651: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8652: /* including age products which are counted in cptcovage.
8653: * but the covariates which are products must be treated
8654: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8655: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8656: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8657:
8658:
1.187 brouard 8659: /* Design
8660: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8661: * < ncovcol=8 >
8662: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8663: * k= 1 2 3 4 5 6 7 8
8664: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8665: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8666: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8667: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8668: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8669: * Tage[++cptcovage]=k
8670: * if products, new covar are created after ncovcol with k1
8671: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8672: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8673: * 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
8674: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8675: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8676: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8677: * < ncovcol=8 >
8678: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8679: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8680: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8681: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8682: * p Tprod[1]@2={ 6, 5}
8683: *p Tvard[1][1]@4= {7, 8, 5, 6}
8684: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8685: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8686: *How to reorganize?
8687: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8688: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8689: * {2, 1, 4, 8, 5, 6, 3, 7}
8690: * Struct []
8691: */
1.225 brouard 8692:
1.187 brouard 8693: /* This loop fills the array Tvar from the string 'model'.*/
8694: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8695: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8696: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8697: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8698: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8699: /* k=1 Tvar[1]=2 (from V2) */
8700: /* k=5 Tvar[5] */
8701: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8702: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8703: /* } */
1.198 brouard 8704: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8705: /*
8706: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8707: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8708: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8709: }
1.187 brouard 8710: cptcovage=0;
8711: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8712: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8713: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8714: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8715: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8716: /*scanf("%d",i);*/
8717: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8718: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8719: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8720: /* covar is not filled and then is empty */
8721: cptcovprod--;
8722: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8723: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8724: Typevar[k]=1; /* 1 for age product */
8725: cptcovage++; /* Sums the number of covariates which include age as a product */
8726: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8727: /*printf("stre=%s ", stre);*/
8728: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8729: cptcovprod--;
8730: cutl(stre,strb,strc,'V');
8731: Tvar[k]=atoi(stre);
8732: Typevar[k]=1; /* 1 for age product */
8733: cptcovage++;
8734: Tage[cptcovage]=k;
8735: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8736: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8737: cptcovn++;
8738: cptcovprodnoage++;k1++;
8739: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8740: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8741: because this model-covariate is a construction we invent a new column
8742: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8743: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8744: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8745: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8746: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8747: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8748: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8749: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8750: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8751: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8752: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8753: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8754: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8755: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8756: for (i=1; i<=lastobs;i++){
8757: /* Computes the new covariate which is a product of
8758: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8759: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8760: }
8761: } /* End age is not in the model */
8762: } /* End if model includes a product */
8763: else { /* no more sum */
8764: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8765: /* scanf("%d",i);*/
8766: cutl(strd,strc,strb,'V');
8767: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8768: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8769: Tvar[k]=atoi(strd);
8770: Typevar[k]=0; /* 0 for simple covariates */
8771: }
8772: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8773: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8774: scanf("%d",i);*/
1.187 brouard 8775: } /* end of loop + on total covariates */
8776: } /* end if strlen(modelsave == 0) age*age might exist */
8777: } /* end if strlen(model == 0) */
1.136 brouard 8778:
8779: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8780: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8781:
1.136 brouard 8782: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8783: printf("cptcovprod=%d ", cptcovprod);
8784: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8785: scanf("%d ",i);*/
8786:
8787:
1.230 brouard 8788: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8789: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8790: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8791: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8792: k = 1 2 3 4 5 6 7 8 9
8793: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8794: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8795: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8796: Dummy[k] 1 0 0 0 3 1 1 2 3
8797: Tmodelind[combination of covar]=k;
1.225 brouard 8798: */
8799: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8800: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8801: /* 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 8802: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8803: printf("Model=%s\n\
8804: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8805: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8806: 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);
8807: fprintf(ficlog,"Model=%s\n\
8808: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8809: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8810: 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 8811: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8812: 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 */
8813: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8814: Fixed[k]= 0;
8815: Dummy[k]= 0;
1.225 brouard 8816: ncoveff++;
1.232 brouard 8817: ncovf++;
1.234 brouard 8818: nsd++;
8819: modell[k].maintype= FTYPE;
8820: TvarsD[nsd]=Tvar[k];
8821: TvarsDind[nsd]=k;
8822: TvarF[ncovf]=Tvar[k];
8823: TvarFind[ncovf]=k;
8824: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8825: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8826: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8827: Fixed[k]= 0;
8828: Dummy[k]= 0;
8829: ncoveff++;
8830: ncovf++;
8831: modell[k].maintype= FTYPE;
8832: TvarF[ncovf]=Tvar[k];
8833: TvarFind[ncovf]=k;
1.230 brouard 8834: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8835: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8836: }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 8837: Fixed[k]= 0;
8838: Dummy[k]= 1;
1.230 brouard 8839: nqfveff++;
1.234 brouard 8840: modell[k].maintype= FTYPE;
8841: modell[k].subtype= FQ;
8842: nsq++;
8843: TvarsQ[nsq]=Tvar[k];
8844: TvarsQind[nsq]=k;
1.232 brouard 8845: ncovf++;
1.234 brouard 8846: TvarF[ncovf]=Tvar[k];
8847: TvarFind[ncovf]=k;
1.231 brouard 8848: 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 8849: 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 8850: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8851: Fixed[k]= 1;
8852: Dummy[k]= 0;
1.225 brouard 8853: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8854: modell[k].maintype= VTYPE;
8855: modell[k].subtype= VD;
8856: nsd++;
8857: TvarsD[nsd]=Tvar[k];
8858: TvarsDind[nsd]=k;
8859: ncovv++; /* Only simple time varying variables */
8860: TvarV[ncovv]=Tvar[k];
1.242 brouard 8861: 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 8862: 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 */
8863: 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 8864: 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);
8865: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8866: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8867: Fixed[k]= 1;
8868: Dummy[k]= 1;
8869: nqtveff++;
8870: modell[k].maintype= VTYPE;
8871: modell[k].subtype= VQ;
8872: ncovv++; /* Only simple time varying variables */
8873: nsq++;
8874: TvarsQ[nsq]=Tvar[k];
8875: TvarsQind[nsq]=k;
8876: TvarV[ncovv]=Tvar[k];
1.242 brouard 8877: 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 8878: 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 */
8879: 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 8880: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8881: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8882: 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 8883: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8884: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8885: ncova++;
8886: TvarA[ncova]=Tvar[k];
8887: TvarAind[ncova]=k;
1.231 brouard 8888: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8889: Fixed[k]= 2;
8890: Dummy[k]= 2;
8891: modell[k].maintype= ATYPE;
8892: modell[k].subtype= APFD;
8893: /* ncoveff++; */
1.227 brouard 8894: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8895: Fixed[k]= 2;
8896: Dummy[k]= 3;
8897: modell[k].maintype= ATYPE;
8898: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8899: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8900: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8901: Fixed[k]= 3;
8902: Dummy[k]= 2;
8903: modell[k].maintype= ATYPE;
8904: modell[k].subtype= APVD; /* Product age * varying dummy */
8905: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8906: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8907: Fixed[k]= 3;
8908: Dummy[k]= 3;
8909: modell[k].maintype= ATYPE;
8910: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8911: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8912: }
8913: }else if (Typevar[k] == 2) { /* product without age */
8914: k1=Tposprod[k];
8915: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8916: if(Tvard[k1][2] <=ncovcol){
8917: Fixed[k]= 1;
8918: Dummy[k]= 0;
8919: modell[k].maintype= FTYPE;
8920: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8921: ncovf++; /* Fixed variables without age */
8922: TvarF[ncovf]=Tvar[k];
8923: TvarFind[ncovf]=k;
8924: }else if(Tvard[k1][2] <=ncovcol+nqv){
8925: Fixed[k]= 0; /* or 2 ?*/
8926: Dummy[k]= 1;
8927: modell[k].maintype= FTYPE;
8928: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8929: ncovf++; /* Varying variables without age */
8930: TvarF[ncovf]=Tvar[k];
8931: TvarFind[ncovf]=k;
8932: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8933: Fixed[k]= 1;
8934: Dummy[k]= 0;
8935: modell[k].maintype= VTYPE;
8936: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8937: ncovv++; /* Varying variables without age */
8938: TvarV[ncovv]=Tvar[k];
8939: TvarVind[ncovv]=k;
8940: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8941: Fixed[k]= 1;
8942: Dummy[k]= 1;
8943: modell[k].maintype= VTYPE;
8944: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8945: ncovv++; /* Varying variables without age */
8946: TvarV[ncovv]=Tvar[k];
8947: TvarVind[ncovv]=k;
8948: }
1.227 brouard 8949: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8950: if(Tvard[k1][2] <=ncovcol){
8951: Fixed[k]= 0; /* or 2 ?*/
8952: Dummy[k]= 1;
8953: modell[k].maintype= FTYPE;
8954: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8955: ncovf++; /* Fixed variables without age */
8956: TvarF[ncovf]=Tvar[k];
8957: TvarFind[ncovf]=k;
8958: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8959: Fixed[k]= 1;
8960: Dummy[k]= 1;
8961: modell[k].maintype= VTYPE;
8962: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8963: ncovv++; /* Varying variables without age */
8964: TvarV[ncovv]=Tvar[k];
8965: TvarVind[ncovv]=k;
8966: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8967: Fixed[k]= 1;
8968: Dummy[k]= 1;
8969: modell[k].maintype= VTYPE;
8970: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8971: ncovv++; /* Varying variables without age */
8972: TvarV[ncovv]=Tvar[k];
8973: TvarVind[ncovv]=k;
8974: ncovv++; /* Varying variables without age */
8975: TvarV[ncovv]=Tvar[k];
8976: TvarVind[ncovv]=k;
8977: }
1.227 brouard 8978: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8979: if(Tvard[k1][2] <=ncovcol){
8980: Fixed[k]= 1;
8981: Dummy[k]= 1;
8982: modell[k].maintype= VTYPE;
8983: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8984: ncovv++; /* Varying variables without age */
8985: TvarV[ncovv]=Tvar[k];
8986: TvarVind[ncovv]=k;
8987: }else if(Tvard[k1][2] <=ncovcol+nqv){
8988: Fixed[k]= 1;
8989: Dummy[k]= 1;
8990: modell[k].maintype= VTYPE;
8991: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8992: ncovv++; /* Varying variables without age */
8993: TvarV[ncovv]=Tvar[k];
8994: TvarVind[ncovv]=k;
8995: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8996: Fixed[k]= 1;
8997: Dummy[k]= 0;
8998: modell[k].maintype= VTYPE;
8999: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9000: ncovv++; /* Varying variables without age */
9001: TvarV[ncovv]=Tvar[k];
9002: TvarVind[ncovv]=k;
9003: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9004: Fixed[k]= 1;
9005: Dummy[k]= 1;
9006: modell[k].maintype= VTYPE;
9007: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9008: ncovv++; /* Varying variables without age */
9009: TvarV[ncovv]=Tvar[k];
