Annotation of imach/src/imach.c, revision 1.253
1.253 ! brouard 1: /* $Id: imach.c,v 1.252 2016/09/15 21:15:37 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.253 ! brouard 4: Revision 1.252 2016/09/15 21:15:37 brouard
! 5: *** empty log message ***
! 6:
1.252 brouard 7: Revision 1.251 2016/09/15 15:01:13 brouard
8: Summary: not working
9:
1.251 brouard 10: Revision 1.250 2016/09/08 16:07:27 brouard
11: Summary: continue
12:
1.250 brouard 13: Revision 1.249 2016/09/07 17:14:18 brouard
14: Summary: Starting values from frequencies
15:
1.249 brouard 16: Revision 1.248 2016/09/07 14:10:18 brouard
17: *** empty log message ***
18:
1.248 brouard 19: Revision 1.247 2016/09/02 11:11:21 brouard
20: *** empty log message ***
21:
1.247 brouard 22: Revision 1.246 2016/09/02 08:49:22 brouard
23: *** empty log message ***
24:
1.246 brouard 25: Revision 1.245 2016/09/02 07:25:01 brouard
26: *** empty log message ***
27:
1.245 brouard 28: Revision 1.244 2016/09/02 07:17:34 brouard
29: *** empty log message ***
30:
1.244 brouard 31: Revision 1.243 2016/09/02 06:45:35 brouard
32: *** empty log message ***
33:
1.243 brouard 34: Revision 1.242 2016/08/30 15:01:20 brouard
35: Summary: Fixing a lots
36:
1.242 brouard 37: Revision 1.241 2016/08/29 17:17:25 brouard
38: Summary: gnuplot problem in Back projection to fix
39:
1.241 brouard 40: Revision 1.240 2016/08/29 07:53:18 brouard
41: Summary: Better
42:
1.240 brouard 43: Revision 1.239 2016/08/26 15:51:03 brouard
44: Summary: Improvement in Powell output in order to copy and paste
45:
46: Author:
47:
1.239 brouard 48: Revision 1.238 2016/08/26 14:23:35 brouard
49: Summary: Starting tests of 0.99
50:
1.238 brouard 51: Revision 1.237 2016/08/26 09:20:19 brouard
52: Summary: to valgrind
53:
1.237 brouard 54: Revision 1.236 2016/08/25 10:50:18 brouard
55: *** empty log message ***
56:
1.236 brouard 57: Revision 1.235 2016/08/25 06:59:23 brouard
58: *** empty log message ***
59:
1.235 brouard 60: Revision 1.234 2016/08/23 16:51:20 brouard
61: *** empty log message ***
62:
1.234 brouard 63: Revision 1.233 2016/08/23 07:40:50 brouard
64: Summary: not working
65:
1.233 brouard 66: Revision 1.232 2016/08/22 14:20:21 brouard
67: Summary: not working
68:
1.232 brouard 69: Revision 1.231 2016/08/22 07:17:15 brouard
70: Summary: not working
71:
1.231 brouard 72: Revision 1.230 2016/08/22 06:55:53 brouard
73: Summary: Not working
74:
1.230 brouard 75: Revision 1.229 2016/07/23 09:45:53 brouard
76: Summary: Completing for func too
77:
1.229 brouard 78: Revision 1.228 2016/07/22 17:45:30 brouard
79: Summary: Fixing some arrays, still debugging
80:
1.227 brouard 81: Revision 1.226 2016/07/12 18:42:34 brouard
82: Summary: temp
83:
1.226 brouard 84: Revision 1.225 2016/07/12 08:40:03 brouard
85: Summary: saving but not running
86:
1.225 brouard 87: Revision 1.224 2016/07/01 13:16:01 brouard
88: Summary: Fixes
89:
1.224 brouard 90: Revision 1.223 2016/02/19 09:23:35 brouard
91: Summary: temporary
92:
1.223 brouard 93: Revision 1.222 2016/02/17 08:14:50 brouard
94: Summary: Probably last 0.98 stable version 0.98r6
95:
1.222 brouard 96: Revision 1.221 2016/02/15 23:35:36 brouard
97: Summary: minor bug
98:
1.220 brouard 99: Revision 1.219 2016/02/15 00:48:12 brouard
100: *** empty log message ***
101:
1.219 brouard 102: Revision 1.218 2016/02/12 11:29:23 brouard
103: Summary: 0.99 Back projections
104:
1.218 brouard 105: Revision 1.217 2015/12/23 17:18:31 brouard
106: Summary: Experimental backcast
107:
1.217 brouard 108: Revision 1.216 2015/12/18 17:32:11 brouard
109: Summary: 0.98r4 Warning and status=-2
110:
111: Version 0.98r4 is now:
112: - displaying an error when status is -1, date of interview unknown and date of death known;
113: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
114: Older changes concerning s=-2, dating from 2005 have been supersed.
115:
1.216 brouard 116: Revision 1.215 2015/12/16 08:52:24 brouard
117: Summary: 0.98r4 working
118:
1.215 brouard 119: Revision 1.214 2015/12/16 06:57:54 brouard
120: Summary: temporary not working
121:
1.214 brouard 122: Revision 1.213 2015/12/11 18:22:17 brouard
123: Summary: 0.98r4
124:
1.213 brouard 125: Revision 1.212 2015/11/21 12:47:24 brouard
126: Summary: minor typo
127:
1.212 brouard 128: Revision 1.211 2015/11/21 12:41:11 brouard
129: Summary: 0.98r3 with some graph of projected cross-sectional
130:
131: Author: Nicolas Brouard
132:
1.211 brouard 133: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 134: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 135: Summary: Adding ftolpl parameter
136: Author: N Brouard
137:
138: We had difficulties to get smoothed confidence intervals. It was due
139: to the period prevalence which wasn't computed accurately. The inner
140: parameter ftolpl is now an outer parameter of the .imach parameter
141: file after estepm. If ftolpl is small 1.e-4 and estepm too,
142: computation are long.
143:
1.209 brouard 144: Revision 1.208 2015/11/17 14:31:57 brouard
145: Summary: temporary
146:
1.208 brouard 147: Revision 1.207 2015/10/27 17:36:57 brouard
148: *** empty log message ***
149:
1.207 brouard 150: Revision 1.206 2015/10/24 07:14:11 brouard
151: *** empty log message ***
152:
1.206 brouard 153: Revision 1.205 2015/10/23 15:50:53 brouard
154: Summary: 0.98r3 some clarification for graphs on likelihood contributions
155:
1.205 brouard 156: Revision 1.204 2015/10/01 16:20:26 brouard
157: Summary: Some new graphs of contribution to likelihood
158:
1.204 brouard 159: Revision 1.203 2015/09/30 17:45:14 brouard
160: Summary: looking at better estimation of the hessian
161:
162: Also a better criteria for convergence to the period prevalence And
163: therefore adding the number of years needed to converge. (The
164: prevalence in any alive state shold sum to one
165:
1.203 brouard 166: Revision 1.202 2015/09/22 19:45:16 brouard
167: Summary: Adding some overall graph on contribution to likelihood. Might change
168:
1.202 brouard 169: Revision 1.201 2015/09/15 17:34:58 brouard
170: Summary: 0.98r0
171:
172: - Some new graphs like suvival functions
173: - Some bugs fixed like model=1+age+V2.
174:
1.201 brouard 175: Revision 1.200 2015/09/09 16:53:55 brouard
176: Summary: Big bug thanks to Flavia
177:
178: Even model=1+age+V2. did not work anymore
179:
1.200 brouard 180: Revision 1.199 2015/09/07 14:09:23 brouard
181: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
182:
1.199 brouard 183: Revision 1.198 2015/09/03 07:14:39 brouard
184: Summary: 0.98q5 Flavia
185:
1.198 brouard 186: Revision 1.197 2015/09/01 18:24:39 brouard
187: *** empty log message ***
188:
1.197 brouard 189: Revision 1.196 2015/08/18 23:17:52 brouard
190: Summary: 0.98q5
191:
1.196 brouard 192: Revision 1.195 2015/08/18 16:28:39 brouard
193: Summary: Adding a hack for testing purpose
194:
195: After reading the title, ftol and model lines, if the comment line has
196: a q, starting with #q, the answer at the end of the run is quit. It
197: permits to run test files in batch with ctest. The former workaround was
198: $ echo q | imach foo.imach
199:
1.195 brouard 200: Revision 1.194 2015/08/18 13:32:00 brouard
201: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
202:
1.194 brouard 203: Revision 1.193 2015/08/04 07:17:42 brouard
204: Summary: 0.98q4
205:
1.193 brouard 206: Revision 1.192 2015/07/16 16:49:02 brouard
207: Summary: Fixing some outputs
208:
1.192 brouard 209: Revision 1.191 2015/07/14 10:00:33 brouard
210: Summary: Some fixes
211:
1.191 brouard 212: Revision 1.190 2015/05/05 08:51:13 brouard
213: Summary: Adding digits in output parameters (7 digits instead of 6)
214:
215: Fix 1+age+.
216:
1.190 brouard 217: Revision 1.189 2015/04/30 14:45:16 brouard
218: Summary: 0.98q2
219:
1.189 brouard 220: Revision 1.188 2015/04/30 08:27:53 brouard
221: *** empty log message ***
222:
1.188 brouard 223: Revision 1.187 2015/04/29 09:11:15 brouard
224: *** empty log message ***
225:
1.187 brouard 226: Revision 1.186 2015/04/23 12:01:52 brouard
227: Summary: V1*age is working now, version 0.98q1
228:
229: Some codes had been disabled in order to simplify and Vn*age was
230: working in the optimization phase, ie, giving correct MLE parameters,
231: but, as usual, outputs were not correct and program core dumped.
232:
1.186 brouard 233: Revision 1.185 2015/03/11 13:26:42 brouard
234: Summary: Inclusion of compile and links command line for Intel Compiler
235:
1.185 brouard 236: Revision 1.184 2015/03/11 11:52:39 brouard
237: Summary: Back from Windows 8. Intel Compiler
238:
1.184 brouard 239: Revision 1.183 2015/03/10 20:34:32 brouard
240: Summary: 0.98q0, trying with directest, mnbrak fixed
241:
242: We use directest instead of original Powell test; probably no
243: incidence on the results, but better justifications;
244: We fixed Numerical Recipes mnbrak routine which was wrong and gave
245: wrong results.
246:
1.183 brouard 247: Revision 1.182 2015/02/12 08:19:57 brouard
248: Summary: Trying to keep directest which seems simpler and more general
249: Author: Nicolas Brouard
250:
1.182 brouard 251: Revision 1.181 2015/02/11 23:22:24 brouard
252: Summary: Comments on Powell added
253:
254: Author:
255:
1.181 brouard 256: Revision 1.180 2015/02/11 17:33:45 brouard
257: Summary: Finishing move from main to function (hpijx and prevalence_limit)
258:
1.180 brouard 259: Revision 1.179 2015/01/04 09:57:06 brouard
260: Summary: back to OS/X
261:
1.179 brouard 262: Revision 1.178 2015/01/04 09:35:48 brouard
263: *** empty log message ***
264:
1.178 brouard 265: Revision 1.177 2015/01/03 18:40:56 brouard
266: Summary: Still testing ilc32 on OSX
267:
1.177 brouard 268: Revision 1.176 2015/01/03 16:45:04 brouard
269: *** empty log message ***
270:
1.176 brouard 271: Revision 1.175 2015/01/03 16:33:42 brouard
272: *** empty log message ***
273:
1.175 brouard 274: Revision 1.174 2015/01/03 16:15:49 brouard
275: Summary: Still in cross-compilation
276:
1.174 brouard 277: Revision 1.173 2015/01/03 12:06:26 brouard
278: Summary: trying to detect cross-compilation
279:
1.173 brouard 280: Revision 1.172 2014/12/27 12:07:47 brouard
281: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
282:
1.172 brouard 283: Revision 1.171 2014/12/23 13:26:59 brouard
284: Summary: Back from Visual C
285:
286: Still problem with utsname.h on Windows
287:
1.171 brouard 288: Revision 1.170 2014/12/23 11:17:12 brouard
289: Summary: Cleaning some \%% back to %%
290:
291: The escape was mandatory for a specific compiler (which one?), but too many warnings.
292:
1.170 brouard 293: Revision 1.169 2014/12/22 23:08:31 brouard
294: Summary: 0.98p
295:
296: Outputs some informations on compiler used, OS etc. Testing on different platforms.
297:
1.169 brouard 298: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 299: Summary: update
1.169 brouard 300:
1.168 brouard 301: Revision 1.167 2014/12/22 13:50:56 brouard
302: Summary: Testing uname and compiler version and if compiled 32 or 64
303:
304: Testing on Linux 64
305:
1.167 brouard 306: Revision 1.166 2014/12/22 11:40:47 brouard
307: *** empty log message ***
308:
1.166 brouard 309: Revision 1.165 2014/12/16 11:20:36 brouard
310: Summary: After compiling on Visual C
311:
312: * imach.c (Module): Merging 1.61 to 1.162
313:
1.165 brouard 314: Revision 1.164 2014/12/16 10:52:11 brouard
315: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
316:
317: * imach.c (Module): Merging 1.61 to 1.162
318:
1.164 brouard 319: Revision 1.163 2014/12/16 10:30:11 brouard
320: * imach.c (Module): Merging 1.61 to 1.162
321:
1.163 brouard 322: Revision 1.162 2014/09/25 11:43:39 brouard
323: Summary: temporary backup 0.99!
324:
1.162 brouard 325: Revision 1.1 2014/09/16 11:06:58 brouard
326: Summary: With some code (wrong) for nlopt
327:
328: Author:
329:
330: Revision 1.161 2014/09/15 20:41:41 brouard
331: Summary: Problem with macro SQR on Intel compiler
332:
1.161 brouard 333: Revision 1.160 2014/09/02 09:24:05 brouard
334: *** empty log message ***
335:
1.160 brouard 336: Revision 1.159 2014/09/01 10:34:10 brouard
337: Summary: WIN32
338: Author: Brouard
339:
1.159 brouard 340: Revision 1.158 2014/08/27 17:11:51 brouard
341: *** empty log message ***
342:
1.158 brouard 343: Revision 1.157 2014/08/27 16:26:55 brouard
344: Summary: Preparing windows Visual studio version
345: Author: Brouard
346:
347: In order to compile on Visual studio, time.h is now correct and time_t
348: and tm struct should be used. difftime should be used but sometimes I
349: just make the differences in raw time format (time(&now).
350: Trying to suppress #ifdef LINUX
351: Add xdg-open for __linux in order to open default browser.
352:
1.157 brouard 353: Revision 1.156 2014/08/25 20:10:10 brouard
354: *** empty log message ***
355:
1.156 brouard 356: Revision 1.155 2014/08/25 18:32:34 brouard
357: Summary: New compile, minor changes
358: Author: Brouard
359:
1.155 brouard 360: Revision 1.154 2014/06/20 17:32:08 brouard
361: Summary: Outputs now all graphs of convergence to period prevalence
362:
1.154 brouard 363: Revision 1.153 2014/06/20 16:45:46 brouard
364: Summary: If 3 live state, convergence to period prevalence on same graph
365: Author: Brouard
366:
1.153 brouard 367: Revision 1.152 2014/06/18 17:54:09 brouard
368: Summary: open browser, use gnuplot on same dir than imach if not found in the path
369:
1.152 brouard 370: Revision 1.151 2014/06/18 16:43:30 brouard
371: *** empty log message ***
372:
1.151 brouard 373: Revision 1.150 2014/06/18 16:42:35 brouard
374: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
375: Author: brouard
376:
1.150 brouard 377: Revision 1.149 2014/06/18 15:51:14 brouard
378: Summary: Some fixes in parameter files errors
379: Author: Nicolas Brouard
380:
1.149 brouard 381: Revision 1.148 2014/06/17 17:38:48 brouard
382: Summary: Nothing new
383: Author: Brouard
384:
385: Just a new packaging for OS/X version 0.98nS
386:
1.148 brouard 387: Revision 1.147 2014/06/16 10:33:11 brouard
388: *** empty log message ***
389:
1.147 brouard 390: Revision 1.146 2014/06/16 10:20:28 brouard
391: Summary: Merge
392: Author: Brouard
393:
394: Merge, before building revised version.
395:
1.146 brouard 396: Revision 1.145 2014/06/10 21:23:15 brouard
397: Summary: Debugging with valgrind
398: Author: Nicolas Brouard
399:
400: Lot of changes in order to output the results with some covariates
401: After the Edimburgh REVES conference 2014, it seems mandatory to
402: improve the code.
403: No more memory valgrind error but a lot has to be done in order to
404: continue the work of splitting the code into subroutines.
405: Also, decodemodel has been improved. Tricode is still not
406: optimal. nbcode should be improved. Documentation has been added in
407: the source code.
408:
1.144 brouard 409: Revision 1.143 2014/01/26 09:45:38 brouard
410: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
411:
412: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
413: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
414:
1.143 brouard 415: Revision 1.142 2014/01/26 03:57:36 brouard
416: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
417:
418: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
419:
1.142 brouard 420: Revision 1.141 2014/01/26 02:42:01 brouard
421: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
422:
1.141 brouard 423: Revision 1.140 2011/09/02 10:37:54 brouard
424: Summary: times.h is ok with mingw32 now.
425:
1.140 brouard 426: Revision 1.139 2010/06/14 07:50:17 brouard
427: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
428: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
429:
1.139 brouard 430: Revision 1.138 2010/04/30 18:19:40 brouard
431: *** empty log message ***
432:
1.138 brouard 433: Revision 1.137 2010/04/29 18:11:38 brouard
434: (Module): Checking covariates for more complex models
435: than V1+V2. A lot of change to be done. Unstable.
436:
1.137 brouard 437: Revision 1.136 2010/04/26 20:30:53 brouard
438: (Module): merging some libgsl code. Fixing computation
439: of likelione (using inter/intrapolation if mle = 0) in order to
440: get same likelihood as if mle=1.
441: Some cleaning of code and comments added.
442:
1.136 brouard 443: Revision 1.135 2009/10/29 15:33:14 brouard
444: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
445:
1.135 brouard 446: Revision 1.134 2009/10/29 13:18:53 brouard
447: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
448:
1.134 brouard 449: Revision 1.133 2009/07/06 10:21:25 brouard
450: just nforces
451:
1.133 brouard 452: Revision 1.132 2009/07/06 08:22:05 brouard
453: Many tings
454:
1.132 brouard 455: Revision 1.131 2009/06/20 16:22:47 brouard
456: Some dimensions resccaled
457:
1.131 brouard 458: Revision 1.130 2009/05/26 06:44:34 brouard
459: (Module): Max Covariate is now set to 20 instead of 8. A
460: lot of cleaning with variables initialized to 0. Trying to make
461: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
462:
1.130 brouard 463: Revision 1.129 2007/08/31 13:49:27 lievre
464: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
465:
1.129 lievre 466: Revision 1.128 2006/06/30 13:02:05 brouard
467: (Module): Clarifications on computing e.j
468:
1.128 brouard 469: Revision 1.127 2006/04/28 18:11:50 brouard
470: (Module): Yes the sum of survivors was wrong since
471: imach-114 because nhstepm was no more computed in the age
472: loop. Now we define nhstepma in the age loop.
473: (Module): In order to speed up (in case of numerous covariates) we
474: compute health expectancies (without variances) in a first step
475: and then all the health expectancies with variances or standard
476: deviation (needs data from the Hessian matrices) which slows the
477: computation.
478: In the future we should be able to stop the program is only health
479: expectancies and graph are needed without standard deviations.
480:
1.127 brouard 481: Revision 1.126 2006/04/28 17:23:28 brouard
482: (Module): Yes the sum of survivors was wrong since
483: imach-114 because nhstepm was no more computed in the age
484: loop. Now we define nhstepma in the age loop.
485: Version 0.98h
486:
1.126 brouard 487: Revision 1.125 2006/04/04 15:20:31 lievre
488: Errors in calculation of health expectancies. Age was not initialized.
489: Forecasting file added.
490:
491: Revision 1.124 2006/03/22 17:13:53 lievre
492: Parameters are printed with %lf instead of %f (more numbers after the comma).
493: The log-likelihood is printed in the log file
494:
495: Revision 1.123 2006/03/20 10:52:43 brouard
496: * imach.c (Module): <title> changed, corresponds to .htm file
497: name. <head> headers where missing.
498:
499: * imach.c (Module): Weights can have a decimal point as for
500: English (a comma might work with a correct LC_NUMERIC environment,
501: otherwise the weight is truncated).
502: Modification of warning when the covariates values are not 0 or
503: 1.
504: Version 0.98g
505:
506: Revision 1.122 2006/03/20 09:45:41 brouard
507: (Module): Weights can have a decimal point as for
508: English (a comma might work with a correct LC_NUMERIC environment,
509: otherwise the weight is truncated).
510: Modification of warning when the covariates values are not 0 or
511: 1.
512: Version 0.98g
513:
514: Revision 1.121 2006/03/16 17:45:01 lievre
515: * imach.c (Module): Comments concerning covariates added
516:
517: * imach.c (Module): refinements in the computation of lli if
518: status=-2 in order to have more reliable computation if stepm is
519: not 1 month. Version 0.98f
520:
521: Revision 1.120 2006/03/16 15:10:38 lievre
522: (Module): refinements in the computation of lli if
523: status=-2 in order to have more reliable computation if stepm is
524: not 1 month. Version 0.98f
525:
526: Revision 1.119 2006/03/15 17:42:26 brouard
527: (Module): Bug if status = -2, the loglikelihood was
528: computed as likelihood omitting the logarithm. Version O.98e
529:
530: Revision 1.118 2006/03/14 18:20:07 brouard
531: (Module): varevsij Comments added explaining the second
532: table of variances if popbased=1 .
533: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
534: (Module): Function pstamp added
535: (Module): Version 0.98d
536:
537: Revision 1.117 2006/03/14 17:16:22 brouard
538: (Module): varevsij Comments added explaining the second
539: table of variances if popbased=1 .
540: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
541: (Module): Function pstamp added
542: (Module): Version 0.98d
543:
544: Revision 1.116 2006/03/06 10:29:27 brouard
545: (Module): Variance-covariance wrong links and
546: varian-covariance of ej. is needed (Saito).
547:
548: Revision 1.115 2006/02/27 12:17:45 brouard
549: (Module): One freematrix added in mlikeli! 0.98c
550:
551: Revision 1.114 2006/02/26 12:57:58 brouard
552: (Module): Some improvements in processing parameter
553: filename with strsep.
554:
555: Revision 1.113 2006/02/24 14:20:24 brouard
556: (Module): Memory leaks checks with valgrind and:
557: datafile was not closed, some imatrix were not freed and on matrix
558: allocation too.
559:
560: Revision 1.112 2006/01/30 09:55:26 brouard
561: (Module): Back to gnuplot.exe instead of wgnuplot.exe
562:
563: Revision 1.111 2006/01/25 20:38:18 brouard
564: (Module): Lots of cleaning and bugs added (Gompertz)
565: (Module): Comments can be added in data file. Missing date values
566: can be a simple dot '.'.
567:
568: Revision 1.110 2006/01/25 00:51:50 brouard
569: (Module): Lots of cleaning and bugs added (Gompertz)
570:
571: Revision 1.109 2006/01/24 19:37:15 brouard
572: (Module): Comments (lines starting with a #) are allowed in data.
573:
574: Revision 1.108 2006/01/19 18:05:42 lievre
575: Gnuplot problem appeared...
576: To be fixed
577:
578: Revision 1.107 2006/01/19 16:20:37 brouard
579: Test existence of gnuplot in imach path
580:
581: Revision 1.106 2006/01/19 13:24:36 brouard
582: Some cleaning and links added in html output
583:
584: Revision 1.105 2006/01/05 20:23:19 lievre
585: *** empty log message ***
586:
587: Revision 1.104 2005/09/30 16:11:43 lievre
588: (Module): sump fixed, loop imx fixed, and simplifications.
589: (Module): If the status is missing at the last wave but we know
590: that the person is alive, then we can code his/her status as -2
591: (instead of missing=-1 in earlier versions) and his/her
592: contributions to the likelihood is 1 - Prob of dying from last
593: health status (= 1-p13= p11+p12 in the easiest case of somebody in
594: the healthy state at last known wave). Version is 0.98
595:
596: Revision 1.103 2005/09/30 15:54:49 lievre
597: (Module): sump fixed, loop imx fixed, and simplifications.
598:
599: Revision 1.102 2004/09/15 17:31:30 brouard
600: Add the possibility to read data file including tab characters.
601:
602: Revision 1.101 2004/09/15 10:38:38 brouard
603: Fix on curr_time
604:
605: Revision 1.100 2004/07/12 18:29:06 brouard
606: Add version for Mac OS X. Just define UNIX in Makefile
607:
608: Revision 1.99 2004/06/05 08:57:40 brouard
609: *** empty log message ***
610:
611: Revision 1.98 2004/05/16 15:05:56 brouard
612: New version 0.97 . First attempt to estimate force of mortality
613: directly from the data i.e. without the need of knowing the health
614: state at each age, but using a Gompertz model: log u =a + b*age .
615: This is the basic analysis of mortality and should be done before any
616: other analysis, in order to test if the mortality estimated from the
617: cross-longitudinal survey is different from the mortality estimated
618: from other sources like vital statistic data.
619:
620: The same imach parameter file can be used but the option for mle should be -3.
621:
1.133 brouard 622: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 623: former routines in order to include the new code within the former code.
624:
625: The output is very simple: only an estimate of the intercept and of
626: the slope with 95% confident intervals.
627:
628: Current limitations:
629: A) Even if you enter covariates, i.e. with the
630: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
631: B) There is no computation of Life Expectancy nor Life Table.
632:
633: Revision 1.97 2004/02/20 13:25:42 lievre
634: Version 0.96d. Population forecasting command line is (temporarily)
635: suppressed.
636:
637: Revision 1.96 2003/07/15 15:38:55 brouard
638: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
639: rewritten within the same printf. Workaround: many printfs.
640:
641: Revision 1.95 2003/07/08 07:54:34 brouard
642: * imach.c (Repository):
643: (Repository): Using imachwizard code to output a more meaningful covariance
644: matrix (cov(a12,c31) instead of numbers.
645:
646: Revision 1.94 2003/06/27 13:00:02 brouard
647: Just cleaning
648:
649: Revision 1.93 2003/06/25 16:33:55 brouard
650: (Module): On windows (cygwin) function asctime_r doesn't
651: exist so I changed back to asctime which exists.
652: (Module): Version 0.96b
653:
654: Revision 1.92 2003/06/25 16:30:45 brouard
655: (Module): On windows (cygwin) function asctime_r doesn't
656: exist so I changed back to asctime which exists.
657:
658: Revision 1.91 2003/06/25 15:30:29 brouard
659: * imach.c (Repository): Duplicated warning errors corrected.
660: (Repository): Elapsed time after each iteration is now output. It
661: helps to forecast when convergence will be reached. Elapsed time
662: is stamped in powell. We created a new html file for the graphs
663: concerning matrix of covariance. It has extension -cov.htm.
664:
665: Revision 1.90 2003/06/24 12:34:15 brouard
666: (Module): Some bugs corrected for windows. Also, when
667: mle=-1 a template is output in file "or"mypar.txt with the design
668: of the covariance matrix to be input.
669:
670: Revision 1.89 2003/06/24 12:30:52 brouard
671: (Module): Some bugs corrected for windows. Also, when
672: mle=-1 a template is output in file "or"mypar.txt with the design
673: of the covariance matrix to be input.
674:
675: Revision 1.88 2003/06/23 17:54:56 brouard
676: * 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.
677:
678: Revision 1.87 2003/06/18 12:26:01 brouard
679: Version 0.96
680:
681: Revision 1.86 2003/06/17 20:04:08 brouard
682: (Module): Change position of html and gnuplot routines and added
683: routine fileappend.
684:
685: Revision 1.85 2003/06/17 13:12:43 brouard
686: * imach.c (Repository): Check when date of death was earlier that
687: current date of interview. It may happen when the death was just
688: prior to the death. In this case, dh was negative and likelihood
689: was wrong (infinity). We still send an "Error" but patch by
690: assuming that the date of death was just one stepm after the
691: interview.
692: (Repository): Because some people have very long ID (first column)
693: we changed int to long in num[] and we added a new lvector for
694: memory allocation. But we also truncated to 8 characters (left
695: truncation)
696: (Repository): No more line truncation errors.
697:
698: Revision 1.84 2003/06/13 21:44:43 brouard
699: * imach.c (Repository): Replace "freqsummary" at a correct
700: place. It differs from routine "prevalence" which may be called
701: many times. Probs is memory consuming and must be used with
702: parcimony.
703: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
704:
705: Revision 1.83 2003/06/10 13:39:11 lievre
706: *** empty log message ***
707:
708: Revision 1.82 2003/06/05 15:57:20 brouard
709: Add log in imach.c and fullversion number is now printed.
710:
711: */
712: /*
713: Interpolated Markov Chain
714:
715: Short summary of the programme:
716:
1.227 brouard 717: This program computes Healthy Life Expectancies or State-specific
718: (if states aren't health statuses) Expectancies from
719: cross-longitudinal data. Cross-longitudinal data consist in:
720:
721: -1- a first survey ("cross") where individuals from different ages
722: are interviewed on their health status or degree of disability (in
723: the case of a health survey which is our main interest)
724:
725: -2- at least a second wave of interviews ("longitudinal") which
726: measure each change (if any) in individual health status. Health
727: expectancies are computed from the time spent in each health state
728: according to a model. More health states you consider, more time is
729: necessary to reach the Maximum Likelihood of the parameters involved
730: in the model. The simplest model is the multinomial logistic model
731: where pij is the probability to be observed in state j at the second
732: wave conditional to be observed in state i at the first
733: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
734: etc , where 'age' is age and 'sex' is a covariate. If you want to
735: have a more complex model than "constant and age", you should modify
736: the program where the markup *Covariates have to be included here
737: again* invites you to do it. More covariates you add, slower the
1.126 brouard 738: convergence.
739:
740: The advantage of this computer programme, compared to a simple
741: multinomial logistic model, is clear when the delay between waves is not
742: identical for each individual. Also, if a individual missed an
743: intermediate interview, the information is lost, but taken into
744: account using an interpolation or extrapolation.
745:
746: hPijx is the probability to be observed in state i at age x+h
747: conditional to the observed state i at age x. The delay 'h' can be
748: split into an exact number (nh*stepm) of unobserved intermediate
749: states. This elementary transition (by month, quarter,
750: semester or year) is modelled as a multinomial logistic. The hPx
751: matrix is simply the matrix product of nh*stepm elementary matrices
752: and the contribution of each individual to the likelihood is simply
753: hPijx.
754:
755: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 756: of the life expectancies. It also computes the period (stable) prevalence.
757:
758: Back prevalence and projections:
1.227 brouard 759:
760: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
761: double agemaxpar, double ftolpl, int *ncvyearp, double
762: dateprev1,double dateprev2, int firstpass, int lastpass, int
763: mobilavproj)
764:
765: Computes the back prevalence limit for any combination of
766: covariate values k at any age between ageminpar and agemaxpar and
767: returns it in **bprlim. In the loops,
768:
769: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
770: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
771:
772: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 773: Computes for any combination of covariates k and any age between bage and fage
774: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
775: oldm=oldms;savm=savms;
1.227 brouard 776:
777: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 778: Computes the transition matrix starting at age 'age' over
779: 'nhstepm*hstepm*stepm' months (i.e. until
780: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 781: nhstepm*hstepm matrices.
782:
783: Returns p3mat[i][j][h] after calling
784: p3mat[i][j][h]=matprod2(newm,
785: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
786: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
787: oldm);
1.226 brouard 788:
789: Important routines
790:
791: - func (or funcone), computes logit (pij) distinguishing
792: o fixed variables (single or product dummies or quantitative);
793: o varying variables by:
794: (1) wave (single, product dummies, quantitative),
795: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
796: % fixed dummy (treated) or quantitative (not done because time-consuming);
797: % varying dummy (not done) or quantitative (not done);
798: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
799: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
800: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
801: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
802: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 803:
1.226 brouard 804:
805:
1.133 brouard 806: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
807: Institut national d'études démographiques, Paris.
1.126 brouard 808: This software have been partly granted by Euro-REVES, a concerted action
809: from the European Union.
810: It is copyrighted identically to a GNU software product, ie programme and
811: software can be distributed freely for non commercial use. Latest version
812: can be accessed at http://euroreves.ined.fr/imach .
813:
814: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
815: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
816:
817: **********************************************************************/
818: /*
819: main
820: read parameterfile
821: read datafile
822: concatwav
823: freqsummary
824: if (mle >= 1)
825: mlikeli
826: print results files
827: if mle==1
828: computes hessian
829: read end of parameter file: agemin, agemax, bage, fage, estepm
830: begin-prev-date,...
831: open gnuplot file
832: open html file
1.145 brouard 833: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
834: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
835: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
836: freexexit2 possible for memory heap.