9010: TvarVind[ncovv]=k;
9011: }
1.227 brouard 9012: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9013: if(Tvard[k1][2] <=ncovcol){
9014: Fixed[k]= 1;
9015: Dummy[k]= 1;
9016: modell[k].maintype= VTYPE;
9017: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9018: ncovv++; /* Varying variables without age */
9019: TvarV[ncovv]=Tvar[k];
9020: TvarVind[ncovv]=k;
9021: }else if(Tvard[k1][2] <=ncovcol+nqv){
9022: Fixed[k]= 1;
9023: Dummy[k]= 1;
9024: modell[k].maintype= VTYPE;
9025: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9026: ncovv++; /* Varying variables without age */
9027: TvarV[ncovv]=Tvar[k];
9028: TvarVind[ncovv]=k;
9029: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9030: Fixed[k]= 1;
9031: Dummy[k]= 1;
9032: modell[k].maintype= VTYPE;
9033: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9034: ncovv++; /* Varying variables without age */
9035: TvarV[ncovv]=Tvar[k];
9036: TvarVind[ncovv]=k;
9037: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9038: Fixed[k]= 1;
9039: Dummy[k]= 1;
9040: modell[k].maintype= VTYPE;
9041: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9042: ncovv++; /* Varying variables without age */
9043: TvarV[ncovv]=Tvar[k];
9044: TvarVind[ncovv]=k;
9045: }
1.227 brouard 9046: }else{
1.240 brouard 9047: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9048: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9049: } /*end k1*/
1.225 brouard 9050: }else{
1.226 brouard 9051: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9052: 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 9053: }
1.227 brouard 9054: 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 9055: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9056: 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]);
9057: }
9058: /* Searching for doublons in the model */
9059: for(k1=1; k1<= cptcovt;k1++){
9060: for(k2=1; k2 <k1;k2++){
9061: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9062: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9063: if(Tvar[k1]==Tvar[k2]){
9064: 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]]);
9065: 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);
9066: return(1);
9067: }
9068: }else if (Typevar[k1] ==2){
9069: k3=Tposprod[k1];
9070: k4=Tposprod[k2];
9071: 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])) ){
9072: 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]]);
9073: 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);
9074: return(1);
9075: }
9076: }
1.227 brouard 9077: }
9078: }
1.225 brouard 9079: }
9080: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9081: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9082: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9083: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9084: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9085: /*endread:*/
1.225 brouard 9086: printf("Exiting decodemodel: ");
9087: return (1);
1.136 brouard 9088: }
9089:
1.169 brouard 9090: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9091: {/* Check ages at death */
1.136 brouard 9092: int i, m;
1.218 brouard 9093: int firstone=0;
9094:
1.136 brouard 9095: for (i=1; i<=imx; i++) {
9096: for(m=2; (m<= maxwav); m++) {
9097: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9098: anint[m][i]=9999;
1.216 brouard 9099: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9100: s[m][i]=-1;
1.136 brouard 9101: }
9102: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 9103: *nberr = *nberr + 1;
1.218 brouard 9104: if(firstone == 0){
9105: firstone=1;
9106: printf("Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\nOther similar cases in log file\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
9107: }
9108: fprintf(ficlog,"Error! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown, you must set an arbitrary year of death or he/she is skipped and results can be biased (%d) because status is a death state %d at wave %d. Wave dropped.\n",(int)moisdc[i],(int)andc[i],num[i],i, *nberr,s[m][i],m);
1.136 brouard 9109: s[m][i]=-1;
9110: }
9111: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9112: (*nberr)++;
1.136 brouard 9113: printf("Error! Month of death of individual %ld on line %d was unknown %2d, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,(int)moisdc[i]);
9114: fprintf(ficlog,"Error! Month of death of individual %ld on line %d was unknown %f, you should set it otherwise the information on the death is skipped and results are biased.\n",num[i],i,moisdc[i]);
9115: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
9116: }
9117: }
9118: }
9119:
9120: for (i=1; i<=imx; i++) {
9121: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9122: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9123: 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 9124: if (s[m][i] >= nlstate+1) {
1.169 brouard 9125: if(agedc[i]>0){
9126: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9127: agev[m][i]=agedc[i];
1.214 brouard 9128: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9129: }else {
1.136 brouard 9130: if ((int)andc[i]!=9999){
9131: nbwarn++;
9132: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9133: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9134: agev[m][i]=-1;
9135: }
9136: }
1.169 brouard 9137: } /* agedc > 0 */
1.214 brouard 9138: } /* end if */
1.136 brouard 9139: else if(s[m][i] !=9){ /* Standard case, age in fractional
9140: years but with the precision of a month */
9141: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9142: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9143: agev[m][i]=1;
9144: else if(agev[m][i] < *agemin){
9145: *agemin=agev[m][i];
9146: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9147: }
9148: else if(agev[m][i] >*agemax){
9149: *agemax=agev[m][i];
1.156 brouard 9150: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9151: }
9152: /*agev[m][i]=anint[m][i]-annais[i];*/
9153: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9154: } /* en if 9*/
1.136 brouard 9155: else { /* =9 */
1.214 brouard 9156: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9157: agev[m][i]=1;
9158: s[m][i]=-1;
9159: }
9160: }
1.214 brouard 9161: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9162: agev[m][i]=1;
1.214 brouard 9163: else{
9164: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9165: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9166: agev[m][i]=0;
9167: }
9168: } /* End for lastpass */
9169: }
1.136 brouard 9170:
9171: for (i=1; i<=imx; i++) {
9172: for(m=firstpass; (m<=lastpass); m++){
9173: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9174: (*nberr)++;
1.136 brouard 9175: 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);
9176: 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);
9177: return 1;
9178: }
9179: }
9180: }
9181:
9182: /*for (i=1; i<=imx; i++){
9183: for (m=firstpass; (m<lastpass); m++){
9184: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9185: }
9186:
9187: }*/
9188:
9189:
1.139 brouard 9190: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9191: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9192:
9193: return (0);
1.164 brouard 9194: /* endread:*/
1.136 brouard 9195: printf("Exiting calandcheckages: ");
9196: return (1);
9197: }
9198:
1.172 brouard 9199: #if defined(_MSC_VER)
9200: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9201: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9202: //#include "stdafx.h"
9203: //#include <stdio.h>
9204: //#include <tchar.h>
9205: //#include <windows.h>
9206: //#include <iostream>
9207: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9208:
9209: LPFN_ISWOW64PROCESS fnIsWow64Process;
9210:
9211: BOOL IsWow64()
9212: {
9213: BOOL bIsWow64 = FALSE;
9214:
9215: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9216: // (HANDLE, PBOOL);
9217:
9218: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9219:
9220: HMODULE module = GetModuleHandle(_T("kernel32"));
9221: const char funcName[] = "IsWow64Process";
9222: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9223: GetProcAddress(module, funcName);
9224:
9225: if (NULL != fnIsWow64Process)
9226: {
9227: if (!fnIsWow64Process(GetCurrentProcess(),
9228: &bIsWow64))
9229: //throw std::exception("Unknown error");
9230: printf("Unknown error\n");
9231: }
9232: return bIsWow64 != FALSE;
9233: }
9234: #endif
1.177 brouard 9235:
1.191 brouard 9236: void syscompilerinfo(int logged)
1.167 brouard 9237: {
9238: /* #include "syscompilerinfo.h"*/
1.185 brouard 9239: /* command line Intel compiler 32bit windows, XP compatible:*/
9240: /* /GS /W3 /Gy
9241: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9242: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9243: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9244: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9245: */
9246: /* 64 bits */
1.185 brouard 9247: /*
9248: /GS /W3 /Gy
9249: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9250: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9251: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9252: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9253: /* Optimization are useless and O3 is slower than O2 */
9254: /*
9255: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9256: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9257: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9258: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9259: */
1.186 brouard 9260: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9261: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9262: /PDB:"visual studio
9263: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9264: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9265: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9266: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9267: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9268: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9269: uiAccess='false'"
9270: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9271: /NOLOGO /TLBID:1
9272: */
1.177 brouard 9273: #if defined __INTEL_COMPILER
1.178 brouard 9274: #if defined(__GNUC__)
9275: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9276: #endif
1.177 brouard 9277: #elif defined(__GNUC__)
1.179 brouard 9278: #ifndef __APPLE__
1.174 brouard 9279: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9280: #endif
1.177 brouard 9281: struct utsname sysInfo;
1.178 brouard 9282: int cross = CROSS;
9283: if (cross){
9284: printf("Cross-");
1.191 brouard 9285: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9286: }
1.174 brouard 9287: #endif
9288:
1.171 brouard 9289: #include <stdint.h>
1.178 brouard 9290:
1.191 brouard 9291: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9292: #if defined(__clang__)
1.191 brouard 9293: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9294: #endif
9295: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9296: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9297: #endif
9298: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9299: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9300: #endif
9301: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9302: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9303: #endif
9304: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9305: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9306: #endif
9307: #if defined(_MSC_VER)
1.191 brouard 9308: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9309: #endif
9310: #if defined(__PGI)
1.191 brouard 9311: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9312: #endif
9313: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9314: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9315: #endif
1.191 brouard 9316: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9317:
1.167 brouard 9318: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9319: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9320: // Windows (x64 and x86)
1.191 brouard 9321: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9322: #elif __unix__ // all unices, not all compilers
9323: // Unix
1.191 brouard 9324: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9325: #elif __linux__
9326: // linux
1.191 brouard 9327: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9328: #elif __APPLE__
1.174 brouard 9329: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9330: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9331: #endif
9332:
9333: /* __MINGW32__ */
9334: /* __CYGWIN__ */
9335: /* __MINGW64__ */
9336: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9337: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9338: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9339: /* _WIN64 // Defined for applications for Win64. */
9340: /* _M_X64 // Defined for compilations that target x64 processors. */
9341: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9342:
1.167 brouard 9343: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9344: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9345: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9346: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9347: #else
1.191 brouard 9348: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9349: #endif
9350:
1.169 brouard 9351: #if defined(__GNUC__)
9352: # if defined(__GNUC_PATCHLEVEL__)
9353: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9354: + __GNUC_MINOR__ * 100 \
9355: + __GNUC_PATCHLEVEL__)
9356: # else
9357: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9358: + __GNUC_MINOR__ * 100)
9359: # endif
1.174 brouard 9360: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9361: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9362:
9363: if (uname(&sysInfo) != -1) {
9364: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9365: 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 9366: }
9367: else
9368: perror("uname() error");
1.179 brouard 9369: //#ifndef __INTEL_COMPILER
9370: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9371: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9372: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9373: #endif
1.169 brouard 9374: #endif
1.172 brouard 9375:
9376: // void main()
9377: // {
1.169 brouard 9378: #if defined(_MSC_VER)
1.174 brouard 9379: if (IsWow64()){
1.191 brouard 9380: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9381: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9382: }
9383: else{
1.191 brouard 9384: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9385: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9386: }
1.172 brouard 9387: // printf("\nPress Enter to continue...");
9388: // getchar();
9389: // }
9390:
1.169 brouard 9391: #endif
9392:
1.167 brouard 9393:
1.219 brouard 9394: }
1.136 brouard 9395:
1.219 brouard 9396: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9397: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9398: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9399: /* double ftolpl = 1.e-10; */
1.180 brouard 9400: double age, agebase, agelim;
1.203 brouard 9401: double tot;
1.180 brouard 9402:
1.202 brouard 9403: strcpy(filerespl,"PL_");
9404: strcat(filerespl,fileresu);
9405: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9406: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9407: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9408: }
1.227 brouard 9409: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9410: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9411: pstamp(ficrespl);
1.203 brouard 9412: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9413: fprintf(ficrespl,"#Age ");
9414: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9415: fprintf(ficrespl,"\n");
1.180 brouard 9416:
1.219 brouard 9417: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9418:
1.219 brouard 9419: agebase=ageminpar;
9420: agelim=agemaxpar;
1.180 brouard 9421:
1.227 brouard 9422: /* i1=pow(2,ncoveff); */
1.234 brouard 9423: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9424: if (cptcovn < 1){i1=1;}
1.180 brouard 9425:
1.238 brouard 9426: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9427: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9428: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9429: continue;
1.235 brouard 9430:
1.238 brouard 9431: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9432: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9433: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9434: /* k=k+1; */
9435: /* to clean */
9436: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9437: fprintf(ficrespl,"#******");
9438: printf("#******");
9439: fprintf(ficlog,"#******");
9440: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9441: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9442: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9443: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9444: }
9445: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9446: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9447: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9448: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9449: }
9450: fprintf(ficrespl,"******\n");
9451: printf("******\n");
9452: fprintf(ficlog,"******\n");
9453: if(invalidvarcomb[k]){
9454: printf("\nCombination (%d) ignored because no case \n",k);
9455: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9456: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9457: continue;
9458: }
1.219 brouard 9459:
1.238 brouard 9460: fprintf(ficrespl,"#Age ");
9461: for(j=1;j<=cptcoveff;j++) {
9462: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9463: }
9464: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9465: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9466:
1.238 brouard 9467: for (age=agebase; age<=agelim; age++){
9468: /* for (age=agebase; age<=agebase; age++){ */
9469: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9470: fprintf(ficrespl,"%.0f ",age );
9471: for(j=1;j<=cptcoveff;j++)
9472: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9473: tot=0.;
9474: for(i=1; i<=nlstate;i++){
9475: tot += prlim[i][i];
9476: fprintf(ficrespl," %.5f", prlim[i][i]);
9477: }
9478: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9479: } /* Age */
9480: /* was end of cptcod */
9481: } /* cptcov */
9482: } /* nres */
1.219 brouard 9483: return 0;
1.180 brouard 9484: }
9485:
1.218 brouard 9486: 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){
9487: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9488:
9489: /* Computes the back prevalence limit for any combination of covariate values
9490: * at any age between ageminpar and agemaxpar
9491: */
1.235 brouard 9492: int i, j, k, i1, nres=0 ;
1.217 brouard 9493: /* double ftolpl = 1.e-10; */
9494: double age, agebase, agelim;
9495: double tot;
1.218 brouard 9496: /* double ***mobaverage; */
9497: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9498:
9499: strcpy(fileresplb,"PLB_");
9500: strcat(fileresplb,fileresu);
9501: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9502: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9503: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9504: }
9505: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9506: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9507: pstamp(ficresplb);
9508: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9509: fprintf(ficresplb,"#Age ");
9510: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9511: fprintf(ficresplb,"\n");
9512:
1.218 brouard 9513:
9514: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9515:
9516: agebase=ageminpar;
9517: agelim=agemaxpar;
9518:
9519:
1.227 brouard 9520: i1=pow(2,cptcoveff);
1.218 brouard 9521: if (cptcovn < 1){i1=1;}
1.227 brouard 9522:
1.238 brouard 9523: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9524: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9525: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9526: continue;
9527: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9528: fprintf(ficresplb,"#******");
9529: printf("#******");
9530: fprintf(ficlog,"#******");
9531: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9532: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9533: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9534: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9535: }
9536: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9537: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9538: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9539: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9540: }
9541: fprintf(ficresplb,"******\n");
9542: printf("******\n");
9543: fprintf(ficlog,"******\n");
9544: if(invalidvarcomb[k]){
9545: printf("\nCombination (%d) ignored because no cases \n",k);
9546: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9547: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9548: continue;
9549: }
1.218 brouard 9550:
1.238 brouard 9551: fprintf(ficresplb,"#Age ");
9552: for(j=1;j<=cptcoveff;j++) {
9553: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9554: }
9555: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9556: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9557:
9558:
1.238 brouard 9559: for (age=agebase; age<=agelim; age++){
9560: /* for (age=agebase; age<=agebase; age++){ */
9561: if(mobilavproj > 0){
9562: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9563: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9564: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9565: }else if (mobilavproj == 0){
9566: 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);
9567: 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);
9568: exit(1);
9569: }else{
9570: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9571: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9572: }
9573: fprintf(ficresplb,"%.0f ",age );
9574: for(j=1;j<=cptcoveff;j++)
9575: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9576: tot=0.;
9577: for(i=1; i<=nlstate;i++){
9578: tot += bprlim[i][i];
9579: fprintf(ficresplb," %.5f", bprlim[i][i]);
9580: }
9581: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9582: } /* Age */
9583: /* was end of cptcod */
1.255 brouard 9584: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9585: } /* end of any combination */
9586: } /* end of nres */
1.218 brouard 9587: /* hBijx(p, bage, fage); */
9588: /* fclose(ficrespijb); */
9589:
9590: return 0;
1.217 brouard 9591: }
1.218 brouard 9592:
1.180 brouard 9593: int hPijx(double *p, int bage, int fage){
9594: /*------------- h Pij x at various ages ------------*/
9595:
9596: int stepsize;
9597: int agelim;
9598: int hstepm;
9599: int nhstepm;
1.235 brouard 9600: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9601:
9602: double agedeb;
9603: double ***p3mat;
9604:
1.201 brouard 9605: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9606: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9607: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9608: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9609: }
9610: printf("Computing pij: result on file '%s' \n", filerespij);
9611: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9612:
9613: stepsize=(int) (stepm+YEARM-1)/YEARM;
9614: /*if (stepm<=24) stepsize=2;*/
9615:
9616: agelim=AGESUP;
9617: hstepm=stepsize*YEARM; /* Every year of age */
9618: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9619:
1.180 brouard 9620: /* hstepm=1; aff par mois*/
9621: pstamp(ficrespij);
9622: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9623: i1= pow(2,cptcoveff);
1.218 brouard 9624: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9625: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9626: /* k=k+1; */
1.235 brouard 9627: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9628: for(k=1; k<=i1;k++){
1.253 brouard 9629: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9630: continue;
1.183 brouard 9631: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9632: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9633: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9634: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9635: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9636: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9637: }
1.183 brouard 9638: fprintf(ficrespij,"******\n");
9639:
9640: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9641: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9642: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9643:
9644: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9645:
1.183 brouard 9646: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9647: oldm=oldms;savm=savms;
1.235 brouard 9648: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9649: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9650: for(i=1; i<=nlstate;i++)
9651: for(j=1; j<=nlstate+ndeath;j++)
9652: fprintf(ficrespij," %1d-%1d",i,j);
9653: fprintf(ficrespij,"\n");
9654: for (h=0; h<=nhstepm; h++){
9655: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9656: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9657: for(i=1; i<=nlstate;i++)
9658: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9659: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9660: fprintf(ficrespij,"\n");
9661: }
1.183 brouard 9662: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9663: fprintf(ficrespij,"\n");
9664: }
1.180 brouard 9665: /*}*/
9666: }
1.218 brouard 9667: return 0;
1.180 brouard 9668: }
1.218 brouard 9669:
9670: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9671: /*------------- h Bij x at various ages ------------*/
9672:
9673: int stepsize;
1.218 brouard 9674: /* int agelim; */
9675: int ageminl;
1.217 brouard 9676: int hstepm;
9677: int nhstepm;
1.238 brouard 9678: int h, i, i1, j, k, nres;
1.218 brouard 9679:
1.217 brouard 9680: double agedeb;
9681: double ***p3mat;
1.218 brouard 9682:
9683: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9684: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9685: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9686: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9687: }
9688: printf("Computing pij back: result on file '%s' \n", filerespijb);
9689: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9690:
9691: stepsize=(int) (stepm+YEARM-1)/YEARM;
9692: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9693:
1.218 brouard 9694: /* agelim=AGESUP; */
9695: ageminl=30;
9696: hstepm=stepsize*YEARM; /* Every year of age */
9697: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9698:
9699: /* hstepm=1; aff par mois*/
9700: pstamp(ficrespijb);
1.255 brouard 9701: 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 9702: i1= pow(2,cptcoveff);
1.218 brouard 9703: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9704: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9705: /* k=k+1; */
1.238 brouard 9706: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9707: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9708: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9709: continue;
9710: fprintf(ficrespijb,"\n#****** ");
9711: for(j=1;j<=cptcoveff;j++)
9712: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9713: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9714: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9715: }
9716: fprintf(ficrespijb,"******\n");
9717: if(invalidvarcomb[k]){
9718: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9719: continue;
9720: }
9721:
9722: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9723: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9724: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9725: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9726: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9727:
9728: /* nhstepm=nhstepm*YEARM; aff par mois*/
9729:
9730: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9731: /* oldm=oldms;savm=savms; */
9732: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9733: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9734: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 9735: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 9736: for(i=1; i<=nlstate;i++)
9737: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9738: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9739: fprintf(ficrespijb,"\n");
1.238 brouard 9740: for (h=0; h<=nhstepm; h++){
9741: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9742: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9743: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9744: for(i=1; i<=nlstate;i++)
9745: for(j=1; j<=nlstate+ndeath;j++)
9746: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9747: fprintf(ficrespijb,"\n");
9748: }
9749: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9750: fprintf(ficrespijb,"\n");
9751: } /* end age deb */
9752: } /* end combination */
9753: } /* end nres */
1.218 brouard 9754: return 0;
9755: } /* hBijx */
1.217 brouard 9756:
1.180 brouard 9757:
1.136 brouard 9758: /***********************************************/
9759: /**************** Main Program *****************/
9760: /***********************************************/
9761:
9762: int main(int argc, char *argv[])
9763: {
9764: #ifdef GSL
9765: const gsl_multimin_fminimizer_type *T;
9766: size_t iteri = 0, it;
9767: int rval = GSL_CONTINUE;
9768: int status = GSL_SUCCESS;
9769: double ssval;
9770: #endif
9771: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9772: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9773: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9774: int jj, ll, li, lj, lk;
1.136 brouard 9775: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9776: int num_filled;
1.136 brouard 9777: int itimes;
9778: int NDIM=2;
9779: int vpopbased=0;
1.235 brouard 9780: int nres=0;
1.258 ! brouard 9781: int endishere=0;
1.136 brouard 9782:
1.164 brouard 9783: char ca[32], cb[32];
1.136 brouard 9784: /* FILE *fichtm; *//* Html File */
9785: /* FILE *ficgp;*/ /*Gnuplot File */
9786: struct stat info;