837:
838: h Pij x | pij_nom ficrestpij
839: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
840: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
841: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
842:
843: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
844: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
845: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
846: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
847: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
848:
1.126 brouard 849: forecasting if prevfcast==1 prevforecast call prevalence()
850: health expectancies
851: Variance-covariance of DFLE
852: prevalence()
853: movingaverage()
854: varevsij()
855: if popbased==1 varevsij(,popbased)
856: total life expectancies
857: Variance of period (stable) prevalence
858: end
859: */
860:
1.187 brouard 861: /* #define DEBUG */
862: /* #define DEBUGBRENT */
1.203 brouard 863: /* #define DEBUGLINMIN */
864: /* #define DEBUGHESS */
865: #define DEBUGHESSIJ
1.224 brouard 866: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 867: #define POWELL /* Instead of NLOPT */
1.224 brouard 868: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 869: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
870: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 871:
872: #include <math.h>
873: #include <stdio.h>
874: #include <stdlib.h>
875: #include <string.h>
1.226 brouard 876: #include <ctype.h>
1.159 brouard 877:
878: #ifdef _WIN32
879: #include <io.h>
1.172 brouard 880: #include <windows.h>
881: #include <tchar.h>
1.159 brouard 882: #else
1.126 brouard 883: #include <unistd.h>
1.159 brouard 884: #endif
1.126 brouard 885:
886: #include <limits.h>
887: #include <sys/types.h>
1.171 brouard 888:
889: #if defined(__GNUC__)
890: #include <sys/utsname.h> /* Doesn't work on Windows */
891: #endif
892:
1.126 brouard 893: #include <sys/stat.h>
894: #include <errno.h>
1.159 brouard 895: /* extern int errno; */
1.126 brouard 896:
1.157 brouard 897: /* #ifdef LINUX */
898: /* #include <time.h> */
899: /* #include "timeval.h" */
900: /* #else */
901: /* #include <sys/time.h> */
902: /* #endif */
903:
1.126 brouard 904: #include <time.h>
905:
1.136 brouard 906: #ifdef GSL
907: #include <gsl/gsl_errno.h>
908: #include <gsl/gsl_multimin.h>
909: #endif
910:
1.167 brouard 911:
1.162 brouard 912: #ifdef NLOPT
913: #include <nlopt.h>
914: typedef struct {
915: double (* function)(double [] );
916: } myfunc_data ;
917: #endif
918:
1.126 brouard 919: /* #include <libintl.h> */
920: /* #define _(String) gettext (String) */
921:
1.251 brouard 922: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 923:
924: #define GNUPLOTPROGRAM "gnuplot"
925: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
926: #define FILENAMELENGTH 132
927:
928: #define GLOCK_ERROR_NOPATH -1 /* empty path */
929: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
930:
1.144 brouard 931: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
932: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 933:
934: #define NINTERVMAX 8
1.144 brouard 935: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
936: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
937: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 938: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 939: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
940: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 941: #define MAXN 20000
1.144 brouard 942: #define YEARM 12. /**< Number of months per year */
1.218 brouard 943: /* #define AGESUP 130 */
944: #define AGESUP 150
945: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 946: #define AGEBASE 40
1.194 brouard 947: #define AGEOVERFLOW 1.e20
1.164 brouard 948: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 949: #ifdef _WIN32
950: #define DIRSEPARATOR '\\'
951: #define CHARSEPARATOR "\\"
952: #define ODIRSEPARATOR '/'
953: #else
1.126 brouard 954: #define DIRSEPARATOR '/'
955: #define CHARSEPARATOR "/"
956: #define ODIRSEPARATOR '\\'
957: #endif
958:
1.253 ! brouard 959: /* $Id: imach.c,v 1.252 2016/09/15 21:15:37 brouard Exp $ */
1.126 brouard 960: /* $State: Exp $ */
1.196 brouard 961: #include "version.h"
962: char version[]=__IMACH_VERSION__;
1.224 brouard 963: 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";
1.253 ! brouard 964: char fullversion[]="$Revision: 1.252 $ $Date: 2016/09/15 21:15:37 $";
1.126 brouard 965: char strstart[80];
966: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 967: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 968: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 969: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
970: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
971: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 972: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
973: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 974: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
975: int cptcovprodnoage=0; /**< Number of covariate products without age */
976: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 977: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
978: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 979: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 980: int nsd=0; /**< Total number of single dummy variables (output) */
981: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 982: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 983: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 984: int ntveff=0; /**< ntveff number of effective time varying variables */
985: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 986: int cptcov=0; /* Working variable */
1.218 brouard 987: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 988: int npar=NPARMAX;
989: int nlstate=2; /* Number of live states */
990: int ndeath=1; /* Number of dead states */
1.130 brouard 991: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 992: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 993: int popbased=0;
994:
995: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 996: int maxwav=0; /* Maxim number of waves */
997: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
998: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
999: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1000: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1001: int mle=1, weightopt=0;
1.126 brouard 1002: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1003: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1004: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1005: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1006: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1007: int selected(int kvar); /* Is covariate kvar selected for printing results */
1008:
1.130 brouard 1009: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1010: double **matprod2(); /* test */
1.126 brouard 1011: double **oldm, **newm, **savm; /* Working pointers to matrices */
1012: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1013: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1014:
1.136 brouard 1015: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1016: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1017: FILE *ficlog, *ficrespow;
1.130 brouard 1018: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1019: double fretone; /* Only one call to likelihood */
1.130 brouard 1020: long ipmx=0; /* Number of contributions */
1.126 brouard 1021: double sw; /* Sum of weights */
1022: char filerespow[FILENAMELENGTH];
1023: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1024: FILE *ficresilk;
1025: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1026: FILE *ficresprobmorprev;
1027: FILE *fichtm, *fichtmcov; /* Html File */
1028: FILE *ficreseij;
1029: char filerese[FILENAMELENGTH];
1030: FILE *ficresstdeij;
1031: char fileresstde[FILENAMELENGTH];
1032: FILE *ficrescveij;
1033: char filerescve[FILENAMELENGTH];
1034: FILE *ficresvij;
1035: char fileresv[FILENAMELENGTH];
1036: FILE *ficresvpl;
1037: char fileresvpl[FILENAMELENGTH];
1038: char title[MAXLINE];
1.234 brouard 1039: char model[MAXLINE]; /**< The model line */
1.217 brouard 1040: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1041: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1042: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1043: char command[FILENAMELENGTH];
1044: int outcmd=0;
1045:
1.217 brouard 1046: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1047: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1048: char filelog[FILENAMELENGTH]; /* Log file */
1049: char filerest[FILENAMELENGTH];
1050: char fileregp[FILENAMELENGTH];
1051: char popfile[FILENAMELENGTH];
1052:
1053: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1054:
1.157 brouard 1055: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1056: /* struct timezone tzp; */
1057: /* extern int gettimeofday(); */
1058: struct tm tml, *gmtime(), *localtime();
1059:
1060: extern time_t time();
1061:
1062: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1063: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1064: struct tm tm;
1065:
1.126 brouard 1066: char strcurr[80], strfor[80];
1067:
1068: char *endptr;
1069: long lval;
1070: double dval;
1071:
1072: #define NR_END 1
1073: #define FREE_ARG char*
1074: #define FTOL 1.0e-10
1075:
1076: #define NRANSI
1.240 brouard 1077: #define ITMAX 200
1078: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1079:
1080: #define TOL 2.0e-4
1081:
1082: #define CGOLD 0.3819660
1083: #define ZEPS 1.0e-10
1084: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1085:
1086: #define GOLD 1.618034
1087: #define GLIMIT 100.0
1088: #define TINY 1.0e-20
1089:
1090: static double maxarg1,maxarg2;
1091: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1092: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1093:
1094: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1095: #define rint(a) floor(a+0.5)
1.166 brouard 1096: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1097: #define mytinydouble 1.0e-16
1.166 brouard 1098: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1099: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1100: /* static double dsqrarg; */
1101: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1102: static double sqrarg;
1103: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1104: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1105: int agegomp= AGEGOMP;
1106:
1107: int imx;
1108: int stepm=1;
1109: /* Stepm, step in month: minimum step interpolation*/
1110:
1111: int estepm;
1112: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1113:
1114: int m,nb;
1115: long *num;
1.197 brouard 1116: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1117: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1118: covariate for which somebody answered excluding
1119: undefined. Usually 2: 0 and 1. */
1120: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1121: covariate for which somebody answered including
1122: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1123: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1124: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1125: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1126: double *ageexmed,*agecens;
1127: double dateintmean=0;
1128:
1129: double *weight;
1130: int **s; /* Status */
1.141 brouard 1131: double *agedc;
1.145 brouard 1132: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1133: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1134: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1135: double **coqvar; /* Fixed quantitative covariate iqv */
1136: double ***cotvar; /* Time varying covariate itv */
1137: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1138: double idx;
1139: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1140: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1141: /*k 1 2 3 4 5 6 7 8 9 */
1142: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1143: /* Tndvar[k] 1 2 3 4 5 */
1144: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1145: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1146: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1147: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1148: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1149: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1150: /* Tprod[i]=k 4 7 */
1151: /* Tage[i]=k 5 8 */
1152: /* */
1153: /* Type */
1154: /* V 1 2 3 4 5 */
1155: /* F F V V V */
1156: /* D Q D D Q */
1157: /* */
1158: int *TvarsD;
1159: int *TvarsDind;
1160: int *TvarsQ;
1161: int *TvarsQind;
1162:
1.235 brouard 1163: #define MAXRESULTLINES 10
1164: int nresult=0;
1165: int TKresult[MAXRESULTLINES];
1.237 brouard 1166: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1167: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1168: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1169: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1170: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1171: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1172:
1.234 brouard 1173: /* 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 1174: 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 */
1175: 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 */
1176: 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 */
1177: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1178: 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 */
1179: 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 1180: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1181: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1182: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1183: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1184: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1185: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1186: 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 */
1187: 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 */
1188:
1.230 brouard 1189: int *Tvarsel; /**< Selected covariates for output */
1190: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1191: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1192: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1193: 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 1194: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1195: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1196: int *Tage;
1.227 brouard 1197: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1198: 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 1199: 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*/
1200: 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 1201: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1202: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1203: int **Tvard;
1204: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1205: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1206: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1207: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1208: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1209: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1210: double *lsurv, *lpop, *tpop;
1211:
1.231 brouard 1212: #define FD 1; /* Fixed dummy covariate */
1213: #define FQ 2; /* Fixed quantitative covariate */
1214: #define FP 3; /* Fixed product covariate */
1215: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1216: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1217: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1218: #define VD 10; /* Varying dummy covariate */
1219: #define VQ 11; /* Varying quantitative covariate */
1220: #define VP 12; /* Varying product covariate */
1221: #define VPDD 13; /* Varying product dummy*dummy covariate */
1222: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1223: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1224: #define APFD 16; /* Age product * fixed dummy covariate */
1225: #define APFQ 17; /* Age product * fixed quantitative covariate */
1226: #define APVD 18; /* Age product * varying dummy covariate */
1227: #define APVQ 19; /* Age product * varying quantitative covariate */
1228:
1229: #define FTYPE 1; /* Fixed covariate */
1230: #define VTYPE 2; /* Varying covariate (loop in wave) */
1231: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1232:
1233: struct kmodel{
1234: int maintype; /* main type */
1235: int subtype; /* subtype */
1236: };
1237: struct kmodel modell[NCOVMAX];
1238:
1.143 brouard 1239: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1240: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1241:
1242: /**************** split *************************/
1243: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1244: {
1245: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1246: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1247: */
1248: char *ss; /* pointer */
1.186 brouard 1249: int l1=0, l2=0; /* length counters */
1.126 brouard 1250:
1251: l1 = strlen(path ); /* length of path */
1252: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1253: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1254: if ( ss == NULL ) { /* no directory, so determine current directory */
1255: strcpy( name, path ); /* we got the fullname name because no directory */
1256: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1257: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1258: /* get current working directory */
1259: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1260: #ifdef WIN32
1261: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1262: #else
1263: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1264: #endif
1.126 brouard 1265: return( GLOCK_ERROR_GETCWD );
1266: }
1267: /* got dirc from getcwd*/
1268: printf(" DIRC = %s \n",dirc);
1.205 brouard 1269: } else { /* strip directory from path */
1.126 brouard 1270: ss++; /* after this, the filename */
1271: l2 = strlen( ss ); /* length of filename */
1272: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1273: strcpy( name, ss ); /* save file name */
1274: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1275: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1276: printf(" DIRC2 = %s \n",dirc);
1277: }
1278: /* We add a separator at the end of dirc if not exists */
1279: l1 = strlen( dirc ); /* length of directory */
1280: if( dirc[l1-1] != DIRSEPARATOR ){
1281: dirc[l1] = DIRSEPARATOR;
1282: dirc[l1+1] = 0;
1283: printf(" DIRC3 = %s \n",dirc);
1284: }
1285: ss = strrchr( name, '.' ); /* find last / */
1286: if (ss >0){
1287: ss++;
1288: strcpy(ext,ss); /* save extension */
1289: l1= strlen( name);
1290: l2= strlen(ss)+1;
1291: strncpy( finame, name, l1-l2);
1292: finame[l1-l2]= 0;
1293: }
1294:
1295: return( 0 ); /* we're done */
1296: }
1297:
1298:
1299: /******************************************/
1300:
1301: void replace_back_to_slash(char *s, char*t)
1302: {
1303: int i;
1304: int lg=0;
1305: i=0;
1306: lg=strlen(t);
1307: for(i=0; i<= lg; i++) {
1308: (s[i] = t[i]);
1309: if (t[i]== '\\') s[i]='/';
1310: }
1311: }
1312:
1.132 brouard 1313: char *trimbb(char *out, char *in)
1.137 brouard 1314: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1315: char *s;
1316: s=out;
1317: while (*in != '\0'){
1.137 brouard 1318: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1319: in++;
1320: }
1321: *out++ = *in++;
1322: }
1323: *out='\0';
1324: return s;
1325: }
1326:
1.187 brouard 1327: /* char *substrchaine(char *out, char *in, char *chain) */
1328: /* { */
1329: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1330: /* char *s, *t; */
1331: /* t=in;s=out; */
1332: /* while ((*in != *chain) && (*in != '\0')){ */
1333: /* *out++ = *in++; */
1334: /* } */
1335:
1336: /* /\* *in matches *chain *\/ */
1337: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1338: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1339: /* } */
1340: /* in--; chain--; */
1341: /* while ( (*in != '\0')){ */
1342: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1343: /* *out++ = *in++; */
1344: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1345: /* } */
1346: /* *out='\0'; */
1347: /* out=s; */
1348: /* return out; */
1349: /* } */
1350: char *substrchaine(char *out, char *in, char *chain)
1351: {
1352: /* Substract chain 'chain' from 'in', return and output 'out' */
1353: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1354:
1355: char *strloc;
1356:
1357: strcpy (out, in);
1358: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1359: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1360: if(strloc != NULL){
1361: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1362: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1363: /* strcpy (strloc, strloc +strlen(chain));*/
1364: }
1365: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1366: return out;
1367: }
1368:
1369:
1.145 brouard 1370: char *cutl(char *blocc, char *alocc, char *in, char occ)
1371: {
1.187 brouard 1372: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1373: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1374: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1375: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1376: */
1.160 brouard 1377: char *s, *t;
1.145 brouard 1378: t=in;s=in;
1379: while ((*in != occ) && (*in != '\0')){
1380: *alocc++ = *in++;
1381: }
1382: if( *in == occ){
1383: *(alocc)='\0';
1384: s=++in;
1385: }
1386:
1387: if (s == t) {/* occ not found */
1388: *(alocc-(in-s))='\0';
1389: in=s;
1390: }
1391: while ( *in != '\0'){
1392: *blocc++ = *in++;
1393: }
1394:
1395: *blocc='\0';
1396: return t;
1397: }
1.137 brouard 1398: char *cutv(char *blocc, char *alocc, char *in, char occ)
1399: {
1.187 brouard 1400: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1401: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1402: gives blocc="abcdef2ghi" and alocc="j".
1403: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1404: */
1405: char *s, *t;
1406: t=in;s=in;
1407: while (*in != '\0'){
1408: while( *in == occ){
1409: *blocc++ = *in++;
1410: s=in;
1411: }
1412: *blocc++ = *in++;
1413: }
1414: if (s == t) /* occ not found */
1415: *(blocc-(in-s))='\0';
1416: else
1417: *(blocc-(in-s)-1)='\0';
1418: in=s;
1419: while ( *in != '\0'){
1420: *alocc++ = *in++;
1421: }
1422:
1423: *alocc='\0';
1424: return s;
1425: }
1426:
1.126 brouard 1427: int nbocc(char *s, char occ)
1428: {
1429: int i,j=0;
1430: int lg=20;
1431: i=0;
1432: lg=strlen(s);
1433: for(i=0; i<= lg; i++) {
1.234 brouard 1434: if (s[i] == occ ) j++;
1.126 brouard 1435: }
1436: return j;
1437: }
1438:
1.137 brouard 1439: /* void cutv(char *u,char *v, char*t, char occ) */
1440: /* { */
1441: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1442: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1443: /* gives u="abcdef2ghi" and v="j" *\/ */
1444: /* int i,lg,j,p=0; */
1445: /* i=0; */
1446: /* lg=strlen(t); */
1447: /* for(j=0; j<=lg-1; j++) { */
1448: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1449: /* } */
1.126 brouard 1450:
1.137 brouard 1451: /* for(j=0; j<p; j++) { */
1452: /* (u[j] = t[j]); */
1453: /* } */
1454: /* u[p]='\0'; */
1.126 brouard 1455:
1.137 brouard 1456: /* for(j=0; j<= lg; j++) { */
1457: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1458: /* } */
1459: /* } */
1.126 brouard 1460:
1.160 brouard 1461: #ifdef _WIN32
1462: char * strsep(char **pp, const char *delim)
1463: {
1464: char *p, *q;
1465:
1466: if ((p = *pp) == NULL)
1467: return 0;
1468: if ((q = strpbrk (p, delim)) != NULL)
1469: {
1470: *pp = q + 1;
1471: *q = '\0';
1472: }
1473: else
1474: *pp = 0;
1475: return p;
1476: }
1477: #endif
1478:
1.126 brouard 1479: /********************** nrerror ********************/
1480:
1481: void nrerror(char error_text[])
1482: {
1483: fprintf(stderr,"ERREUR ...\n");
1484: fprintf(stderr,"%s\n",error_text);
1485: exit(EXIT_FAILURE);
1486: }
1487: /*********************** vector *******************/
1488: double *vector(int nl, int nh)
1489: {
1490: double *v;
1491: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1492: if (!v) nrerror("allocation failure in vector");
1493: return v-nl+NR_END;
1494: }
1495:
1496: /************************ free vector ******************/
1497: void free_vector(double*v, int nl, int nh)
1498: {
1499: free((FREE_ARG)(v+nl-NR_END));
1500: }
1501:
1502: /************************ivector *******************************/
1503: int *ivector(long nl,long nh)
1504: {
1505: int *v;
1506: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1507: if (!v) nrerror("allocation failure in ivector");
1508: return v-nl+NR_END;
1509: }
1510:
1511: /******************free ivector **************************/
1512: void free_ivector(int *v, long nl, long nh)
1513: {
1514: free((FREE_ARG)(v+nl-NR_END));
1515: }
1516:
1517: /************************lvector *******************************/
1518: long *lvector(long nl,long nh)
1519: {
1520: long *v;
1521: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1522: if (!v) nrerror("allocation failure in ivector");
1523: return v-nl+NR_END;
1524: }
1525:
1526: /******************free lvector **************************/
1527: void free_lvector(long *v, long nl, long nh)
1528: {
1529: free((FREE_ARG)(v+nl-NR_END));
1530: }
1531:
1532: /******************* imatrix *******************************/
1533: int **imatrix(long nrl, long nrh, long ncl, long nch)
1534: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1535: {
1536: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1537: int **m;
1538:
1539: /* allocate pointers to rows */
1540: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1541: if (!m) nrerror("allocation failure 1 in matrix()");
1542: m += NR_END;
1543: m -= nrl;
1544:
1545:
1546: /* allocate rows and set pointers to them */
1547: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1548: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1549: m[nrl] += NR_END;
1550: m[nrl] -= ncl;
1551:
1552: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1553:
1554: /* return pointer to array of pointers to rows */
1555: return m;
1556: }
1557:
1558: /****************** free_imatrix *************************/
1559: void free_imatrix(m,nrl,nrh,ncl,nch)
1560: int **m;
1561: long nch,ncl,nrh,nrl;
1562: /* free an int matrix allocated by imatrix() */
1563: {
1564: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1565: free((FREE_ARG) (m+nrl-NR_END));
1566: }
1567:
1568: /******************* matrix *******************************/
1569: double **matrix(long nrl, long nrh, long ncl, long nch)
1570: {
1571: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1572: double **m;
1573:
1574: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1575: if (!m) nrerror("allocation failure 1 in matrix()");
1576: m += NR_END;
1577: m -= nrl;
1578:
1579: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1580: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1581: m[nrl] += NR_END;
1582: m[nrl] -= ncl;
1583:
1584: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1585: return m;
1.145 brouard 1586: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1587: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1588: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1589: */
1590: }
1591:
1592: /*************************free matrix ************************/
1593: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1594: {
1595: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1596: free((FREE_ARG)(m+nrl-NR_END));
1597: }
1598:
1599: /******************* ma3x *******************************/
1600: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1601: {
1602: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1603: double ***m;
1604:
1605: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1606: if (!m) nrerror("allocation failure 1 in matrix()");
1607: m += NR_END;
1608: m -= nrl;
1609:
1610: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1611: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1612: m[nrl] += NR_END;
1613: m[nrl] -= ncl;
1614:
1615: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1616:
1617: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1618: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1619: m[nrl][ncl] += NR_END;
1620: m[nrl][ncl] -= nll;
1621: for (j=ncl+1; j<=nch; j++)
1622: m[nrl][j]=m[nrl][j-1]+nlay;
1623:
1624: for (i=nrl+1; i<=nrh; i++) {
1625: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1626: for (j=ncl+1; j<=nch; j++)
1627: m[i][j]=m[i][j-1]+nlay;
1628: }
1629: return m;
1630: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1631: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1632: */
1633: }
1634:
1635: /*************************free ma3x ************************/
1636: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1637: {
1638: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1639: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1640: free((FREE_ARG)(m+nrl-NR_END));
1641: }
1642:
1643: /*************** function subdirf ***********/
1644: char *subdirf(char fileres[])
1645: {
1646: /* Caution optionfilefiname is hidden */
1647: strcpy(tmpout,optionfilefiname);
1648: strcat(tmpout,"/"); /* Add to the right */
1649: strcat(tmpout,fileres);
1650: return tmpout;
1651: }
1652:
1653: /*************** function subdirf2 ***********/
1654: char *subdirf2(char fileres[], char *preop)
1655: {
1656:
1657: /* Caution optionfilefiname is hidden */
1658: strcpy(tmpout,optionfilefiname);
1659: strcat(tmpout,"/");
1660: strcat(tmpout,preop);
1661: strcat(tmpout,fileres);
1662: return tmpout;
1663: }
1664:
1665: /*************** function subdirf3 ***********/
1666: char *subdirf3(char fileres[], char *preop, char *preop2)
1667: {
1668:
1669: /* Caution optionfilefiname is hidden */
1670: strcpy(tmpout,optionfilefiname);
1671: strcat(tmpout,"/");
1672: strcat(tmpout,preop);
1673: strcat(tmpout,preop2);
1674: strcat(tmpout,fileres);
1675: return tmpout;
1676: }
1.213 brouard 1677:
1678: /*************** function subdirfext ***********/
1679: char *subdirfext(char fileres[], char *preop, char *postop)
1680: {
1681:
1682: strcpy(tmpout,preop);
1683: strcat(tmpout,fileres);
1684: strcat(tmpout,postop);
1685: return tmpout;
1686: }
1.126 brouard 1687:
1.213 brouard 1688: /*************** function subdirfext3 ***********/
1689: char *subdirfext3(char fileres[], char *preop, char *postop)
1690: {
1691:
1692: /* Caution optionfilefiname is hidden */
1693: strcpy(tmpout,optionfilefiname);
1694: strcat(tmpout,"/");
1695: strcat(tmpout,preop);
1696: strcat(tmpout,fileres);
1697: strcat(tmpout,postop);
1698: return tmpout;
1699: }
1700:
1.162 brouard 1701: char *asc_diff_time(long time_sec, char ascdiff[])
1702: {
1703: long sec_left, days, hours, minutes;
1704: days = (time_sec) / (60*60*24);
1705: sec_left = (time_sec) % (60*60*24);
1706: hours = (sec_left) / (60*60) ;
1707: sec_left = (sec_left) %(60*60);
1708: minutes = (sec_left) /60;
1709: sec_left = (sec_left) % (60);
1710: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1711: return ascdiff;
1712: }
1713:
1.126 brouard 1714: /***************** f1dim *************************/
1715: extern int ncom;
1716: extern double *pcom,*xicom;
1717: extern double (*nrfunc)(double []);
1718:
1719: double f1dim(double x)
1720: {
1721: int j;
1722: double f;
1723: double *xt;
1724:
1725: xt=vector(1,ncom);
1726: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1727: f=(*nrfunc)(xt);
1728: free_vector(xt,1,ncom);
1729: return f;
1730: }
1731:
1732: /*****************brent *************************/
1733: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1734: {
1735: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1736: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1737: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1738: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1739: * returned function value.
1740: */
1.126 brouard 1741: int iter;
1742: double a,b,d,etemp;
1.159 brouard 1743: double fu=0,fv,fw,fx;
1.164 brouard 1744: double ftemp=0.;
1.126 brouard 1745: double p,q,r,tol1,tol2,u,v,w,x,xm;
1746: double e=0.0;
1747:
1748: a=(ax < cx ? ax : cx);
1749: b=(ax > cx ? ax : cx);
1750: x=w=v=bx;
1751: fw=fv=fx=(*f)(x);
1752: for (iter=1;iter<=ITMAX;iter++) {
1753: xm=0.5*(a+b);
1754: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1755: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1756: printf(".");fflush(stdout);
1757: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1758: #ifdef DEBUGBRENT
1.126 brouard 1759: 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);
1760: 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);
1761: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1762: #endif
1763: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1764: *xmin=x;
1765: return fx;
1766: }
1767: ftemp=fu;
1768: if (fabs(e) > tol1) {
1769: r=(x-w)*(fx-fv);
1770: q=(x-v)*(fx-fw);
1771: p=(x-v)*q-(x-w)*r;
1772: q=2.0*(q-r);
1773: if (q > 0.0) p = -p;
1774: q=fabs(q);
1775: etemp=e;
1776: e=d;
1777: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1778: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1779: else {
1.224 brouard 1780: d=p/q;
1781: u=x+d;
1782: if (u-a < tol2 || b-u < tol2)
1783: d=SIGN(tol1,xm-x);
1.126 brouard 1784: }
1785: } else {
1786: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1787: }
1788: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1789: fu=(*f)(u);
1790: if (fu <= fx) {
1791: if (u >= x) a=x; else b=x;
1792: SHFT(v,w,x,u)
1.183 brouard 1793: SHFT(fv,fw,fx,fu)
1794: } else {
1795: if (u < x) a=u; else b=u;
1796: if (fu <= fw || w == x) {
1.224 brouard 1797: v=w;
1798: w=u;
1799: fv=fw;
1800: fw=fu;
1.183 brouard 1801: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1802: v=u;
1803: fv=fu;
1.183 brouard 1804: }
1805: }
1.126 brouard 1806: }
1807: nrerror("Too many iterations in brent");
1808: *xmin=x;
1809: return fx;
1810: }
1811:
1812: /****************** mnbrak ***********************/
1813:
1814: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1815: double (*func)(double))
1.183 brouard 1816: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1817: the downhill direction (defined by the function as evaluated at the initial points) and returns
1818: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1819: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1820: */
1.126 brouard 1821: double ulim,u,r,q, dum;
1822: double fu;
1.187 brouard 1823:
1824: double scale=10.;
1825: int iterscale=0;
1826:
1827: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1828: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1829:
1830:
1831: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1832: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1833: /* *bx = *ax - (*ax - *bx)/scale; */
1834: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1835: /* } */
1836:
1.126 brouard 1837: if (*fb > *fa) {
1838: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1839: SHFT(dum,*fb,*fa,dum)
1840: }
1.126 brouard 1841: *cx=(*bx)+GOLD*(*bx-*ax);
1842: *fc=(*func)(*cx);
1.183 brouard 1843: #ifdef DEBUG
1.224 brouard 1844: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1845: 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 1846: #endif
1.224 brouard 1847: 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 1848: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1849: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1850: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1851: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1852: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1853: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1854: fu=(*func)(u);
1.163 brouard 1855: #ifdef DEBUG
1856: /* f(x)=A(x-u)**2+f(u) */
1857: double A, fparabu;
1858: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1859: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1860: 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);
1861: 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 1862: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1863: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1864: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1865: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1866: #endif
1.184 brouard 1867: #ifdef MNBRAKORIGINAL
1.183 brouard 1868: #else
1.191 brouard 1869: /* if (fu > *fc) { */
1870: /* #ifdef DEBUG */
1871: /* printf("mnbrak4 fu > fc \n"); */
1872: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1873: /* #endif */
1874: /* /\* 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 *\\/ *\/ */
1875: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1876: /* dum=u; /\* Shifting c and u *\/ */
1877: /* u = *cx; */
1878: /* *cx = dum; */
1879: /* dum = fu; */
1880: /* fu = *fc; */
1881: /* *fc =dum; */
1882: /* } else { /\* end *\/ */
1883: /* #ifdef DEBUG */
1884: /* printf("mnbrak3 fu < fc \n"); */
1885: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1886: /* #endif */
1887: /* dum=u; /\* Shifting c and u *\/ */
1888: /* u = *cx; */
1889: /* *cx = dum; */
1890: /* dum = fu; */
1891: /* fu = *fc; */
1892: /* *fc =dum; */
1893: /* } */
1.224 brouard 1894: #ifdef DEBUGMNBRAK
1895: double A, fparabu;
1896: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1897: fparabu= *fa - A*(*ax-u)*(*ax-u);
1898: 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);
1899: 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 1900: #endif
1.191 brouard 1901: dum=u; /* Shifting c and u */
1902: u = *cx;
1903: *cx = dum;
1904: dum = fu;
1905: fu = *fc;
1906: *fc =dum;
1.183 brouard 1907: #endif
1.162 brouard 1908: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1909: #ifdef DEBUG
1.224 brouard 1910: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1911: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1912: #endif
1.126 brouard 1913: fu=(*func)(u);
1914: if (fu < *fc) {
1.183 brouard 1915: #ifdef DEBUG
1.224 brouard 1916: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1917: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1918: #endif
1919: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1920: SHFT(*fb,*fc,fu,(*func)(u))
1921: #ifdef DEBUG
1922: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1923: #endif
1924: }
1.162 brouard 1925: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1926: #ifdef DEBUG
1.224 brouard 1927: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1928: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1929: #endif
1.126 brouard 1930: u=ulim;
1931: fu=(*func)(u);
1.183 brouard 1932: } else { /* u could be left to b (if r > q parabola has a maximum) */
1933: #ifdef DEBUG
1.224 brouard 1934: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1935: 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 1936: #endif
1.126 brouard 1937: u=(*cx)+GOLD*(*cx-*bx);
1938: fu=(*func)(u);
1.224 brouard 1939: #ifdef DEBUG
1940: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1941: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1942: #endif
1.183 brouard 1943: } /* end tests */
1.126 brouard 1944: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1945: SHFT(*fa,*fb,*fc,fu)
1946: #ifdef DEBUG
1.224 brouard 1947: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1948: 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 1949: #endif
1950: } /* 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 1951: }
1952:
1953: /*************** linmin ************************/
1.162 brouard 1954: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1955: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1956: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1957: the value of func at the returned location p . This is actually all accomplished by calling the
1958: routines mnbrak and brent .*/
1.126 brouard 1959: int ncom;
1960: double *pcom,*xicom;
1961: double (*nrfunc)(double []);
1962:
1.224 brouard 1963: #ifdef LINMINORIGINAL
1.126 brouard 1964: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1965: #else
1966: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1967: #endif
1.126 brouard 1968: {
1969: double brent(double ax, double bx, double cx,
1970: double (*f)(double), double tol, double *xmin);
1971: double f1dim(double x);
1972: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1973: double *fc, double (*func)(double));
1974: int j;
1975: double xx,xmin,bx,ax;
1976: double fx,fb,fa;
1.187 brouard 1977:
1.203 brouard 1978: #ifdef LINMINORIGINAL
1979: #else
1980: double scale=10., axs, xxs; /* Scale added for infinity */
1981: #endif
1982:
1.126 brouard 1983: ncom=n;
1984: pcom=vector(1,n);
1985: xicom=vector(1,n);
1986: nrfunc=func;
1987: for (j=1;j<=n;j++) {
1988: pcom[j]=p[j];
1.202 brouard 1989: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1990: }
1.187 brouard 1991:
1.203 brouard 1992: #ifdef LINMINORIGINAL
1993: xx=1.;
1994: #else
1995: axs=0.0;
1996: xxs=1.;
1997: do{
1998: xx= xxs;
1999: #endif
1.187 brouard 2000: ax=0.;
2001: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2002: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2003: /* 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)) */
2004: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2005: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2006: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2007: /* 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 2008: #ifdef LINMINORIGINAL
2009: #else
2010: if (fx != fx){
1.224 brouard 2011: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2012: printf("|");
2013: fprintf(ficlog,"|");
1.203 brouard 2014: #ifdef DEBUGLINMIN
1.224 brouard 2015: 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 2016: #endif
2017: }
1.224 brouard 2018: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2019: #endif
2020:
1.191 brouard 2021: #ifdef DEBUGLINMIN
2022: 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 2023: 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 2024: #endif
1.224 brouard 2025: #ifdef LINMINORIGINAL
2026: #else
2027: if(fb == fx){ /* Flat function in the direction */
2028: xmin=xx;
2029: *flat=1;
2030: }else{
2031: *flat=0;
2032: #endif
2033: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2034: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2035: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2036: /* fmin = f(p[j] + xmin * xi[j]) */
2037: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2038: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2039: #ifdef DEBUG
1.224 brouard 2040: 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);
2041: 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);
2042: #endif
2043: #ifdef LINMINORIGINAL
2044: #else
2045: }
1.126 brouard 2046: #endif
1.191 brouard 2047: #ifdef DEBUGLINMIN
2048: printf("linmin end ");
1.202 brouard 2049: fprintf(ficlog,"linmin end ");
1.191 brouard 2050: #endif
1.126 brouard 2051: for (j=1;j<=n;j++) {
1.203 brouard 2052: #ifdef LINMINORIGINAL
2053: xi[j] *= xmin;
2054: #else
2055: #ifdef DEBUGLINMIN
2056: if(xxs <1.0)
2057: printf(" before xi[%d]=%12.8f", j,xi[j]);
2058: #endif
2059: 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) */
2060: #ifdef DEBUGLINMIN
2061: if(xxs <1.0)
2062: 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 );
2063: #endif
2064: #endif
1.187 brouard 2065: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2066: }
1.191 brouard 2067: #ifdef DEBUGLINMIN
1.203 brouard 2068: printf("\n");
1.191 brouard 2069: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2070: 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 2071: for (j=1;j<=n;j++) {
1.202 brouard 2072: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2073: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2074: if(j % ncovmodel == 0){
1.191 brouard 2075: printf("\n");
1.202 brouard 2076: fprintf(ficlog,"\n");
2077: }
1.191 brouard 2078: }
1.203 brouard 2079: #else
1.191 brouard 2080: #endif
1.126 brouard 2081: free_vector(xicom,1,n);
2082: free_vector(pcom,1,n);
2083: }
2084:
2085:
2086: /*************** powell ************************/
1.162 brouard 2087: /*
2088: Minimization of a function func of n variables. Input consists of an initial starting point
2089: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2090: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2091: such that failure to decrease by more than this amount on one iteration signals doneness. On
2092: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2093: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2094: */
1.224 brouard 2095: #ifdef LINMINORIGINAL
2096: #else
2097: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2098: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2099: #endif
1.126 brouard 2100: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2101: double (*func)(double []))
2102: {
1.224 brouard 2103: #ifdef LINMINORIGINAL
2104: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2105: double (*func)(double []));
1.224 brouard 2106: #else