1.191 brouard 9787: double agedeb=0.;
1.194 brouard 9788:
9789: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9790: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9791:
1.165 brouard 9792: double fret;
1.191 brouard 9793: double dum=0.; /* Dummy variable */
1.136 brouard 9794: double ***p3mat;
1.218 brouard 9795: /* double ***mobaverage; */
1.164 brouard 9796:
9797: char line[MAXLINE];
1.197 brouard 9798: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9799:
1.234 brouard 9800: char modeltemp[MAXLINE];
1.230 brouard 9801: char resultline[MAXLINE];
9802:
1.136 brouard 9803: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9804: char *tok, *val; /* pathtot */
1.136 brouard 9805: int firstobs=1, lastobs=10;
1.195 brouard 9806: int c, h , cpt, c2;
1.191 brouard 9807: int jl=0;
9808: int i1, j1, jk, stepsize=0;
1.194 brouard 9809: int count=0;
9810:
1.164 brouard 9811: int *tab;
1.136 brouard 9812: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9813: int backcast=0;
1.136 brouard 9814: int mobilav=0,popforecast=0;
1.191 brouard 9815: int hstepm=0, nhstepm=0;
1.136 brouard 9816: int agemortsup;
9817: float sumlpop=0.;
9818: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9819: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9820:
1.191 brouard 9821: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9822: double ftolpl=FTOL;
9823: double **prlim;
1.217 brouard 9824: double **bprlim;
1.136 brouard 9825: double ***param; /* Matrix of parameters */
1.251 brouard 9826: double ***paramstart; /* Matrix of starting parameter values */
9827: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9828: double **matcov; /* Matrix of covariance */
1.203 brouard 9829: double **hess; /* Hessian matrix */
1.136 brouard 9830: double ***delti3; /* Scale */
9831: double *delti; /* Scale */
9832: double ***eij, ***vareij;
9833: double **varpl; /* Variances of prevalence limits by age */
9834: double *epj, vepp;
1.164 brouard 9835:
1.136 brouard 9836: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9837: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9838:
1.136 brouard 9839: double **ximort;
1.145 brouard 9840: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9841: int *dcwave;
9842:
1.164 brouard 9843: char z[1]="c";
1.136 brouard 9844:
9845: /*char *strt;*/
9846: char strtend[80];
1.126 brouard 9847:
1.164 brouard 9848:
1.126 brouard 9849: /* setlocale (LC_ALL, ""); */
9850: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9851: /* textdomain (PACKAGE); */
9852: /* setlocale (LC_CTYPE, ""); */
9853: /* setlocale (LC_MESSAGES, ""); */
9854:
9855: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9856: rstart_time = time(NULL);
9857: /* (void) gettimeofday(&start_time,&tzp);*/
9858: start_time = *localtime(&rstart_time);
1.126 brouard 9859: curr_time=start_time;
1.157 brouard 9860: /*tml = *localtime(&start_time.tm_sec);*/
9861: /* strcpy(strstart,asctime(&tml)); */
9862: strcpy(strstart,asctime(&start_time));
1.126 brouard 9863:
9864: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9865: /* tp.tm_sec = tp.tm_sec +86400; */
9866: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9867: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9868: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9869: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9870: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9871: /* strt=asctime(&tmg); */
9872: /* printf("Time(after) =%s",strstart); */
9873: /* (void) time (&time_value);
9874: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9875: * tm = *localtime(&time_value);
9876: * strstart=asctime(&tm);
9877: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9878: */
9879:
9880: nberr=0; /* Number of errors and warnings */
9881: nbwarn=0;
1.184 brouard 9882: #ifdef WIN32
9883: _getcwd(pathcd, size);
9884: #else
1.126 brouard 9885: getcwd(pathcd, size);
1.184 brouard 9886: #endif
1.191 brouard 9887: syscompilerinfo(0);
1.196 brouard 9888: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9889: if(argc <=1){
9890: printf("\nEnter the parameter file name: ");
1.205 brouard 9891: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9892: printf("ERROR Empty parameter file name\n");
9893: goto end;
9894: }
1.126 brouard 9895: i=strlen(pathr);
9896: if(pathr[i-1]=='\n')
9897: pathr[i-1]='\0';
1.156 brouard 9898: i=strlen(pathr);
1.205 brouard 9899: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9900: pathr[i-1]='\0';
1.205 brouard 9901: }
9902: i=strlen(pathr);
9903: if( i==0 ){
9904: printf("ERROR Empty parameter file name\n");
9905: goto end;
9906: }
9907: for (tok = pathr; tok != NULL; ){
1.126 brouard 9908: printf("Pathr |%s|\n",pathr);
9909: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9910: printf("val= |%s| pathr=%s\n",val,pathr);
9911: strcpy (pathtot, val);
9912: if(pathr[0] == '\0') break; /* Dirty */
9913: }
9914: }
9915: else{
9916: strcpy(pathtot,argv[1]);
9917: }
9918: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9919: /*cygwin_split_path(pathtot,path,optionfile);
9920: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9921: /* cutv(path,optionfile,pathtot,'\\');*/
9922:
9923: /* Split argv[0], imach program to get pathimach */
9924: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9925: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9926: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9927: /* strcpy(pathimach,argv[0]); */
9928: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9929: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9930: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9931: #ifdef WIN32
9932: _chdir(path); /* Can be a relative path */
9933: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9934: #else
1.126 brouard 9935: chdir(path); /* Can be a relative path */
1.184 brouard 9936: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9937: #endif
9938: printf("Current directory %s!\n",pathcd);
1.126 brouard 9939: strcpy(command,"mkdir ");
9940: strcat(command,optionfilefiname);
9941: if((outcmd=system(command)) != 0){
1.169 brouard 9942: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9943: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9944: /* fclose(ficlog); */
9945: /* exit(1); */
9946: }
9947: /* if((imk=mkdir(optionfilefiname))<0){ */
9948: /* perror("mkdir"); */
9949: /* } */
9950:
9951: /*-------- arguments in the command line --------*/
9952:
1.186 brouard 9953: /* Main Log file */
1.126 brouard 9954: strcat(filelog, optionfilefiname);
9955: strcat(filelog,".log"); /* */
9956: if((ficlog=fopen(filelog,"w"))==NULL) {
9957: printf("Problem with logfile %s\n",filelog);
9958: goto end;
9959: }
9960: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9961: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9962: fprintf(ficlog,"\nEnter the parameter file name: \n");
9963: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9964: path=%s \n\
9965: optionfile=%s\n\
9966: optionfilext=%s\n\
1.156 brouard 9967: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9968:
1.197 brouard 9969: syscompilerinfo(1);
1.167 brouard 9970:
1.126 brouard 9971: printf("Local time (at start):%s",strstart);
9972: fprintf(ficlog,"Local time (at start): %s",strstart);
9973: fflush(ficlog);
9974: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9975: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9976:
9977: /* */
9978: strcpy(fileres,"r");
9979: strcat(fileres, optionfilefiname);
1.201 brouard 9980: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9981: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9982: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9983:
1.186 brouard 9984: /* Main ---------arguments file --------*/
1.126 brouard 9985:
9986: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9987: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9988: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9989: fflush(ficlog);
1.149 brouard 9990: /* goto end; */
9991: exit(70);
1.126 brouard 9992: }
9993:
9994:
9995:
9996: strcpy(filereso,"o");
1.201 brouard 9997: strcat(filereso,fileresu);
1.126 brouard 9998: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9999: printf("Problem with Output resultfile: %s\n", filereso);
10000: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10001: fflush(ficlog);
10002: goto end;
10003: }
10004:
10005: /* Reads comments: lines beginning with '#' */
10006: numlinepar=0;
1.197 brouard 10007:
10008: /* First parameter line */
10009: while(fgets(line, MAXLINE, ficpar)) {
10010: /* If line starts with a # it is a comment */
10011: if (line[0] == '#') {
10012: numlinepar++;
10013: fputs(line,stdout);
10014: fputs(line,ficparo);
10015: fputs(line,ficlog);
10016: continue;
10017: }else
10018: break;
10019: }
10020: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10021: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10022: if (num_filled != 5) {
10023: printf("Should be 5 parameters\n");
10024: }
1.126 brouard 10025: numlinepar++;
1.197 brouard 10026: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10027: }
10028: /* Second parameter line */
10029: while(fgets(line, MAXLINE, ficpar)) {
10030: /* If line starts with a # it is a comment */
10031: if (line[0] == '#') {
10032: numlinepar++;
10033: fputs(line,stdout);
10034: fputs(line,ficparo);
10035: fputs(line,ficlog);
10036: continue;
10037: }else
10038: break;
10039: }
1.223 brouard 10040: 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", \
10041: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10042: if (num_filled != 11) {
10043: 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 10044: printf("but line=%s\n",line);
1.197 brouard 10045: }
1.223 brouard 10046: 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 10047: }
1.203 brouard 10048: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10049: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10050: /* Third parameter line */
10051: while(fgets(line, MAXLINE, ficpar)) {
10052: /* If line starts with a # it is a comment */
10053: if (line[0] == '#') {
10054: numlinepar++;
10055: fputs(line,stdout);
10056: fputs(line,ficparo);
10057: fputs(line,ficlog);
10058: continue;
10059: }else
10060: break;
10061: }
1.201 brouard 10062: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10063: if (num_filled == 0)
10064: model[0]='\0';
10065: else if (num_filled != 1){
1.197 brouard 10066: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10067: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10068: model[0]='\0';
10069: goto end;
10070: }
10071: else{
10072: if (model[0]=='+'){
10073: for(i=1; i<=strlen(model);i++)
10074: modeltemp[i-1]=model[i];
1.201 brouard 10075: strcpy(model,modeltemp);
1.197 brouard 10076: }
10077: }
1.199 brouard 10078: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10079: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10080: }
10081: /* 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); */
10082: /* numlinepar=numlinepar+3; /\* In general *\/ */
10083: /* 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 10084: 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);
10085: 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 10086: fflush(ficlog);
1.190 brouard 10087: /* if(model[0]=='#'|| model[0]== '\0'){ */
10088: if(model[0]=='#'){
1.187 brouard 10089: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10090: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10091: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10092: if(mle != -1){
10093: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10094: exit(1);
10095: }
10096: }
1.126 brouard 10097: while((c=getc(ficpar))=='#' && c!= EOF){
10098: ungetc(c,ficpar);
10099: fgets(line, MAXLINE, ficpar);
10100: numlinepar++;
1.195 brouard 10101: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10102: z[0]=line[1];
10103: }
10104: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10105: fputs(line, stdout);
10106: //puts(line);
1.126 brouard 10107: fputs(line,ficparo);
10108: fputs(line,ficlog);
10109: }
10110: ungetc(c,ficpar);
10111:
10112:
1.145 brouard 10113: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10114: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10115: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10116: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10117: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10118: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10119: v1+v2*age+v2*v3 makes cptcovn = 3
10120: */
10121: if (strlen(model)>1)
1.187 brouard 10122: 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 10123: else
1.187 brouard 10124: ncovmodel=2; /* Constant and age */
1.133 brouard 10125: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10126: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10127: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10128: 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);
10129: 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);
10130: fflush(stdout);
10131: fclose (ficlog);
10132: goto end;
10133: }
1.126 brouard 10134: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10135: delti=delti3[1][1];
10136: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10137: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10138: /* We could also provide initial parameters values giving by simple logistic regression
10139: * only one way, that is without matrix product. We will have nlstate maximizations */
10140: /* for(i=1;i<nlstate;i++){ */
10141: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10142: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10143: /* } */
1.126 brouard 10144: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10145: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10146: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10147: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10148: fclose (ficparo);
10149: fclose (ficlog);
10150: goto end;
10151: exit(0);
1.220 brouard 10152: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10153: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10154: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10155: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10156: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10157: matcov=matrix(1,npar,1,npar);
1.203 brouard 10158: hess=matrix(1,npar,1,npar);
1.220 brouard 10159: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10160: /* Read guessed parameters */
1.126 brouard 10161: /* Reads comments: lines beginning with '#' */
10162: while((c=getc(ficpar))=='#' && c!= EOF){
10163: ungetc(c,ficpar);
10164: fgets(line, MAXLINE, ficpar);
10165: numlinepar++;
1.141 brouard 10166: fputs(line,stdout);
1.126 brouard 10167: fputs(line,ficparo);
10168: fputs(line,ficlog);
10169: }
10170: ungetc(c,ficpar);
10171:
10172: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10173: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10174: for(i=1; i <=nlstate; i++){
1.234 brouard 10175: j=0;
1.126 brouard 10176: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10177: if(jj==i) continue;
10178: j++;
10179: fscanf(ficpar,"%1d%1d",&i1,&j1);
10180: if ((i1 != i) || (j1 != jj)){
10181: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10182: It might be a problem of design; if ncovcol and the model are correct\n \
10183: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10184: exit(1);
10185: }
10186: fprintf(ficparo,"%1d%1d",i1,j1);
10187: if(mle==1)
10188: printf("%1d%1d",i,jj);
10189: fprintf(ficlog,"%1d%1d",i,jj);
10190: for(k=1; k<=ncovmodel;k++){
10191: fscanf(ficpar," %lf",¶m[i][j][k]);
10192: if(mle==1){
10193: printf(" %lf",param[i][j][k]);
10194: fprintf(ficlog," %lf",param[i][j][k]);
10195: }
10196: else
10197: fprintf(ficlog," %lf",param[i][j][k]);
10198: fprintf(ficparo," %lf",param[i][j][k]);
10199: }
10200: fscanf(ficpar,"\n");
10201: numlinepar++;
10202: if(mle==1)
10203: printf("\n");
10204: fprintf(ficlog,"\n");
10205: fprintf(ficparo,"\n");
1.126 brouard 10206: }
10207: }
10208: fflush(ficlog);
1.234 brouard 10209:
1.251 brouard 10210: /* Reads parameters values */
1.126 brouard 10211: p=param[1][1];
1.251 brouard 10212: pstart=paramstart[1][1];
1.126 brouard 10213:
10214: /* Reads comments: lines beginning with '#' */
10215: while((c=getc(ficpar))=='#' && c!= EOF){
10216: ungetc(c,ficpar);
10217: fgets(line, MAXLINE, ficpar);
10218: numlinepar++;
1.141 brouard 10219: fputs(line,stdout);
1.126 brouard 10220: fputs(line,ficparo);
10221: fputs(line,ficlog);
10222: }
10223: ungetc(c,ficpar);
10224:
10225: for(i=1; i <=nlstate; i++){
10226: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10227: fscanf(ficpar,"%1d%1d",&i1,&j1);
10228: if ( (i1-i) * (j1-j) != 0){
10229: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10230: exit(1);
10231: }
10232: printf("%1d%1d",i,j);
10233: fprintf(ficparo,"%1d%1d",i1,j1);
10234: fprintf(ficlog,"%1d%1d",i1,j1);
10235: for(k=1; k<=ncovmodel;k++){
10236: fscanf(ficpar,"%le",&delti3[i][j][k]);
10237: printf(" %le",delti3[i][j][k]);
10238: fprintf(ficparo," %le",delti3[i][j][k]);
10239: fprintf(ficlog," %le",delti3[i][j][k]);
10240: }
10241: fscanf(ficpar,"\n");
10242: numlinepar++;
10243: printf("\n");
10244: fprintf(ficparo,"\n");
10245: fprintf(ficlog,"\n");
1.126 brouard 10246: }
10247: }
10248: fflush(ficlog);
1.234 brouard 10249:
1.145 brouard 10250: /* Reads covariance matrix */
1.126 brouard 10251: delti=delti3[1][1];
1.220 brouard 10252:
10253:
1.126 brouard 10254: /* 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 10255:
1.126 brouard 10256: /* Reads comments: lines beginning with '#' */
10257: while((c=getc(ficpar))=='#' && c!= EOF){
10258: ungetc(c,ficpar);
10259: fgets(line, MAXLINE, ficpar);
10260: numlinepar++;
1.141 brouard 10261: fputs(line,stdout);
1.126 brouard 10262: fputs(line,ficparo);
10263: fputs(line,ficlog);
10264: }
10265: ungetc(c,ficpar);
1.220 brouard 10266:
1.126 brouard 10267: matcov=matrix(1,npar,1,npar);
1.203 brouard 10268: hess=matrix(1,npar,1,npar);
1.131 brouard 10269: for(i=1; i <=npar; i++)
10270: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10271:
1.194 brouard 10272: /* Scans npar lines */
1.126 brouard 10273: for(i=1; i <=npar; i++){
1.226 brouard 10274: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10275: if(count != 3){
1.226 brouard 10276: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10277: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10278: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10279: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10280: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10281: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10282: exit(1);
1.220 brouard 10283: }else{
1.226 brouard 10284: if(mle==1)
10285: printf("%1d%1d%d",i1,j1,jk);
10286: }
10287: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10288: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10289: for(j=1; j <=i; j++){
1.226 brouard 10290: fscanf(ficpar," %le",&matcov[i][j]);
10291: if(mle==1){
10292: printf(" %.5le",matcov[i][j]);
10293: }
10294: fprintf(ficlog," %.5le",matcov[i][j]);
10295: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10296: }
10297: fscanf(ficpar,"\n");
10298: numlinepar++;
10299: if(mle==1)
1.220 brouard 10300: printf("\n");
1.126 brouard 10301: fprintf(ficlog,"\n");
10302: fprintf(ficparo,"\n");
10303: }
1.194 brouard 10304: /* End of read covariance matrix npar lines */
1.126 brouard 10305: for(i=1; i <=npar; i++)
10306: for(j=i+1;j<=npar;j++)
1.226 brouard 10307: matcov[i][j]=matcov[j][i];
1.126 brouard 10308:
10309: if(mle==1)
10310: printf("\n");
10311: fprintf(ficlog,"\n");
10312:
10313: fflush(ficlog);
10314:
10315: /*-------- Rewriting parameter file ----------*/
10316: strcpy(rfileres,"r"); /* "Rparameterfile */
10317: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10318: strcat(rfileres,"."); /* */
10319: strcat(rfileres,optionfilext); /* Other files have txt extension */
10320: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10321: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10322: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10323: }
10324: fprintf(ficres,"#%s\n",version);
10325: } /* End of mle != -3 */
1.218 brouard 10326:
1.186 brouard 10327: /* Main data
10328: */
1.126 brouard 10329: n= lastobs;
10330: num=lvector(1,n);
10331: moisnais=vector(1,n);
10332: annais=vector(1,n);
10333: moisdc=vector(1,n);
10334: andc=vector(1,n);
1.220 brouard 10335: weight=vector(1,n);
1.126 brouard 10336: agedc=vector(1,n);
10337: cod=ivector(1,n);
1.220 brouard 10338: for(i=1;i<=n;i++){
1.234 brouard 10339: num[i]=0;
10340: moisnais[i]=0;
10341: annais[i]=0;
10342: moisdc[i]=0;
10343: andc[i]=0;
10344: agedc[i]=0;
10345: cod[i]=0;
10346: weight[i]=1.0; /* Equal weights, 1 by default */
10347: }
1.126 brouard 10348: mint=matrix(1,maxwav,1,n);
10349: anint=matrix(1,maxwav,1,n);
1.131 brouard 10350: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10351: tab=ivector(1,NCOVMAX);
1.144 brouard 10352: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10353: 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 10354:
1.136 brouard 10355: /* Reads data from file datafile */
10356: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10357: goto end;
10358:
10359: /* Calculation of the number of parameters from char model */
1.234 brouard 10360: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10361: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10362: k=3 V4 Tvar[k=3]= 4 (from V4)
10363: k=2 V1 Tvar[k=2]= 1 (from V1)
10364: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10365: */
10366:
10367: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10368: TvarsDind=ivector(1,NCOVMAX); /* */
10369: TvarsD=ivector(1,NCOVMAX); /* */
10370: TvarsQind=ivector(1,NCOVMAX); /* */
10371: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10372: TvarF=ivector(1,NCOVMAX); /* */
10373: TvarFind=ivector(1,NCOVMAX); /* */
10374: TvarV=ivector(1,NCOVMAX); /* */
10375: TvarVind=ivector(1,NCOVMAX); /* */
10376: TvarA=ivector(1,NCOVMAX); /* */
10377: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10378: TvarFD=ivector(1,NCOVMAX); /* */
10379: TvarFDind=ivector(1,NCOVMAX); /* */
10380: TvarFQ=ivector(1,NCOVMAX); /* */
10381: TvarFQind=ivector(1,NCOVMAX); /* */
10382: TvarVD=ivector(1,NCOVMAX); /* */
10383: TvarVDind=ivector(1,NCOVMAX); /* */
10384: TvarVQ=ivector(1,NCOVMAX); /* */
10385: TvarVQind=ivector(1,NCOVMAX); /* */
10386:
1.230 brouard 10387: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10388: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10389: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10390: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10391: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10392: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10393: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10394: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10395: */
10396: /* For model-covariate k tells which data-covariate to use but
10397: because this model-covariate is a construction we invent a new column
10398: ncovcol + k1
10399: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10400: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10401: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10402: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10403: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10404: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10405: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10406: */
1.145 brouard 10407: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10408: 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 10409: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10410: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10411: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10412: 4 covariates (3 plus signs)
10413: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10414: */
1.230 brouard 10415: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10416: * individual dummy, fixed or varying:
10417: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10418: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10419: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10420: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10421: * Tmodelind[1]@9={9,0,3,2,}*/
10422: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10423: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10424: * individual quantitative, fixed or varying:
10425: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10426: * 3, 1, 0, 0, 0, 0, 0, 0},
10427: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10428: /* Main decodemodel */
10429:
1.187 brouard 10430:
1.223 brouard 10431: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10432: goto end;
10433:
1.137 brouard 10434: if((double)(lastobs-imx)/(double)imx > 1.10){
10435: nbwarn++;
10436: 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);
10437: 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);
10438: }
1.136 brouard 10439: /* if(mle==1){*/
1.137 brouard 10440: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10441: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10442: }
10443:
10444: /*-calculation of age at interview from date of interview and age at death -*/
10445: agev=matrix(1,maxwav,1,imx);
10446:
10447: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10448: goto end;
10449:
1.126 brouard 10450:
1.136 brouard 10451: agegomp=(int)agemin;
10452: free_vector(moisnais,1,n);
10453: free_vector(annais,1,n);
1.126 brouard 10454: /* free_matrix(mint,1,maxwav,1,n);
10455: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10456: /* free_vector(moisdc,1,n); */
10457: /* free_vector(andc,1,n); */
1.145 brouard 10458: /* */
10459:
1.126 brouard 10460: wav=ivector(1,imx);
1.214 brouard 10461: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10462: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10463: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10464: 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.*/
10465: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10466: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10467:
10468: /* Concatenates waves */
1.214 brouard 10469: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10470: Death is a valid wave (if date is known).
10471: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10472: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10473: and mw[mi+1][i]. dh depends on stepm.
10474: */
10475:
1.126 brouard 10476: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10477: /* Concatenates waves */
1.145 brouard 10478:
1.215 brouard 10479: free_vector(moisdc,1,n);
10480: free_vector(andc,1,n);
10481:
1.126 brouard 10482: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10483: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10484: ncodemax[1]=1;
1.145 brouard 10485: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10486: cptcoveff=0;
1.220 brouard 10487: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10488: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10489: }
10490:
10491: ncovcombmax=pow(2,cptcoveff);
10492: invalidvarcomb=ivector(1, ncovcombmax);
10493: for(i=1;i<ncovcombmax;i++)
10494: invalidvarcomb[i]=0;
10495:
1.211 brouard 10496: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10497: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10498: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10499:
1.200 brouard 10500: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10501: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10502: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10503: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10504: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10505: * (currently 0 or 1) in the data.
10506: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10507: * corresponding modality (h,j).
10508: */
10509:
1.145 brouard 10510: h=0;
10511: /*if (cptcovn > 0) */
1.126 brouard 10512: m=pow(2,cptcoveff);
10513:
1.144 brouard 10514: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10515: * For k=4 covariates, h goes from 1 to m=2**k
10516: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10517: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10518: * h\k 1 2 3 4
1.143 brouard 10519: *______________________________
10520: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10521: * 2 2 1 1 1
10522: * 3 i=2 1 2 1 1
10523: * 4 2 2 1 1
10524: * 5 i=3 1 i=2 1 2 1
10525: * 6 2 1 2 1
10526: * 7 i=4 1 2 2 1
10527: * 8 2 2 2 1
1.197 brouard 10528: * 9 i=5 1 i=3 1 i=2 1 2
10529: * 10 2 1 1 2
10530: * 11 i=6 1 2 1 2
10531: * 12 2 2 1 2
10532: * 13 i=7 1 i=4 1 2 2
10533: * 14 2 1 2 2
10534: * 15 i=8 1 2 2 2
10535: * 16 2 2 2 2
1.143 brouard 10536: */
1.212 brouard 10537: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10538: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10539: * and the value of each covariate?
10540: * V1=1, V2=1, V3=2, V4=1 ?
10541: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10542: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10543: * In order to get the real value in the data, we use nbcode
10544: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10545: * We are keeping this crazy system in order to be able (in the future?)
10546: * to have more than 2 values (0 or 1) for a covariate.
10547: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10548: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10549: * bbbbbbbb
10550: * 76543210
10551: * h-1 00000101 (6-1=5)
1.219 brouard 10552: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10553: * &
10554: * 1 00000001 (1)
1.219 brouard 10555: * 00000000 = 1 & ((h-1) >> (k-1))
10556: * +1= 00000001 =1
1.211 brouard 10557: *
10558: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10559: * h' 1101 =2^3+2^2+0x2^1+2^0
10560: * >>k' 11
10561: * & 00000001
10562: * = 00000001
10563: * +1 = 00000010=2 = codtabm(14,3)
10564: * Reverse h=6 and m=16?