1.241 brouard 2107: void linmin(double p[], double xi[], int n, double *fret,
2108: double (*func)(double []),int *flat);
1.224 brouard 2109: #endif
1.239 brouard 2110: int i,ibig,j,jk,k;
1.126 brouard 2111: double del,t,*pt,*ptt,*xit;
1.181 brouard 2112: double directest;
1.126 brouard 2113: double fp,fptt;
2114: double *xits;
2115: int niterf, itmp;
1.224 brouard 2116: #ifdef LINMINORIGINAL
2117: #else
2118:
2119: flatdir=ivector(1,n);
2120: for (j=1;j<=n;j++) flatdir[j]=0;
2121: #endif
1.126 brouard 2122:
2123: pt=vector(1,n);
2124: ptt=vector(1,n);
2125: xit=vector(1,n);
2126: xits=vector(1,n);
2127: *fret=(*func)(p);
2128: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2129: rcurr_time = time(NULL);
1.126 brouard 2130: for (*iter=1;;++(*iter)) {
1.187 brouard 2131: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2132: ibig=0;
2133: del=0.0;
1.157 brouard 2134: rlast_time=rcurr_time;
2135: /* (void) gettimeofday(&curr_time,&tzp); */
2136: rcurr_time = time(NULL);
2137: curr_time = *localtime(&rcurr_time);
2138: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2139: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2140: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2141: for (i=1;i<=n;i++) {
1.126 brouard 2142: fprintf(ficrespow," %.12lf", p[i]);
2143: }
1.239 brouard 2144: fprintf(ficrespow,"\n");fflush(ficrespow);
2145: printf("\n#model= 1 + age ");
2146: fprintf(ficlog,"\n#model= 1 + age ");
2147: if(nagesqr==1){
1.241 brouard 2148: printf(" + age*age ");
2149: fprintf(ficlog," + age*age ");
1.239 brouard 2150: }
2151: for(j=1;j <=ncovmodel-2;j++){
2152: if(Typevar[j]==0) {
2153: printf(" + V%d ",Tvar[j]);
2154: fprintf(ficlog," + V%d ",Tvar[j]);
2155: }else if(Typevar[j]==1) {
2156: printf(" + V%d*age ",Tvar[j]);
2157: fprintf(ficlog," + V%d*age ",Tvar[j]);
2158: }else if(Typevar[j]==2) {
2159: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2160: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2161: }
2162: }
1.126 brouard 2163: printf("\n");
1.239 brouard 2164: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2165: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2166: fprintf(ficlog,"\n");
1.239 brouard 2167: for(i=1,jk=1; i <=nlstate; i++){
2168: for(k=1; k <=(nlstate+ndeath); k++){
2169: if (k != i) {
2170: printf("%d%d ",i,k);
2171: fprintf(ficlog,"%d%d ",i,k);
2172: for(j=1; j <=ncovmodel; j++){
2173: printf("%12.7f ",p[jk]);
2174: fprintf(ficlog,"%12.7f ",p[jk]);
2175: jk++;
2176: }
2177: printf("\n");
2178: fprintf(ficlog,"\n");
2179: }
2180: }
2181: }
1.241 brouard 2182: if(*iter <=3 && *iter >1){
1.157 brouard 2183: tml = *localtime(&rcurr_time);
2184: strcpy(strcurr,asctime(&tml));
2185: rforecast_time=rcurr_time;
1.126 brouard 2186: itmp = strlen(strcurr);
2187: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2188: strcurr[itmp-1]='\0';
1.162 brouard 2189: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2190: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2191: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2192: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2193: forecast_time = *localtime(&rforecast_time);
2194: strcpy(strfor,asctime(&forecast_time));
2195: itmp = strlen(strfor);
2196: if(strfor[itmp-1]=='\n')
2197: strfor[itmp-1]='\0';
2198: 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);
2199: 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 2200: }
2201: }
1.187 brouard 2202: for (i=1;i<=n;i++) { /* For each direction i */
2203: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2204: fptt=(*fret);
2205: #ifdef DEBUG
1.203 brouard 2206: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2207: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2208: #endif
1.203 brouard 2209: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2210: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2211: #ifdef LINMINORIGINAL
1.188 brouard 2212: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2213: #else
2214: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2215: flatdir[i]=flat; /* Function is vanishing in that direction i */
2216: #endif
2217: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2218: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2219: /* because that direction will be replaced unless the gain del is small */
2220: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2221: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2222: /* with the new direction. */
2223: del=fabs(fptt-(*fret));
2224: ibig=i;
1.126 brouard 2225: }
2226: #ifdef DEBUG
2227: printf("%d %.12e",i,(*fret));
2228: fprintf(ficlog,"%d %.12e",i,(*fret));
2229: for (j=1;j<=n;j++) {
1.224 brouard 2230: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2231: printf(" x(%d)=%.12e",j,xit[j]);
2232: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2233: }
2234: for(j=1;j<=n;j++) {
1.225 brouard 2235: printf(" p(%d)=%.12e",j,p[j]);
2236: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2237: }
2238: printf("\n");
2239: fprintf(ficlog,"\n");
2240: #endif
1.187 brouard 2241: } /* end loop on each direction i */
2242: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2243: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2244: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2245: for(j=1;j<=n;j++) {
1.225 brouard 2246: if(flatdir[j] >0){
2247: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2248: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2249: }
2250: /* printf("\n"); */
2251: /* fprintf(ficlog,"\n"); */
2252: }
1.243 brouard 2253: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2254: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2255: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2256: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2257: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2258: /* decreased of more than 3.84 */
2259: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2260: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2261: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2262:
1.188 brouard 2263: /* Starting the program with initial values given by a former maximization will simply change */
2264: /* the scales of the directions and the directions, because the are reset to canonical directions */
2265: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2266: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2267: #ifdef DEBUG
2268: int k[2],l;
2269: k[0]=1;
2270: k[1]=-1;
2271: printf("Max: %.12e",(*func)(p));
2272: fprintf(ficlog,"Max: %.12e",(*func)(p));
2273: for (j=1;j<=n;j++) {
2274: printf(" %.12e",p[j]);
2275: fprintf(ficlog," %.12e",p[j]);
2276: }
2277: printf("\n");
2278: fprintf(ficlog,"\n");
2279: for(l=0;l<=1;l++) {
2280: for (j=1;j<=n;j++) {
2281: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2282: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2283: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2284: }
2285: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2286: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2287: }
2288: #endif
2289:
1.224 brouard 2290: #ifdef LINMINORIGINAL
2291: #else
2292: free_ivector(flatdir,1,n);
2293: #endif
1.126 brouard 2294: free_vector(xit,1,n);
2295: free_vector(xits,1,n);
2296: free_vector(ptt,1,n);
2297: free_vector(pt,1,n);
2298: return;
1.192 brouard 2299: } /* enough precision */
1.240 brouard 2300: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2301: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2302: ptt[j]=2.0*p[j]-pt[j];
2303: xit[j]=p[j]-pt[j];
2304: pt[j]=p[j];
2305: }
1.181 brouard 2306: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2307: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2308: if (*iter <=4) {
1.225 brouard 2309: #else
2310: #endif
1.224 brouard 2311: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2312: #else
1.161 brouard 2313: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2314: #endif
1.162 brouard 2315: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2316: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2317: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2318: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2319: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2320: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2321: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2322: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2323: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2324: /* Even if f3 <f1, directest can be negative and t >0 */
2325: /* mu² and del² are equal when f3=f1 */
2326: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2327: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2328: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2329: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2330: #ifdef NRCORIGINAL
2331: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2332: #else
2333: 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 2334: t= t- del*SQR(fp-fptt);
1.183 brouard 2335: #endif
1.202 brouard 2336: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2337: #ifdef DEBUG
1.181 brouard 2338: 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);
2339: 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 2340: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2341: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2342: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2343: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2344: 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);
2345: 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);
2346: #endif
1.183 brouard 2347: #ifdef POWELLORIGINAL
2348: if (t < 0.0) { /* Then we use it for new direction */
2349: #else
1.182 brouard 2350: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2351: 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 2352: 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 2353: 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 2354: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2355: }
1.181 brouard 2356: if (directest < 0.0) { /* Then we use it for new direction */
2357: #endif
1.191 brouard 2358: #ifdef DEBUGLINMIN
1.234 brouard 2359: printf("Before linmin in direction P%d-P0\n",n);
2360: for (j=1;j<=n;j++) {
2361: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2362: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2363: if(j % ncovmodel == 0){
2364: printf("\n");
2365: fprintf(ficlog,"\n");
2366: }
2367: }
1.224 brouard 2368: #endif
2369: #ifdef LINMINORIGINAL
1.234 brouard 2370: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2371: #else
1.234 brouard 2372: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2373: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2374: #endif
1.234 brouard 2375:
1.191 brouard 2376: #ifdef DEBUGLINMIN
1.234 brouard 2377: for (j=1;j<=n;j++) {
2378: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2379: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2380: if(j % ncovmodel == 0){
2381: printf("\n");
2382: fprintf(ficlog,"\n");
2383: }
2384: }
1.224 brouard 2385: #endif
1.234 brouard 2386: for (j=1;j<=n;j++) {
2387: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2388: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2389: }
1.224 brouard 2390: #ifdef LINMINORIGINAL
2391: #else
1.234 brouard 2392: for (j=1, flatd=0;j<=n;j++) {
2393: if(flatdir[j]>0)
2394: flatd++;
2395: }
2396: if(flatd >0){
2397: printf("%d flat directions\n",flatd);
2398: fprintf(ficlog,"%d flat directions\n",flatd);
2399: for (j=1;j<=n;j++) {
2400: if(flatdir[j]>0){
2401: printf("%d ",j);
2402: fprintf(ficlog,"%d ",j);
2403: }
2404: }
2405: printf("\n");
2406: fprintf(ficlog,"\n");
2407: }
1.191 brouard 2408: #endif
1.234 brouard 2409: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2410: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2411:
1.126 brouard 2412: #ifdef DEBUG
1.234 brouard 2413: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2414: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2415: for(j=1;j<=n;j++){
2416: printf(" %lf",xit[j]);
2417: fprintf(ficlog," %lf",xit[j]);
2418: }
2419: printf("\n");
2420: fprintf(ficlog,"\n");
1.126 brouard 2421: #endif
1.192 brouard 2422: } /* end of t or directest negative */
1.224 brouard 2423: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2424: #else
1.234 brouard 2425: } /* end if (fptt < fp) */
1.192 brouard 2426: #endif
1.225 brouard 2427: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2428: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2429: #else
1.224 brouard 2430: #endif
1.234 brouard 2431: } /* loop iteration */
1.126 brouard 2432: }
1.234 brouard 2433:
1.126 brouard 2434: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2435:
1.235 brouard 2436: 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 2437: {
1.235 brouard 2438: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2439: (and selected quantitative values in nres)
2440: by left multiplying the unit
1.234 brouard 2441: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2442: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2443: /* Wx is row vector: population in state 1, population in state 2, population dead */
2444: /* or prevalence in state 1, prevalence in state 2, 0 */
2445: /* newm is the matrix after multiplications, its rows are identical at a factor */
2446: /* Initial matrix pimij */
2447: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2448: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2449: /* 0, 0 , 1} */
2450: /*
2451: * and after some iteration: */
2452: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2453: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2454: /* 0, 0 , 1} */
2455: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2456: /* {0.51571254859325999, 0.4842874514067399, */
2457: /* 0.51326036147820708, 0.48673963852179264} */
2458: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2459:
1.126 brouard 2460: int i, ii,j,k;
1.209 brouard 2461: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2462: /* double **matprod2(); */ /* test */
1.218 brouard 2463: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2464: double **newm;
1.209 brouard 2465: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2466: int ncvloop=0;
1.169 brouard 2467:
1.209 brouard 2468: min=vector(1,nlstate);
2469: max=vector(1,nlstate);
2470: meandiff=vector(1,nlstate);
2471:
1.218 brouard 2472: /* Starting with matrix unity */
1.126 brouard 2473: for (ii=1;ii<=nlstate+ndeath;ii++)
2474: for (j=1;j<=nlstate+ndeath;j++){
2475: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2476: }
1.169 brouard 2477:
2478: cov[1]=1.;
2479:
2480: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2481: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2482: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2483: ncvloop++;
1.126 brouard 2484: newm=savm;
2485: /* Covariates have to be included here again */
1.138 brouard 2486: cov[2]=agefin;
1.187 brouard 2487: if(nagesqr==1)
2488: cov[3]= agefin*agefin;;
1.234 brouard 2489: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2490: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2491: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2492: /* 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 2493: }
2494: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2495: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2496: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2497: /* 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 2498: }
1.237 brouard 2499: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2500: if(Dummy[Tvar[Tage[k]]]){
2501: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2502: } else{
1.235 brouard 2503: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2504: }
1.235 brouard 2505: /* 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 2506: }
1.237 brouard 2507: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2508: /* 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 2509: if(Dummy[Tvard[k][1]==0]){
2510: if(Dummy[Tvard[k][2]==0]){
2511: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2512: }else{
2513: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2514: }
2515: }else{
2516: if(Dummy[Tvard[k][2]==0]){
2517: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2518: }else{
2519: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2520: }
2521: }
1.234 brouard 2522: }
1.138 brouard 2523: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2524: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2525: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2526: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2527: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2528: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2529: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2530:
1.126 brouard 2531: savm=oldm;
2532: oldm=newm;
1.209 brouard 2533:
2534: for(j=1; j<=nlstate; j++){
2535: max[j]=0.;
2536: min[j]=1.;
2537: }
2538: for(i=1;i<=nlstate;i++){
2539: sumnew=0;
2540: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2541: for(j=1; j<=nlstate; j++){
2542: prlim[i][j]= newm[i][j]/(1-sumnew);
2543: max[j]=FMAX(max[j],prlim[i][j]);
2544: min[j]=FMIN(min[j],prlim[i][j]);
2545: }
2546: }
2547:
1.126 brouard 2548: maxmax=0.;
1.209 brouard 2549: for(j=1; j<=nlstate; j++){
2550: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2551: maxmax=FMAX(maxmax,meandiff[j]);
2552: /* 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 2553: } /* j loop */
1.203 brouard 2554: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2555: /* 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 2556: if(maxmax < ftolpl){
1.209 brouard 2557: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2558: free_vector(min,1,nlstate);
2559: free_vector(max,1,nlstate);
2560: free_vector(meandiff,1,nlstate);
1.126 brouard 2561: return prlim;
2562: }
1.169 brouard 2563: } /* age loop */
1.208 brouard 2564: /* After some age loop it doesn't converge */
1.209 brouard 2565: 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 2566: 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 2567: /* 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); */
2568: free_vector(min,1,nlstate);
2569: free_vector(max,1,nlstate);
2570: free_vector(meandiff,1,nlstate);
1.208 brouard 2571:
1.169 brouard 2572: return prlim; /* should not reach here */
1.126 brouard 2573: }
2574:
1.217 brouard 2575:
2576: /**** Back Prevalence limit (stable or period prevalence) ****************/
2577:
1.218 brouard 2578: /* 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) */
2579: /* 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 2580: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2581: {
1.218 brouard 2582: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2583: matrix by transitions matrix until convergence is reached with precision ftolpl */
2584: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2585: /* Wx is row vector: population in state 1, population in state 2, population dead */
2586: /* or prevalence in state 1, prevalence in state 2, 0 */
2587: /* newm is the matrix after multiplications, its rows are identical at a factor */
2588: /* Initial matrix pimij */
2589: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2590: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2591: /* 0, 0 , 1} */
2592: /*
2593: * and after some iteration: */
2594: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2595: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2596: /* 0, 0 , 1} */
2597: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2598: /* {0.51571254859325999, 0.4842874514067399, */
2599: /* 0.51326036147820708, 0.48673963852179264} */
2600: /* If we start from prlim again, prlim tends to a constant matrix */
2601:
2602: int i, ii,j,k;
1.247 brouard 2603: int first=0;
1.217 brouard 2604: double *min, *max, *meandiff, maxmax,sumnew=0.;
2605: /* double **matprod2(); */ /* test */
2606: double **out, cov[NCOVMAX+1], **bmij();
2607: double **newm;
1.218 brouard 2608: double **dnewm, **doldm, **dsavm; /* for use */
2609: double **oldm, **savm; /* for use */
2610:
1.217 brouard 2611: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2612: int ncvloop=0;
2613:
2614: min=vector(1,nlstate);
2615: max=vector(1,nlstate);
2616: meandiff=vector(1,nlstate);
2617:
1.218 brouard 2618: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2619: oldm=oldms; savm=savms;
2620:
2621: /* Starting with matrix unity */
2622: for (ii=1;ii<=nlstate+ndeath;ii++)
2623: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2624: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2625: }
2626:
2627: cov[1]=1.;
2628:
2629: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2630: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2631: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2632: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2633: ncvloop++;
1.218 brouard 2634: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2635: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2636: /* Covariates have to be included here again */
2637: cov[2]=agefin;
2638: if(nagesqr==1)
2639: cov[3]= agefin*agefin;;
1.242 brouard 2640: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2641: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2642: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2643: /* 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)); */
2644: }
2645: /* for (k=1; k<=cptcovn;k++) { */
2646: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2647: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2648: /* /\* 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])]); *\/ */
2649: /* } */
2650: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2651: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2652: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2653: /* 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]); */
2654: }
2655: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2656: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2657: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2658: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2659: for (k=1; k<=cptcovage;k++){ /* For product with age */
2660: if(Dummy[Tvar[Tage[k]]]){
2661: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2662: } else{
2663: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2664: }
2665: /* 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]); */
2666: }
2667: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2668: /* 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]); */
2669: if(Dummy[Tvard[k][1]==0]){
2670: if(Dummy[Tvard[k][2]==0]){
2671: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2672: }else{
2673: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2674: }
2675: }else{
2676: if(Dummy[Tvard[k][2]==0]){
2677: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2678: }else{
2679: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2680: }
2681: }
1.217 brouard 2682: }
2683:
2684: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2685: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2686: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2687: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2688: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2689: /* ij should be linked to the correct index of cov */
2690: /* age and covariate values ij are in 'cov', but we need to pass
2691: * ij for the observed prevalence at age and status and covariate
2692: * number: prevacurrent[(int)agefin][ii][ij]
2693: */
2694: /* 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 *\/ */
2695: /* 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 *\/ */
2696: 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 2697: savm=oldm;
2698: oldm=newm;
2699: for(j=1; j<=nlstate; j++){
2700: max[j]=0.;
2701: min[j]=1.;
2702: }
2703: for(j=1; j<=nlstate; j++){
2704: for(i=1;i<=nlstate;i++){
1.234 brouard 2705: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2706: bprlim[i][j]= newm[i][j];
2707: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2708: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2709: }
2710: }
1.218 brouard 2711:
1.217 brouard 2712: maxmax=0.;
2713: for(i=1; i<=nlstate; i++){
2714: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2715: maxmax=FMAX(maxmax,meandiff[i]);
2716: /* 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); */
2717: } /* j loop */
2718: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2719: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2720: if(maxmax < ftolpl){
1.220 brouard 2721: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2722: free_vector(min,1,nlstate);
2723: free_vector(max,1,nlstate);
2724: free_vector(meandiff,1,nlstate);
2725: return bprlim;
2726: }
2727: } /* age loop */
2728: /* After some age loop it doesn't converge */
1.247 brouard 2729: if(first){
2730: first=1;
2731: 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\
2732: 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);
2733: }
2734: 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 2735: 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);
2736: /* 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); */
2737: free_vector(min,1,nlstate);
2738: free_vector(max,1,nlstate);
2739: free_vector(meandiff,1,nlstate);
2740:
2741: return bprlim; /* should not reach here */
2742: }
2743:
1.126 brouard 2744: /*************** transition probabilities ***************/
2745:
2746: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2747: {
1.138 brouard 2748: /* According to parameters values stored in x and the covariate's values stored in cov,
2749: computes the probability to be observed in state j being in state i by appying the
2750: model to the ncovmodel covariates (including constant and age).
2751: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2752: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2753: ncth covariate in the global vector x is given by the formula:
2754: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2755: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2756: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2757: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2758: Outputs ps[i][j] the probability to be observed in j being in j according to
2759: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2760: */
2761: double s1, lnpijopii;
1.126 brouard 2762: /*double t34;*/
1.164 brouard 2763: int i,j, nc, ii, jj;
1.126 brouard 2764:
1.223 brouard 2765: for(i=1; i<= nlstate; i++){
2766: for(j=1; j<i;j++){
2767: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2768: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2769: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2770: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2771: }
2772: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2773: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2774: }
2775: for(j=i+1; j<=nlstate+ndeath;j++){
2776: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2777: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2778: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2779: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2780: }
2781: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2782: }
2783: }
1.218 brouard 2784:
1.223 brouard 2785: for(i=1; i<= nlstate; i++){
2786: s1=0;
2787: for(j=1; j<i; j++){
2788: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2789: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2790: }
2791: for(j=i+1; j<=nlstate+ndeath; j++){
2792: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2793: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2794: }
2795: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2796: ps[i][i]=1./(s1+1.);
2797: /* Computing other pijs */
2798: for(j=1; j<i; j++)
2799: ps[i][j]= exp(ps[i][j])*ps[i][i];
2800: for(j=i+1; j<=nlstate+ndeath; j++)
2801: ps[i][j]= exp(ps[i][j])*ps[i][i];
2802: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2803: } /* end i */
1.218 brouard 2804:
1.223 brouard 2805: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2806: for(jj=1; jj<= nlstate+ndeath; jj++){
2807: ps[ii][jj]=0;
2808: ps[ii][ii]=1;
2809: }
2810: }
1.218 brouard 2811:
2812:
1.223 brouard 2813: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2814: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2815: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2816: /* } */
2817: /* printf("\n "); */
2818: /* } */
2819: /* printf("\n ");printf("%lf ",cov[2]);*/
2820: /*
2821: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2822: goto end;*/
1.223 brouard 2823: return ps;
1.126 brouard 2824: }
2825:
1.218 brouard 2826: /*************** backward transition probabilities ***************/
2827:
2828: /* 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 ) */
2829: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2830: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2831: {
1.222 brouard 2832: /* Computes the backward probability at age agefin and covariate ij
2833: * and returns in **ps as well as **bmij.
2834: */
1.218 brouard 2835: int i, ii, j,k;
1.222 brouard 2836:
2837: double **out, **pmij();
2838: double sumnew=0.;
1.218 brouard 2839: double agefin;
1.222 brouard 2840:
2841: double **dnewm, **dsavm, **doldm;
2842: double **bbmij;
2843:
1.218 brouard 2844: doldm=ddoldms; /* global pointers */
1.222 brouard 2845: dnewm=ddnewms;
2846: dsavm=ddsavms;
2847:
2848: agefin=cov[2];
2849: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2850: the observed prevalence (with this covariate ij) */
2851: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2852: /* We do have the matrix Px in savm and we need pij */
2853: for (j=1;j<=nlstate+ndeath;j++){
2854: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2855: for (ii=1;ii<=nlstate;ii++){
2856: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2857: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2858: for (ii=1;ii<=nlstate+ndeath;ii++){
2859: if(sumnew >= 1.e-10){
2860: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2861: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2862: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2863: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2864: /* }else */
2865: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2866: }else{
1.242 brouard 2867: ;
2868: /* 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 2869: }
2870: } /*End ii */
2871: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2872: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2873: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2874: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2875: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2876: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2877: /* left Product of this matrix by diag matrix of prevalences (savm) */
2878: for (j=1;j<=nlstate+ndeath;j++){
2879: for (ii=1;ii<=nlstate+ndeath;ii++){
2880: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2881: }
2882: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2883: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2884: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2885: /* end bmij */
2886: return ps;
1.218 brouard 2887: }
1.217 brouard 2888: /*************** transition probabilities ***************/
2889:
1.218 brouard 2890: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2891: {
2892: /* According to parameters values stored in x and the covariate's values stored in cov,
2893: computes the probability to be observed in state j being in state i by appying the
2894: model to the ncovmodel covariates (including constant and age).
2895: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2896: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2897: ncth covariate in the global vector x is given by the formula:
2898: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2899: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2900: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2901: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2902: Outputs ps[i][j] the probability to be observed in j being in j according to
2903: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2904: */
2905: double s1, lnpijopii;
2906: /*double t34;*/
2907: int i,j, nc, ii, jj;
2908:
1.234 brouard 2909: for(i=1; i<= nlstate; i++){
2910: for(j=1; j<i;j++){
2911: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2912: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2913: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2914: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2915: }
2916: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2917: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2918: }
2919: for(j=i+1; j<=nlstate+ndeath;j++){
2920: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2921: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2922: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2923: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2924: }
2925: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2926: }
2927: }
2928:
2929: for(i=1; i<= nlstate; i++){
2930: s1=0;
2931: for(j=1; j<i; j++){
2932: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2933: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2934: }
2935: for(j=i+1; j<=nlstate+ndeath; j++){
2936: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2937: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2938: }
2939: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2940: ps[i][i]=1./(s1+1.);
2941: /* Computing other pijs */
2942: for(j=1; j<i; j++)
2943: ps[i][j]= exp(ps[i][j])*ps[i][i];
2944: for(j=i+1; j<=nlstate+ndeath; j++)
2945: ps[i][j]= exp(ps[i][j])*ps[i][i];
2946: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2947: } /* end i */
2948:
2949: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2950: for(jj=1; jj<= nlstate+ndeath; jj++){
2951: ps[ii][jj]=0;
2952: ps[ii][ii]=1;
2953: }
2954: }
2955: /* Added for backcast */ /* Transposed matrix too */
2956: for(jj=1; jj<= nlstate+ndeath; jj++){
2957: s1=0.;
2958: for(ii=1; ii<= nlstate+ndeath; ii++){
2959: s1+=ps[ii][jj];
2960: }
2961: for(ii=1; ii<= nlstate; ii++){
2962: ps[ii][jj]=ps[ii][jj]/s1;
2963: }
2964: }
2965: /* Transposition */
2966: for(jj=1; jj<= nlstate+ndeath; jj++){
2967: for(ii=jj; ii<= nlstate+ndeath; ii++){
2968: s1=ps[ii][jj];
2969: ps[ii][jj]=ps[jj][ii];
2970: ps[jj][ii]=s1;
2971: }
2972: }
2973: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2974: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2975: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2976: /* } */
2977: /* printf("\n "); */
2978: /* } */
2979: /* printf("\n ");printf("%lf ",cov[2]);*/
2980: /*
2981: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2982: goto end;*/
2983: return ps;
1.217 brouard 2984: }
2985:
2986:
1.126 brouard 2987: /**************** Product of 2 matrices ******************/
2988:
1.145 brouard 2989: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2990: {
2991: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2992: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2993: /* in, b, out are matrice of pointers which should have been initialized
2994: before: only the contents of out is modified. The function returns
2995: a pointer to pointers identical to out */
1.145 brouard 2996: int i, j, k;
1.126 brouard 2997: for(i=nrl; i<= nrh; i++)
1.145 brouard 2998: for(k=ncolol; k<=ncoloh; k++){
2999: out[i][k]=0.;
3000: for(j=ncl; j<=nch; j++)
3001: out[i][k] +=in[i][j]*b[j][k];
3002: }
1.126 brouard 3003: return out;
3004: }
3005:
3006:
3007: /************* Higher Matrix Product ***************/
3008:
1.235 brouard 3009: 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 3010: {
1.218 brouard 3011: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3012: 'nhstepm*hstepm*stepm' months (i.e. until
3013: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3014: nhstepm*hstepm matrices.
3015: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3016: (typically every 2 years instead of every month which is too big
3017: for the memory).
3018: Model is determined by parameters x and covariates have to be
3019: included manually here.
3020:
3021: */
3022:
3023: int i, j, d, h, k;
1.131 brouard 3024: double **out, cov[NCOVMAX+1];
1.126 brouard 3025: double **newm;
1.187 brouard 3026: double agexact;
1.214 brouard 3027: double agebegin, ageend;
1.126 brouard 3028:
3029: /* Hstepm could be zero and should return the unit matrix */
3030: for (i=1;i<=nlstate+ndeath;i++)
3031: for (j=1;j<=nlstate+ndeath;j++){
3032: oldm[i][j]=(i==j ? 1.0 : 0.0);
3033: po[i][j][0]=(i==j ? 1.0 : 0.0);
3034: }
3035: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3036: for(h=1; h <=nhstepm; h++){
3037: for(d=1; d <=hstepm; d++){
3038: newm=savm;
3039: /* Covariates have to be included here again */
3040: cov[1]=1.;
1.214 brouard 3041: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3042: cov[2]=agexact;
3043: if(nagesqr==1)
1.227 brouard 3044: cov[3]= agexact*agexact;
1.235 brouard 3045: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3046: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3047: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3048: /* 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)); */
3049: }
3050: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3051: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3052: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3053: /* 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]); */
3054: }
3055: for (k=1; k<=cptcovage;k++){
3056: if(Dummy[Tvar[Tage[k]]]){
3057: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3058: } else{
3059: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3060: }
3061: /* 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]); */
3062: }
3063: for (k=1; k<=cptcovprod;k++){ /* */
3064: /* 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]); */
3065: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3066: }
3067: /* for (k=1; k<=cptcovn;k++) */
3068: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3069: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3070: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3071: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3072: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3073:
3074:
1.126 brouard 3075: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3076: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3077: /* right multiplication of oldm by the current matrix */
1.126 brouard 3078: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3079: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3080: /* if((int)age == 70){ */
3081: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3082: /* for(i=1; i<=nlstate+ndeath; i++) { */
3083: /* printf("%d pmmij ",i); */
3084: /* for(j=1;j<=nlstate+ndeath;j++) { */
3085: /* printf("%f ",pmmij[i][j]); */
3086: /* } */
3087: /* printf(" oldm "); */
3088: /* for(j=1;j<=nlstate+ndeath;j++) { */
3089: /* printf("%f ",oldm[i][j]); */
3090: /* } */
3091: /* printf("\n"); */
3092: /* } */
3093: /* } */
1.126 brouard 3094: savm=oldm;
3095: oldm=newm;
3096: }
3097: for(i=1; i<=nlstate+ndeath; i++)
3098: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3099: po[i][j][h]=newm[i][j];
3100: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3101: }
1.128 brouard 3102: /*printf("h=%d ",h);*/
1.126 brouard 3103: } /* end h */
1.218 brouard 3104: /* printf("\n H=%d \n",h); */
1.126 brouard 3105: return po;
3106: }
3107:
1.217 brouard 3108: /************* Higher Back Matrix Product ***************/
1.218 brouard 3109: /* 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 3110: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3111: {
1.218 brouard 3112: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3113: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3114: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3115: nhstepm*hstepm matrices.
3116: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3117: (typically every 2 years instead of every month which is too big
1.217 brouard 3118: for the memory).
1.218 brouard 3119: Model is determined by parameters x and covariates have to be
3120: included manually here.
1.217 brouard 3121:
1.222 brouard 3122: */
1.217 brouard 3123:
3124: int i, j, d, h, k;
3125: double **out, cov[NCOVMAX+1];
3126: double **newm;
3127: double agexact;
3128: double agebegin, ageend;
1.222 brouard 3129: double **oldm, **savm;
1.217 brouard 3130:
1.222 brouard 3131: oldm=oldms;savm=savms;
1.217 brouard 3132: /* Hstepm could be zero and should return the unit matrix */
3133: for (i=1;i<=nlstate+ndeath;i++)
3134: for (j=1;j<=nlstate+ndeath;j++){
3135: oldm[i][j]=(i==j ? 1.0 : 0.0);
3136: po[i][j][0]=(i==j ? 1.0 : 0.0);
3137: }
3138: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3139: for(h=1; h <=nhstepm; h++){
3140: for(d=1; d <=hstepm; d++){
3141: newm=savm;
3142: /* Covariates have to be included here again */
3143: cov[1]=1.;
3144: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3145: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3146: cov[2]=agexact;
3147: if(nagesqr==1)
1.222 brouard 3148: cov[3]= agexact*agexact;
1.218 brouard 3149: for (k=1; k<=cptcovn;k++)
1.222 brouard 3150: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3151: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3152: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3153: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3154: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3155: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3156: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3157: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3158: /* 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 3159:
3160:
1.217 brouard 3161: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3162: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3163: /* Careful transposed matrix */
1.222 brouard 3164: /* age is in cov[2] */
1.218 brouard 3165: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3166: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3167: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3168: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3169: /* if((int)age == 70){ */
3170: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3171: /* for(i=1; i<=nlstate+ndeath; i++) { */
3172: /* printf("%d pmmij ",i); */
3173: /* for(j=1;j<=nlstate+ndeath;j++) { */
3174: /* printf("%f ",pmmij[i][j]); */
3175: /* } */
3176: /* printf(" oldm "); */
3177: /* for(j=1;j<=nlstate+ndeath;j++) { */
3178: /* printf("%f ",oldm[i][j]); */
3179: /* } */
3180: /* printf("\n"); */
3181: /* } */
3182: /* } */
3183: savm=oldm;
3184: oldm=newm;
3185: }
3186: for(i=1; i<=nlstate+ndeath; i++)
3187: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3188: po[i][j][h]=newm[i][j];
3189: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3190: }
3191: /*printf("h=%d ",h);*/
3192: } /* end h */
1.222 brouard 3193: /* printf("\n H=%d \n",h); */
1.217 brouard 3194: return po;
3195: }
3196:
3197:
1.162 brouard 3198: #ifdef NLOPT
3199: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3200: double fret;
3201: double *xt;
3202: int j;
3203: myfunc_data *d2 = (myfunc_data *) pd;
3204: /* xt = (p1-1); */
3205: xt=vector(1,n);
3206: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3207:
3208: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3209: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3210: printf("Function = %.12lf ",fret);
3211: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3212: printf("\n");
3213: free_vector(xt,1,n);
3214: return fret;
3215: }
3216: #endif
1.126 brouard 3217:
3218: /*************** log-likelihood *************/
3219: double func( double *x)
3220: {
1.226 brouard 3221: int i, ii, j, k, mi, d, kk;
3222: int ioffset=0;
3223: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3224: double **out;
3225: double lli; /* Individual log likelihood */
3226: int s1, s2;
1.228 brouard 3227: 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 3228: double bbh, survp;
3229: long ipmx;
3230: double agexact;
3231: /*extern weight */
3232: /* We are differentiating ll according to initial status */
3233: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3234: /*for(i=1;i<imx;i++)
3235: printf(" %d\n",s[4][i]);
3236: */
1.162 brouard 3237:
1.226 brouard 3238: ++countcallfunc;
1.162 brouard 3239:
1.226 brouard 3240: cov[1]=1.;
1.126 brouard 3241:
1.226 brouard 3242: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3243: ioffset=0;
1.226 brouard 3244: if(mle==1){
3245: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3246: /* Computes the values of the ncovmodel covariates of the model
3247: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3248: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3249: to be observed in j being in i according to the model.