10565: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10566: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10567: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10568: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10569: * V3=decodtabm(14,3,2**4)=2
10570: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10571: *(h-1) >> (j-1) 0011 =13 >> 2
10572: * &1 000000001
10573: * = 000000001
10574: * +1= 000000010 =2
10575: * 2211
10576: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10577: * V3=2
1.220 brouard 10578: * codtabm and decodtabm are identical
1.211 brouard 10579: */
10580:
1.145 brouard 10581:
10582: free_ivector(Ndum,-1,NCOVMAX);
10583:
10584:
1.126 brouard 10585:
1.186 brouard 10586: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10587: strcpy(optionfilegnuplot,optionfilefiname);
10588: if(mle==-3)
1.201 brouard 10589: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10590: strcat(optionfilegnuplot,".gp");
10591:
10592: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10593: printf("Problem with file %s",optionfilegnuplot);
10594: }
10595: else{
1.204 brouard 10596: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10597: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10598: //fprintf(ficgp,"set missing 'NaNq'\n");
10599: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10600: }
10601: /* fclose(ficgp);*/
1.186 brouard 10602:
10603:
10604: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10605:
10606: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10607: if(mle==-3)
1.201 brouard 10608: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10609: strcat(optionfilehtm,".htm");
10610: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10611: printf("Problem with %s \n",optionfilehtm);
10612: exit(0);
1.126 brouard 10613: }
10614:
10615: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10616: strcat(optionfilehtmcov,"-cov.htm");
10617: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10618: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10619: }
10620: else{
10621: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10622: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10623: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10624: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10625: }
10626:
1.213 brouard 10627: 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 10628: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10629: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10630: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10631: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10632: \n\
10633: <hr size=\"2\" color=\"#EC5E5E\">\
10634: <ul><li><h4>Parameter files</h4>\n\
10635: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10636: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10637: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10638: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10639: - Date and time at start: %s</ul>\n",\
10640: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10641: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10642: fileres,fileres,\
10643: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10644: fflush(fichtm);
10645:
10646: strcpy(pathr,path);
10647: strcat(pathr,optionfilefiname);
1.184 brouard 10648: #ifdef WIN32
10649: _chdir(optionfilefiname); /* Move to directory named optionfile */
10650: #else
1.126 brouard 10651: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10652: #endif
10653:
1.126 brouard 10654:
1.220 brouard 10655: /* Calculates basic frequencies. Computes observed prevalence at single age
10656: and for any valid combination of covariates
1.126 brouard 10657: and prints on file fileres'p'. */
1.251 brouard 10658: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10659: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10660:
10661: fprintf(fichtm,"\n");
10662: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10663: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10664: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10665: imx,agemin,agemax,jmin,jmax,jmean);
10666: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10667: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10668: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10669: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10670: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10671:
1.126 brouard 10672: /* For Powell, parameters are in a vector p[] starting at p[1]
10673: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10674: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10675:
10676: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10677: /* For mortality only */
1.126 brouard 10678: if (mle==-3){
1.136 brouard 10679: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10680: for(i=1;i<=NDIM;i++)
10681: for(j=1;j<=NDIM;j++)
10682: ximort[i][j]=0.;
1.186 brouard 10683: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10684: cens=ivector(1,n);
10685: ageexmed=vector(1,n);
10686: agecens=vector(1,n);
10687: dcwave=ivector(1,n);
1.223 brouard 10688:
1.126 brouard 10689: for (i=1; i<=imx; i++){
10690: dcwave[i]=-1;
10691: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10692: if (s[m][i]>nlstate) {
10693: dcwave[i]=m;
10694: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10695: break;
10696: }
1.126 brouard 10697: }
1.226 brouard 10698:
1.126 brouard 10699: for (i=1; i<=imx; i++) {
10700: if (wav[i]>0){
1.226 brouard 10701: ageexmed[i]=agev[mw[1][i]][i];
10702: j=wav[i];
10703: agecens[i]=1.;
10704:
10705: if (ageexmed[i]> 1 && wav[i] > 0){
10706: agecens[i]=agev[mw[j][i]][i];
10707: cens[i]= 1;
10708: }else if (ageexmed[i]< 1)
10709: cens[i]= -1;
10710: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10711: cens[i]=0 ;
1.126 brouard 10712: }
10713: else cens[i]=-1;
10714: }
10715:
10716: for (i=1;i<=NDIM;i++) {
10717: for (j=1;j<=NDIM;j++)
1.226 brouard 10718: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10719: }
10720:
1.145 brouard 10721: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10722: /*printf("%lf %lf", p[1], p[2]);*/
10723:
10724:
1.136 brouard 10725: #ifdef GSL
10726: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10727: #else
1.126 brouard 10728: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10729: #endif
1.201 brouard 10730: strcpy(filerespow,"POW-MORT_");
10731: strcat(filerespow,fileresu);
1.126 brouard 10732: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10733: printf("Problem with resultfile: %s\n", filerespow);
10734: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10735: }
1.136 brouard 10736: #ifdef GSL
10737: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10738: #else
1.126 brouard 10739: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10740: #endif
1.126 brouard 10741: /* for (i=1;i<=nlstate;i++)
10742: for(j=1;j<=nlstate+ndeath;j++)
10743: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10744: */
10745: fprintf(ficrespow,"\n");
1.136 brouard 10746: #ifdef GSL
10747: /* gsl starts here */
10748: T = gsl_multimin_fminimizer_nmsimplex;
10749: gsl_multimin_fminimizer *sfm = NULL;
10750: gsl_vector *ss, *x;
10751: gsl_multimin_function minex_func;
10752:
10753: /* Initial vertex size vector */
10754: ss = gsl_vector_alloc (NDIM);
10755:
10756: if (ss == NULL){
10757: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10758: }
10759: /* Set all step sizes to 1 */
10760: gsl_vector_set_all (ss, 0.001);
10761:
10762: /* Starting point */
1.126 brouard 10763:
1.136 brouard 10764: x = gsl_vector_alloc (NDIM);
10765:
10766: if (x == NULL){
10767: gsl_vector_free(ss);
10768: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10769: }
10770:
10771: /* Initialize method and iterate */
10772: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10773: /* gsl_vector_set(x, 0, 0.0268); */
10774: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10775: gsl_vector_set(x, 0, p[1]);
10776: gsl_vector_set(x, 1, p[2]);
10777:
10778: minex_func.f = &gompertz_f;
10779: minex_func.n = NDIM;
10780: minex_func.params = (void *)&p; /* ??? */
10781:
10782: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10783: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10784:
10785: printf("Iterations beginning .....\n\n");
10786: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10787:
10788: iteri=0;
10789: while (rval == GSL_CONTINUE){
10790: iteri++;
10791: status = gsl_multimin_fminimizer_iterate(sfm);
10792:
10793: if (status) printf("error: %s\n", gsl_strerror (status));
10794: fflush(0);
10795:
10796: if (status)
10797: break;
10798:
10799: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10800: ssval = gsl_multimin_fminimizer_size (sfm);
10801:
10802: if (rval == GSL_SUCCESS)
10803: printf ("converged to a local maximum at\n");
10804:
10805: printf("%5d ", iteri);
10806: for (it = 0; it < NDIM; it++){
10807: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10808: }
10809: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10810: }
10811:
10812: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10813:
10814: gsl_vector_free(x); /* initial values */
10815: gsl_vector_free(ss); /* inital step size */
10816: for (it=0; it<NDIM; it++){
10817: p[it+1]=gsl_vector_get(sfm->x,it);
10818: fprintf(ficrespow," %.12lf", p[it]);
10819: }
10820: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10821: #endif
10822: #ifdef POWELL
10823: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10824: #endif
1.126 brouard 10825: fclose(ficrespow);
10826:
1.203 brouard 10827: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10828:
10829: for(i=1; i <=NDIM; i++)
10830: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10831: matcov[i][j]=matcov[j][i];
1.126 brouard 10832:
10833: printf("\nCovariance matrix\n ");
1.203 brouard 10834: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10835: for(i=1; i <=NDIM; i++) {
10836: for(j=1;j<=NDIM;j++){
1.220 brouard 10837: printf("%f ",matcov[i][j]);
10838: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10839: }
1.203 brouard 10840: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10841: }
10842:
10843: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10844: for (i=1;i<=NDIM;i++) {
1.126 brouard 10845: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10846: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10847: }
1.126 brouard 10848: lsurv=vector(1,AGESUP);
10849: lpop=vector(1,AGESUP);
10850: tpop=vector(1,AGESUP);
10851: lsurv[agegomp]=100000;
10852:
10853: for (k=agegomp;k<=AGESUP;k++) {
10854: agemortsup=k;
10855: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10856: }
10857:
10858: for (k=agegomp;k<agemortsup;k++)
10859: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10860:
10861: for (k=agegomp;k<agemortsup;k++){
10862: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10863: sumlpop=sumlpop+lpop[k];
10864: }
10865:
10866: tpop[agegomp]=sumlpop;
10867: for (k=agegomp;k<(agemortsup-3);k++){
10868: /* tpop[k+1]=2;*/
10869: tpop[k+1]=tpop[k]-lpop[k];
10870: }
10871:
10872:
10873: printf("\nAge lx qx dx Lx Tx e(x)\n");
10874: for (k=agegomp;k<(agemortsup-2);k++)
10875: 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]);
10876:
10877:
10878: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10879: ageminpar=50;
10880: agemaxpar=100;
1.194 brouard 10881: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10882: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10883: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10884: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10885: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10886: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10887: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10888: }else{
10889: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10890: 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 10891: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10892: }
1.201 brouard 10893: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10894: stepm, weightopt,\
10895: model,imx,p,matcov,agemortsup);
10896:
10897: free_vector(lsurv,1,AGESUP);
10898: free_vector(lpop,1,AGESUP);
10899: free_vector(tpop,1,AGESUP);
1.220 brouard 10900: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10901: free_ivector(cens,1,n);
10902: free_vector(agecens,1,n);
10903: free_ivector(dcwave,1,n);
1.220 brouard 10904: #ifdef GSL
1.136 brouard 10905: #endif
1.186 brouard 10906: } /* Endof if mle==-3 mortality only */
1.205 brouard 10907: /* Standard */
10908: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10909: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10910: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10911: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10912: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10913: for (k=1; k<=npar;k++)
10914: printf(" %d %8.5f",k,p[k]);
10915: printf("\n");
1.205 brouard 10916: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10917: /* mlikeli uses func not funcone */
1.247 brouard 10918: /* for(i=1;i<nlstate;i++){ */
10919: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10920: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10921: /* } */
1.205 brouard 10922: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10923: }
10924: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10925: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10926: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10927: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10928: }
10929: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10930: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10931: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10932: for (k=1; k<=npar;k++)
10933: printf(" %d %8.5f",k,p[k]);
10934: printf("\n");
10935:
10936: /*--------- results files --------------*/
1.224 brouard 10937: 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 10938:
10939:
10940: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10941: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10942: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10943: for(i=1,jk=1; i <=nlstate; i++){
10944: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10945: if (k != i) {
10946: printf("%d%d ",i,k);
10947: fprintf(ficlog,"%d%d ",i,k);
10948: fprintf(ficres,"%1d%1d ",i,k);
10949: for(j=1; j <=ncovmodel; j++){
10950: printf("%12.7f ",p[jk]);
10951: fprintf(ficlog,"%12.7f ",p[jk]);
10952: fprintf(ficres,"%12.7f ",p[jk]);
10953: jk++;
10954: }
10955: printf("\n");
10956: fprintf(ficlog,"\n");
10957: fprintf(ficres,"\n");
10958: }
1.126 brouard 10959: }
10960: }
1.203 brouard 10961: if(mle != 0){
10962: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10963: ftolhess=ftol; /* Usually correct */
1.203 brouard 10964: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10965: 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");
10966: 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");
10967: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10968: for(k=1; k <=(nlstate+ndeath); k++){
10969: if (k != i) {
10970: printf("%d%d ",i,k);
10971: fprintf(ficlog,"%d%d ",i,k);
10972: for(j=1; j <=ncovmodel; j++){
10973: 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]));
10974: 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]));
10975: jk++;
10976: }
10977: printf("\n");
10978: fprintf(ficlog,"\n");
10979: }
10980: }
1.193 brouard 10981: }
1.203 brouard 10982: } /* end of hesscov and Wald tests */
1.225 brouard 10983:
1.203 brouard 10984: /* */
1.126 brouard 10985: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10986: printf("# Scales (for hessian or gradient estimation)\n");
10987: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10988: for(i=1,jk=1; i <=nlstate; i++){
10989: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10990: if (j!=i) {
10991: fprintf(ficres,"%1d%1d",i,j);
10992: printf("%1d%1d",i,j);
10993: fprintf(ficlog,"%1d%1d",i,j);
10994: for(k=1; k<=ncovmodel;k++){
10995: printf(" %.5e",delti[jk]);
10996: fprintf(ficlog," %.5e",delti[jk]);
10997: fprintf(ficres," %.5e",delti[jk]);
10998: jk++;
10999: }
11000: printf("\n");
11001: fprintf(ficlog,"\n");
11002: fprintf(ficres,"\n");
11003: }
1.126 brouard 11004: }
11005: }
11006:
11007: 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 11008: if(mle >= 1) /* To big for the screen */
1.126 brouard 11009: 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");
11010: 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");
11011: /* # 121 Var(a12)\n\ */
11012: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11013: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11014: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11015: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11016: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11017: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11018: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11019:
11020:
11021: /* Just to have a covariance matrix which will be more understandable
11022: even is we still don't want to manage dictionary of variables
11023: */
11024: for(itimes=1;itimes<=2;itimes++){
11025: jj=0;
11026: for(i=1; i <=nlstate; i++){
1.225 brouard 11027: for(j=1; j <=nlstate+ndeath; j++){
11028: if(j==i) continue;
11029: for(k=1; k<=ncovmodel;k++){
11030: jj++;
11031: ca[0]= k+'a'-1;ca[1]='\0';
11032: if(itimes==1){
11033: if(mle>=1)
11034: printf("#%1d%1d%d",i,j,k);
11035: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11036: fprintf(ficres,"#%1d%1d%d",i,j,k);
11037: }else{
11038: if(mle>=1)
11039: printf("%1d%1d%d",i,j,k);
11040: fprintf(ficlog,"%1d%1d%d",i,j,k);
11041: fprintf(ficres,"%1d%1d%d",i,j,k);
11042: }
11043: ll=0;
11044: for(li=1;li <=nlstate; li++){
11045: for(lj=1;lj <=nlstate+ndeath; lj++){
11046: if(lj==li) continue;
11047: for(lk=1;lk<=ncovmodel;lk++){
11048: ll++;
11049: if(ll<=jj){
11050: cb[0]= lk +'a'-1;cb[1]='\0';
11051: if(ll<jj){
11052: if(itimes==1){
11053: if(mle>=1)
11054: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11055: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11056: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11057: }else{
11058: if(mle>=1)
11059: printf(" %.5e",matcov[jj][ll]);
11060: fprintf(ficlog," %.5e",matcov[jj][ll]);
11061: fprintf(ficres," %.5e",matcov[jj][ll]);
11062: }
11063: }else{
11064: if(itimes==1){
11065: if(mle>=1)
11066: printf(" Var(%s%1d%1d)",ca,i,j);
11067: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11068: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11069: }else{
11070: if(mle>=1)
11071: printf(" %.7e",matcov[jj][ll]);
11072: fprintf(ficlog," %.7e",matcov[jj][ll]);
11073: fprintf(ficres," %.7e",matcov[jj][ll]);
11074: }
11075: }
11076: }
11077: } /* end lk */
11078: } /* end lj */
11079: } /* end li */
11080: if(mle>=1)
11081: printf("\n");
11082: fprintf(ficlog,"\n");
11083: fprintf(ficres,"\n");
11084: numlinepar++;
11085: } /* end k*/
11086: } /*end j */
1.126 brouard 11087: } /* end i */
11088: } /* end itimes */
11089:
11090: fflush(ficlog);
11091: fflush(ficres);
1.225 brouard 11092: while(fgets(line, MAXLINE, ficpar)) {
11093: /* If line starts with a # it is a comment */
11094: if (line[0] == '#') {
11095: numlinepar++;
11096: fputs(line,stdout);
11097: fputs(line,ficparo);
11098: fputs(line,ficlog);
11099: continue;
11100: }else
11101: break;
11102: }
11103:
1.209 brouard 11104: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11105: /* ungetc(c,ficpar); */
11106: /* fgets(line, MAXLINE, ficpar); */
11107: /* fputs(line,stdout); */
11108: /* fputs(line,ficparo); */
11109: /* } */
11110: /* ungetc(c,ficpar); */
1.126 brouard 11111:
11112: estepm=0;
1.209 brouard 11113: 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 11114:
11115: if (num_filled != 6) {
11116: 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);
11117: 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);
11118: goto end;
11119: }
11120: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11121: }
11122: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11123: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11124:
1.209 brouard 11125: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11126: if (estepm==0 || estepm < stepm) estepm=stepm;
11127: if (fage <= 2) {
11128: bage = ageminpar;
11129: fage = agemaxpar;
11130: }
11131:
11132: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11133: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11134: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11135:
1.186 brouard 11136: /* Other stuffs, more or less useful */
1.254 brouard 11137: while(fgets(line, MAXLINE, ficpar)) {
11138: /* If line starts with a # it is a comment */
11139: if (line[0] == '#') {
11140: numlinepar++;
11141: fputs(line,stdout);
11142: fputs(line,ficparo);
11143: fputs(line,ficlog);
11144: continue;
11145: }else
11146: break;
11147: }
11148:
11149: 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){
11150:
11151: if (num_filled != 7) {
11152: 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);
11153: 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);
11154: goto end;
11155: }
11156: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11157: 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);
11158: 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);
11159: 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 11160: }
1.254 brouard 11161:
11162: while(fgets(line, MAXLINE, ficpar)) {
11163: /* If line starts with a # it is a comment */
11164: if (line[0] == '#') {
11165: numlinepar++;
11166: fputs(line,stdout);
11167: fputs(line,ficparo);
11168: fputs(line,ficlog);
11169: continue;
11170: }else
11171: break;
1.126 brouard 11172: }
11173:
11174:
11175: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11176: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11177:
1.254 brouard 11178: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11179: if (num_filled != 1) {
11180: 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);
11181: 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);
11182: goto end;
11183: }
11184: printf("pop_based=%d\n",popbased);
11185: fprintf(ficlog,"pop_based=%d\n",popbased);
11186: fprintf(ficparo,"pop_based=%d\n",popbased);
11187: fprintf(ficres,"pop_based=%d\n",popbased);
11188: }
11189:
1.258 ! brouard 11190: /* Results */
! 11191: nresult=0;
! 11192: do{
! 11193: if(!fgets(line, MAXLINE, ficpar)){
! 11194: endishere=1;
! 11195: parameterline=14;
! 11196: }else if (line[0] == '#') {
! 11197: /* If line starts with a # it is a comment */
1.254 brouard 11198: numlinepar++;
11199: fputs(line,stdout);
11200: fputs(line,ficparo);
11201: fputs(line,ficlog);
11202: continue;
1.258 ! brouard 11203: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
! 11204: parameterline=11;
! 11205: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
! 11206: parameterline=12;
! 11207: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
! 11208: parameterline=13;
! 11209: else{
! 11210: parameterline=14;
1.254 brouard 11211: }
1.258 ! brouard 11212: switch (parameterline){
! 11213: case 11:
! 11214: 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){
! 11215: if (num_filled != 8) {
! 11216: 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);
! 11217: 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);
! 11218: goto end;
! 11219: }
! 11220: 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);
! 11221: 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);
! 11222: 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);
! 11223: 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);
! 11224: /* day and month of proj2 are not used but only year anproj2.*/
! 11225: }
1.254 brouard 11226: break;
1.258 ! brouard 11227: case 12:
! 11228: /*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);*/
! 11229: 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){
! 11230: if (num_filled != 8) {
! 11231: 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);
! 11232: 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);
! 11233: goto end;
! 11234: }
! 11235: 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);
! 11236: 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);
! 11237: 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);
! 11238: 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);
! 11239: /* day and month of proj2 are not used but only year anproj2.*/
! 11240: }
1.230 brouard 11241: break;
1.258 ! brouard 11242: case 13:
! 11243: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
! 11244: if (num_filled == 0){
! 11245: resultline[0]='\0';
! 11246: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
! 11247: 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);
! 11248: break;
! 11249: } else if (num_filled != 1){
! 11250: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
! 11251: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
! 11252: }
! 11253: nresult++; /* Sum of resultlines */
! 11254: printf("Result %d: result=%s\n",nresult, resultline);
! 11255: if(nresult > MAXRESULTLINES){
! 11256: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
! 11257: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
! 11258: goto end;
! 11259: }
! 11260: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
! 11261: fprintf(ficparo,"result: %s\n",resultline);
! 11262: fprintf(ficres,"result: %s\n",resultline);
! 11263: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11264: break;
1.258 ! brouard 11265: case 14:
! 11266: if(ncovmodel >2){
! 11267: printf("ERROR: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
! 11268: goto end;
! 11269: }
! 11270: default:
! 11271: nresult=1;
! 11272: decoderesult(".",nresult ); /* No covariate */
! 11273: }
! 11274: } /* End switch parameterline */
! 11275: }while(endishere==0); /* End do */
1.126 brouard 11276:
1.230 brouard 11277: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11278: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11279:
11280: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11281: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11282: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11283: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11284: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11285: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11286: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11287: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11288: }else{
1.218 brouard 11289: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11290: }
11291: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 ! brouard 11292: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11293: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11294:
1.225 brouard 11295: /*------------ free_vector -------------*/
11296: /* chdir(path); */
1.220 brouard 11297:
1.215 brouard 11298: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11299: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11300: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11301: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11302: free_lvector(num,1,n);
11303: free_vector(agedc,1,n);
11304: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11305: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11306: fclose(ficparo);
11307: fclose(ficres);
1.220 brouard 11308:
11309:
1.186 brouard 11310: /* Other results (useful)*/
1.220 brouard 11311:
11312:
1.126 brouard 11313: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11314: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11315: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11316: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11317: fclose(ficrespl);
11318:
11319: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11320: /*#include "hpijx.h"*/
11321: hPijx(p, bage, fage);
1.145 brouard 11322: fclose(ficrespij);
1.227 brouard 11323:
1.220 brouard 11324: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11325: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11326: k=1;
1.126 brouard 11327: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11328:
1.219 brouard 11329: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11330: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11331: for(i=1;i<=AGESUP;i++)
1.219 brouard 11332: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11333: for(k=1;k<=ncovcombmax;k++)
11334: probs[i][j][k]=0.;
1.219 brouard 11335: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11336: if (mobilav!=0 ||mobilavproj !=0 ) {
11337: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11338: for(i=1;i<=AGESUP;i++)
11339: for(j=1;j<=nlstate;j++)
11340: for(k=1;k<=ncovcombmax;k++)
11341: mobaverages[i][j][k]=0.;
1.219 brouard 11342: mobaverage=mobaverages;
11343: if (mobilav!=0) {
1.235 brouard 11344: printf("Movingaveraging observed prevalence\n");
1.258 ! brouard 11345: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11346: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11347: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11348: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11349: }
1.219 brouard 11350: }
11351: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11352: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11353: else if (mobilavproj !=0) {
1.235 brouard 11354: printf("Movingaveraging projected observed prevalence\n");
1.258 ! brouard 11355: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11356: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11357: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11358: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11359: }
1.219 brouard 11360: }
11361: }/* end if moving average */
1.227 brouard 11362:
1.126 brouard 11363: /*---------- Forecasting ------------------*/
11364: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11365: if(prevfcast==1){
11366: /* if(stepm ==1){*/
1.225 brouard 11367: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11368: }
1.217 brouard 11369: if(backcast==1){
1.219 brouard 11370: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11371: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11372: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11373:
11374: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11375:
11376: bprlim=matrix(1,nlstate,1,nlstate);
11377: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11378: fclose(ficresplb);
11379:
1.222 brouard 11380: hBijx(p, bage, fage, mobaverage);
11381: fclose(ficrespijb);
1.219 brouard 11382: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11383:
11384: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11385: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11386: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11387: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11388: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11389: }
1.217 brouard 11390:
1.186 brouard 11391:
11392: /* ------ Other prevalence ratios------------ */
1.126 brouard 11393:
1.215 brouard 11394: free_ivector(wav,1,imx);
11395: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11396: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11397: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11398:
11399:
1.127 brouard 11400: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11401:
1.201 brouard 11402: strcpy(filerese,"E_");
11403: strcat(filerese,fileresu);
1.126 brouard 11404: if((ficreseij=fopen(filerese,"w"))==NULL) {
11405: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11406: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11407: }
1.208 brouard 11408: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11409: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11410:
11411: pstamp(ficreseij);
1.219 brouard 11412:
1.235 brouard 11413: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11414: if (cptcovn < 1){i1=1;}
11415:
11416: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11417: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11418: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11419: continue;
1.219 brouard 11420: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11421: printf("\n#****** ");
1.225 brouard 11422: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11423: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11424: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11425: }
11426: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11427: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11428: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11429: }
11430: fprintf(ficreseij,"******\n");
1.235 brouard 11431: printf("******\n");
1.219 brouard 11432:
11433: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11434: oldm=oldms;savm=savms;
1.235 brouard 11435: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11436:
1.219 brouard 11437: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11438: }
11439: fclose(ficreseij);
1.208 brouard 11440: printf("done evsij\n");fflush(stdout);
11441: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11442:
1.227 brouard 11443: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11444:
11445:
1.201 brouard 11446: strcpy(filerest,"T_");
11447: strcat(filerest,fileresu);