3250: */
1.243 brouard 3251: ioffset=2+nagesqr ;
1.233 brouard 3252: /* Fixed */
1.234 brouard 3253: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3254: 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)*/
3255: }
1.226 brouard 3256: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3257: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3258: has been calculated etc */
3259: /* For an individual i, wav[i] gives the number of effective waves */
3260: /* We compute the contribution to Likelihood of each effective transition
3261: mw[mi][i] is real wave of the mi th effectve wave */
3262: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3263: s2=s[mw[mi+1][i]][i];
3264: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3265: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3266: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3267: */
3268: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3269: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3270: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3271: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3272: }
3273: for (ii=1;ii<=nlstate+ndeath;ii++)
3274: for (j=1;j<=nlstate+ndeath;j++){
3275: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3276: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3277: }
3278: for(d=0; d<dh[mi][i]; d++){
3279: newm=savm;
3280: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3281: cov[2]=agexact;
3282: if(nagesqr==1)
3283: cov[3]= agexact*agexact; /* Should be changed here */
3284: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3285: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3286: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3287: else
3288: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3289: }
3290: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3291: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3292: savm=oldm;
3293: oldm=newm;
3294: } /* end mult */
3295:
3296: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3297: /* But now since version 0.9 we anticipate for bias at large stepm.
3298: * If stepm is larger than one month (smallest stepm) and if the exact delay
3299: * (in months) between two waves is not a multiple of stepm, we rounded to
3300: * the nearest (and in case of equal distance, to the lowest) interval but now
3301: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3302: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3303: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3304: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3305: * -stepm/2 to stepm/2 .
3306: * For stepm=1 the results are the same as for previous versions of Imach.
3307: * For stepm > 1 the results are less biased than in previous versions.
3308: */
1.234 brouard 3309: s1=s[mw[mi][i]][i];
3310: s2=s[mw[mi+1][i]][i];
3311: bbh=(double)bh[mi][i]/(double)stepm;
3312: /* bias bh is positive if real duration
3313: * is higher than the multiple of stepm and negative otherwise.
3314: */
3315: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3316: if( s2 > nlstate){
3317: /* i.e. if s2 is a death state and if the date of death is known
3318: then the contribution to the likelihood is the probability to
3319: die between last step unit time and current step unit time,
3320: which is also equal to probability to die before dh
3321: minus probability to die before dh-stepm .
3322: In version up to 0.92 likelihood was computed
3323: as if date of death was unknown. Death was treated as any other
3324: health state: the date of the interview describes the actual state
3325: and not the date of a change in health state. The former idea was
3326: to consider that at each interview the state was recorded
3327: (healthy, disable or death) and IMaCh was corrected; but when we
3328: introduced the exact date of death then we should have modified
3329: the contribution of an exact death to the likelihood. This new
3330: contribution is smaller and very dependent of the step unit
3331: stepm. It is no more the probability to die between last interview
3332: and month of death but the probability to survive from last
3333: interview up to one month before death multiplied by the
3334: probability to die within a month. Thanks to Chris
3335: Jackson for correcting this bug. Former versions increased
3336: mortality artificially. The bad side is that we add another loop
3337: which slows down the processing. The difference can be up to 10%
3338: lower mortality.
3339: */
3340: /* If, at the beginning of the maximization mostly, the
3341: cumulative probability or probability to be dead is
3342: constant (ie = 1) over time d, the difference is equal to
3343: 0. out[s1][3] = savm[s1][3]: probability, being at state
3344: s1 at precedent wave, to be dead a month before current
3345: wave is equal to probability, being at state s1 at
3346: precedent wave, to be dead at mont of the current
3347: wave. Then the observed probability (that this person died)
3348: is null according to current estimated parameter. In fact,
3349: it should be very low but not zero otherwise the log go to
3350: infinity.
3351: */
1.183 brouard 3352: /* #ifdef INFINITYORIGINAL */
3353: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3354: /* #else */
3355: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3356: /* lli=log(mytinydouble); */
3357: /* else */
3358: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3359: /* #endif */
1.226 brouard 3360: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3361:
1.226 brouard 3362: } else if ( s2==-1 ) { /* alive */
3363: for (j=1,survp=0. ; j<=nlstate; j++)
3364: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3365: /*survp += out[s1][j]; */
3366: lli= log(survp);
3367: }
3368: else if (s2==-4) {
3369: for (j=3,survp=0. ; j<=nlstate; j++)
3370: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3371: lli= log(survp);
3372: }
3373: else if (s2==-5) {
3374: for (j=1,survp=0. ; j<=2; j++)
3375: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3376: lli= log(survp);
3377: }
3378: else{
3379: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3380: /* 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 */
3381: }
3382: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3383: /*if(lli ==000.0)*/
3384: /*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); */
3385: ipmx +=1;
3386: sw += weight[i];
3387: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3388: /* if (lli < log(mytinydouble)){ */
3389: /* 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); */
3390: /* 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]); */
3391: /* } */
3392: } /* end of wave */
3393: } /* end of individual */
3394: } else if(mle==2){
3395: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3396: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3397: for(mi=1; mi<= wav[i]-1; mi++){
3398: for (ii=1;ii<=nlstate+ndeath;ii++)
3399: for (j=1;j<=nlstate+ndeath;j++){
3400: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3401: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3402: }
3403: for(d=0; d<=dh[mi][i]; d++){
3404: newm=savm;
3405: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3406: cov[2]=agexact;
3407: if(nagesqr==1)
3408: cov[3]= agexact*agexact;
3409: for (kk=1; kk<=cptcovage;kk++) {
3410: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3411: }
3412: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3413: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3414: savm=oldm;
3415: oldm=newm;
3416: } /* end mult */
3417:
3418: s1=s[mw[mi][i]][i];
3419: s2=s[mw[mi+1][i]][i];
3420: bbh=(double)bh[mi][i]/(double)stepm;
3421: 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 */
3422: ipmx +=1;
3423: sw += weight[i];
3424: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3425: } /* end of wave */
3426: } /* end of individual */
3427: } else if(mle==3){ /* exponential inter-extrapolation */
3428: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3429: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3430: for(mi=1; mi<= wav[i]-1; mi++){
3431: for (ii=1;ii<=nlstate+ndeath;ii++)
3432: for (j=1;j<=nlstate+ndeath;j++){
3433: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3434: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3435: }
3436: for(d=0; d<dh[mi][i]; d++){
3437: newm=savm;
3438: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3439: cov[2]=agexact;
3440: if(nagesqr==1)
3441: cov[3]= agexact*agexact;
3442: for (kk=1; kk<=cptcovage;kk++) {
3443: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3444: }
3445: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3446: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3447: savm=oldm;
3448: oldm=newm;
3449: } /* end mult */
3450:
3451: s1=s[mw[mi][i]][i];
3452: s2=s[mw[mi+1][i]][i];
3453: bbh=(double)bh[mi][i]/(double)stepm;
3454: lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */
3455: ipmx +=1;
3456: sw += weight[i];
3457: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3458: } /* end of wave */
3459: } /* end of individual */
3460: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3461: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3462: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3463: for(mi=1; mi<= wav[i]-1; mi++){
3464: for (ii=1;ii<=nlstate+ndeath;ii++)
3465: for (j=1;j<=nlstate+ndeath;j++){
3466: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3467: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3468: }
3469: for(d=0; d<dh[mi][i]; d++){
3470: newm=savm;
3471: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3472: cov[2]=agexact;
3473: if(nagesqr==1)
3474: cov[3]= agexact*agexact;
3475: for (kk=1; kk<=cptcovage;kk++) {
3476: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3477: }
1.126 brouard 3478:
1.226 brouard 3479: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3480: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3481: savm=oldm;
3482: oldm=newm;
3483: } /* end mult */
3484:
3485: s1=s[mw[mi][i]][i];
3486: s2=s[mw[mi+1][i]][i];
3487: if( s2 > nlstate){
3488: lli=log(out[s1][s2] - savm[s1][s2]);
3489: } else if ( s2==-1 ) { /* alive */
3490: for (j=1,survp=0. ; j<=nlstate; j++)
3491: survp += out[s1][j];
3492: lli= log(survp);
3493: }else{
3494: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3495: }
3496: ipmx +=1;
3497: sw += weight[i];
3498: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3499: /* 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 3500: } /* end of wave */
3501: } /* end of individual */
3502: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3503: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3504: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3505: for(mi=1; mi<= wav[i]-1; mi++){
3506: for (ii=1;ii<=nlstate+ndeath;ii++)
3507: for (j=1;j<=nlstate+ndeath;j++){
3508: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3509: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3510: }
3511: for(d=0; d<dh[mi][i]; d++){
3512: newm=savm;
3513: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3514: cov[2]=agexact;
3515: if(nagesqr==1)
3516: cov[3]= agexact*agexact;
3517: for (kk=1; kk<=cptcovage;kk++) {
3518: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3519: }
1.126 brouard 3520:
1.226 brouard 3521: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3522: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3523: savm=oldm;
3524: oldm=newm;
3525: } /* end mult */
3526:
3527: s1=s[mw[mi][i]][i];
3528: s2=s[mw[mi+1][i]][i];
3529: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3530: ipmx +=1;
3531: sw += weight[i];
3532: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3533: /*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]);*/
3534: } /* end of wave */
3535: } /* end of individual */
3536: } /* End of if */
3537: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3538: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3539: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3540: return -l;
1.126 brouard 3541: }
3542:
3543: /*************** log-likelihood *************/
3544: double funcone( double *x)
3545: {
1.228 brouard 3546: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3547: int i, ii, j, k, mi, d, kk;
1.228 brouard 3548: int ioffset=0;
1.131 brouard 3549: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3550: double **out;
3551: double lli; /* Individual log likelihood */
3552: double llt;
3553: int s1, s2;
1.228 brouard 3554: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3555:
1.126 brouard 3556: double bbh, survp;
1.187 brouard 3557: double agexact;
1.214 brouard 3558: double agebegin, ageend;
1.126 brouard 3559: /*extern weight */
3560: /* We are differentiating ll according to initial status */
3561: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3562: /*for(i=1;i<imx;i++)
3563: printf(" %d\n",s[4][i]);
3564: */
3565: cov[1]=1.;
3566:
3567: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3568: ioffset=0;
3569: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3570: /* ioffset=2+nagesqr+cptcovage; */
3571: ioffset=2+nagesqr;
1.232 brouard 3572: /* Fixed */
1.224 brouard 3573: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3574: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3575: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3576: 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)*/
3577: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3578: /* cov[2+6]=covar[Tvar[6]][i]; */
3579: /* cov[2+6]=covar[2][i]; V2 */
3580: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3581: /* cov[2+7]=covar[Tvar[7]][i]; */
3582: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3583: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3584: /* cov[2+9]=covar[Tvar[9]][i]; */
3585: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3586: }
1.232 brouard 3587: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3588: /* 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?)*\/ */
3589: /* } */
1.231 brouard 3590: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3591: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3592: /* } */
1.225 brouard 3593:
1.233 brouard 3594:
3595: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3596: /* Wave varying (but not age varying) */
3597: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3598: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3599: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3600: }
1.232 brouard 3601: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3602: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3603: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3604: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3605: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3606: /* 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 3607: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3608: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3609: /* /\* 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]); *\/ */
3610: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3611: /* } */
1.126 brouard 3612: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3613: for (j=1;j<=nlstate+ndeath;j++){
3614: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3615: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3616: }
1.214 brouard 3617:
3618: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3619: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3620: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3621: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3622: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3623: and mw[mi+1][i]. dh depends on stepm.*/
3624: newm=savm;
1.247 brouard 3625: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3626: cov[2]=agexact;
3627: if(nagesqr==1)
3628: cov[3]= agexact*agexact;
3629: for (kk=1; kk<=cptcovage;kk++) {
3630: if(!FixedV[Tvar[Tage[kk]]])
3631: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3632: else
3633: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3634: }
3635: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3636: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3637: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3638: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3639: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3640: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3641: savm=oldm;
3642: oldm=newm;
1.126 brouard 3643: } /* end mult */
3644:
3645: s1=s[mw[mi][i]][i];
3646: s2=s[mw[mi+1][i]][i];
1.217 brouard 3647: /* if(s2==-1){ */
3648: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3649: /* /\* exit(1); *\/ */
3650: /* } */
1.126 brouard 3651: bbh=(double)bh[mi][i]/(double)stepm;
3652: /* bias is positive if real duration
3653: * is higher than the multiple of stepm and negative otherwise.
3654: */
3655: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3656: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3657: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3658: for (j=1,survp=0. ; j<=nlstate; j++)
3659: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3660: lli= log(survp);
1.126 brouard 3661: }else if (mle==1){
1.242 brouard 3662: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3663: } else if(mle==2){
1.242 brouard 3664: 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 3665: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3666: 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 3667: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3668: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3669: } else{ /* mle=0 back to 1 */
1.242 brouard 3670: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3671: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3672: } /* End of if */
3673: ipmx +=1;
3674: sw += weight[i];
3675: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3676: /*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 3677: if(globpr){
1.246 brouard 3678: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3679: %11.6f %11.6f %11.6f ", \
1.242 brouard 3680: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3681: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3682: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3683: llt +=ll[k]*gipmx/gsw;
3684: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3685: }
3686: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3687: }
1.232 brouard 3688: } /* end of wave */
3689: } /* end of individual */
3690: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3691: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3692: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3693: if(globpr==0){ /* First time we count the contributions and weights */
3694: gipmx=ipmx;
3695: gsw=sw;
3696: }
3697: return -l;
1.126 brouard 3698: }
3699:
3700:
3701: /*************** function likelione ***********/
3702: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3703: {
3704: /* This routine should help understanding what is done with
3705: the selection of individuals/waves and
3706: to check the exact contribution to the likelihood.
3707: Plotting could be done.
3708: */
3709: int k;
3710:
3711: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3712: strcpy(fileresilk,"ILK_");
1.202 brouard 3713: strcat(fileresilk,fileresu);
1.126 brouard 3714: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3715: printf("Problem with resultfile: %s\n", fileresilk);
3716: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3717: }
1.214 brouard 3718: 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");
3719: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3720: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3721: for(k=1; k<=nlstate; k++)
3722: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3723: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3724: }
3725:
3726: *fretone=(*funcone)(p);
3727: if(*globpri !=0){
3728: fclose(ficresilk);
1.205 brouard 3729: if (mle ==0)
3730: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3731: else if(mle >=1)
3732: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3733: 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 3734:
1.208 brouard 3735:
3736: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3737: 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 3738: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3739: }
1.207 brouard 3740: 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 3741: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3742: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3743: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3744: fflush(fichtm);
1.205 brouard 3745: }
1.126 brouard 3746: return;
3747: }
3748:
3749:
3750: /*********** Maximum Likelihood Estimation ***************/
3751:
3752: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3753: {
1.165 brouard 3754: int i,j, iter=0;
1.126 brouard 3755: double **xi;
3756: double fret;
3757: double fretone; /* Only one call to likelihood */
3758: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3759:
3760: #ifdef NLOPT
3761: int creturn;
3762: nlopt_opt opt;
3763: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3764: double *lb;
3765: double minf; /* the minimum objective value, upon return */
3766: double * p1; /* Shifted parameters from 0 instead of 1 */
3767: myfunc_data dinst, *d = &dinst;
3768: #endif
3769:
3770:
1.126 brouard 3771: xi=matrix(1,npar,1,npar);
3772: for (i=1;i<=npar;i++)
3773: for (j=1;j<=npar;j++)
3774: xi[i][j]=(i==j ? 1.0 : 0.0);
3775: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3776: strcpy(filerespow,"POW_");
1.126 brouard 3777: strcat(filerespow,fileres);
3778: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3779: printf("Problem with resultfile: %s\n", filerespow);
3780: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3781: }
3782: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3783: for (i=1;i<=nlstate;i++)
3784: for(j=1;j<=nlstate+ndeath;j++)
3785: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3786: fprintf(ficrespow,"\n");
1.162 brouard 3787: #ifdef POWELL
1.126 brouard 3788: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3789: #endif
1.126 brouard 3790:
1.162 brouard 3791: #ifdef NLOPT
3792: #ifdef NEWUOA
3793: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3794: #else
3795: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3796: #endif
3797: lb=vector(0,npar-1);
3798: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3799: nlopt_set_lower_bounds(opt, lb);
3800: nlopt_set_initial_step1(opt, 0.1);
3801:
3802: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3803: d->function = func;
3804: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3805: nlopt_set_min_objective(opt, myfunc, d);
3806: nlopt_set_xtol_rel(opt, ftol);
3807: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3808: printf("nlopt failed! %d\n",creturn);
3809: }
3810: else {
3811: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3812: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3813: iter=1; /* not equal */
3814: }
3815: nlopt_destroy(opt);
3816: #endif
1.126 brouard 3817: free_matrix(xi,1,npar,1,npar);
3818: fclose(ficrespow);
1.203 brouard 3819: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3820: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3821: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3822:
3823: }
3824:
3825: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3826: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3827: {
3828: double **a,**y,*x,pd;
1.203 brouard 3829: /* double **hess; */
1.164 brouard 3830: int i, j;
1.126 brouard 3831: int *indx;
3832:
3833: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3834: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3835: void lubksb(double **a, int npar, int *indx, double b[]) ;
3836: void ludcmp(double **a, int npar, int *indx, double *d) ;
3837: double gompertz(double p[]);
1.203 brouard 3838: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3839:
3840: printf("\nCalculation of the hessian matrix. Wait...\n");
3841: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3842: for (i=1;i<=npar;i++){
1.203 brouard 3843: printf("%d-",i);fflush(stdout);
3844: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3845:
3846: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3847:
3848: /* printf(" %f ",p[i]);
3849: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3850: }
3851:
3852: for (i=1;i<=npar;i++) {
3853: for (j=1;j<=npar;j++) {
3854: if (j>i) {
1.203 brouard 3855: printf(".%d-%d",i,j);fflush(stdout);
3856: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3857: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3858:
3859: hess[j][i]=hess[i][j];
3860: /*printf(" %lf ",hess[i][j]);*/
3861: }
3862: }
3863: }
3864: printf("\n");
3865: fprintf(ficlog,"\n");
3866:
3867: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3868: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3869:
3870: a=matrix(1,npar,1,npar);
3871: y=matrix(1,npar,1,npar);
3872: x=vector(1,npar);
3873: indx=ivector(1,npar);
3874: for (i=1;i<=npar;i++)
3875: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3876: ludcmp(a,npar,indx,&pd);
3877:
3878: for (j=1;j<=npar;j++) {
3879: for (i=1;i<=npar;i++) x[i]=0;
3880: x[j]=1;
3881: lubksb(a,npar,indx,x);
3882: for (i=1;i<=npar;i++){
3883: matcov[i][j]=x[i];
3884: }
3885: }
3886:
3887: printf("\n#Hessian matrix#\n");
3888: fprintf(ficlog,"\n#Hessian matrix#\n");
3889: for (i=1;i<=npar;i++) {
3890: for (j=1;j<=npar;j++) {
1.203 brouard 3891: printf("%.6e ",hess[i][j]);
3892: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3893: }
3894: printf("\n");
3895: fprintf(ficlog,"\n");
3896: }
3897:
1.203 brouard 3898: /* printf("\n#Covariance matrix#\n"); */
3899: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3900: /* for (i=1;i<=npar;i++) { */
3901: /* for (j=1;j<=npar;j++) { */
3902: /* printf("%.6e ",matcov[i][j]); */
3903: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3904: /* } */
3905: /* printf("\n"); */
3906: /* fprintf(ficlog,"\n"); */
3907: /* } */
3908:
1.126 brouard 3909: /* Recompute Inverse */
1.203 brouard 3910: /* for (i=1;i<=npar;i++) */
3911: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3912: /* ludcmp(a,npar,indx,&pd); */
3913:
3914: /* printf("\n#Hessian matrix recomputed#\n"); */
3915:
3916: /* for (j=1;j<=npar;j++) { */
3917: /* for (i=1;i<=npar;i++) x[i]=0; */
3918: /* x[j]=1; */
3919: /* lubksb(a,npar,indx,x); */
3920: /* for (i=1;i<=npar;i++){ */
3921: /* y[i][j]=x[i]; */
3922: /* printf("%.3e ",y[i][j]); */
3923: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3924: /* } */
3925: /* printf("\n"); */
3926: /* fprintf(ficlog,"\n"); */
3927: /* } */
3928:
3929: /* Verifying the inverse matrix */
3930: #ifdef DEBUGHESS
3931: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3932:
1.203 brouard 3933: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3934: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3935:
3936: for (j=1;j<=npar;j++) {
3937: for (i=1;i<=npar;i++){
1.203 brouard 3938: printf("%.2f ",y[i][j]);
3939: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3940: }
3941: printf("\n");
3942: fprintf(ficlog,"\n");
3943: }
1.203 brouard 3944: #endif
1.126 brouard 3945:
3946: free_matrix(a,1,npar,1,npar);
3947: free_matrix(y,1,npar,1,npar);
3948: free_vector(x,1,npar);
3949: free_ivector(indx,1,npar);
1.203 brouard 3950: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3951:
3952:
3953: }
3954:
3955: /*************** hessian matrix ****************/
3956: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3957: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3958: int i;
3959: int l=1, lmax=20;
1.203 brouard 3960: double k1,k2, res, fx;
1.132 brouard 3961: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3962: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3963: int k=0,kmax=10;
3964: double l1;
3965:
3966: fx=func(x);
3967: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3968: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3969: l1=pow(10,l);
3970: delts=delt;
3971: for(k=1 ; k <kmax; k=k+1){
3972: delt = delta*(l1*k);
3973: p2[theta]=x[theta] +delt;
1.145 brouard 3974: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3975: p2[theta]=x[theta]-delt;
3976: k2=func(p2)-fx;
3977: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3978: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3979:
1.203 brouard 3980: #ifdef DEBUGHESSII
1.126 brouard 3981: 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);
3982: 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);
3983: #endif
3984: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3985: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3986: k=kmax;
3987: }
3988: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3989: k=kmax; l=lmax*10;
1.126 brouard 3990: }
3991: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3992: delts=delt;
3993: }
1.203 brouard 3994: } /* End loop k */
1.126 brouard 3995: }
3996: delti[theta]=delts;
3997: return res;
3998:
3999: }
4000:
1.203 brouard 4001: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4002: {
4003: int i;
1.164 brouard 4004: int l=1, lmax=20;
1.126 brouard 4005: double k1,k2,k3,k4,res,fx;
1.132 brouard 4006: double p2[MAXPARM+1];
1.203 brouard 4007: int k, kmax=1;
4008: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4009:
4010: int firstime=0;
1.203 brouard 4011:
1.126 brouard 4012: fx=func(x);
1.203 brouard 4013: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4014: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4015: p2[thetai]=x[thetai]+delti[thetai]*k;
4016: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4017: k1=func(p2)-fx;
4018:
1.203 brouard 4019: p2[thetai]=x[thetai]+delti[thetai]*k;
4020: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4021: k2=func(p2)-fx;
4022:
1.203 brouard 4023: p2[thetai]=x[thetai]-delti[thetai]*k;
4024: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4025: k3=func(p2)-fx;
4026:
1.203 brouard 4027: p2[thetai]=x[thetai]-delti[thetai]*k;
4028: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4029: k4=func(p2)-fx;
1.203 brouard 4030: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4031: if(k1*k2*k3*k4 <0.){
1.208 brouard 4032: firstime=1;
1.203 brouard 4033: kmax=kmax+10;
1.208 brouard 4034: }
4035: if(kmax >=10 || firstime ==1){
1.246 brouard 4036: 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);
4037: 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 4038: 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);
4039: 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);
4040: }
4041: #ifdef DEBUGHESSIJ
4042: v1=hess[thetai][thetai];
4043: v2=hess[thetaj][thetaj];
4044: cv12=res;
4045: /* Computing eigen value of Hessian matrix */
4046: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4047: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4048: if ((lc2 <0) || (lc1 <0) ){
4049: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4050: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4051: 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);
4052: 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);
4053: }
1.126 brouard 4054: #endif
4055: }
4056: return res;
4057: }
4058:
1.203 brouard 4059: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4060: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4061: /* { */
4062: /* int i; */
4063: /* int l=1, lmax=20; */
4064: /* double k1,k2,k3,k4,res,fx; */
4065: /* double p2[MAXPARM+1]; */
4066: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4067: /* int k=0,kmax=10; */
4068: /* double l1; */
4069:
4070: /* fx=func(x); */
4071: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4072: /* l1=pow(10,l); */
4073: /* delts=delt; */
4074: /* for(k=1 ; k <kmax; k=k+1){ */
4075: /* delt = delti*(l1*k); */
4076: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4077: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4078: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4079: /* k1=func(p2)-fx; */
4080:
4081: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4082: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4083: /* k2=func(p2)-fx; */
4084:
4085: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4086: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4087: /* k3=func(p2)-fx; */
4088:
4089: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4090: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4091: /* k4=func(p2)-fx; */
4092: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4093: /* #ifdef DEBUGHESSIJ */
4094: /* 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); */
4095: /* 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); */
4096: /* #endif */
4097: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4098: /* k=kmax; */
4099: /* } */
4100: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4101: /* k=kmax; l=lmax*10; */
4102: /* } */
4103: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4104: /* delts=delt; */
4105: /* } */
4106: /* } /\* End loop k *\/ */
4107: /* } */
4108: /* delti[theta]=delts; */
4109: /* return res; */
4110: /* } */
4111:
4112:
1.126 brouard 4113: /************** Inverse of matrix **************/
4114: void ludcmp(double **a, int n, int *indx, double *d)
4115: {
4116: int i,imax,j,k;
4117: double big,dum,sum,temp;
4118: double *vv;
4119:
4120: vv=vector(1,n);
4121: *d=1.0;
4122: for (i=1;i<=n;i++) {
4123: big=0.0;
4124: for (j=1;j<=n;j++)
4125: if ((temp=fabs(a[i][j])) > big) big=temp;
4126: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4127: vv[i]=1.0/big;
4128: }
4129: for (j=1;j<=n;j++) {
4130: for (i=1;i<j;i++) {
4131: sum=a[i][j];
4132: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4133: a[i][j]=sum;
4134: }
4135: big=0.0;
4136: for (i=j;i<=n;i++) {
4137: sum=a[i][j];
4138: for (k=1;k<j;k++)
4139: sum -= a[i][k]*a[k][j];
4140: a[i][j]=sum;
4141: if ( (dum=vv[i]*fabs(sum)) >= big) {
4142: big=dum;
4143: imax=i;
4144: }
4145: }
4146: if (j != imax) {
4147: for (k=1;k<=n;k++) {
4148: dum=a[imax][k];
4149: a[imax][k]=a[j][k];
4150: a[j][k]=dum;
4151: }
4152: *d = -(*d);
4153: vv[imax]=vv[j];
4154: }
4155: indx[j]=imax;
4156: if (a[j][j] == 0.0) a[j][j]=TINY;
4157: if (j != n) {
4158: dum=1.0/(a[j][j]);
4159: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4160: }
4161: }
4162: free_vector(vv,1,n); /* Doesn't work */
4163: ;
4164: }
4165:
4166: void lubksb(double **a, int n, int *indx, double b[])
4167: {
4168: int i,ii=0,ip,j;
4169: double sum;
4170:
4171: for (i=1;i<=n;i++) {
4172: ip=indx[i];
4173: sum=b[ip];
4174: b[ip]=b[i];
4175: if (ii)
4176: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4177: else if (sum) ii=i;
4178: b[i]=sum;
4179: }
4180: for (i=n;i>=1;i--) {
4181: sum=b[i];
4182: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4183: b[i]=sum/a[i][i];
4184: }
4185: }
4186:
4187: void pstamp(FILE *fichier)
4188: {
1.196 brouard 4189: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4190: }
4191:
1.253 ! brouard 4192: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
! 4193:
! 4194: /* y=a+bx regression */
! 4195: double sumx = 0.0; /* sum of x */
! 4196: double sumx2 = 0.0; /* sum of x**2 */
! 4197: double sumxy = 0.0; /* sum of x * y */
! 4198: double sumy = 0.0; /* sum of y */
! 4199: double sumy2 = 0.0; /* sum of y**2 */
! 4200: double sume2; /* sum of square or residuals */
! 4201: double yhat;
! 4202:
! 4203: double denom=0;
! 4204: int i;
! 4205: int ne=*no;
! 4206:
! 4207: for ( i=ifi, ne=0;i<=ila;i++) {
! 4208: if(!isfinite(x[i]) || !isfinite(y[i])){
! 4209: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
! 4210: continue;
! 4211: }
! 4212: ne=ne+1;
! 4213: sumx += x[i];
! 4214: sumx2 += x[i]*x[i];
! 4215: sumxy += x[i] * y[i];
! 4216: sumy += y[i];
! 4217: sumy2 += y[i]*y[i];
! 4218: denom = (ne * sumx2 - sumx*sumx);
! 4219: /* 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); */
! 4220: }
! 4221:
! 4222: denom = (ne * sumx2 - sumx*sumx);
! 4223: if (denom == 0) {
! 4224: // vertical, slope m is infinity
! 4225: *b = INFINITY;
! 4226: *a = 0;
! 4227: if (r) *r = 0;
! 4228: return 1;
! 4229: }
! 4230:
! 4231: *b = (ne * sumxy - sumx * sumy) / denom;
! 4232: *a = (sumy * sumx2 - sumx * sumxy) / denom;
! 4233: if (r!=NULL) {
! 4234: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
! 4235: sqrt((sumx2 - sumx*sumx/ne) *
! 4236: (sumy2 - sumy*sumy/ne));
! 4237: }
! 4238: *no=ne;
! 4239: for ( i=ifi, ne=0;i<=ila;i++) {
! 4240: if(!isfinite(x[i]) || !isfinite(y[i])){
! 4241: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
! 4242: continue;
! 4243: }
! 4244: ne=ne+1;
! 4245: yhat = y[i] - *a -*b* x[i];
! 4246: sume2 += yhat * yhat ;
! 4247:
! 4248: denom = (ne * sumx2 - sumx*sumx);
! 4249: /* 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); */
! 4250: }
! 4251: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
! 4252: *sa= *sb * sqrt(sumx2/ne);
! 4253:
! 4254: return 0;
! 4255: }
! 4256:
1.126 brouard 4257: /************ Frequencies ********************/
1.251 brouard 4258: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4259: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4260: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4261: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4262:
1.253 ! brouard 4263: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0;
1.226 brouard 4264: int iind=0, iage=0;
4265: int mi; /* Effective wave */
4266: int first;
4267: double ***freq; /* Frequencies */
1.253 ! brouard 4268: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
! 4269: int no;
1.226 brouard 4270: double *meanq;
4271: double **meanqt;
4272: double *pp, **prop, *posprop, *pospropt;
4273: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4274: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4275: double agebegin, ageend;
4276:
4277: pp=vector(1,nlstate);
1.251 brouard 4278: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4279: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4280: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4281: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4282: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4283: meanqt=matrix(1,lastpass,1,nqtveff);
4284: strcpy(fileresp,"P_");
4285: strcat(fileresp,fileresu);
4286: /*strcat(fileresphtm,fileresu);*/
4287: if((ficresp=fopen(fileresp,"w"))==NULL) {
4288: printf("Problem with prevalence resultfile: %s\n", fileresp);
4289: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4290: exit(0);
4291: }
1.240 brouard 4292:
1.226 brouard 4293: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4294: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4295: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4296: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4297: fflush(ficlog);
4298: exit(70);
4299: }
4300: else{
4301: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4302: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4303: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4304: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4305: }
1.237 brouard 4306: 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 4307:
1.226 brouard 4308: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4309: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4310: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4311: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4312: fflush(ficlog);
4313: exit(70);
1.240 brouard 4314: } else{
1.226 brouard 4315: 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 4316: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4317: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4318: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4319: }
1.240 brouard 4320: 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);
4321:
1.253 ! brouard 4322: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
! 4323: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4324: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4325: j1=0;
1.126 brouard 4326:
1.227 brouard 4327: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4328: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4329: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4330:
4331:
1.226 brouard 4332: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4333: reference=low_education V1=0,V2=0
4334: med_educ V1=1 V2=0,
4335: high_educ V1=0 V2=1
4336: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4337: */
1.249 brouard 4338: dateintsum=0;
4339: k2cpt=0;
4340:
1.253 ! brouard 4341: if(cptcoveff == 0 )
! 4342: nl=1; /* Constant model only */
! 4343: else
! 4344: nl=2;
! 4345: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
! 4346: if(nj==1)
! 4347: j=0; /* First pass for the constant */
! 4348: else
! 4349: j=cptcoveff; /* Other passes for the covariate values */
1.251 brouard 4350: first=1;
4351: 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 */
4352: posproptt=0.;
4353: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4354: scanf("%d", i);*/
4355: for (i=-5; i<=nlstate+ndeath; i++)
4356: for (jk=-5; jk<=nlstate+ndeath; jk++)
4357: for(m=iagemin; m <= iagemax+3; m++)
4358: freq[i][jk][m]=0;
4359:
4360: for (i=1; i<=nlstate; i++) {
1.240 brouard 4361: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4362: prop[i][m]=0;
4363: posprop[i]=0;
4364: pospropt[i]=0;
4365: }
4366: /* for (z1=1; z1<= nqfveff; z1++) { */
4367: /* meanq[z1]+=0.; */
4368: /* for(m=1;m<=lastpass;m++){ */
4369: /* meanqt[m][z1]=0.; */
4370: /* } */
4371: /* } */
4372:
4373: /* dateintsum=0; */
4374: /* k2cpt=0; */
4375:
4376: /* For that combination of covariate j1, we count and print the frequencies in one pass */
4377: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4378: bool=1;
4379: if(j !=0){
4380: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4381: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4382: /* for (z1=1; z1<= nqfveff; z1++) { */
4383: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4384: /* } */
4385: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4386: /* if(Tvaraff[z1] ==-20){ */
4387: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4388: /* }else if(Tvaraff[z1] ==-10){ */
4389: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4390: /* }else */
4391: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
4392: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
4393: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4394: /* 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",
4395: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4396: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4397: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4398: } /* Onlyf fixed */
4399: } /* end z1 */
4400: } /* cptcovn > 0 */
4401: } /* end any */
4402: }/* end j==0 */
4403: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
4404: /* for(m=firstpass; m<=lastpass; m++){ */
4405: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4406: m=mw[mi][iind];
4407: if(j!=0){
4408: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4409: for (z1=1; z1<=cptcoveff; z1++) {
4410: if( Fixed[Tmodelind[z1]]==1){
4411: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4412: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4413: value is -1, we don't select. It differs from the
4414: constant and age model which counts them. */
4415: bool=0; /* not selected */
4416: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4417: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4418: bool=0;
4419: }
4420: }
4421: }
4422: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4423: } /* end j==0 */
4424: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4425: if(bool==1){
4426: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4427: and mw[mi+1][iind]. dh depends on stepm. */
4428: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4429: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4430: if(m >=firstpass && m <=lastpass){
4431: k2=anint[m][iind]+(mint[m][iind]/12.);
4432: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4433: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4434: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4435: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4436: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4437: if (m<lastpass) {
4438: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4439: /* 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]); */
4440: if(s[m][iind]==-1)
4441: 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.));
4442: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4443: /* if((int)agev[m][iind] == 55) */
4444: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4445: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4446: 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 4447: }
1.251 brouard 4448: } /* end if between passes */
4449: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4450: dateintsum=dateintsum+k2; /* on all covariates ?*/
4451: k2cpt++;
4452: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4453: }
1.251 brouard 4454: }else{
4455: bool=1;
4456: }/* end bool 2 */
4457: } /* end m */
4458: } /* end bool */
4459: } /* end iind = 1 to imx */
4460: /* prop[s][age] is feeded for any initial and valid live state as well as
4461: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4462:
4463:
4464: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4465: pstamp(ficresp);
4466: if (cptcoveff>0 && j!=0){
4467: printf( "\n#********** Variable ");
4468: fprintf(ficresp, "\n#********** Variable ");
4469: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4470: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4471: fprintf(ficlog, "\n#********** Variable ");
4472: for (z1=1; z1<=cptcoveff; z1++){
4473: if(!FixedV[Tvaraff[z1]]){
4474: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4475: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4476: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4477: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4478: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4479: }else{
1.251 brouard 4480: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4481: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4482: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4483: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4484: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4485: }
4486: }
4487: printf( "**********\n#");
4488: fprintf(ficresp, "**********\n#");
4489: fprintf(ficresphtm, "**********</h3>\n");
4490: fprintf(ficresphtmfr, "**********</h3>\n");
4491: fprintf(ficlog, "**********\n");
4492: }
4493: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4494: for(i=1; i<=nlstate;i++) {
4495: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
4496: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4497: }
4498: fprintf(ficresp, "\n");
4499: fprintf(ficresphtm, "\n");
4500:
4501: /* Header of frequency table by age */
4502: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4503: fprintf(ficresphtmfr,"<th>Age</th> ");
4504: for(jk=-1; jk <=nlstate+ndeath; jk++){
4505: for(m=-1; m <=nlstate+ndeath; m++){
4506: if(jk!=0 && m!=0)
4507: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.240 brouard 4508: }
1.226 brouard 4509: }
1.251 brouard 4510: fprintf(ficresphtmfr, "\n");
4511:
4512: /* For each age */
4513: for(iage=iagemin; iage <= iagemax+3; iage++){
4514: fprintf(ficresphtm,"<tr>");
4515: if(iage==iagemax+1){
4516: fprintf(ficlog,"1");
4517: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4518: }else if(iage==iagemax+2){
4519: fprintf(ficlog,"0");
4520: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4521: }else if(iage==iagemax+3){
4522: fprintf(ficlog,"Total");
4523: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4524: }else{
1.240 brouard 4525: if(first==1){
1.251 brouard 4526: first=0;
4527: printf("See log file for details...\n");
4528: }
4529: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4530: fprintf(ficlog,"Age %d", iage);
4531: }
4532: for(jk=1; jk <=nlstate ; jk++){
4533: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4534: pp[jk] += freq[jk][m][iage];
4535: }
4536: for(jk=1; jk <=nlstate ; jk++){
4537: for(m=-1, pos=0; m <=0 ; m++)
4538: pos += freq[jk][m][iage];
4539: if(pp[jk]>=1.e-10){
4540: if(first==1){
4541: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4542: }
4543: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4544: }else{
4545: if(first==1)
4546: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4547: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
1.240 brouard 4548: }
4549: }
4550:
1.251 brouard 4551: for(jk=1; jk <=nlstate ; jk++){
4552: /* posprop[jk]=0; */
4553: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4554: pp[jk] += freq[jk][m][iage];
4555: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
4556:
4557: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
4558: pos += pp[jk]; /* pos is the total number of transitions until this age */
4559: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4560: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4561: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
1.240 brouard 4562: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4563: }
1.251 brouard 4564: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4565: if(pos>=1.e-5){
1.251 brouard 4566: if(first==1)
4567: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4568: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4569: }else{
4570: if(first==1)
4571: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4572: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4573: }
4574: if( iage <= iagemax){
4575: if(pos>=1.e-5){
4576: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4577: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4578: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4579: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4580: }
4581: else{
4582: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4583: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4584: }
1.240 brouard 4585: }
1.251 brouard 4586: pospropt[jk] +=posprop[jk];
4587: } /* end loop jk */
4588: /* pospropt=0.; */
4589: for(jk=-1; jk <=nlstate+ndeath; jk++){
4590: for(m=-1; m <=nlstate+ndeath; m++){
4591: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4592: if(first==1){
4593: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4594: }
1.253 ! brouard 4595: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]); */
1.251 brouard 4596: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4597: }
4598: if(jk!=0 && m!=0)
4599: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
1.240 brouard 4600: }
1.251 brouard 4601: } /* end loop jk */
4602: posproptt=0.;
4603: for(jk=1; jk <=nlstate; jk++){
4604: posproptt += pospropt[jk];
4605: }
4606: fprintf(ficresphtmfr,"</tr>\n ");
4607: if(iage <= iagemax){
4608: fprintf(ficresp,"\n");
4609: fprintf(ficresphtm,"</tr>\n");
1.240 brouard 4610: }
1.251 brouard 4611: if(first==1)
4612: printf("Others in log...\n");
4613: fprintf(ficlog,"\n");
4614: } /* end loop age iage */
4615: fprintf(ficresphtm,"<tr><th>Tot</th>");
4616: for(jk=1; jk <=nlstate ; jk++){
4617: if(posproptt < 1.e-5){
4618: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
4619: }else{
4620: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.240 brouard 4621: }
1.226 brouard 4622: }
1.251 brouard 4623: fprintf(ficresphtm,"</tr>\n");
4624: fprintf(ficresphtm,"</table>\n");
4625: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4626: if(posproptt < 1.e-5){
1.251 brouard 4627: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4628: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4629: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4630: invalidvarcomb[j1]=1;
1.226 brouard 4631: }else{
1.251 brouard 4632: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4633: invalidvarcomb[j1]=0;
1.226 brouard 4634: }
1.251 brouard 4635: fprintf(ficresphtmfr,"</table>\n");
4636: fprintf(ficlog,"\n");
4637: if(j!=0){
4638: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
4639: for(i=1,jk=1; i <=nlstate; i++){
4640: for(k=1; k <=(nlstate+ndeath); k++){
4641: if (k != i) {
4642: for(jj=1; jj <=ncovmodel; jj++){ /* For counting jk */
1.253 ! brouard 4643: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4644: if(j1==1){ /* All dummy covariates to zero */
4645: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4646: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4647: printf("%d%d ",i,k);
4648: fprintf(ficlog,"%d%d ",i,k);
4649: 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]));
4650: 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]));
4651: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4652: }
1.253 ! brouard 4653: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
! 4654: for(iage=iagemin; iage <= iagemax+3; iage++){
! 4655: x[iage]= (double)iage;
! 4656: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
! 4657: /* printf("i=%d, k=%d, jk=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,jk,j1,jj, iage, y[iage]); */
! 4658: }
! 4659: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
! 4660: pstart[jk]=b;
! 4661: pstart[jk-1]=a;
1.252 brouard 4662: }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 */
4663: 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]);
4664: 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 4665: 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 4666: printf("%d%d ",i,k);
4667: fprintf(ficlog,"%d%d ",i,k);
1.251 brouard 4668: 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]));
4669: }else{ /* Other cases, like quantitative fixed or varying covariates */
4670: ;
4671: }
4672: /* printf("%12.7f )", param[i][jj][k]); */
4673: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4674: jk++;
4675: } /* end jj */
4676: } /* end k!= i */
4677: } /* end k */
4678: } /* end i, jk */
4679: } /* end j !=0 */
4680: } /* end selected combination of covariate j1 */
4681: if(j==0){ /* We can estimate starting values from the occurences in each case */
4682: printf("#Freqsummary: Starting values for the constants:\n");
4683: fprintf(ficlog,"\n");
4684: for(i=1,jk=1; i <=nlstate; i++){
4685: for(k=1; k <=(nlstate+ndeath); k++){
4686: if (k != i) {
4687: printf("%d%d ",i,k);
4688: fprintf(ficlog,"%d%d ",i,k);
4689: for(jj=1; jj <=ncovmodel; jj++){
1.253 ! brouard 4690: pstart[jk]=p[jk]; /* Setting pstart to p values by default */
! 4691: if(jj==1){ /* Age has to be done */
! 4692: pstart[jk]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4693: 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]));
4694: 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]));
4695: }
4696: /* printf("%12.7f )", param[i][jj][k]); */
4697: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
4698: jk++;
1.250 brouard 4699: }
1.251 brouard 4700: printf("\n");
4701: fprintf(ficlog,"\n");
1.250 brouard 4702: }
4703: }
4704: }
1.251 brouard 4705: printf("#Freqsummary\n");
4706: fprintf(ficlog,"\n");
4707: for(jk=-1; jk <=nlstate+ndeath; jk++){
4708: for(m=-1; m <=nlstate+ndeath; m++){
4709: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 brouard 4710: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
4711: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
1.251 brouard 4712: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
4713: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4714: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
4715: /* } */
4716: }
4717: } /* end loop jk */
4718:
4719: printf("\n");
4720: fprintf(ficlog,"\n");
4721: } /* end j=0 */
1.249 brouard 4722: } /* end j */
1.252 brouard 4723:
1.253 ! brouard 4724: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4725: for(i=1, jk=1; i <=nlstate; i++){
4726: for(j=1; j <=nlstate+ndeath; j++){
4727: if(j!=i){
4728: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4729: printf("%1d%1d",i,j);
4730: fprintf(ficparo,"%1d%1d",i,j);
4731: for(k=1; k<=ncovmodel;k++){
4732: /* printf(" %lf",param[i][j][k]); */
4733: /* fprintf(ficparo," %lf",param[i][j][k]); */
4734: p[jk]=pstart[jk];
4735: printf(" %f ",pstart[jk]);
4736: fprintf(ficparo," %f ",pstart[jk]);
4737: jk++;
4738: }
4739: printf("\n");
4740: fprintf(ficparo,"\n");
4741: }
4742: }
4743: }
4744: } /* end mle=-2 */
1.226 brouard 4745: dateintmean=dateintsum/k2cpt;
1.240 brouard 4746:
1.226 brouard 4747: fclose(ficresp);
4748: fclose(ficresphtm);
4749: fclose(ficresphtmfr);
4750: free_vector(meanq,1,nqfveff);
4751: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 ! brouard 4752: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
! 4753: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4754: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4755: free_vector(pospropt,1,nlstate);
4756: free_vector(posprop,1,nlstate);
1.251 brouard 4757: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4758: free_vector(pp,1,nlstate);
4759: /* End of freqsummary */
4760: }
1.126 brouard 4761:
4762: /************ Prevalence ********************/
1.227 brouard 4763: 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)
4764: {
4765: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4766: in each health status at the date of interview (if between dateprev1 and dateprev2).
4767: We still use firstpass and lastpass as another selection.
4768: */
1.126 brouard 4769:
1.227 brouard 4770: int i, m, jk, j1, bool, z1,j, iv;
4771: int mi; /* Effective wave */
4772: int iage;
4773: double agebegin, ageend;
4774:
4775: double **prop;
4776: double posprop;
4777: double y2; /* in fractional years */
4778: int iagemin, iagemax;
4779: int first; /** to stop verbosity which is redirected to log file */
4780:
4781: iagemin= (int) agemin;
4782: iagemax= (int) agemax;
4783: /*pp=vector(1,nlstate);*/
1.251 brouard 4784: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4785: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4786: j1=0;
1.222 brouard 4787:
1.227 brouard 4788: /*j=cptcoveff;*/
4789: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4790:
1.227 brouard 4791: first=1;
4792: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4793: for (i=1; i<=nlstate; i++)
1.251 brouard 4794: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4795: prop[i][iage]=0.0;
4796: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4797: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4798: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4799:
4800: for (i=1; i<=imx; i++) { /* Each individual */
4801: bool=1;
4802: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4803: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4804: m=mw[mi][i];
4805: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4806: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4807: for (z1=1; z1<=cptcoveff; z1++){
4808: if( Fixed[Tmodelind[z1]]==1){
4809: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4810: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4811: bool=0;
4812: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4813: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4814: bool=0;
4815: }
4816: }
4817: if(bool==1){ /* Otherwise we skip that wave/person */
4818: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4819: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4820: if(m >=firstpass && m <=lastpass){
4821: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4822: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4823: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4824: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4825: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4826: 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);
4827: exit(1);
4828: }
4829: if (s[m][i]>0 && s[m][i]<=nlstate) {
4830: /*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]]);*/
4831: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4832: prop[s[m][i]][iagemax+3] += weight[i];
4833: } /* end valid statuses */
4834: } /* end selection of dates */
4835: } /* end selection of waves */
4836: } /* end bool */
4837: } /* end wave */
4838: } /* end individual */
4839: for(i=iagemin; i <= iagemax+3; i++){
4840: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4841: posprop += prop[jk][i];
4842: }
4843:
4844: for(jk=1; jk <=nlstate ; jk++){
4845: if( i <= iagemax){
4846: if(posprop>=1.e-5){
4847: probs[i][jk][j1]= prop[jk][i]/posprop;
4848: } else{
4849: if(first==1){
4850: first=0;
4851: 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]);
4852: }
4853: }
4854: }
4855: }/* end jk */
4856: }/* end i */
1.222 brouard 4857: /*} *//* end i1 */
1.227 brouard 4858: } /* end j1 */
1.222 brouard 4859:
1.227 brouard 4860: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4861: /*free_vector(pp,1,nlstate);*/
1.251 brouard 4862: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4863: } /* End of prevalence */
1.126 brouard 4864:
4865: /************* Waves Concatenation ***************/
4866:
4867: 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)
4868: {
4869: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4870: Death is a valid wave (if date is known).
4871: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4872: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4873: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4874: */
1.126 brouard 4875:
1.224 brouard 4876: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4877: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4878: double sum=0., jmean=0.;*/
1.224 brouard 4879: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4880: int j, k=0,jk, ju, jl;
4881: double sum=0.;
4882: first=0;
1.214 brouard 4883: firstwo=0;
1.217 brouard 4884: firsthree=0;
1.218 brouard 4885: firstfour=0;
1.164 brouard 4886: jmin=100000;
1.126 brouard 4887: jmax=-1;
4888: jmean=0.;
1.224 brouard 4889:
4890: /* Treating live states */
1.214 brouard 4891: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4892: mi=0; /* First valid wave */
1.227 brouard 4893: mli=0; /* Last valid wave */
1.126 brouard 4894: m=firstpass;
1.214 brouard 4895: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4896: 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 */
4897: mli=m-1;/* mw[++mi][i]=m-1; */
4898: }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 */
4899: mw[++mi][i]=m;
4900: mli=m;
1.224 brouard 4901: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4902: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4903: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4904: }
1.227 brouard 4905: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4906: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4907: break;
1.224 brouard 4908: #else
1.227 brouard 4909: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4910: if(firsthree == 0){
4911: 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);
4912: firsthree=1;
4913: }
4914: 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);
4915: mw[++mi][i]=m;
4916: mli=m;
4917: }
4918: if(s[m][i]==-2){ /* Vital status is really unknown */
4919: nbwarn++;
4920: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4921: 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);
4922: 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);
4923: }
4924: break;
4925: }
4926: break;
1.224 brouard 4927: #endif
1.227 brouard 4928: }/* End m >= lastpass */
1.126 brouard 4929: }/* end while */
1.224 brouard 4930:
1.227 brouard 4931: /* 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 4932: /* After last pass */
1.224 brouard 4933: /* Treating death states */
1.214 brouard 4934: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4935: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4936: /* } */
1.126 brouard 4937: mi++; /* Death is another wave */
4938: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4939: /* Only death is a correct wave */
1.126 brouard 4940: mw[mi][i]=m;
1.224 brouard 4941: }
4942: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4943: else if ((int) andc[i] != 9999) { /* Status is negative. A death occured after lastpass, we can't take it into account because of potential bias */
1.216 brouard 4944: /* m++; */
4945: /* mi++; */
4946: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4947: /* mw[mi][i]=m; */
1.218 brouard 4948: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4949: 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 */
4950: nbwarn++;
4951: if(firstfiv==0){
4952: 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 );
4953: firstfiv=1;
4954: }else{
4955: 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 );
4956: }
4957: }else{ /* Death occured afer last wave potential bias */
4958: nberr++;
4959: if(firstwo==0){
4960: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4961: firstwo=1;
4962: }
4963: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
4964: }
1.218 brouard 4965: }else{ /* end date of interview is known */
1.227 brouard 4966: /* death is known but not confirmed by death status at any wave */
4967: if(firstfour==0){
4968: 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 );
4969: firstfour=1;
4970: }
4971: 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 4972: }
1.224 brouard 4973: } /* end if date of death is known */
4974: #endif
4975: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4976: /* wav[i]=mw[mi][i]; */
1.126 brouard 4977: if(mi==0){
4978: nbwarn++;
4979: if(first==0){
1.227 brouard 4980: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4981: first=1;
1.126 brouard 4982: }
4983: if(first==1){
1.227 brouard 4984: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4985: }
4986: } /* end mi==0 */
4987: } /* End individuals */
1.214 brouard 4988: /* wav and mw are no more changed */
1.223 brouard 4989:
1.214 brouard 4990:
1.126 brouard 4991: for(i=1; i<=imx; i++){
4992: for(mi=1; mi<wav[i];mi++){
4993: if (stepm <=0)
1.227 brouard 4994: dh[mi][i]=1;
1.126 brouard 4995: else{
1.227 brouard 4996: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4997: if (agedc[i] < 2*AGESUP) {
4998: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4999: if(j==0) j=1; /* Survives at least one month after exam */
5000: else if(j<0){
5001: nberr++;
5002: 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]);
5003: j=1; /* Temporary Dangerous patch */
5004: 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);
5005: 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]);
5006: 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);
5007: }
5008: k=k+1;
5009: if (j >= jmax){
5010: jmax=j;
5011: ijmax=i;
5012: }
5013: if (j <= jmin){
5014: jmin=j;
5015: ijmin=i;
5016: }
5017: sum=sum+j;
5018: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5019: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5020: }
5021: }
5022: else{
5023: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5024: /* 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 5025:
1.227 brouard 5026: k=k+1;
5027: if (j >= jmax) {
5028: jmax=j;
5029: ijmax=i;
5030: }
5031: else if (j <= jmin){
5032: jmin=j;
5033: ijmin=i;
5034: }
5035: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5036: /*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]);*/
5037: if(j<0){
5038: nberr++;
5039: 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]);
5040: 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]);
5041: }
5042: sum=sum+j;
5043: }
5044: jk= j/stepm;
5045: jl= j -jk*stepm;
5046: ju= j -(jk+1)*stepm;
5047: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5048: if(jl==0){
5049: dh[mi][i]=jk;
5050: bh[mi][i]=0;
5051: }else{ /* We want a negative bias in order to only have interpolation ie
5052: * to avoid the price of an extra matrix product in likelihood */
5053: dh[mi][i]=jk+1;
5054: bh[mi][i]=ju;
5055: }
5056: }else{
5057: if(jl <= -ju){
5058: dh[mi][i]=jk;
5059: bh[mi][i]=jl; /* bias is positive if real duration
5060: * is higher than the multiple of stepm and negative otherwise.
5061: */
5062: }
5063: else{
5064: dh[mi][i]=jk+1;
5065: bh[mi][i]=ju;
5066: }
5067: if(dh[mi][i]==0){
5068: dh[mi][i]=1; /* At least one step */
5069: bh[mi][i]=ju; /* At least one step */
5070: /* 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);*/
5071: }
5072: } /* end if mle */
1.126 brouard 5073: }
5074: } /* end wave */
5075: }
5076: jmean=sum/k;
5077: 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 5078: 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 5079: }
1.126 brouard 5080:
5081: /*********** Tricode ****************************/
1.220 brouard 5082: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5083: {
5084: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5085: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5086: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5087: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5088: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5089: */
1.130 brouard 5090:
1.242 brouard 5091: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5092: int modmaxcovj=0; /* Modality max of covariates j */
5093: int cptcode=0; /* Modality max of covariates j */
5094: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5095:
5096:
1.242 brouard 5097: /* cptcoveff=0; */
5098: /* *cptcov=0; */
1.126 brouard 5099:
1.242 brouard 5100: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5101:
1.242 brouard 5102: /* Loop on covariates without age and products and no quantitative variable */
5103: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5104: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5105: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5106: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5107: switch(Fixed[k]) {
5108: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5109: 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*/
5110: ij=(int)(covar[Tvar[k]][i]);
5111: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5112: * If product of Vn*Vm, still boolean *:
5113: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5114: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5115: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5116: modality of the nth covariate of individual i. */
5117: if (ij > modmaxcovj)
5118: modmaxcovj=ij;
5119: else if (ij < modmincovj)
5120: modmincovj=ij;
5121: if ((ij < -1) && (ij > NCOVMAX)){
5122: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5123: exit(1);
5124: }else
5125: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5126: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5127: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5128: /* getting the maximum value of the modality of the covariate
5129: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5130: female ies 1, then modmaxcovj=1.
5131: */
5132: } /* end for loop on individuals i */
5133: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5134: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5135: cptcode=modmaxcovj;
5136: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5137: /*for (i=0; i<=cptcode; i++) {*/
5138: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5139: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5140: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5141: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5142: if( j != -1){
5143: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5144: covariate for which somebody answered excluding
5145: undefined. Usually 2: 0 and 1. */
5146: }
5147: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5148: covariate for which somebody answered including
5149: undefined. Usually 3: -1, 0 and 1. */
5150: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5151: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5152: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5153:
1.242 brouard 5154: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5155: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5156: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5157: /* modmincovj=3; modmaxcovj = 7; */
5158: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5159: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5160: /* defining two dummy variables: variables V1_1 and V1_2.*/
5161: /* nbcode[Tvar[j]][ij]=k; */
5162: /* nbcode[Tvar[j]][1]=0; */
5163: /* nbcode[Tvar[j]][2]=1; */
5164: /* nbcode[Tvar[j]][3]=2; */
5165: /* To be continued (not working yet). */
5166: ij=0; /* ij is similar to i but can jump over null modalities */
5167: 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*/
5168: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5169: break;
5170: }
5171: ij++;
5172: 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*/
5173: cptcode = ij; /* New max modality for covar j */
5174: } /* end of loop on modality i=-1 to 1 or more */
5175: break;
5176: case 1: /* Testing on varying covariate, could be simple and
5177: * should look at waves or product of fixed *
5178: * varying. No time to test -1, assuming 0 and 1 only */
5179: ij=0;
5180: for(i=0; i<=1;i++){
5181: nbcode[Tvar[k]][++ij]=i;
5182: }
5183: break;
5184: default:
5185: break;
5186: } /* end switch */
5187: } /* end dummy test */
5188:
5189: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5190: /* /\*recode from 0 *\/ */
5191: /* k is a modality. If we have model=V1+V1*sex */
5192: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5193: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5194: /* } */
5195: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5196: /* if (ij > ncodemax[j]) { */
5197: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5198: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5199: /* break; */
5200: /* } */
5201: /* } /\* end of loop on modality k *\/ */
5202: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5203:
5204: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5205: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5206: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5207: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5208: 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 */
5209: 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 */
5210: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5211: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5212:
5213: ij=0;
5214: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5215: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5216: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5217: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5218: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5219: /* If product not in single variable we don't print results */
5220: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5221: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5222: 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*/
5223: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5224: 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 */
5225: if(Fixed[k]!=0)
5226: anyvaryingduminmodel=1;
5227: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5228: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5229: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5230: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5231: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5232: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5233: }
5234: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5235: /* ij--; */
5236: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5237: *cptcov=ij; /*Number of total real effective covariates: effective
5238: * because they can be excluded from the model and real
5239: * if in the model but excluded because missing values, but how to get k from ij?*/
5240: for(j=ij+1; j<= cptcovt; j++){
5241: Tvaraff[j]=0;
5242: Tmodelind[j]=0;
5243: }
5244: for(j=ntveff+1; j<= cptcovt; j++){
5245: TmodelInvind[j]=0;
5246: }
5247: /* To be sorted */
5248: ;
5249: }
1.126 brouard 5250:
1.145 brouard 5251:
1.126 brouard 5252: /*********** Health Expectancies ****************/
5253:
1.235 brouard 5254: 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 5255:
5256: {
5257: /* Health expectancies, no variances */
1.164 brouard 5258: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5259: int nhstepma, nstepma; /* Decreasing with age */
5260: double age, agelim, hf;
5261: double ***p3mat;
5262: double eip;
5263:
1.238 brouard 5264: /* pstamp(ficreseij); */
1.126 brouard 5265: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5266: fprintf(ficreseij,"# Age");
5267: for(i=1; i<=nlstate;i++){
5268: for(j=1; j<=nlstate;j++){
5269: fprintf(ficreseij," e%1d%1d ",i,j);
5270: }
5271: fprintf(ficreseij," e%1d. ",i);
5272: }
5273: fprintf(ficreseij,"\n");
5274:
5275:
5276: if(estepm < stepm){
5277: printf ("Problem %d lower than %d\n",estepm, stepm);
5278: }
5279: else hstepm=estepm;
5280: /* We compute the life expectancy from trapezoids spaced every estepm months
5281: * This is mainly to measure the difference between two models: for example
5282: * if stepm=24 months pijx are given only every 2 years and by summing them
5283: * we are calculating an estimate of the Life Expectancy assuming a linear
5284: * progression in between and thus overestimating or underestimating according
5285: * to the curvature of the survival function. If, for the same date, we
5286: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5287: * to compare the new estimate of Life expectancy with the same linear
5288: * hypothesis. A more precise result, taking into account a more precise
5289: * curvature will be obtained if estepm is as small as stepm. */
5290:
5291: /* For example we decided to compute the life expectancy with the smallest unit */
5292: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5293: nhstepm is the number of hstepm from age to agelim
5294: nstepm is the number of stepm from age to agelin.
5295: Look at hpijx to understand the reason of that which relies in memory size
5296: and note for a fixed period like estepm months */
5297: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5298: survival function given by stepm (the optimization length). Unfortunately it
5299: means that if the survival funtion is printed only each two years of age and if
5300: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5301: results. So we changed our mind and took the option of the best precision.
5302: */
5303: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5304:
5305: agelim=AGESUP;
5306: /* If stepm=6 months */
5307: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5308: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5309:
5310: /* nhstepm age range expressed in number of stepm */
5311: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5312: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5313: /* if (stepm >= YEARM) hstepm=1;*/
5314: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5315: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5316:
5317: for (age=bage; age<=fage; age ++){
5318: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5319: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5320: /* if (stepm >= YEARM) hstepm=1;*/
5321: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5322:
5323: /* If stepm=6 months */
5324: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5325: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5326:
1.235 brouard 5327: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5328:
5329: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5330:
5331: printf("%d|",(int)age);fflush(stdout);
5332: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5333:
5334: /* Computing expectancies */
5335: for(i=1; i<=nlstate;i++)
5336: for(j=1; j<=nlstate;j++)
5337: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5338: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5339:
5340: /* 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]);*/
5341:
5342: }
5343:
5344: fprintf(ficreseij,"%3.0f",age );
5345: for(i=1; i<=nlstate;i++){
5346: eip=0;
5347: for(j=1; j<=nlstate;j++){
5348: eip +=eij[i][j][(int)age];
5349: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5350: }
5351: fprintf(ficreseij,"%9.4f", eip );
5352: }
5353: fprintf(ficreseij,"\n");
5354:
5355: }
5356: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5357: printf("\n");
5358: fprintf(ficlog,"\n");
5359:
5360: }
5361:
1.235 brouard 5362: 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 5363:
5364: {
5365: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5366: to initial status i, ei. .
1.126 brouard 5367: */
5368: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5369: int nhstepma, nstepma; /* Decreasing with age */
5370: double age, agelim, hf;
5371: double ***p3matp, ***p3matm, ***varhe;
5372: double **dnewm,**doldm;
5373: double *xp, *xm;
5374: double **gp, **gm;
5375: double ***gradg, ***trgradg;
5376: int theta;
5377:
5378: double eip, vip;
5379:
5380: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5381: xp=vector(1,npar);
5382: xm=vector(1,npar);
5383: dnewm=matrix(1,nlstate*nlstate,1,npar);
5384: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5385:
5386: pstamp(ficresstdeij);
5387: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5388: fprintf(ficresstdeij,"# Age");
5389: for(i=1; i<=nlstate;i++){
5390: for(j=1; j<=nlstate;j++)
5391: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5392: fprintf(ficresstdeij," e%1d. ",i);
5393: }
5394: fprintf(ficresstdeij,"\n");
5395:
5396: pstamp(ficrescveij);
5397: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5398: fprintf(ficrescveij,"# Age");
5399: for(i=1; i<=nlstate;i++)
5400: for(j=1; j<=nlstate;j++){
5401: cptj= (j-1)*nlstate+i;
5402: for(i2=1; i2<=nlstate;i2++)
5403: for(j2=1; j2<=nlstate;j2++){
5404: cptj2= (j2-1)*nlstate+i2;
5405: if(cptj2 <= cptj)
5406: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5407: }
5408: }
5409: fprintf(ficrescveij,"\n");
5410:
5411: if(estepm < stepm){
5412: printf ("Problem %d lower than %d\n",estepm, stepm);
5413: }
5414: else hstepm=estepm;
5415: /* We compute the life expectancy from trapezoids spaced every estepm months
5416: * This is mainly to measure the difference between two models: for example
5417: * if stepm=24 months pijx are given only every 2 years and by summing them
5418: * we are calculating an estimate of the Life Expectancy assuming a linear
5419: * progression in between and thus overestimating or underestimating according
5420: * to the curvature of the survival function. If, for the same date, we
5421: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5422: * to compare the new estimate of Life expectancy with the same linear
5423: * hypothesis. A more precise result, taking into account a more precise
5424: * curvature will be obtained if estepm is as small as stepm. */
5425:
5426: /* For example we decided to compute the life expectancy with the smallest unit */
5427: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5428: nhstepm is the number of hstepm from age to agelim
5429: nstepm is the number of stepm from age to agelin.
5430: Look at hpijx to understand the reason of that which relies in memory size
5431: and note for a fixed period like estepm months */
5432: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5433: survival function given by stepm (the optimization length). Unfortunately it
5434: means that if the survival funtion is printed only each two years of age and if
5435: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5436: results. So we changed our mind and took the option of the best precision.
5437: */
5438: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5439:
5440: /* If stepm=6 months */
5441: /* nhstepm age range expressed in number of stepm */
5442: agelim=AGESUP;
5443: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5444: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5445: /* if (stepm >= YEARM) hstepm=1;*/
5446: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5447:
5448: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5449: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5450: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5451: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5452: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5453: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5454:
5455: for (age=bage; age<=fage; age ++){
5456: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5457: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5458: /* if (stepm >= YEARM) hstepm=1;*/
5459: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5460:
1.126 brouard 5461: /* If stepm=6 months */
5462: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5463: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5464:
5465: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5466:
1.126 brouard 5467: /* Computing Variances of health expectancies */
5468: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5469: decrease memory allocation */
5470: for(theta=1; theta <=npar; theta++){
5471: for(i=1; i<=npar; i++){
1.222 brouard 5472: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5473: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5474: }
1.235 brouard 5475: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5476: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5477:
1.126 brouard 5478: for(j=1; j<= nlstate; j++){
1.222 brouard 5479: for(i=1; i<=nlstate; i++){
5480: for(h=0; h<=nhstepm-1; h++){
5481: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5482: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5483: }
5484: }
1.126 brouard 5485: }
1.218 brouard 5486:
1.126 brouard 5487: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5488: for(h=0; h<=nhstepm-1; h++){
5489: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5490: }
1.126 brouard 5491: }/* End theta */
5492:
5493:
5494: for(h=0; h<=nhstepm-1; h++)
5495: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5496: for(theta=1; theta <=npar; theta++)
5497: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5498:
1.218 brouard 5499:
1.222 brouard 5500: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5501: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5502: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5503:
1.222 brouard 5504: printf("%d|",(int)age);fflush(stdout);
5505: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5506: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5507: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5508: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5509: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5510: for(ij=1;ij<=nlstate*nlstate;ij++)
5511: for(ji=1;ji<=nlstate*nlstate;ji++)
5512: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5513: }
5514: }
1.218 brouard 5515:
1.126 brouard 5516: /* Computing expectancies */
1.235 brouard 5517: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5518: for(i=1; i<=nlstate;i++)
5519: for(j=1; j<=nlstate;j++)
1.222 brouard 5520: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5521: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5522:
1.222 brouard 5523: /* 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 5524:
1.222 brouard 5525: }
1.218 brouard 5526:
1.126 brouard 5527: fprintf(ficresstdeij,"%3.0f",age );
5528: for(i=1; i<=nlstate;i++){
5529: eip=0.;
5530: vip=0.;
5531: for(j=1; j<=nlstate;j++){
1.222 brouard 5532: eip += eij[i][j][(int)age];
5533: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5534: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5535: 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 5536: }
5537: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5538: }
5539: fprintf(ficresstdeij,"\n");
1.218 brouard 5540:
1.126 brouard 5541: fprintf(ficrescveij,"%3.0f",age );
5542: for(i=1; i<=nlstate;i++)
5543: for(j=1; j<=nlstate;j++){
1.222 brouard 5544: cptj= (j-1)*nlstate+i;
5545: for(i2=1; i2<=nlstate;i2++)
5546: for(j2=1; j2<=nlstate;j2++){
5547: cptj2= (j2-1)*nlstate+i2;
5548: if(cptj2 <= cptj)
5549: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5550: }
1.126 brouard 5551: }
5552: fprintf(ficrescveij,"\n");
1.218 brouard 5553:
1.126 brouard 5554: }
5555: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5556: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5557: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5558: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5559: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5560: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5561: printf("\n");
5562: fprintf(ficlog,"\n");
1.218 brouard 5563:
1.126 brouard 5564: free_vector(xm,1,npar);
5565: free_vector(xp,1,npar);
5566: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5567: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5568: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5569: }
1.218 brouard 5570:
1.126 brouard 5571: /************ Variance ******************/
1.235 brouard 5572: 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 5573: {
5574: /* Variance of health expectancies */
5575: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5576: /* double **newm;*/
5577: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5578:
5579: /* int movingaverage(); */
5580: double **dnewm,**doldm;
5581: double **dnewmp,**doldmp;
5582: int i, j, nhstepm, hstepm, h, nstepm ;
5583: int k;
5584: double *xp;
5585: double **gp, **gm; /* for var eij */
5586: double ***gradg, ***trgradg; /*for var eij */
5587: double **gradgp, **trgradgp; /* for var p point j */
5588: double *gpp, *gmp; /* for var p point j */
5589: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5590: double ***p3mat;
5591: double age,agelim, hf;
5592: /* double ***mobaverage; */
5593: int theta;
5594: char digit[4];
5595: char digitp[25];
5596:
5597: char fileresprobmorprev[FILENAMELENGTH];
5598:
5599: if(popbased==1){
5600: if(mobilav!=0)
5601: strcpy(digitp,"-POPULBASED-MOBILAV_");
5602: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5603: }
5604: else
5605: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5606:
1.218 brouard 5607: /* if (mobilav!=0) { */
5608: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5609: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5610: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5611: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5612: /* } */
5613: /* } */
5614:
5615: strcpy(fileresprobmorprev,"PRMORPREV-");
5616: sprintf(digit,"%-d",ij);
5617: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5618: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5619: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5620: strcat(fileresprobmorprev,fileresu);
5621: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5622: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5623: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5624: }
5625: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5626: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5627: pstamp(ficresprobmorprev);
5628: 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 5629: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5630: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5631: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5632: }
5633: for(j=1;j<=cptcoveff;j++)
5634: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5635: fprintf(ficresprobmorprev,"\n");
5636:
1.218 brouard 5637: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5638: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5639: fprintf(ficresprobmorprev," p.%-d SE",j);
5640: for(i=1; i<=nlstate;i++)
5641: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5642: }
5643: fprintf(ficresprobmorprev,"\n");
5644:
5645: fprintf(ficgp,"\n# Routine varevsij");
5646: fprintf(ficgp,"\nunset title \n");
5647: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5648: 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");
5649: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5650: /* } */
5651: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5652: pstamp(ficresvij);
5653: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5654: if(popbased==1)
5655: 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);
5656: else
5657: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5658: fprintf(ficresvij,"# Age");
5659: for(i=1; i<=nlstate;i++)
5660: for(j=1; j<=nlstate;j++)
5661: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5662: fprintf(ficresvij,"\n");
5663:
5664: xp=vector(1,npar);
5665: dnewm=matrix(1,nlstate,1,npar);
5666: doldm=matrix(1,nlstate,1,nlstate);
5667: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5668: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5669:
5670: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5671: gpp=vector(nlstate+1,nlstate+ndeath);
5672: gmp=vector(nlstate+1,nlstate+ndeath);
5673: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5674:
1.218 brouard 5675: if(estepm < stepm){
5676: printf ("Problem %d lower than %d\n",estepm, stepm);
5677: }
5678: else hstepm=estepm;
5679: /* For example we decided to compute the life expectancy with the smallest unit */
5680: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5681: nhstepm is the number of hstepm from age to agelim
5682: nstepm is the number of stepm from age to agelim.