1.127 brouard 11448: if((ficrest=fopen(filerest,"w"))==NULL) {
11449: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11450: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11451: }
1.208 brouard 11452: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11453: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11454:
1.126 brouard 11455:
1.201 brouard 11456: strcpy(fileresstde,"STDE_");
11457: strcat(fileresstde,fileresu);
1.126 brouard 11458: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11459: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11460: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11461: }
1.227 brouard 11462: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11463: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11464:
1.201 brouard 11465: strcpy(filerescve,"CVE_");
11466: strcat(filerescve,fileresu);
1.126 brouard 11467: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11468: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11469: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11470: }
1.227 brouard 11471: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11472: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11473:
1.201 brouard 11474: strcpy(fileresv,"V_");
11475: strcat(fileresv,fileresu);
1.126 brouard 11476: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11477: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11478: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11479: }
1.227 brouard 11480: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11481: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11482:
1.145 brouard 11483: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11484: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11485:
1.235 brouard 11486: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11487: if (cptcovn < 1){i1=1;}
11488:
11489: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11490: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11491: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11492: continue;
1.242 brouard 11493: printf("\n#****** Result for:");
11494: fprintf(ficrest,"\n#****** Result for:");
11495: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11496: for(j=1;j<=cptcoveff;j++){
11497: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11498: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11499: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11500: }
1.235 brouard 11501: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11502: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11503: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11504: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11505: }
1.208 brouard 11506: fprintf(ficrest,"******\n");
1.227 brouard 11507: fprintf(ficlog,"******\n");
11508: printf("******\n");
1.208 brouard 11509:
11510: fprintf(ficresstdeij,"\n#****** ");
11511: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11512: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11513: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11514: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11515: }
1.235 brouard 11516: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11517: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11518: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11519: }
1.208 brouard 11520: fprintf(ficresstdeij,"******\n");
11521: fprintf(ficrescveij,"******\n");
11522:
11523: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11524: /* pstamp(ficresvij); */
1.225 brouard 11525: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11526: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11527: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11528: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11529: }
1.208 brouard 11530: fprintf(ficresvij,"******\n");
11531:
11532: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11533: oldm=oldms;savm=savms;
1.235 brouard 11534: printf(" cvevsij ");
11535: fprintf(ficlog, " cvevsij ");
11536: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11537: printf(" end cvevsij \n ");
11538: fprintf(ficlog, " end cvevsij \n ");
11539:
11540: /*
11541: */
11542: /* goto endfree; */
11543:
11544: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11545: pstamp(ficrest);
11546:
11547:
11548: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11549: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11550: cptcod= 0; /* To be deleted */
11551: printf("varevsij vpopbased=%d \n",vpopbased);
11552: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11553: 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 11554: 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 ");
11555: if(vpopbased==1)
11556: 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);
11557: else
11558: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11559: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11560: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11561: fprintf(ficrest,"\n");
11562: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11563: epj=vector(1,nlstate+1);
11564: printf("Computing age specific period (stable) prevalences in each health state \n");
11565: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11566: for(age=bage; age <=fage ;age++){
1.235 brouard 11567: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11568: if (vpopbased==1) {
11569: if(mobilav ==0){
11570: for(i=1; i<=nlstate;i++)
11571: prlim[i][i]=probs[(int)age][i][k];
11572: }else{ /* mobilav */
11573: for(i=1; i<=nlstate;i++)
11574: prlim[i][i]=mobaverage[(int)age][i][k];
11575: }
11576: }
1.219 brouard 11577:
1.227 brouard 11578: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11579: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11580: /* printf(" age %4.0f ",age); */
11581: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11582: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11583: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11584: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11585: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11586: }
11587: epj[nlstate+1] +=epj[j];
11588: }
11589: /* printf(" age %4.0f \n",age); */
1.219 brouard 11590:
1.227 brouard 11591: for(i=1, vepp=0.;i <=nlstate;i++)
11592: for(j=1;j <=nlstate;j++)
11593: vepp += vareij[i][j][(int)age];
11594: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11595: for(j=1;j <=nlstate;j++){
11596: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11597: }
11598: fprintf(ficrest,"\n");
11599: }
1.208 brouard 11600: } /* End vpopbased */
11601: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11602: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11603: free_vector(epj,1,nlstate+1);
1.235 brouard 11604: printf("done selection\n");fflush(stdout);
11605: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11606:
1.145 brouard 11607: /*}*/
1.235 brouard 11608: } /* End k selection */
1.227 brouard 11609:
11610: printf("done State-specific expectancies\n");fflush(stdout);
11611: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11612:
1.126 brouard 11613: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11614:
1.201 brouard 11615: strcpy(fileresvpl,"VPL_");
11616: strcat(fileresvpl,fileresu);
1.126 brouard 11617: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11618: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11619: exit(0);
11620: }
1.208 brouard 11621: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11622: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11623:
1.145 brouard 11624: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11625: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11626:
1.235 brouard 11627: i1=pow(2,cptcoveff);
11628: if (cptcovn < 1){i1=1;}
11629:
11630: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11631: for(k=1; k<=i1;k++){
1.253 brouard 11632: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11633: continue;
1.227 brouard 11634: fprintf(ficresvpl,"\n#****** ");
11635: printf("\n#****** ");
11636: fprintf(ficlog,"\n#****** ");
11637: for(j=1;j<=cptcoveff;j++) {
11638: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11639: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11640: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11641: }
1.235 brouard 11642: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11643: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11644: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11645: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11646: }
1.227 brouard 11647: fprintf(ficresvpl,"******\n");
11648: printf("******\n");
11649: fprintf(ficlog,"******\n");
11650:
11651: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11652: oldm=oldms;savm=savms;
1.235 brouard 11653: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11654: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11655: /*}*/
1.126 brouard 11656: }
1.227 brouard 11657:
1.126 brouard 11658: fclose(ficresvpl);
1.208 brouard 11659: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11660: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11661:
11662: free_vector(weight,1,n);
11663: free_imatrix(Tvard,1,NCOVMAX,1,2);
11664: free_imatrix(s,1,maxwav+1,1,n);
11665: free_matrix(anint,1,maxwav,1,n);
11666: free_matrix(mint,1,maxwav,1,n);
11667: free_ivector(cod,1,n);
11668: free_ivector(tab,1,NCOVMAX);
11669: fclose(ficresstdeij);
11670: fclose(ficrescveij);
11671: fclose(ficresvij);
11672: fclose(ficrest);
11673: fclose(ficpar);
11674:
11675:
1.126 brouard 11676: /*---------- End : free ----------------*/
1.219 brouard 11677: if (mobilav!=0 ||mobilavproj !=0)
11678: 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 11679: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11680: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11681: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11682: } /* mle==-3 arrives here for freeing */
1.227 brouard 11683: /* endfree:*/
11684: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11685: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11686: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11687: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11688: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11689: free_matrix(coqvar,1,maxwav,1,n);
11690: free_matrix(covar,0,NCOVMAX,1,n);
11691: free_matrix(matcov,1,npar,1,npar);
11692: free_matrix(hess,1,npar,1,npar);
11693: /*free_vector(delti,1,npar);*/
11694: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11695: free_matrix(agev,1,maxwav,1,imx);
11696: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11697:
11698: free_ivector(ncodemax,1,NCOVMAX);
11699: free_ivector(ncodemaxwundef,1,NCOVMAX);
11700: free_ivector(Dummy,-1,NCOVMAX);
11701: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11702: free_ivector(DummyV,1,NCOVMAX);
11703: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11704: free_ivector(Typevar,-1,NCOVMAX);
11705: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11706: free_ivector(TvarsQ,1,NCOVMAX);
11707: free_ivector(TvarsQind,1,NCOVMAX);
11708: free_ivector(TvarsD,1,NCOVMAX);
11709: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11710: free_ivector(TvarFD,1,NCOVMAX);
11711: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11712: free_ivector(TvarF,1,NCOVMAX);
11713: free_ivector(TvarFind,1,NCOVMAX);
11714: free_ivector(TvarV,1,NCOVMAX);
11715: free_ivector(TvarVind,1,NCOVMAX);
11716: free_ivector(TvarA,1,NCOVMAX);
11717: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11718: free_ivector(TvarFQ,1,NCOVMAX);
11719: free_ivector(TvarFQind,1,NCOVMAX);
11720: free_ivector(TvarVD,1,NCOVMAX);
11721: free_ivector(TvarVDind,1,NCOVMAX);
11722: free_ivector(TvarVQ,1,NCOVMAX);
11723: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11724: free_ivector(Tvarsel,1,NCOVMAX);
11725: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11726: free_ivector(Tposprod,1,NCOVMAX);
11727: free_ivector(Tprod,1,NCOVMAX);
11728: free_ivector(Tvaraff,1,NCOVMAX);
11729: free_ivector(invalidvarcomb,1,ncovcombmax);
11730: free_ivector(Tage,1,NCOVMAX);
11731: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11732: free_ivector(TmodelInvind,1,NCOVMAX);
11733: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11734:
11735: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11736: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11737: fflush(fichtm);
11738: fflush(ficgp);
11739:
1.227 brouard 11740:
1.126 brouard 11741: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11742: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11743: 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 11744: }else{
11745: printf("End of Imach\n");
11746: fprintf(ficlog,"End of Imach\n");
11747: }
11748: printf("See log file on %s\n",filelog);
11749: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11750: /*(void) gettimeofday(&end_time,&tzp);*/
11751: rend_time = time(NULL);
11752: end_time = *localtime(&rend_time);
11753: /* tml = *localtime(&end_time.tm_sec); */
11754: strcpy(strtend,asctime(&end_time));
1.126 brouard 11755: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11756: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11757: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11758:
1.157 brouard 11759: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11760: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11761: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11762: /* printf("Total time was %d uSec.\n", total_usecs);*/
11763: /* if(fileappend(fichtm,optionfilehtm)){ */
11764: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11765: fclose(fichtm);
11766: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11767: fclose(fichtmcov);
11768: fclose(ficgp);
11769: fclose(ficlog);
11770: /*------ End -----------*/
1.227 brouard 11771:
11772:
11773: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11774: #ifdef WIN32
1.227 brouard 11775: if (_chdir(pathcd) != 0)
11776: printf("Can't move to directory %s!\n",path);
11777: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11778: #else
1.227 brouard 11779: if(chdir(pathcd) != 0)
11780: printf("Can't move to directory %s!\n", path);
11781: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11782: #endif
1.126 brouard 11783: printf("Current directory %s!\n",pathcd);
11784: /*strcat(plotcmd,CHARSEPARATOR);*/
11785: sprintf(plotcmd,"gnuplot");
1.157 brouard 11786: #ifdef _WIN32
1.126 brouard 11787: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11788: #endif
11789: if(!stat(plotcmd,&info)){
1.158 brouard 11790: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11791: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11792: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11793: }else
11794: strcpy(pplotcmd,plotcmd);
1.157 brouard 11795: #ifdef __unix
1.126 brouard 11796: strcpy(plotcmd,GNUPLOTPROGRAM);
11797: if(!stat(plotcmd,&info)){
1.158 brouard 11798: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11799: }else
11800: strcpy(pplotcmd,plotcmd);
11801: #endif
11802: }else
11803: strcpy(pplotcmd,plotcmd);
11804:
11805: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11806: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11807:
1.126 brouard 11808: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11809: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11810: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11811: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11812: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11813: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11814: }
1.158 brouard 11815: printf(" Successful, please wait...");
1.126 brouard 11816: while (z[0] != 'q') {
11817: /* chdir(path); */
1.154 brouard 11818: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11819: scanf("%s",z);
11820: /* if (z[0] == 'c') system("./imach"); */
11821: if (z[0] == 'e') {
1.158 brouard 11822: #ifdef __APPLE__
1.152 brouard 11823: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11824: #elif __linux
11825: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11826: #else
1.152 brouard 11827: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11828: #endif
11829: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11830: system(pplotcmd);
1.126 brouard 11831: }
11832: else if (z[0] == 'g') system(plotcmd);
11833: else if (z[0] == 'q') exit(0);
11834: }
1.227 brouard 11835: end:
1.126 brouard 11836: while (z[0] != 'q') {
1.195 brouard 11837: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11838: scanf("%s",z);
11839: }
11840: }
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