5683: Look at function hpijx to understand why because of memory size limitations,
5684: we decided (b) to get a life expectancy respecting the most precise curvature of the
5685: survival function given by stepm (the optimization length). Unfortunately it
5686: means that if the survival funtion is printed every two years of age and if
5687: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5688: results. So we changed our mind and took the option of the best precision.
5689: */
5690: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5691: agelim = AGESUP;
5692: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5693: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5694: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5695: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5696: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5697: gp=matrix(0,nhstepm,1,nlstate);
5698: gm=matrix(0,nhstepm,1,nlstate);
5699:
5700:
5701: for(theta=1; theta <=npar; theta++){
5702: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5703: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5704: }
5705:
1.242 brouard 5706: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5707:
5708: if (popbased==1) {
5709: if(mobilav ==0){
5710: for(i=1; i<=nlstate;i++)
5711: prlim[i][i]=probs[(int)age][i][ij];
5712: }else{ /* mobilav */
5713: for(i=1; i<=nlstate;i++)
5714: prlim[i][i]=mobaverage[(int)age][i][ij];
5715: }
5716: }
5717:
1.235 brouard 5718: 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 5719: for(j=1; j<= nlstate; j++){
5720: for(h=0; h<=nhstepm; h++){
5721: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5722: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5723: }
5724: }
5725: /* Next for computing probability of death (h=1 means
5726: computed over hstepm matrices product = hstepm*stepm months)
5727: as a weighted average of prlim.
5728: */
5729: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5730: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5731: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5732: }
5733: /* end probability of death */
5734:
5735: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5736: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5737:
1.242 brouard 5738: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5739:
5740: if (popbased==1) {
5741: if(mobilav ==0){
5742: for(i=1; i<=nlstate;i++)
5743: prlim[i][i]=probs[(int)age][i][ij];
5744: }else{ /* mobilav */
5745: for(i=1; i<=nlstate;i++)
5746: prlim[i][i]=mobaverage[(int)age][i][ij];
5747: }
5748: }
5749:
1.235 brouard 5750: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5751:
5752: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5753: for(h=0; h<=nhstepm; h++){
5754: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5755: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5756: }
5757: }
5758: /* This for computing probability of death (h=1 means
5759: computed over hstepm matrices product = hstepm*stepm months)
5760: as a weighted average of prlim.
5761: */
5762: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5763: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5764: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5765: }
5766: /* end probability of death */
5767:
5768: for(j=1; j<= nlstate; j++) /* vareij */
5769: for(h=0; h<=nhstepm; h++){
5770: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5771: }
5772:
5773: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5774: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5775: }
5776:
5777: } /* End theta */
5778:
5779: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5780:
5781: for(h=0; h<=nhstepm; h++) /* veij */
5782: for(j=1; j<=nlstate;j++)
5783: for(theta=1; theta <=npar; theta++)
5784: trgradg[h][j][theta]=gradg[h][theta][j];
5785:
5786: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5787: for(theta=1; theta <=npar; theta++)
5788: trgradgp[j][theta]=gradgp[theta][j];
5789:
5790:
5791: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5792: for(i=1;i<=nlstate;i++)
5793: for(j=1;j<=nlstate;j++)
5794: vareij[i][j][(int)age] =0.;
5795:
5796: for(h=0;h<=nhstepm;h++){
5797: for(k=0;k<=nhstepm;k++){
5798: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5799: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5800: for(i=1;i<=nlstate;i++)
5801: for(j=1;j<=nlstate;j++)
5802: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5803: }
5804: }
5805:
5806: /* pptj */
5807: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5808: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5809: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5810: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5811: varppt[j][i]=doldmp[j][i];
5812: /* end ppptj */
5813: /* x centered again */
5814:
1.242 brouard 5815: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5816:
5817: if (popbased==1) {
5818: if(mobilav ==0){
5819: for(i=1; i<=nlstate;i++)
5820: prlim[i][i]=probs[(int)age][i][ij];
5821: }else{ /* mobilav */
5822: for(i=1; i<=nlstate;i++)
5823: prlim[i][i]=mobaverage[(int)age][i][ij];
5824: }
5825: }
5826:
5827: /* This for computing probability of death (h=1 means
5828: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5829: as a weighted average of prlim.
5830: */
1.235 brouard 5831: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5832: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5833: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5834: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5835: }
5836: /* end probability of death */
5837:
5838: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5839: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5840: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5841: for(i=1; i<=nlstate;i++){
5842: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5843: }
5844: }
5845: fprintf(ficresprobmorprev,"\n");
5846:
5847: fprintf(ficresvij,"%.0f ",age );
5848: for(i=1; i<=nlstate;i++)
5849: for(j=1; j<=nlstate;j++){
5850: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5851: }
5852: fprintf(ficresvij,"\n");
5853: free_matrix(gp,0,nhstepm,1,nlstate);
5854: free_matrix(gm,0,nhstepm,1,nlstate);
5855: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5856: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5857: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5858: } /* End age */
5859: free_vector(gpp,nlstate+1,nlstate+ndeath);
5860: free_vector(gmp,nlstate+1,nlstate+ndeath);
5861: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5862: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5863: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5864: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5865: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5866: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5867: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5868: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5869: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5870: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5871: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5872: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5873: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5874: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5875: 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);
5876: /* 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 5877: */
1.218 brouard 5878: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5879: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5880:
1.218 brouard 5881: free_vector(xp,1,npar);
5882: free_matrix(doldm,1,nlstate,1,nlstate);
5883: free_matrix(dnewm,1,nlstate,1,npar);
5884: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5885: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5886: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5887: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5888: fclose(ficresprobmorprev);
5889: fflush(ficgp);
5890: fflush(fichtm);
5891: } /* end varevsij */
1.126 brouard 5892:
5893: /************ Variance of prevlim ******************/
1.235 brouard 5894: 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 5895: {
1.205 brouard 5896: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5897: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5898:
1.126 brouard 5899: double **dnewm,**doldm;
5900: int i, j, nhstepm, hstepm;
5901: double *xp;
5902: double *gp, *gm;
5903: double **gradg, **trgradg;
1.208 brouard 5904: double **mgm, **mgp;
1.126 brouard 5905: double age,agelim;
5906: int theta;
5907:
5908: pstamp(ficresvpl);
5909: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5910: fprintf(ficresvpl,"# Age ");
5911: if(nresult >=1)
5912: fprintf(ficresvpl," Result# ");
1.126 brouard 5913: for(i=1; i<=nlstate;i++)
5914: fprintf(ficresvpl," %1d-%1d",i,i);
5915: fprintf(ficresvpl,"\n");
5916:
5917: xp=vector(1,npar);
5918: dnewm=matrix(1,nlstate,1,npar);
5919: doldm=matrix(1,nlstate,1,nlstate);
5920:
5921: hstepm=1*YEARM; /* Every year of age */
5922: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5923: agelim = AGESUP;
5924: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5925: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5926: if (stepm >= YEARM) hstepm=1;
5927: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5928: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5929: mgp=matrix(1,npar,1,nlstate);
5930: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5931: gp=vector(1,nlstate);
5932: gm=vector(1,nlstate);
5933:
5934: for(theta=1; theta <=npar; theta++){
5935: for(i=1; i<=npar; i++){ /* Computes gradient */
5936: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5937: }
1.209 brouard 5938: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5939: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5940: else
1.235 brouard 5941: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5942: for(i=1;i<=nlstate;i++){
1.126 brouard 5943: gp[i] = prlim[i][i];
1.208 brouard 5944: mgp[theta][i] = prlim[i][i];
5945: }
1.126 brouard 5946: for(i=1; i<=npar; i++) /* Computes gradient */
5947: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5948: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5949: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5950: else
1.235 brouard 5951: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5952: for(i=1;i<=nlstate;i++){
1.126 brouard 5953: gm[i] = prlim[i][i];
1.208 brouard 5954: mgm[theta][i] = prlim[i][i];
5955: }
1.126 brouard 5956: for(i=1;i<=nlstate;i++)
5957: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5958: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5959: } /* End theta */
5960:
5961: trgradg =matrix(1,nlstate,1,npar);
5962:
5963: for(j=1; j<=nlstate;j++)
5964: for(theta=1; theta <=npar; theta++)
5965: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5966: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5967: /* printf("\nmgm mgp %d ",(int)age); */
5968: /* for(j=1; j<=nlstate;j++){ */
5969: /* printf(" %d ",j); */
5970: /* for(theta=1; theta <=npar; theta++) */
5971: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5972: /* printf("\n "); */
5973: /* } */
5974: /* } */
5975: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5976: /* printf("\n gradg %d ",(int)age); */
5977: /* for(j=1; j<=nlstate;j++){ */
5978: /* printf("%d ",j); */
5979: /* for(theta=1; theta <=npar; theta++) */
5980: /* printf("%d %lf ",theta,gradg[theta][j]); */
5981: /* printf("\n "); */
5982: /* } */
5983: /* } */
1.126 brouard 5984:
5985: for(i=1;i<=nlstate;i++)
5986: varpl[i][(int)age] =0.;
1.209 brouard 5987: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5988: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5989: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5990: }else{
1.126 brouard 5991: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5992: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5993: }
1.126 brouard 5994: for(i=1;i<=nlstate;i++)
5995: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5996:
5997: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5998: if(nresult >=1)
5999: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6000: for(i=1; i<=nlstate;i++)
6001: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6002: fprintf(ficresvpl,"\n");
6003: free_vector(gp,1,nlstate);
6004: free_vector(gm,1,nlstate);
1.208 brouard 6005: free_matrix(mgm,1,npar,1,nlstate);
6006: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6007: free_matrix(gradg,1,npar,1,nlstate);
6008: free_matrix(trgradg,1,nlstate,1,npar);
6009: } /* End age */
6010:
6011: free_vector(xp,1,npar);
6012: free_matrix(doldm,1,nlstate,1,npar);
6013: free_matrix(dnewm,1,nlstate,1,nlstate);
6014:
6015: }
6016:
6017: /************ Variance of one-step probabilities ******************/
6018: 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 6019: {
6020: int i, j=0, k1, l1, tj;
6021: int k2, l2, j1, z1;
6022: int k=0, l;
6023: int first=1, first1, first2;
6024: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6025: double **dnewm,**doldm;
6026: double *xp;
6027: double *gp, *gm;
6028: double **gradg, **trgradg;
6029: double **mu;
6030: double age, cov[NCOVMAX+1];
6031: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6032: int theta;
6033: char fileresprob[FILENAMELENGTH];
6034: char fileresprobcov[FILENAMELENGTH];
6035: char fileresprobcor[FILENAMELENGTH];
6036: double ***varpij;
6037:
6038: strcpy(fileresprob,"PROB_");
6039: strcat(fileresprob,fileres);
6040: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6041: printf("Problem with resultfile: %s\n", fileresprob);
6042: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6043: }
6044: strcpy(fileresprobcov,"PROBCOV_");
6045: strcat(fileresprobcov,fileresu);
6046: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6047: printf("Problem with resultfile: %s\n", fileresprobcov);
6048: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6049: }
6050: strcpy(fileresprobcor,"PROBCOR_");
6051: strcat(fileresprobcor,fileresu);
6052: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6053: printf("Problem with resultfile: %s\n", fileresprobcor);
6054: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6055: }
6056: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6057: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6058: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6059: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6060: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6061: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6062: pstamp(ficresprob);
6063: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6064: fprintf(ficresprob,"# Age");
6065: pstamp(ficresprobcov);
6066: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6067: fprintf(ficresprobcov,"# Age");
6068: pstamp(ficresprobcor);
6069: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6070: fprintf(ficresprobcor,"# Age");
1.126 brouard 6071:
6072:
1.222 brouard 6073: for(i=1; i<=nlstate;i++)
6074: for(j=1; j<=(nlstate+ndeath);j++){
6075: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6076: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6077: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6078: }
6079: /* fprintf(ficresprob,"\n");
6080: fprintf(ficresprobcov,"\n");
6081: fprintf(ficresprobcor,"\n");
6082: */
6083: xp=vector(1,npar);
6084: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6085: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6086: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6087: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6088: first=1;
6089: fprintf(ficgp,"\n# Routine varprob");
6090: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6091: fprintf(fichtm,"\n");
6092:
6093: 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);
6094: 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);
6095: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6096: and drawn. It helps understanding how is the covariance between two incidences.\
6097: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6098: 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 6099: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6100: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6101: standard deviations wide on each axis. <br>\
6102: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6103: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6104: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6105:
1.222 brouard 6106: cov[1]=1;
6107: /* tj=cptcoveff; */
1.225 brouard 6108: tj = (int) pow(2,cptcoveff);
1.222 brouard 6109: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6110: j1=0;
1.224 brouard 6111: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6112: if (cptcovn>0) {
6113: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6114: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6115: fprintf(ficresprob, "**********\n#\n");
6116: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6117: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6118: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6119:
1.222 brouard 6120: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6121: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6122: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6123:
6124:
1.222 brouard 6125: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6126: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6127: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6128:
1.222 brouard 6129: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6130: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6131: fprintf(ficresprobcor, "**********\n#");
6132: if(invalidvarcomb[j1]){
6133: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6134: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6135: continue;
6136: }
6137: }
6138: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6139: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6140: gp=vector(1,(nlstate)*(nlstate+ndeath));
6141: gm=vector(1,(nlstate)*(nlstate+ndeath));
6142: for (age=bage; age<=fage; age ++){
6143: cov[2]=age;
6144: if(nagesqr==1)
6145: cov[3]= age*age;
6146: for (k=1; k<=cptcovn;k++) {
6147: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6148: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6149: * 1 1 1 1 1
6150: * 2 2 1 1 1
6151: * 3 1 2 1 1
6152: */
6153: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6154: }
6155: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6156: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6157: for (k=1; k<=cptcovprod;k++)
6158: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6159:
6160:
1.222 brouard 6161: for(theta=1; theta <=npar; theta++){
6162: for(i=1; i<=npar; i++)
6163: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6164:
1.222 brouard 6165: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6166:
1.222 brouard 6167: k=0;
6168: for(i=1; i<= (nlstate); i++){
6169: for(j=1; j<=(nlstate+ndeath);j++){
6170: k=k+1;
6171: gp[k]=pmmij[i][j];
6172: }
6173: }
1.220 brouard 6174:
1.222 brouard 6175: for(i=1; i<=npar; i++)
6176: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6177:
1.222 brouard 6178: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6179: k=0;
6180: for(i=1; i<=(nlstate); i++){
6181: for(j=1; j<=(nlstate+ndeath);j++){
6182: k=k+1;
6183: gm[k]=pmmij[i][j];
6184: }
6185: }
1.220 brouard 6186:
1.222 brouard 6187: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6188: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6189: }
1.126 brouard 6190:
1.222 brouard 6191: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6192: for(theta=1; theta <=npar; theta++)
6193: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6194:
1.222 brouard 6195: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6196: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6197:
1.222 brouard 6198: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6199:
1.222 brouard 6200: k=0;
6201: for(i=1; i<=(nlstate); i++){
6202: for(j=1; j<=(nlstate+ndeath);j++){
6203: k=k+1;
6204: mu[k][(int) age]=pmmij[i][j];
6205: }
6206: }
6207: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6208: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6209: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6210:
1.222 brouard 6211: /*printf("\n%d ",(int)age);
6212: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6213: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6214: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6215: }*/
1.220 brouard 6216:
1.222 brouard 6217: fprintf(ficresprob,"\n%d ",(int)age);
6218: fprintf(ficresprobcov,"\n%d ",(int)age);
6219: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6220:
1.222 brouard 6221: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6222: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6223: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6224: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6225: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6226: }
6227: i=0;
6228: for (k=1; k<=(nlstate);k++){
6229: for (l=1; l<=(nlstate+ndeath);l++){
6230: i++;
6231: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6232: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6233: for (j=1; j<=i;j++){
6234: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6235: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6236: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6237: }
6238: }
6239: }/* end of loop for state */
6240: } /* end of loop for age */
6241: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6242: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6243: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6244: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6245:
6246: /* Confidence intervalle of pij */
6247: /*
6248: fprintf(ficgp,"\nunset parametric;unset label");
6249: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6250: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6251: 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);
6252: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6253: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6254: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6255: */
6256:
6257: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6258: first1=1;first2=2;
6259: for (k2=1; k2<=(nlstate);k2++){
6260: for (l2=1; l2<=(nlstate+ndeath);l2++){
6261: if(l2==k2) continue;
6262: j=(k2-1)*(nlstate+ndeath)+l2;
6263: for (k1=1; k1<=(nlstate);k1++){
6264: for (l1=1; l1<=(nlstate+ndeath);l1++){
6265: if(l1==k1) continue;
6266: i=(k1-1)*(nlstate+ndeath)+l1;
6267: if(i<=j) continue;
6268: for (age=bage; age<=fage; age ++){
6269: if ((int)age %5==0){
6270: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6271: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6272: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6273: mu1=mu[i][(int) age]/stepm*YEARM ;
6274: mu2=mu[j][(int) age]/stepm*YEARM;
6275: c12=cv12/sqrt(v1*v2);
6276: /* Computing eigen value of matrix of covariance */
6277: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6278: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6279: if ((lc2 <0) || (lc1 <0) ){
6280: if(first2==1){
6281: first1=0;
6282: 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);
6283: }
6284: 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);
6285: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6286: /* lc2=fabs(lc2); */
6287: }
1.220 brouard 6288:
1.222 brouard 6289: /* Eigen vectors */
6290: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6291: /*v21=sqrt(1.-v11*v11); *//* error */
6292: v21=(lc1-v1)/cv12*v11;
6293: v12=-v21;
6294: v22=v11;
6295: tnalp=v21/v11;
6296: if(first1==1){
6297: first1=0;
6298: 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);
6299: }
6300: 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);
6301: /*printf(fignu*/
6302: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6303: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6304: if(first==1){
6305: first=0;
6306: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6307: fprintf(ficgp,"\nset parametric;unset label");
6308: 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);
6309: fprintf(ficgp,"\nset ter svg size 640, 480");
6310: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6311: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6312: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6313: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6314: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6315: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6316: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6317: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6318: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6319: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6320: 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", \
6321: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6322: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6323: }else{
6324: first=0;
6325: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6326: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6327: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6328: 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", \
6329: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6330: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6331: }/* if first */
6332: } /* age mod 5 */
6333: } /* end loop age */
6334: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6335: first=1;
6336: } /*l12 */
6337: } /* k12 */
6338: } /*l1 */
6339: }/* k1 */
6340: } /* loop on combination of covariates j1 */
6341: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6342: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6343: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6344: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6345: free_vector(xp,1,npar);
6346: fclose(ficresprob);
6347: fclose(ficresprobcov);
6348: fclose(ficresprobcor);
6349: fflush(ficgp);
6350: fflush(fichtmcov);
6351: }
1.126 brouard 6352:
6353:
6354: /******************* Printing html file ***********/
1.201 brouard 6355: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6356: int lastpass, int stepm, int weightopt, char model[],\
6357: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6358: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6359: double jprev1, double mprev1,double anprev1, double dateprev1, \
6360: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6361: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6362:
6363: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6364: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6365: </ul>");
1.237 brouard 6366: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6367: </ul>", model);
1.214 brouard 6368: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6369: 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",
6370: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6371: 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 6372: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6373: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6374: fprintf(fichtm,"\
6375: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6376: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6377: fprintf(fichtm,"\
1.217 brouard 6378: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6379: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6380: fprintf(fichtm,"\
1.126 brouard 6381: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6382: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6383: fprintf(fichtm,"\
1.217 brouard 6384: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6385: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6386: fprintf(fichtm,"\
1.211 brouard 6387: - (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 6388: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6389: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6390: if(prevfcast==1){
6391: fprintf(fichtm,"\
6392: - Prevalence projections by age and states: \
1.201 brouard 6393: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6394: }
1.126 brouard 6395:
1.222 brouard 6396: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6397:
1.225 brouard 6398: m=pow(2,cptcoveff);
1.222 brouard 6399: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6400:
1.222 brouard 6401: jj1=0;
1.237 brouard 6402:
6403: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6404: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 ! brouard 6405: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6406: continue;
1.220 brouard 6407:
1.222 brouard 6408: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6409: jj1++;
6410: if (cptcovn > 0) {
6411: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6412: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6413: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6414: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6415: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6416: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6417: }
1.237 brouard 6418: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6419: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6420: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6421: }
6422:
1.230 brouard 6423: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6424: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6425: if(invalidvarcomb[k1]){
6426: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6427: printf("\nCombination (%d) ignored because no cases \n",k1);
6428: continue;
6429: }
6430: }
6431: /* aij, bij */
1.241 brouard 6432: 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> \
6433: <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 6434: /* Pij */
1.241 brouard 6435: 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> \
6436: <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 6437: /* Quasi-incidences */
6438: 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 6439: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6440: 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 6441: 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> \
6442: <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 6443: /* Survival functions (period) in state j */
6444: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6445: 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> \
6446: <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 6447: }
6448: /* State specific survival functions (period) */
6449: for(cpt=1; cpt<=nlstate;cpt++){
6450: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6451: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6452: <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 6453: }
6454: /* Period (stable) prevalence in each health state */
6455: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6456: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability to be in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6457: <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 6458: }
6459: if(backcast==1){
6460: /* Period (stable) back prevalence in each health state */
6461: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6462: fprintf(fichtm,"<br>\n- Convergence to period (stable) back prevalence in state %d. Or probability to be in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6463: <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 6464: }
1.217 brouard 6465: }
1.222 brouard 6466: if(prevfcast==1){
6467: /* Projection of prevalence up to period (stable) prevalence in each health state */
6468: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6469: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f) up to period (stable) prevalence in state %d. Or probability to be in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6470: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6471: }
6472: }
1.220 brouard 6473:
1.222 brouard 6474: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6475: 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> \
6476: <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 6477: }
6478: /* } /\* end i1 *\/ */
6479: }/* End k1 */
6480: fprintf(fichtm,"</ul>");
1.126 brouard 6481:
1.222 brouard 6482: fprintf(fichtm,"\
1.126 brouard 6483: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6484: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6485: - 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 6486: But because parameters are usually highly correlated (a higher incidence of disability \
6487: and a higher incidence of recovery can give very close observed transition) it might \
6488: be very useful to look not only at linear confidence intervals estimated from the \
6489: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6490: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6491: covariance matrix of the one-step probabilities. \
6492: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6493:
1.222 brouard 6494: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6495: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6496: fprintf(fichtm,"\
1.126 brouard 6497: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6498: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6499:
1.222 brouard 6500: fprintf(fichtm,"\
1.126 brouard 6501: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6502: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6503: fprintf(fichtm,"\
1.126 brouard 6504: - 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): \
6505: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6506: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6507: fprintf(fichtm,"\
1.126 brouard 6508: - (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): \
6509: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6510: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6511: fprintf(fichtm,"\
1.128 brouard 6512: - 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 6513: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6514: fprintf(fichtm,"\
1.128 brouard 6515: - 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 6516: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6517: fprintf(fichtm,"\
1.126 brouard 6518: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6519: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6520:
6521: /* if(popforecast==1) fprintf(fichtm,"\n */
6522: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6523: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6524: /* <br>",fileres,fileres,fileres,fileres); */
6525: /* else */
6526: /* 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 6527: fflush(fichtm);
6528: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6529:
1.225 brouard 6530: m=pow(2,cptcoveff);
1.222 brouard 6531: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6532:
1.222 brouard 6533: jj1=0;
1.237 brouard 6534:
1.241 brouard 6535: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6536: for(k1=1; k1<=m;k1++){
1.253 ! brouard 6537: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6538: continue;
1.222 brouard 6539: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6540: jj1++;
1.126 brouard 6541: if (cptcovn > 0) {
6542: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6543: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6544: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6545: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6546: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6547: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6548: }
6549:
1.126 brouard 6550: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6551:
1.222 brouard 6552: if(invalidvarcomb[k1]){
6553: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6554: continue;
6555: }
1.126 brouard 6556: }
6557: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6558: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6559: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6560: <img src=\"%s_%d-%d-%d.svg\">",cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 6561: }
6562: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6563: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6564: true period expectancies (those weighted with period prevalences are also\
6565: drawn in addition to the population based expectancies computed using\
1.241 brouard 6566: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6567: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6568: /* } /\* end i1 *\/ */
6569: }/* End k1 */
1.241 brouard 6570: }/* End nres */
1.222 brouard 6571: fprintf(fichtm,"</ul>");
6572: fflush(fichtm);
1.126 brouard 6573: }
6574:
6575: /******************* Gnuplot file **************/
1.223 brouard 6576: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6577:
6578: char dirfileres[132],optfileres[132];
1.223 brouard 6579: char gplotcondition[132];
1.237 brouard 6580: 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 6581: int lv=0, vlv=0, kl=0;
1.130 brouard 6582: int ng=0;
1.201 brouard 6583: int vpopbased;
1.223 brouard 6584: int ioffset; /* variable offset for columns */
1.235 brouard 6585: int nres=0; /* Index of resultline */
1.219 brouard 6586:
1.126 brouard 6587: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6588: /* printf("Problem with file %s",optionfilegnuplot); */
6589: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6590: /* } */
6591:
6592: /*#ifdef windows */
6593: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6594: /*#endif */
1.225 brouard 6595: m=pow(2,cptcoveff);
1.126 brouard 6596:
1.202 brouard 6597: /* Contribution to likelihood */
6598: /* Plot the probability implied in the likelihood */
1.223 brouard 6599: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6600: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6601: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6602: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6603: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6604: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6605: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6606: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6607: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6608: 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));
6609: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6610: 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));
6611: for (i=1; i<= nlstate ; i ++) {
6612: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6613: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6614: 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);
6615: for (j=2; j<= nlstate+ndeath ; j ++) {
6616: 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);
6617: }
6618: fprintf(ficgp,";\nset out; unset ylabel;\n");
6619: }
6620: /* 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 */
6621: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6622: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6623: fprintf(ficgp,"\nset out;unset log\n");
6624: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6625:
1.126 brouard 6626: strcpy(dirfileres,optionfilefiname);
6627: strcpy(optfileres,"vpl");
1.223 brouard 6628: /* 1eme*/
1.238 brouard 6629: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6630: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6631: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6632: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 ! brouard 6633: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6634: continue;
6635: /* We are interested in selected combination by the resultline */
1.246 brouard 6636: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6637: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6638: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6639: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6640: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6641: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6642: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6643: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6644: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6645: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6646: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6647: }
6648: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6649: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6650: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6651: }
1.246 brouard 6652: /* printf("\n#\n"); */
1.238 brouard 6653: fprintf(ficgp,"\n#\n");
6654: if(invalidvarcomb[k1]){
6655: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6656: continue;
6657: }
1.235 brouard 6658:
1.241 brouard 6659: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6660: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6661: 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 6662:
1.238 brouard 6663: for (i=1; i<= nlstate ; i ++) {
6664: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6665: else fprintf(ficgp," %%*lf (%%*lf)");
6666: }
1.242 brouard 6667: 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 6668: for (i=1; i<= nlstate ; i ++) {
6669: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6670: else fprintf(ficgp," %%*lf (%%*lf)");
6671: }
1.242 brouard 6672: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6673: for (i=1; i<= nlstate ; i ++) {
6674: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6675: else fprintf(ficgp," %%*lf (%%*lf)");
6676: }
6677: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6678: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6679: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6680: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6681: if(cptcoveff ==0){
1.245 brouard 6682: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6683: }else{
6684: kl=0;
6685: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6686: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6687: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6688: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6689: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6690: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6691: kl++;
1.238 brouard 6692: /* 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 *\/ */
6693: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6694: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6695: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6696: if(k==cptcoveff){
1.245 brouard 6697: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6698: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6699: }else{
6700: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6701: kl++;
6702: }
6703: } /* end covariate */
6704: } /* end if no covariate */
6705: } /* end if backcast */
6706: fprintf(ficgp,"\nset out \n");
6707: } /* nres */
1.201 brouard 6708: } /* k1 */
6709: } /* cpt */
1.235 brouard 6710:
6711:
1.126 brouard 6712: /*2 eme*/
1.238 brouard 6713: for (k1=1; k1<= m ; k1 ++){
6714: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6715: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6716: continue;
6717: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6718: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6719: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6720: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6721: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6722: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6723: vlv= nbcode[Tvaraff[k]][lv];
6724: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6725: }
1.237 brouard 6726: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6727: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6728: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6729: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6730: }
1.211 brouard 6731: fprintf(ficgp,"\n#\n");
1.223 brouard 6732: if(invalidvarcomb[k1]){
6733: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6734: continue;
6735: }
1.219 brouard 6736:
1.241 brouard 6737: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6738: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6739: if(vpopbased==0)
6740: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6741: else
6742: fprintf(ficgp,"\nreplot ");
6743: for (i=1; i<= nlstate+1 ; i ++) {
6744: k=2*i;
6745: 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);
6746: for (j=1; j<= nlstate+1 ; j ++) {
6747: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6748: else fprintf(ficgp," %%*lf (%%*lf)");
6749: }
6750: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6751: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6752: 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);
6753: for (j=1; j<= nlstate+1 ; j ++) {
6754: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6755: else fprintf(ficgp," %%*lf (%%*lf)");
6756: }
6757: fprintf(ficgp,"\" t\"\" w l lt 0,");
6758: 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);
6759: for (j=1; j<= nlstate+1 ; j ++) {
6760: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6761: else fprintf(ficgp," %%*lf (%%*lf)");
6762: }
6763: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6764: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6765: } /* state */
6766: } /* vpopbased */
1.244 brouard 6767: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6768: } /* end nres */
6769: } /* k1 end 2 eme*/
6770:
6771:
6772: /*3eme*/
6773: for (k1=1; k1<= m ; k1 ++){
6774: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6775: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6776: continue;
6777:
6778: for (cpt=1; cpt<= nlstate ; cpt ++) {
6779: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6780: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6781: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6782: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6783: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6784: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6785: vlv= nbcode[Tvaraff[k]][lv];
6786: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6787: }
6788: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6789: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6790: }
6791: fprintf(ficgp,"\n#\n");
6792: if(invalidvarcomb[k1]){
6793: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6794: continue;
6795: }
6796:
6797: /* k=2+nlstate*(2*cpt-2); */
6798: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6799: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6800: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6801: 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 6802: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6803: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6804: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6805: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6806: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6807: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6808:
1.238 brouard 6809: */
6810: for (i=1; i< nlstate ; i ++) {
6811: 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);
6812: /* 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 6813:
1.238 brouard 6814: }
6815: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6816: }
6817: } /* end nres */
6818: } /* end kl 3eme */
1.126 brouard 6819:
1.223 brouard 6820: /* 4eme */
1.201 brouard 6821: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6822: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6823: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6824: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 6825: continue;
1.238 brouard 6826: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6827: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6828: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6829: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6830: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6831: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6832: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6833: vlv= nbcode[Tvaraff[k]][lv];
6834: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6835: }
6836: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6837: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6838: }
6839: fprintf(ficgp,"\n#\n");
6840: if(invalidvarcomb[k1]){
6841: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6842: continue;
1.223 brouard 6843: }
1.238 brouard 6844:
1.241 brouard 6845: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6846: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6847: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6848: k=3;
6849: for (i=1; i<= nlstate ; i ++){
6850: if(i==1){
6851: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6852: }else{
6853: fprintf(ficgp,", '' ");
6854: }
6855: l=(nlstate+ndeath)*(i-1)+1;
6856: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6857: for (j=2; j<= nlstate+ndeath ; j ++)
6858: fprintf(ficgp,"+$%d",k+l+j-1);
6859: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6860: } /* nlstate */
6861: fprintf(ficgp,"\nset out\n");
6862: } /* end cpt state*/
6863: } /* end nres */
6864: } /* end covariate k1 */
6865:
1.220 brouard 6866: /* 5eme */
1.201 brouard 6867: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6868: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6869: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6870: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 6871: continue;
1.238 brouard 6872: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6873: 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);
6874: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6875: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6876: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6877: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6878: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6879: vlv= nbcode[Tvaraff[k]][lv];
6880: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6881: }
6882: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6883: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6884: }
6885: fprintf(ficgp,"\n#\n");
6886: if(invalidvarcomb[k1]){
6887: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6888: continue;
6889: }
1.227 brouard 6890:
1.241 brouard 6891: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6892: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6893: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6894: k=3;
6895: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6896: if(j==1)
6897: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6898: else
6899: fprintf(ficgp,", '' ");
6900: l=(nlstate+ndeath)*(cpt-1) +j;
6901: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6902: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6903: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6904: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6905: } /* nlstate */
6906: fprintf(ficgp,", '' ");
6907: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6908: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6909: l=(nlstate+ndeath)*(cpt-1) +j;
6910: if(j < nlstate)
6911: fprintf(ficgp,"$%d +",k+l);
6912: else
6913: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6914: }
6915: fprintf(ficgp,"\nset out\n");
6916: } /* end cpt state*/
6917: } /* end covariate */
6918: } /* end nres */
1.227 brouard 6919:
1.220 brouard 6920: /* 6eme */
1.202 brouard 6921: /* CV preval stable (period) for each covariate */
1.237 brouard 6922: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6923: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6924: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6925: continue;
1.153 brouard 6926: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6927:
1.211 brouard 6928: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6929: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6930: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6931: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6932: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6933: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6934: vlv= nbcode[Tvaraff[k]][lv];
6935: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6936: }
1.237 brouard 6937: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6938: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6939: }
1.211 brouard 6940: fprintf(ficgp,"\n#\n");
1.223 brouard 6941: if(invalidvarcomb[k1]){
1.227 brouard 6942: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6943: continue;
1.223 brouard 6944: }
1.227 brouard 6945:
1.241 brouard 6946: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6947: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6948: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6949: k=3; /* Offset */
1.153 brouard 6950: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6951: if(i==1)
6952: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6953: else
6954: fprintf(ficgp,", '' ");
6955: l=(nlstate+ndeath)*(i-1)+1;
6956: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6957: for (j=2; j<= nlstate ; j ++)
6958: fprintf(ficgp,"+$%d",k+l+j-1);
6959: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6960: } /* nlstate */
1.201 brouard 6961: fprintf(ficgp,"\nset out\n");
1.153 brouard 6962: } /* end cpt state*/
6963: } /* end covariate */
1.227 brouard 6964:
6965:
1.220 brouard 6966: /* 7eme */
1.218 brouard 6967: if(backcast == 1){
1.217 brouard 6968: /* CV back preval stable (period) for each covariate */
1.237 brouard 6969: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6970: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 6971: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6972: continue;
1.218 brouard 6973: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6974: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6975: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6976: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6977: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6978: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6979: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6980: vlv= nbcode[Tvaraff[k]][lv];
6981: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6982: }
1.237 brouard 6983: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6984: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6985: }
1.227 brouard 6986: fprintf(ficgp,"\n#\n");
6987: if(invalidvarcomb[k1]){
6988: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6989: continue;
6990: }
6991:
1.241 brouard 6992: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6993: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6994: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6995: k=3; /* Offset */
6996: for (i=1; i<= nlstate ; i ++){
6997: if(i==1)
6998: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6999: else
7000: fprintf(ficgp,", '' ");
7001: /* l=(nlstate+ndeath)*(i-1)+1; */
7002: l=(nlstate+ndeath)*(cpt-1)+1;
7003: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7004: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
7005: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
7006: /* for (j=2; j<= nlstate ; j ++) */
7007: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7008: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7009: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7010: } /* nlstate */
7011: fprintf(ficgp,"\nset out\n");
1.218 brouard 7012: } /* end cpt state*/
7013: } /* end covariate */
7014: } /* End if backcast */
7015:
1.223 brouard 7016: /* 8eme */
1.218 brouard 7017: if(prevfcast==1){
7018: /* Projection from cross-sectional to stable (period) for each covariate */
7019:
1.237 brouard 7020: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7021: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 7022: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7023: continue;
1.211 brouard 7024: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 7025: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7026: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7027: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7028: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7029: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7030: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7031: vlv= nbcode[Tvaraff[k]][lv];
7032: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7033: }
1.237 brouard 7034: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7035: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7036: }
1.227 brouard 7037: fprintf(ficgp,"\n#\n");
7038: if(invalidvarcomb[k1]){
7039: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7040: continue;
7041: }
7042:
7043: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7044: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 7045: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7046: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7047: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7048: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7049: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7050: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7051: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7052: if(i==1){
7053: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7054: }else{
7055: fprintf(ficgp,",\\\n '' ");
7056: }
7057: if(cptcoveff ==0){ /* No covariate */
7058: ioffset=2; /* Age is in 2 */
7059: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7060: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7061: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7062: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7063: fprintf(ficgp," u %d:(", ioffset);
7064: if(i==nlstate+1)
7065: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
7066: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7067: else
7068: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7069: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7070: }else{ /* more than 2 covariates */
7071: if(cptcoveff ==1){
7072: ioffset=4; /* Age is in 4 */
7073: }else{
7074: ioffset=6; /* Age is in 6 */
7075: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7076: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7077: }
7078: fprintf(ficgp," u %d:(",ioffset);
7079: kl=0;
7080: strcpy(gplotcondition,"(");
7081: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7082: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7083: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7084: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7085: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7086: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7087: kl++;
7088: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7089: kl++;
7090: if(k <cptcoveff && cptcoveff>1)
7091: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7092: }
7093: strcpy(gplotcondition+strlen(gplotcondition),")");
7094: /* 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 *\/ */
7095: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7096: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7097: /* '' 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*/
7098: if(i==nlstate+1){
7099: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
7100: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7101: }else{
7102: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7103: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7104: }
7105: } /* end if covariate */
7106: } /* nlstate */
7107: fprintf(ficgp,"\nset out\n");
1.223 brouard 7108: } /* end cpt state*/
7109: } /* end covariate */
7110: } /* End if prevfcast */
1.227 brouard 7111:
7112:
1.238 brouard 7113: /* 9eme writing MLE parameters */
7114: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7115: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7116: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7117: for(k=1; k <=(nlstate+ndeath); k++){
7118: if (k != i) {
1.227 brouard 7119: fprintf(ficgp,"# current state %d\n",k);
7120: for(j=1; j <=ncovmodel; j++){
7121: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7122: jk++;
7123: }
7124: fprintf(ficgp,"\n");
1.126 brouard 7125: }
7126: }
1.223 brouard 7127: }
1.187 brouard 7128: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7129:
1.145 brouard 7130: /*goto avoid;*/
1.238 brouard 7131: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7132: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7133: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7134: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7135: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7136: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7137: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7138: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7139: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7140: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7141: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7142: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7143: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7144: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7145: fprintf(ficgp,"#\n");
1.223 brouard 7146: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7147: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7148: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7149: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 7150: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7151: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
7152: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 7153: if(m != 1 && TKresult[nres]!= jk)
1.237 brouard 7154: continue;
7155: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
7156: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7157: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7158: }
7159: fprintf(ficgp,"\n#\n");
1.241 brouard 7160: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7161: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7162: if (ng==1){
7163: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7164: fprintf(ficgp,"\nunset log y");
7165: }else if (ng==2){
7166: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7167: fprintf(ficgp,"\nset log y");
7168: }else if (ng==3){
7169: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7170: fprintf(ficgp,"\nset log y");
7171: }else
7172: fprintf(ficgp,"\nunset title ");
7173: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7174: i=1;
7175: for(k2=1; k2<=nlstate; k2++) {
7176: k3=i;
7177: for(k=1; k<=(nlstate+ndeath); k++) {
7178: if (k != k2){
7179: switch( ng) {
7180: case 1:
7181: if(nagesqr==0)
7182: fprintf(ficgp," p%d+p%d*x",i,i+1);
7183: else /* nagesqr =1 */
7184: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7185: break;
7186: case 2: /* ng=2 */
7187: if(nagesqr==0)
7188: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7189: else /* nagesqr =1 */
7190: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7191: break;
7192: case 3:
7193: if(nagesqr==0)
7194: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7195: else /* nagesqr =1 */
7196: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7197: break;
7198: }
7199: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7200: ijp=1; /* product no age */
7201: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7202: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7203: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7204: if(j==Tage[ij]) { /* Product by age */
7205: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7206: if(DummyV[j]==0){
1.237 brouard 7207: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7208: }else{ /* quantitative */
7209: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7210: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7211: }
7212: ij++;
7213: }
7214: }else if(j==Tprod[ijp]) { /* */
7215: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7216: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7217: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7218: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7219: /* 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)]); */
7220: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7221: }else{ /* Vn is dummy and Vm is quanti */
7222: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7223: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7224: }
7225: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7226: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7227: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7228: }else{ /* Both quanti */
7229: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7230: }
7231: }
1.238 brouard 7232: ijp++;
1.237 brouard 7233: }
7234: } else{ /* simple covariate */
7235: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7236: if(Dummy[j]==0){
7237: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7238: }else{ /* quantitative */
7239: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7240: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7241: }
1.237 brouard 7242: } /* end simple */
7243: } /* end j */
1.223 brouard 7244: }else{
7245: i=i-ncovmodel;
7246: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7247: fprintf(ficgp," (1.");
7248: }
1.227 brouard 7249:
1.223 brouard 7250: if(ng != 1){
7251: fprintf(ficgp,")/(1");
1.227 brouard 7252:
1.223 brouard 7253: for(k1=1; k1 <=nlstate; k1++){
7254: if(nagesqr==0)
7255: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7256: else /* nagesqr =1 */
7257: 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 7258:
1.223 brouard 7259: ij=1;
7260: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7261: if((j-2)==Tage[ij]) { /* Bug valgrind */
7262: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7263: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7264: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7265: ij++;
7266: }
7267: }
7268: else
1.225 brouard 7269: 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 7270: }
7271: fprintf(ficgp,")");
7272: }
7273: fprintf(ficgp,")");
7274: if(ng ==2)
7275: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7276: else /* ng= 3 */
7277: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7278: }else{ /* end ng <> 1 */
7279: if( k !=k2) /* logit p11 is hard to draw */
7280: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7281: }
7282: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7283: fprintf(ficgp,",");
7284: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7285: fprintf(ficgp,",");
7286: i=i+ncovmodel;
7287: } /* end k */
7288: } /* end k2 */
7289: fprintf(ficgp,"\n set out\n");
7290: } /* end jk */
7291: } /* end ng */
7292: /* avoid: */
7293: fflush(ficgp);
1.126 brouard 7294: } /* end gnuplot */
7295:
7296:
7297: /*************** Moving average **************/
1.219 brouard 7298: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7299: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7300:
1.222 brouard 7301: int i, cpt, cptcod;
7302: int modcovmax =1;
7303: int mobilavrange, mob;
7304: int iage=0;
7305:
7306: double sum=0.;
7307: double age;
7308: double *sumnewp, *sumnewm;
7309: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7310:
7311:
1.225 brouard 7312: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7313: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7314:
7315: sumnewp = vector(1,ncovcombmax);
7316: sumnewm = vector(1,ncovcombmax);
7317: agemingood = vector(1,ncovcombmax);
7318: agemaxgood = vector(1,ncovcombmax);
7319:
7320: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7321: sumnewm[cptcod]=0.;
7322: sumnewp[cptcod]=0.;
7323: agemingood[cptcod]=0;
7324: agemaxgood[cptcod]=0;
7325: }
7326: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7327:
7328: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7329: if(mobilav==1) mobilavrange=5; /* default */
7330: else mobilavrange=mobilav;
7331: for (age=bage; age<=fage; age++)
7332: for (i=1; i<=nlstate;i++)
7333: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7334: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7335: /* We keep the original values on the extreme ages bage, fage and for
7336: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7337: we use a 5 terms etc. until the borders are no more concerned.
7338: */
7339: for (mob=3;mob <=mobilavrange;mob=mob+2){
7340: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7341: for (i=1; i<=nlstate;i++){
7342: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7343: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7344: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7345: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7346: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7347: }
7348: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7349: }
7350: }
7351: }/* end age */
7352: }/* end mob */
7353: }else
7354: return -1;
7355: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7356: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7357: if(invalidvarcomb[cptcod]){
7358: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7359: continue;
7360: }
1.219 brouard 7361:
1.222 brouard 7362: agemingood[cptcod]=fage-(mob-1)/2;
7363: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7364: sumnewm[cptcod]=0.;
7365: for (i=1; i<=nlstate;i++){
7366: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7367: }
7368: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7369: agemingood[cptcod]=age;
7370: }else{ /* bad */
7371: for (i=1; i<=nlstate;i++){
7372: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7373: } /* i */
7374: } /* end bad */
7375: }/* age */
7376: sum=0.;
7377: for (i=1; i<=nlstate;i++){
7378: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7379: }
7380: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7381: 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);
7382: /* for (i=1; i<=nlstate;i++){ */
7383: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7384: /* } /\* i *\/ */
7385: } /* end bad */
7386: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7387: /* From youngest, finding the oldest wrong */
7388: agemaxgood[cptcod]=bage+(mob-1)/2;
7389: for (age=bage+(mob-1)/2; age<=fage; age++){
7390: sumnewm[cptcod]=0.;
7391: for (i=1; i<=nlstate;i++){
7392: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7393: }
7394: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7395: agemaxgood[cptcod]=age;
7396: }else{ /* bad */
7397: for (i=1; i<=nlstate;i++){
7398: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7399: } /* i */
7400: } /* end bad */
7401: }/* age */
7402: sum=0.;
7403: for (i=1; i<=nlstate;i++){
7404: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7405: }
7406: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7407: 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);
7408: /* for (i=1; i<=nlstate;i++){ */
7409: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7410: /* } /\* i *\/ */
7411: } /* end bad */
7412:
7413: for (age=bage; age<=fage; age++){
1.235 brouard 7414: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7415: sumnewp[cptcod]=0.;
7416: sumnewm[cptcod]=0.;
7417: for (i=1; i<=nlstate;i++){
7418: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7419: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7420: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7421: }
7422: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7423: }
7424: /* printf("\n"); */
7425: /* } */
7426: /* brutal averaging */
7427: for (i=1; i<=nlstate;i++){
7428: for (age=1; age<=bage; age++){
7429: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7430: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7431: }
7432: for (age=fage; age<=AGESUP; age++){
7433: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7434: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7435: }
7436: } /* end i status */
7437: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7438: for (age=1; age<=AGESUP; age++){
7439: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7440: mobaverage[(int)age][i][cptcod]=0.;
7441: }
7442: }
7443: }/* end cptcod */
7444: free_vector(sumnewm,1, ncovcombmax);
7445: free_vector(sumnewp,1, ncovcombmax);
7446: free_vector(agemaxgood,1, ncovcombmax);
7447: free_vector(agemingood,1, ncovcombmax);
7448: return 0;
7449: }/* End movingaverage */
1.218 brouard 7450:
1.126 brouard 7451:
7452: /************** Forecasting ******************/
1.235 brouard 7453: 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 7454: /* proj1, year, month, day of starting projection
7455: agemin, agemax range of age
7456: dateprev1 dateprev2 range of dates during which prevalence is computed
7457: anproj2 year of en of projection (same day and month as proj1).
7458: */
1.235 brouard 7459: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7460: double agec; /* generic age */
7461: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7462: double *popeffectif,*popcount;
7463: double ***p3mat;
1.218 brouard 7464: /* double ***mobaverage; */
1.126 brouard 7465: char fileresf[FILENAMELENGTH];
7466:
7467: agelim=AGESUP;
1.211 brouard 7468: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7469: in each health status at the date of interview (if between dateprev1 and dateprev2).
7470: We still use firstpass and lastpass as another selection.
7471: */
1.214 brouard 7472: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7473: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7474:
1.201 brouard 7475: strcpy(fileresf,"F_");
7476: strcat(fileresf,fileresu);
1.126 brouard 7477: if((ficresf=fopen(fileresf,"w"))==NULL) {
7478: printf("Problem with forecast resultfile: %s\n", fileresf);
7479: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7480: }
1.235 brouard 7481: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7482: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7483:
1.225 brouard 7484: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7485:
7486:
7487: stepsize=(int) (stepm+YEARM-1)/YEARM;
7488: if (stepm<=12) stepsize=1;
7489: if(estepm < stepm){
7490: printf ("Problem %d lower than %d\n",estepm, stepm);
7491: }
7492: else hstepm=estepm;
7493:
7494: hstepm=hstepm/stepm;
7495: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7496: fractional in yp1 */
7497: anprojmean=yp;
7498: yp2=modf((yp1*12),&yp);
7499: mprojmean=yp;
7500: yp1=modf((yp2*30.5),&yp);
7501: jprojmean=yp;
7502: if(jprojmean==0) jprojmean=1;
7503: if(mprojmean==0) jprojmean=1;
7504:
1.227 brouard 7505: i1=pow(2,cptcoveff);
1.126 brouard 7506: if (cptcovn < 1){i1=1;}
7507:
7508: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7509:
7510: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7511:
1.126 brouard 7512: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7513: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7514: for(k=1; k<=i1;k++){
1.253 ! brouard 7515: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7516: continue;
1.227 brouard 7517: if(invalidvarcomb[k]){
7518: printf("\nCombination (%d) projection ignored because no cases \n",k);
7519: continue;
7520: }
7521: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7522: for(j=1;j<=cptcoveff;j++) {
7523: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7524: }
1.235 brouard 7525: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7526: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7527: }
1.227 brouard 7528: fprintf(ficresf," yearproj age");
7529: for(j=1; j<=nlstate+ndeath;j++){
7530: for(i=1; i<=nlstate;i++)
7531: fprintf(ficresf," p%d%d",i,j);
7532: fprintf(ficresf," wp.%d",j);
7533: }
7534: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7535: fprintf(ficresf,"\n");
7536: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7537: for (agec=fage; agec>=(ageminpar-1); agec--){
7538: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7539: nhstepm = nhstepm/hstepm;
7540: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7541: oldm=oldms;savm=savms;
1.235 brouard 7542: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7543:
7544: for (h=0; h<=nhstepm; h++){
7545: if (h*hstepm/YEARM*stepm ==yearp) {
7546: fprintf(ficresf,"\n");
7547: for(j=1;j<=cptcoveff;j++)
7548: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7549: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7550: }
7551: for(j=1; j<=nlstate+ndeath;j++) {
7552: ppij=0.;
7553: for(i=1; i<=nlstate;i++) {
7554: if (mobilav==1)
7555: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7556: else {
7557: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7558: }
7559: if (h*hstepm/YEARM*stepm== yearp) {
7560: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7561: }
7562: } /* end i */
7563: if (h*hstepm/YEARM*stepm==yearp) {
7564: fprintf(ficresf," %.3f", ppij);
7565: }
7566: }/* end j */
7567: } /* end h */
7568: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7569: } /* end agec */
7570: } /* end yearp */
7571: } /* end k */
1.219 brouard 7572:
1.126 brouard 7573: fclose(ficresf);
1.215 brouard 7574: printf("End of Computing forecasting \n");
7575: fprintf(ficlog,"End of Computing forecasting\n");
7576:
1.126 brouard 7577: }
7578:
1.218 brouard 7579: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7580: /* 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 7581: /* /\* back1, year, month, day of starting backection */
7582: /* agemin, agemax range of age */
7583: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7584: /* anback2 year of en of backection (same day and month as back1). */
7585: /* *\/ */
7586: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7587: /* double agec; /\* generic age *\/ */
7588: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7589: /* double *popeffectif,*popcount; */
7590: /* double ***p3mat; */
7591: /* /\* double ***mobaverage; *\/ */
7592: /* char fileresfb[FILENAMELENGTH]; */
7593:
7594: /* agelim=AGESUP; */
7595: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7596: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7597: /* We still use firstpass and lastpass as another selection. */
7598: /* *\/ */
7599: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7600: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7601: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7602:
7603: /* strcpy(fileresfb,"FB_"); */
7604: /* strcat(fileresfb,fileresu); */
7605: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7606: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7607: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7608: /* } */
7609: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7610: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7611:
1.225 brouard 7612: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7613:
7614: /* /\* if (mobilav!=0) { *\/ */
7615: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7616: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7617: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7618: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7619: /* /\* } *\/ */
7620: /* /\* } *\/ */
7621:
7622: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7623: /* if (stepm<=12) stepsize=1; */
7624: /* if(estepm < stepm){ */
7625: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7626: /* } */
7627: /* else hstepm=estepm; */
7628:
7629: /* hstepm=hstepm/stepm; */
7630: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7631: /* fractional in yp1 *\/ */
7632: /* anprojmean=yp; */
7633: /* yp2=modf((yp1*12),&yp); */
7634: /* mprojmean=yp; */
7635: /* yp1=modf((yp2*30.5),&yp); */
7636: /* jprojmean=yp; */
7637: /* if(jprojmean==0) jprojmean=1; */
7638: /* if(mprojmean==0) jprojmean=1; */
7639:
1.225 brouard 7640: /* i1=cptcoveff; */
1.218 brouard 7641: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7642:
1.218 brouard 7643: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7644:
1.218 brouard 7645: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7646:
7647: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7648: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7649: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7650: /* k=k+1; */
7651: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7652: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7653: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7654: /* } */
7655: /* fprintf(ficresfb," yearbproj age"); */
7656: /* for(j=1; j<=nlstate+ndeath;j++){ */
7657: /* for(i=1; i<=nlstate;i++) */
7658: /* fprintf(ficresfb," p%d%d",i,j); */
7659: /* fprintf(ficresfb," p.%d",j); */
7660: /* } */
7661: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7662: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7663: /* fprintf(ficresfb,"\n"); */
7664: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7665: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7666: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7667: /* nhstepm = nhstepm/hstepm; */
7668: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7669: /* oldm=oldms;savm=savms; */
7670: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7671: /* for (h=0; h<=nhstepm; h++){ */
7672: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7673: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7674: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7675: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7676: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7677: /* } */
7678: /* for(j=1; j<=nlstate+ndeath;j++) { */
7679: /* ppij=0.; */
7680: /* for(i=1; i<=nlstate;i++) { */
7681: /* if (mobilav==1) */
7682: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7683: /* else { */
7684: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7685: /* } */
7686: /* if (h*hstepm/YEARM*stepm== yearp) { */
7687: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7688: /* } */
7689: /* } /\* end i *\/ */
7690: /* if (h*hstepm/YEARM*stepm==yearp) { */
7691: /* fprintf(ficresfb," %.3f", ppij); */
7692: /* } */
7693: /* }/\* end j *\/ */
7694: /* } /\* end h *\/ */
7695: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7696: /* } /\* end agec *\/ */
7697: /* } /\* end yearp *\/ */
7698: /* } /\* end cptcod *\/ */
7699: /* } /\* end cptcov *\/ */
7700:
7701: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7702:
7703: /* fclose(ficresfb); */
7704: /* printf("End of Computing Back forecasting \n"); */
7705: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7706:
1.218 brouard 7707: /* } */
1.217 brouard 7708:
1.126 brouard 7709: /************** Forecasting *****not tested NB*************/
1.227 brouard 7710: /* 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 7711:
1.227 brouard 7712: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7713: /* int *popage; */
7714: /* double calagedatem, agelim, kk1, kk2; */
7715: /* double *popeffectif,*popcount; */
7716: /* double ***p3mat,***tabpop,***tabpopprev; */
7717: /* /\* double ***mobaverage; *\/ */
7718: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7719:
1.227 brouard 7720: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7721: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7722: /* agelim=AGESUP; */
7723: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7724:
1.227 brouard 7725: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7726:
7727:
1.227 brouard 7728: /* strcpy(filerespop,"POP_"); */
7729: /* strcat(filerespop,fileresu); */
7730: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7731: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7732: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7733: /* } */
7734: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7735: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7736:
1.227 brouard 7737: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7738:
1.227 brouard 7739: /* /\* if (mobilav!=0) { *\/ */
7740: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7741: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7742: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7743: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7744: /* /\* } *\/ */
7745: /* /\* } *\/ */
1.126 brouard 7746:
1.227 brouard 7747: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7748: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7749:
1.227 brouard 7750: /* agelim=AGESUP; */
1.126 brouard 7751:
1.227 brouard 7752: /* hstepm=1; */
7753: /* hstepm=hstepm/stepm; */
1.218 brouard 7754:
1.227 brouard 7755: /* if (popforecast==1) { */
7756: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7757: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7758: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7759: /* } */
7760: /* popage=ivector(0,AGESUP); */
7761: /* popeffectif=vector(0,AGESUP); */
7762: /* popcount=vector(0,AGESUP); */
1.126 brouard 7763:
1.227 brouard 7764: /* i=1; */
7765: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7766:
1.227 brouard 7767: /* imx=i; */
7768: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7769: /* } */
1.218 brouard 7770:
1.227 brouard 7771: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7772: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7773: /* k=k+1; */
7774: /* fprintf(ficrespop,"\n#******"); */
7775: /* for(j=1;j<=cptcoveff;j++) { */
7776: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7777: /* } */
7778: /* fprintf(ficrespop,"******\n"); */
7779: /* fprintf(ficrespop,"# Age"); */
7780: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7781: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7782:
1.227 brouard 7783: /* for (cpt=0; cpt<=0;cpt++) { */
7784: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7785:
1.227 brouard 7786: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7787: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7788: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7789:
1.227 brouard 7790: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7791: /* oldm=oldms;savm=savms; */
7792: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7793:
1.227 brouard 7794: /* for (h=0; h<=nhstepm; h++){ */
7795: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7796: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7797: /* } */
7798: /* for(j=1; j<=nlstate+ndeath;j++) { */
7799: /* kk1=0.;kk2=0; */
7800: /* for(i=1; i<=nlstate;i++) { */
7801: /* if (mobilav==1) */
7802: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7803: /* else { */
7804: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7805: /* } */
7806: /* } */
7807: /* if (h==(int)(calagedatem+12*cpt)){ */
7808: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7809: /* /\*fprintf(ficrespop," %.3f", kk1); */
7810: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7811: /* } */
7812: /* } */
7813: /* for(i=1; i<=nlstate;i++){ */
7814: /* kk1=0.; */
7815: /* for(j=1; j<=nlstate;j++){ */
7816: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7817: /* } */
7818: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7819: /* } */
1.218 brouard 7820:
1.227 brouard 7821: /* if (h==(int)(calagedatem+12*cpt)) */
7822: /* for(j=1; j<=nlstate;j++) */
7823: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7824: /* } */
7825: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7826: /* } */
7827: /* } */
1.218 brouard 7828:
1.227 brouard 7829: /* /\******\/ */
1.218 brouard 7830:
1.227 brouard 7831: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7832: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7833: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7834: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7835: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7836:
1.227 brouard 7837: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7838: /* oldm=oldms;savm=savms; */
7839: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7840: /* for (h=0; h<=nhstepm; h++){ */
7841: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7842: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7843: /* } */
7844: /* for(j=1; j<=nlstate+ndeath;j++) { */
7845: /* kk1=0.;kk2=0; */
7846: /* for(i=1; i<=nlstate;i++) { */
7847: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7848: /* } */
7849: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7850: /* } */
7851: /* } */
7852: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7853: /* } */
7854: /* } */
7855: /* } */
7856: /* } */
1.218 brouard 7857:
1.227 brouard 7858: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7859:
1.227 brouard 7860: /* if (popforecast==1) { */
7861: /* free_ivector(popage,0,AGESUP); */
7862: /* free_vector(popeffectif,0,AGESUP); */
7863: /* free_vector(popcount,0,AGESUP); */
7864: /* } */
7865: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7866: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7867: /* fclose(ficrespop); */
7868: /* } /\* End of popforecast *\/ */
1.218 brouard 7869:
1.126 brouard 7870: int fileappend(FILE *fichier, char *optionfich)
7871: {
7872: if((fichier=fopen(optionfich,"a"))==NULL) {
7873: printf("Problem with file: %s\n", optionfich);
7874: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7875: return (0);
7876: }
7877: fflush(fichier);
7878: return (1);
7879: }
7880:
7881:
7882: /**************** function prwizard **********************/
7883: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7884: {
7885:
7886: /* Wizard to print covariance matrix template */
7887:
1.164 brouard 7888: char ca[32], cb[32];
7889: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7890: int numlinepar;
7891:
7892: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7893: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7894: for(i=1; i <=nlstate; i++){
7895: jj=0;
7896: for(j=1; j <=nlstate+ndeath; j++){
7897: if(j==i) continue;
7898: jj++;
7899: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7900: printf("%1d%1d",i,j);
7901: fprintf(ficparo,"%1d%1d",i,j);
7902: for(k=1; k<=ncovmodel;k++){
7903: /* printf(" %lf",param[i][j][k]); */
7904: /* fprintf(ficparo," %lf",param[i][j][k]); */
7905: printf(" 0.");
7906: fprintf(ficparo," 0.");
7907: }
7908: printf("\n");
7909: fprintf(ficparo,"\n");
7910: }
7911: }
7912: printf("# Scales (for hessian or gradient estimation)\n");
7913: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7914: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
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: fprintf(ficparo,"%1d%1d",i,j);
7921: printf("%1d%1d",i,j);
7922: fflush(stdout);
7923: for(k=1; k<=ncovmodel;k++){
7924: /* printf(" %le",delti3[i][j][k]); */
7925: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7926: printf(" 0.");
7927: fprintf(ficparo," 0.");
7928: }
7929: numlinepar++;
7930: printf("\n");
7931: fprintf(ficparo,"\n");
7932: }
7933: }
7934: printf("# Covariance matrix\n");
7935: /* # 121 Var(a12)\n\ */
7936: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7937: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7938: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7939: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7940: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7941: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7942: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7943: fflush(stdout);
7944: fprintf(ficparo,"# Covariance matrix\n");
7945: /* # 121 Var(a12)\n\ */
7946: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7947: /* # ...\n\ */
7948: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7949:
7950: for(itimes=1;itimes<=2;itimes++){
7951: jj=0;
7952: for(i=1; i <=nlstate; i++){
7953: for(j=1; j <=nlstate+ndeath; j++){
7954: if(j==i) continue;
7955: for(k=1; k<=ncovmodel;k++){
7956: jj++;
7957: ca[0]= k+'a'-1;ca[1]='\0';
7958: if(itimes==1){
7959: printf("#%1d%1d%d",i,j,k);
7960: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7961: }else{
7962: printf("%1d%1d%d",i,j,k);
7963: fprintf(ficparo,"%1d%1d%d",i,j,k);
7964: /* printf(" %.5le",matcov[i][j]); */
7965: }
7966: ll=0;
7967: for(li=1;li <=nlstate; li++){
7968: for(lj=1;lj <=nlstate+ndeath; lj++){
7969: if(lj==li) continue;
7970: for(lk=1;lk<=ncovmodel;lk++){
7971: ll++;
7972: if(ll<=jj){
7973: cb[0]= lk +'a'-1;cb[1]='\0';
7974: if(ll<jj){
7975: if(itimes==1){
7976: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7977: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7978: }else{
7979: printf(" 0.");
7980: fprintf(ficparo," 0.");
7981: }
7982: }else{
7983: if(itimes==1){
7984: printf(" Var(%s%1d%1d)",ca,i,j);
7985: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7986: }else{
7987: printf(" 0.");
7988: fprintf(ficparo," 0.");
7989: }
7990: }
7991: }
7992: } /* end lk */
7993: } /* end lj */
7994: } /* end li */
7995: printf("\n");
7996: fprintf(ficparo,"\n");
7997: numlinepar++;
7998: } /* end k*/
7999: } /*end j */
8000: } /* end i */
8001: } /* end itimes */
8002:
8003: } /* end of prwizard */
8004: /******************* Gompertz Likelihood ******************************/
8005: double gompertz(double x[])
8006: {
8007: double A,B,L=0.0,sump=0.,num=0.;
8008: int i,n=0; /* n is the size of the sample */
8009:
1.220 brouard 8010: for (i=1;i<=imx ; i++) {
1.126 brouard 8011: sump=sump+weight[i];
8012: /* sump=sump+1;*/
8013: num=num+1;
8014: }
8015:
8016:
8017: /* for (i=0; i<=imx; i++)
8018: 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]);*/
8019:
8020: for (i=1;i<=imx ; i++)
8021: {
8022: if (cens[i] == 1 && wav[i]>1)
8023: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8024:
8025: if (cens[i] == 0 && wav[i]>1)
8026: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8027: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8028:
8029: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8030: if (wav[i] > 1 ) { /* ??? */
8031: L=L+A*weight[i];
8032: /* 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]);*/
8033: }
8034: }
8035:
8036: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8037:
8038: return -2*L*num/sump;
8039: }
8040:
1.136 brouard 8041: #ifdef GSL
8042: /******************* Gompertz_f Likelihood ******************************/
8043: double gompertz_f(const gsl_vector *v, void *params)
8044: {
8045: double A,B,LL=0.0,sump=0.,num=0.;
8046: double *x= (double *) v->data;
8047: int i,n=0; /* n is the size of the sample */
8048:
8049: for (i=0;i<=imx-1 ; i++) {
8050: sump=sump+weight[i];
8051: /* sump=sump+1;*/
8052: num=num+1;
8053: }
8054:
8055:
8056: /* for (i=0; i<=imx; i++)
8057: 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]);*/
8058: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8059: for (i=1;i<=imx ; i++)
8060: {
8061: if (cens[i] == 1 && wav[i]>1)
8062: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8063:
8064: if (cens[i] == 0 && wav[i]>1)
8065: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8066: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8067:
8068: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8069: if (wav[i] > 1 ) { /* ??? */
8070: LL=LL+A*weight[i];
8071: /* 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]);*/
8072: }
8073: }
8074:
8075: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8076: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8077:
8078: return -2*LL*num/sump;
8079: }
8080: #endif
8081:
1.126 brouard 8082: /******************* Printing html file ***********/
1.201 brouard 8083: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8084: int lastpass, int stepm, int weightopt, char model[],\
8085: int imx, double p[],double **matcov,double agemortsup){
8086: int i,k;
8087:
8088: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8089: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8090: for (i=1;i<=2;i++)
8091: 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 8092: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8093: fprintf(fichtm,"</ul>");
8094:
8095: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8096:
8097: 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>");
8098:
8099: for (k=agegomp;k<(agemortsup-2);k++)
8100: 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]);
8101:
8102:
8103: fflush(fichtm);
8104: }
8105:
8106: /******************* Gnuplot file **************/
1.201 brouard 8107: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8108:
8109: char dirfileres[132],optfileres[132];
1.164 brouard 8110:
1.126 brouard 8111: int ng;
8112:
8113:
8114: /*#ifdef windows */
8115: fprintf(ficgp,"cd \"%s\" \n",pathc);
8116: /*#endif */
8117:
8118:
8119: strcpy(dirfileres,optionfilefiname);
8120: strcpy(optfileres,"vpl");
1.199 brouard 8121: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8122: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8123: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8124: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8125: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8126:
8127: }
8128:
1.136 brouard 8129: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8130: {
1.126 brouard 8131:
1.136 brouard 8132: /*-------- data file ----------*/
8133: FILE *fic;
8134: char dummy[]=" ";
1.240 brouard 8135: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8136: int lstra;
1.136 brouard 8137: int linei, month, year,iout;
8138: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8139: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8140: char *stratrunc;
1.223 brouard 8141:
1.240 brouard 8142: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8143: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8144:
1.240 brouard 8145: for(v=1; v <=ncovcol;v++){
8146: DummyV[v]=0;
8147: FixedV[v]=0;
8148: }
8149: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8150: DummyV[v]=1;
8151: FixedV[v]=0;
8152: }
8153: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8154: DummyV[v]=0;
8155: FixedV[v]=1;
8156: }
8157: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8158: DummyV[v]=1;
8159: FixedV[v]=1;
8160: }
8161: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8162: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8163: 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]);
8164: }
1.126 brouard 8165:
1.136 brouard 8166: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8167: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8168: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8169: }
1.126 brouard 8170:
1.136 brouard 8171: i=1;
8172: linei=0;
8173: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8174: linei=linei+1;
8175: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8176: if(line[j] == '\t')
8177: line[j] = ' ';
8178: }
8179: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8180: ;
8181: };
8182: line[j+1]=0; /* Trims blanks at end of line */
8183: if(line[0]=='#'){
8184: fprintf(ficlog,"Comment line\n%s\n",line);
8185: printf("Comment line\n%s\n",line);
8186: continue;
8187: }
8188: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8189: strcpy(line, linetmp);
1.223 brouard 8190:
8191: /* Loops on waves */
8192: for (j=maxwav;j>=1;j--){
8193: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8194: cutv(stra, strb, line, ' ');
8195: if(strb[0]=='.') { /* Missing value */
8196: lval=-1;
8197: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8198: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8199: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8200: 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);
8201: 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);
8202: return 1;
8203: }
8204: }else{
8205: errno=0;
8206: /* what_kind_of_number(strb); */
8207: dval=strtod(strb,&endptr);
8208: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8209: /* if(strb != endptr && *endptr == '\0') */
8210: /* dval=dlval; */
8211: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8212: if( strb[0]=='\0' || (*endptr != '\0')){
8213: 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);
8214: 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);
8215: return 1;
8216: }
8217: cotqvar[j][iv][i]=dval;
8218: cotvar[j][ntv+iv][i]=dval;
8219: }
8220: strcpy(line,stra);
1.223 brouard 8221: }/* end loop ntqv */
1.225 brouard 8222:
1.223 brouard 8223: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8224: cutv(stra, strb, line, ' ');
8225: if(strb[0]=='.') { /* Missing value */
8226: lval=-1;
8227: }else{
8228: errno=0;
8229: lval=strtol(strb,&endptr,10);
8230: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8231: if( strb[0]=='\0' || (*endptr != '\0')){
8232: 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);
8233: 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);
8234: return 1;
8235: }
8236: }
8237: if(lval <-1 || lval >1){
8238: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8239: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8240: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8241: For example, for multinomial values like 1, 2 and 3,\n \
8242: build V1=0 V2=0 for the reference value (1),\n \
8243: V1=1 V2=0 for (2) \n \
1.223 brouard 8244: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8245: output of IMaCh is often meaningless.\n \
1.223 brouard 8246: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8247: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8248: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8249: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8250: For example, for multinomial values like 1, 2 and 3,\n \
8251: build V1=0 V2=0 for the reference value (1),\n \
8252: V1=1 V2=0 for (2) \n \
1.223 brouard 8253: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8254: output of IMaCh is often meaningless.\n \
1.223 brouard 8255: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8256: return 1;
8257: }
8258: cotvar[j][iv][i]=(double)(lval);
8259: strcpy(line,stra);
1.223 brouard 8260: }/* end loop ntv */
1.225 brouard 8261:
1.223 brouard 8262: /* Statuses at wave */
1.137 brouard 8263: cutv(stra, strb, line, ' ');
1.223 brouard 8264: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8265: lval=-1;
1.136 brouard 8266: }else{
1.238 brouard 8267: errno=0;
8268: lval=strtol(strb,&endptr,10);
8269: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8270: if( strb[0]=='\0' || (*endptr != '\0')){
8271: 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);
8272: 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);
8273: return 1;
8274: }
1.136 brouard 8275: }
1.225 brouard 8276:
1.136 brouard 8277: s[j][i]=lval;
1.225 brouard 8278:
1.223 brouard 8279: /* Date of Interview */
1.136 brouard 8280: strcpy(line,stra);
8281: cutv(stra, strb,line,' ');
1.169 brouard 8282: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8283: }
1.169 brouard 8284: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8285: month=99;
8286: year=9999;
1.136 brouard 8287: }else{
1.225 brouard 8288: 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);
8289: 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);
8290: return 1;
1.136 brouard 8291: }
8292: anint[j][i]= (double) year;
8293: mint[j][i]= (double)month;
8294: strcpy(line,stra);
1.223 brouard 8295: } /* End loop on waves */
1.225 brouard 8296:
1.223 brouard 8297: /* Date of death */
1.136 brouard 8298: cutv(stra, strb,line,' ');
1.169 brouard 8299: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8300: }
1.169 brouard 8301: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8302: month=99;
8303: year=9999;
8304: }else{
1.141 brouard 8305: 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 8306: 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);
8307: return 1;
1.136 brouard 8308: }
8309: andc[i]=(double) year;
8310: moisdc[i]=(double) month;
8311: strcpy(line,stra);
8312:
1.223 brouard 8313: /* Date of birth */
1.136 brouard 8314: cutv(stra, strb,line,' ');
1.169 brouard 8315: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8316: }
1.169 brouard 8317: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8318: month=99;
8319: year=9999;
8320: }else{
1.141 brouard 8321: 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);
8322: 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 8323: return 1;
1.136 brouard 8324: }
8325: if (year==9999) {
1.141 brouard 8326: 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);
8327: 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 8328: return 1;
8329:
1.136 brouard 8330: }
8331: annais[i]=(double)(year);
8332: moisnais[i]=(double)(month);
8333: strcpy(line,stra);
1.225 brouard 8334:
1.223 brouard 8335: /* Sample weight */
1.136 brouard 8336: cutv(stra, strb,line,' ');
8337: errno=0;
8338: dval=strtod(strb,&endptr);
8339: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8340: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8341: 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 8342: fflush(ficlog);
8343: return 1;
8344: }
8345: weight[i]=dval;
8346: strcpy(line,stra);
1.225 brouard 8347:
1.223 brouard 8348: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8349: cutv(stra, strb, line, ' ');
8350: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8351: lval=-1;
1.223 brouard 8352: }else{
1.225 brouard 8353: errno=0;
8354: /* what_kind_of_number(strb); */
8355: dval=strtod(strb,&endptr);
8356: /* if(strb != endptr && *endptr == '\0') */
8357: /* dval=dlval; */
8358: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8359: if( strb[0]=='\0' || (*endptr != '\0')){
8360: 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);
8361: 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);
8362: return 1;
8363: }
8364: coqvar[iv][i]=dval;
1.226 brouard 8365: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8366: }
8367: strcpy(line,stra);
8368: }/* end loop nqv */
1.136 brouard 8369:
1.223 brouard 8370: /* Covariate values */
1.136 brouard 8371: for (j=ncovcol;j>=1;j--){
8372: cutv(stra, strb,line,' ');
1.223 brouard 8373: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8374: lval=-1;
1.136 brouard 8375: }else{
1.225 brouard 8376: errno=0;
8377: lval=strtol(strb,&endptr,10);
8378: if( strb[0]=='\0' || (*endptr != '\0')){
8379: 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);
8380: 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);
8381: return 1;
8382: }
1.136 brouard 8383: }
8384: if(lval <-1 || lval >1){
1.225 brouard 8385: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8386: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8387: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8388: For example, for multinomial values like 1, 2 and 3,\n \
8389: build V1=0 V2=0 for the reference value (1),\n \
8390: V1=1 V2=0 for (2) \n \
1.136 brouard 8391: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8392: output of IMaCh is often meaningless.\n \
1.136 brouard 8393: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8394: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8395: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8396: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8397: For example, for multinomial values like 1, 2 and 3,\n \
8398: build V1=0 V2=0 for the reference value (1),\n \
8399: V1=1 V2=0 for (2) \n \
1.136 brouard 8400: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8401: output of IMaCh is often meaningless.\n \
1.136 brouard 8402: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8403: return 1;
1.136 brouard 8404: }
8405: covar[j][i]=(double)(lval);
8406: strcpy(line,stra);
8407: }
8408: lstra=strlen(stra);
1.225 brouard 8409:
1.136 brouard 8410: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8411: stratrunc = &(stra[lstra-9]);
8412: num[i]=atol(stratrunc);
8413: }
8414: else
8415: num[i]=atol(stra);
8416: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8417: 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;}*/
8418:
8419: i=i+1;
8420: } /* End loop reading data */
1.225 brouard 8421:
1.136 brouard 8422: *imax=i-1; /* Number of individuals */
8423: fclose(fic);
1.225 brouard 8424:
1.136 brouard 8425: return (0);
1.164 brouard 8426: /* endread: */
1.225 brouard 8427: printf("Exiting readdata: ");
8428: fclose(fic);
8429: return (1);
1.223 brouard 8430: }
1.126 brouard 8431:
1.234 brouard 8432: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8433: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8434: while (*p2 == ' ')
1.234 brouard 8435: p2++;
8436: /* while ((*p1++ = *p2++) !=0) */
8437: /* ; */
8438: /* do */
8439: /* while (*p2 == ' ') */
8440: /* p2++; */
8441: /* while (*p1++ == *p2++); */
8442: *stri=p2;
1.145 brouard 8443: }
8444:
1.235 brouard 8445: int decoderesult ( char resultline[], int nres)
1.230 brouard 8446: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8447: {
1.235 brouard 8448: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8449: char resultsav[MAXLINE];
1.234 brouard 8450: int resultmodel[MAXLINE];
8451: int modelresult[MAXLINE];
1.230 brouard 8452: char stra[80], strb[80], strc[80], strd[80],stre[80];
8453:
1.234 brouard 8454: removefirstspace(&resultline);
1.233 brouard 8455: printf("decoderesult:%s\n",resultline);
1.230 brouard 8456:
8457: if (strstr(resultline,"v") !=0){
8458: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8459: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8460: return 1;
8461: }
8462: trimbb(resultsav, resultline);
8463: if (strlen(resultsav) >1){
8464: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8465: }
1.253 ! brouard 8466: if(j == 0){ /* Resultline but no = */
! 8467: TKresult[nres]=0; /* Combination for the nresult and the model */
! 8468: return (0);
! 8469: }
! 8470:
1.234 brouard 8471: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8472: 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);
8473: 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);
8474: }
8475: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8476: if(nbocc(resultsav,'=') >1){
8477: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8478: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8479: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8480: }else
8481: cutl(strc,strd,resultsav,'=');
1.230 brouard 8482: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8483:
1.230 brouard 8484: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8485: Tvarsel[k]=atoi(strc);
8486: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8487: /* cptcovsel++; */
8488: if (nbocc(stra,'=') >0)
8489: strcpy(resultsav,stra); /* and analyzes it */
8490: }
1.235 brouard 8491: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8492: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8493: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8494: match=0;
1.236 brouard 8495: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8496: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8497: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8498: match=1;
8499: break;
8500: }
8501: }
8502: if(match == 0){
8503: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8504: }
8505: }
8506: }
1.235 brouard 8507: /* Checking for missing or useless values in comparison of current model needs */
8508: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8509: match=0;
1.235 brouard 8510: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8511: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8512: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8513: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8514: ++match;
8515: }
8516: }
8517: }
8518: if(match == 0){
8519: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8520: }else if(match > 1){
8521: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8522: }
8523: }
1.235 brouard 8524:
1.234 brouard 8525: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8526: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8527: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8528: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8529: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8530: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8531: /* 1 0 0 0 */
8532: /* 2 1 0 0 */
8533: /* 3 0 1 0 */
8534: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8535: /* 5 0 0 1 */
8536: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8537: /* 7 0 1 1 */
8538: /* 8 1 1 1 */
1.237 brouard 8539: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8540: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8541: /* V5*age V5 known which value for nres? */
8542: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8543: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8544: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8545: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8546: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8547: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8548: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8549: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8550: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8551: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8552: k4++;;
8553: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8554: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8555: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8556: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8557: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8558: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8559: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8560: k4q++;;
8561: }
8562: }
1.234 brouard 8563:
1.235 brouard 8564: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8565: return (0);
8566: }
1.235 brouard 8567:
1.230 brouard 8568: int decodemodel( char model[], int lastobs)
8569: /**< This routine decodes the model and returns:
1.224 brouard 8570: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8571: * - nagesqr = 1 if age*age in the model, otherwise 0.
8572: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8573: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8574: * - cptcovage number of covariates with age*products =2
8575: * - cptcovs number of simple covariates
8576: * - 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
8577: * which is a new column after the 9 (ncovcol) variables.
8578: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8579: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8580: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8581: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8582: */
1.136 brouard 8583: {
1.238 brouard 8584: int i, j, k, ks, v;
1.227 brouard 8585: int j1, k1, k2, k3, k4;
1.136 brouard 8586: char modelsav[80];
1.145 brouard 8587: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8588: char *strpt;
1.136 brouard 8589:
1.145 brouard 8590: /*removespace(model);*/
1.136 brouard 8591: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8592: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8593: if (strstr(model,"AGE") !=0){
1.192 brouard 8594: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8595: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8596: return 1;
8597: }
1.141 brouard 8598: if (strstr(model,"v") !=0){
8599: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8600: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8601: return 1;
8602: }
1.187 brouard 8603: strcpy(modelsav,model);
8604: if ((strpt=strstr(model,"age*age")) !=0){
8605: printf(" strpt=%s, model=%s\n",strpt, model);
8606: if(strpt != model){
1.234 brouard 8607: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8608: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8609: corresponding column of parameters.\n",model);
1.234 brouard 8610: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8611: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8612: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8613: return 1;
1.225 brouard 8614: }
1.187 brouard 8615: nagesqr=1;
8616: if (strstr(model,"+age*age") !=0)
1.234 brouard 8617: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8618: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8619: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8620: else
1.234 brouard 8621: substrchaine(modelsav, model, "age*age");
1.187 brouard 8622: }else
8623: nagesqr=0;
8624: if (strlen(modelsav) >1){
8625: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8626: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8627: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8628: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8629: * cst, age and age*age
8630: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8631: /* including age products which are counted in cptcovage.
8632: * but the covariates which are products must be treated
8633: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8634: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8635: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8636:
8637:
1.187 brouard 8638: /* Design
8639: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8640: * < ncovcol=8 >
8641: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8642: * k= 1 2 3 4 5 6 7 8
8643: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8644: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8645: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8646: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8647: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8648: * Tage[++cptcovage]=k
8649: * if products, new covar are created after ncovcol with k1
8650: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8651: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8652: * 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
8653: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8654: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8655: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8656: * < ncovcol=8 >
8657: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8658: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8659: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8660: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8661: * p Tprod[1]@2={ 6, 5}
8662: *p Tvard[1][1]@4= {7, 8, 5, 6}
8663: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8664: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8665: *How to reorganize?
8666: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8667: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8668: * {2, 1, 4, 8, 5, 6, 3, 7}
8669: * Struct []
8670: */
1.225 brouard 8671:
1.187 brouard 8672: /* This loop fills the array Tvar from the string 'model'.*/
8673: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8674: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8675: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8676: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8677: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8678: /* k=1 Tvar[1]=2 (from V2) */
8679: /* k=5 Tvar[5] */
8680: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8681: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8682: /* } */
1.198 brouard 8683: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8684: /*
8685: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8686: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8687: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8688: }
1.187 brouard 8689: cptcovage=0;
8690: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8691: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8692: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8693: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8694: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8695: /*scanf("%d",i);*/
8696: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8697: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8698: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8699: /* covar is not filled and then is empty */
8700: cptcovprod--;
8701: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8702: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8703: Typevar[k]=1; /* 1 for age product */
8704: cptcovage++; /* Sums the number of covariates which include age as a product */
8705: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8706: /*printf("stre=%s ", stre);*/
8707: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8708: cptcovprod--;
8709: cutl(stre,strb,strc,'V');
8710: Tvar[k]=atoi(stre);
8711: Typevar[k]=1; /* 1 for age product */
8712: cptcovage++;
8713: Tage[cptcovage]=k;
8714: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8715: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8716: cptcovn++;
8717: cptcovprodnoage++;k1++;
8718: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8719: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8720: because this model-covariate is a construction we invent a new column
8721: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8722: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8723: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8724: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8725: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8726: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8727: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8728: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8729: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8730: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8731: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8732: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8733: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8734: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8735: for (i=1; i<=lastobs;i++){
8736: /* Computes the new covariate which is a product of
8737: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8738: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8739: }
8740: } /* End age is not in the model */
8741: } /* End if model includes a product */
8742: else { /* no more sum */
8743: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8744: /* scanf("%d",i);*/
8745: cutl(strd,strc,strb,'V');
8746: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8747: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8748: Tvar[k]=atoi(strd);
8749: Typevar[k]=0; /* 0 for simple covariates */
8750: }
8751: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8752: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8753: scanf("%d",i);*/
1.187 brouard 8754: } /* end of loop + on total covariates */
8755: } /* end if strlen(modelsave == 0) age*age might exist */
8756: } /* end if strlen(model == 0) */
1.136 brouard 8757:
8758: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8759: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8760:
1.136 brouard 8761: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8762: printf("cptcovprod=%d ", cptcovprod);
8763: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8764: scanf("%d ",i);*/
8765:
8766:
1.230 brouard 8767: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8768: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8769: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8770: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8771: k = 1 2 3 4 5 6 7 8 9
8772: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8773: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8774: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8775: Dummy[k] 1 0 0 0 3 1 1 2 3
8776: Tmodelind[combination of covar]=k;
1.225 brouard 8777: */
8778: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8779: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8780: /* 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 8781: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8782: printf("Model=%s\n\
8783: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8784: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8785: 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);
8786: fprintf(ficlog,"Model=%s\n\
8787: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8788: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8789: 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 8790: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8791: 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 */
8792: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8793: Fixed[k]= 0;
8794: Dummy[k]= 0;
1.225 brouard 8795: ncoveff++;
1.232 brouard 8796: ncovf++;
1.234 brouard 8797: nsd++;
8798: modell[k].maintype= FTYPE;
8799: TvarsD[nsd]=Tvar[k];
8800: TvarsDind[nsd]=k;
8801: TvarF[ncovf]=Tvar[k];
8802: TvarFind[ncovf]=k;
8803: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8804: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8805: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8806: Fixed[k]= 0;
8807: Dummy[k]= 0;
8808: ncoveff++;
8809: ncovf++;
8810: modell[k].maintype= FTYPE;
8811: TvarF[ncovf]=Tvar[k];
8812: TvarFind[ncovf]=k;
1.230 brouard 8813: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8814: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8815: }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 8816: Fixed[k]= 0;
8817: Dummy[k]= 1;
1.230 brouard 8818: nqfveff++;
1.234 brouard 8819: modell[k].maintype= FTYPE;
8820: modell[k].subtype= FQ;
8821: nsq++;
8822: TvarsQ[nsq]=Tvar[k];
8823: TvarsQind[nsq]=k;
1.232 brouard 8824: ncovf++;
1.234 brouard 8825: TvarF[ncovf]=Tvar[k];
8826: TvarFind[ncovf]=k;
1.231 brouard 8827: 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 8828: 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 8829: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8830: Fixed[k]= 1;
8831: Dummy[k]= 0;
1.225 brouard 8832: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8833: modell[k].maintype= VTYPE;
8834: modell[k].subtype= VD;
8835: nsd++;
8836: TvarsD[nsd]=Tvar[k];
8837: TvarsDind[nsd]=k;
8838: ncovv++; /* Only simple time varying variables */
8839: TvarV[ncovv]=Tvar[k];
1.242 brouard 8840: 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 8841: 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 */
8842: 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 8843: 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);
8844: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8845: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8846: Fixed[k]= 1;
8847: Dummy[k]= 1;
8848: nqtveff++;
8849: modell[k].maintype= VTYPE;
8850: modell[k].subtype= VQ;
8851: ncovv++; /* Only simple time varying variables */
8852: nsq++;
8853: TvarsQ[nsq]=Tvar[k];
8854: TvarsQind[nsq]=k;
8855: TvarV[ncovv]=Tvar[k];
1.242 brouard 8856: 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 8857: 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 */
8858: 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 8859: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8860: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8861: 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 8862: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8863: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8864: ncova++;
8865: TvarA[ncova]=Tvar[k];
8866: TvarAind[ncova]=k;
1.231 brouard 8867: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8868: Fixed[k]= 2;
8869: Dummy[k]= 2;
8870: modell[k].maintype= ATYPE;
8871: modell[k].subtype= APFD;
8872: /* ncoveff++; */
1.227 brouard 8873: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8874: Fixed[k]= 2;
8875: Dummy[k]= 3;
8876: modell[k].maintype= ATYPE;
8877: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8878: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8879: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8880: Fixed[k]= 3;
8881: Dummy[k]= 2;
8882: modell[k].maintype= ATYPE;
8883: modell[k].subtype= APVD; /* Product age * varying dummy */
8884: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8885: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8886: Fixed[k]= 3;
8887: Dummy[k]= 3;
8888: modell[k].maintype= ATYPE;
8889: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8890: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8891: }
8892: }else if (Typevar[k] == 2) { /* product without age */
8893: k1=Tposprod[k];
8894: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8895: if(Tvard[k1][2] <=ncovcol){
8896: Fixed[k]= 1;
8897: Dummy[k]= 0;
8898: modell[k].maintype= FTYPE;
8899: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8900: ncovf++; /* Fixed variables without age */
8901: TvarF[ncovf]=Tvar[k];
8902: TvarFind[ncovf]=k;
8903: }else if(Tvard[k1][2] <=ncovcol+nqv){
8904: Fixed[k]= 0; /* or 2 ?*/
8905: Dummy[k]= 1;
8906: modell[k].maintype= FTYPE;
8907: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8908: ncovf++; /* Varying variables without age */
8909: TvarF[ncovf]=Tvar[k];
8910: TvarFind[ncovf]=k;
8911: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8912: Fixed[k]= 1;
8913: Dummy[k]= 0;
8914: modell[k].maintype= VTYPE;
8915: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8916: ncovv++; /* Varying variables without age */
8917: TvarV[ncovv]=Tvar[k];
8918: TvarVind[ncovv]=k;
8919: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8920: Fixed[k]= 1;
8921: Dummy[k]= 1;
8922: modell[k].maintype= VTYPE;
8923: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8924: ncovv++; /* Varying variables without age */
8925: TvarV[ncovv]=Tvar[k];
8926: TvarVind[ncovv]=k;
8927: }
1.227 brouard 8928: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8929: if(Tvard[k1][2] <=ncovcol){
8930: Fixed[k]= 0; /* or 2 ?*/
8931: Dummy[k]= 1;
8932: modell[k].maintype= FTYPE;
8933: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8934: ncovf++; /* Fixed variables without age */
8935: TvarF[ncovf]=Tvar[k];
8936: TvarFind[ncovf]=k;
8937: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8938: Fixed[k]= 1;
8939: Dummy[k]= 1;
8940: modell[k].maintype= VTYPE;
8941: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8942: ncovv++; /* Varying variables without age */
8943: TvarV[ncovv]=Tvar[k];
8944: TvarVind[ncovv]=k;
8945: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8946: Fixed[k]= 1;
8947: Dummy[k]= 1;
8948: modell[k].maintype= VTYPE;
8949: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8950: ncovv++; /* Varying variables without age */
8951: TvarV[ncovv]=Tvar[k];
8952: TvarVind[ncovv]=k;
8953: ncovv++; /* Varying variables without age */
8954: TvarV[ncovv]=Tvar[k];
8955: TvarVind[ncovv]=k;
8956: }
1.227 brouard 8957: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8958: if(Tvard[k1][2] <=ncovcol){
8959: Fixed[k]= 1;
8960: Dummy[k]= 1;
8961: modell[k].maintype= VTYPE;
8962: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8963: ncovv++; /* Varying variables without age */
8964: TvarV[ncovv]=Tvar[k];
8965: TvarVind[ncovv]=k;
8966: }else if(Tvard[k1][2] <=ncovcol+nqv){
8967: Fixed[k]= 1;
8968: Dummy[k]= 1;
8969: modell[k].maintype= VTYPE;
8970: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8971: ncovv++; /* Varying variables without age */
8972: TvarV[ncovv]=Tvar[k];
8973: TvarVind[ncovv]=k;
8974: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8975: Fixed[k]= 1;
8976: Dummy[k]= 0;
8977: modell[k].maintype= VTYPE;
8978: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8979: ncovv++; /* Varying variables without age */
8980: TvarV[ncovv]=Tvar[k];
8981: TvarVind[ncovv]=k;
8982: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8983: Fixed[k]= 1;
8984: Dummy[k]= 1;
8985: modell[k].maintype= VTYPE;
8986: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8987: ncovv++; /* Varying variables without age */
8988: TvarV[ncovv]=Tvar[k];
8989: TvarVind[ncovv]=k;
8990: }
1.227 brouard 8991: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8992: if(Tvard[k1][2] <=ncovcol){
8993: Fixed[k]= 1;
8994: Dummy[k]= 1;
8995: modell[k].maintype= VTYPE;
8996: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8997: ncovv++; /* Varying variables without age */
8998: TvarV[ncovv]=Tvar[k];
8999: TvarVind[ncovv]=k;
9000: }else if(Tvard[k1][2] <=ncovcol+nqv){
9001: Fixed[k]= 1;
9002: Dummy[k]= 1;
9003: modell[k].maintype= VTYPE;
9004: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9005: ncovv++; /* Varying variables without age */
9006: TvarV[ncovv]=Tvar[k];
9007: TvarVind[ncovv]=k;
9008: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9009: Fixed[k]= 1;
9010: Dummy[k]= 1;
9011: modell[k].maintype= VTYPE;
9012: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9013: ncovv++; /* Varying variables without age */
9014: TvarV[ncovv]=Tvar[k];
9015: TvarVind[ncovv]=k;
9016: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9017: Fixed[k]= 1;
9018: Dummy[k]= 1;
9019: modell[k].maintype= VTYPE;
9020: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9021: ncovv++; /* Varying variables without age */
9022: TvarV[ncovv]=Tvar[k];
9023: TvarVind[ncovv]=k;
9024: }
1.227 brouard 9025: }else{
1.240 brouard 9026: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9027: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9028: } /*end k1*/
1.225 brouard 9029: }else{
1.226 brouard 9030: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9031: 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 9032: }
1.227 brouard 9033: 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 9034: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9035: 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]);
9036: }
9037: /* Searching for doublons in the model */
9038: for(k1=1; k1<= cptcovt;k1++){
9039: for(k2=1; k2 <k1;k2++){
9040: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9041: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9042: if(Tvar[k1]==Tvar[k2]){
9043: 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]]);
9044: 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);
9045: return(1);
9046: }
9047: }else if (Typevar[k1] ==2){
9048: k3=Tposprod[k1];
9049: k4=Tposprod[k2];
9050: 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])) ){
9051: 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]]);
9052: 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);
9053: return(1);
9054: }
9055: }
1.227 brouard 9056: }
9057: }
1.225 brouard 9058: }
9059: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9060: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9061: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9062: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9063: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9064: /*endread:*/
1.225 brouard 9065: printf("Exiting decodemodel: ");
9066: return (1);
1.136 brouard 9067: }
9068:
1.169 brouard 9069: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9070: {/* Check ages at death */
1.136 brouard 9071: int i, m;
1.218 brouard 9072: int firstone=0;
9073:
1.136 brouard 9074: for (i=1; i<=imx; i++) {
9075: for(m=2; (m<= maxwav); m++) {
9076: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9077: anint[m][i]=9999;
1.216 brouard 9078: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9079: s[m][i]=-1;
1.136 brouard 9080: }
9081: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 9082: *nberr = *nberr + 1;
1.218 brouard 9083: if(firstone == 0){
9084: firstone=1;
9085: 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);
9086: }
9087: 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 9088: s[m][i]=-1;
9089: }
9090: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9091: (*nberr)++;
1.136 brouard 9092: 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]);
9093: 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]);
9094: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
9095: }
9096: }
9097: }
9098:
9099: for (i=1; i<=imx; i++) {
9100: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9101: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9102: 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 9103: if (s[m][i] >= nlstate+1) {
1.169 brouard 9104: if(agedc[i]>0){
9105: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9106: agev[m][i]=agedc[i];
1.214 brouard 9107: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9108: }else {
1.136 brouard 9109: if ((int)andc[i]!=9999){
9110: nbwarn++;
9111: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9112: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9113: agev[m][i]=-1;
9114: }
9115: }
1.169 brouard 9116: } /* agedc > 0 */
1.214 brouard 9117: } /* end if */
1.136 brouard 9118: else if(s[m][i] !=9){ /* Standard case, age in fractional
9119: years but with the precision of a month */
9120: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9121: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9122: agev[m][i]=1;
9123: else if(agev[m][i] < *agemin){
9124: *agemin=agev[m][i];
9125: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9126: }
9127: else if(agev[m][i] >*agemax){
9128: *agemax=agev[m][i];
1.156 brouard 9129: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9130: }
9131: /*agev[m][i]=anint[m][i]-annais[i];*/
9132: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9133: } /* en if 9*/
1.136 brouard 9134: else { /* =9 */
1.214 brouard 9135: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9136: agev[m][i]=1;
9137: s[m][i]=-1;
9138: }
9139: }
1.214 brouard 9140: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9141: agev[m][i]=1;
1.214 brouard 9142: else{
9143: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9144: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9145: agev[m][i]=0;
9146: }
9147: } /* End for lastpass */
9148: }
1.136 brouard 9149:
9150: for (i=1; i<=imx; i++) {
9151: for(m=firstpass; (m<=lastpass); m++){
9152: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9153: (*nberr)++;
1.136 brouard 9154: 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);
9155: 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);
9156: return 1;
9157: }
9158: }
9159: }
9160:
9161: /*for (i=1; i<=imx; i++){
9162: for (m=firstpass; (m<lastpass); m++){
9163: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9164: }
9165:
9166: }*/
9167:
9168:
1.139 brouard 9169: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9170: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9171:
9172: return (0);
1.164 brouard 9173: /* endread:*/
1.136 brouard 9174: printf("Exiting calandcheckages: ");
9175: return (1);
9176: }
9177:
1.172 brouard 9178: #if defined(_MSC_VER)
9179: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9180: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9181: //#include "stdafx.h"
9182: //#include <stdio.h>
9183: //#include <tchar.h>
9184: //#include <windows.h>
9185: //#include <iostream>
9186: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9187:
9188: LPFN_ISWOW64PROCESS fnIsWow64Process;
9189:
9190: BOOL IsWow64()
9191: {
9192: BOOL bIsWow64 = FALSE;
9193:
9194: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9195: // (HANDLE, PBOOL);
9196:
9197: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9198:
9199: HMODULE module = GetModuleHandle(_T("kernel32"));
9200: const char funcName[] = "IsWow64Process";
9201: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9202: GetProcAddress(module, funcName);
9203:
9204: if (NULL != fnIsWow64Process)
9205: {
9206: if (!fnIsWow64Process(GetCurrentProcess(),
9207: &bIsWow64))
9208: //throw std::exception("Unknown error");
9209: printf("Unknown error\n");
9210: }
9211: return bIsWow64 != FALSE;
9212: }
9213: #endif
1.177 brouard 9214:
1.191 brouard 9215: void syscompilerinfo(int logged)
1.167 brouard 9216: {
9217: /* #include "syscompilerinfo.h"*/
1.185 brouard 9218: /* command line Intel compiler 32bit windows, XP compatible:*/
9219: /* /GS /W3 /Gy
9220: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9221: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9222: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9223: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9224: */
9225: /* 64 bits */
1.185 brouard 9226: /*
9227: /GS /W3 /Gy
9228: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9229: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9230: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9231: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9232: /* Optimization are useless and O3 is slower than O2 */
9233: /*
9234: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9235: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9236: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9237: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9238: */
1.186 brouard 9239: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9240: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9241: /PDB:"visual studio
9242: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9243: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9244: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9245: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9246: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9247: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9248: uiAccess='false'"
9249: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9250: /NOLOGO /TLBID:1
9251: */
1.177 brouard 9252: #if defined __INTEL_COMPILER
1.178 brouard 9253: #if defined(__GNUC__)
9254: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9255: #endif
1.177 brouard 9256: #elif defined(__GNUC__)
1.179 brouard 9257: #ifndef __APPLE__
1.174 brouard 9258: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9259: #endif
1.177 brouard 9260: struct utsname sysInfo;
1.178 brouard 9261: int cross = CROSS;
9262: if (cross){
9263: printf("Cross-");
1.191 brouard 9264: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9265: }
1.174 brouard 9266: #endif
9267:
1.171 brouard 9268: #include <stdint.h>
1.178 brouard 9269:
1.191 brouard 9270: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9271: #if defined(__clang__)
1.191 brouard 9272: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9273: #endif
9274: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9275: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9276: #endif
9277: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9278: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9279: #endif
9280: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9281: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9282: #endif
9283: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9284: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9285: #endif
9286: #if defined(_MSC_VER)
1.191 brouard 9287: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9288: #endif
9289: #if defined(__PGI)
1.191 brouard 9290: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9291: #endif
9292: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9293: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9294: #endif
1.191 brouard 9295: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9296:
1.167 brouard 9297: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9298: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9299: // Windows (x64 and x86)
1.191 brouard 9300: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9301: #elif __unix__ // all unices, not all compilers
9302: // Unix
1.191 brouard 9303: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9304: #elif __linux__
9305: // linux
1.191 brouard 9306: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9307: #elif __APPLE__
1.174 brouard 9308: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9309: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9310: #endif
9311:
9312: /* __MINGW32__ */
9313: /* __CYGWIN__ */
9314: /* __MINGW64__ */
9315: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9316: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9317: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9318: /* _WIN64 // Defined for applications for Win64. */
9319: /* _M_X64 // Defined for compilations that target x64 processors. */
9320: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9321:
1.167 brouard 9322: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9323: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9324: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9325: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9326: #else
1.191 brouard 9327: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9328: #endif
9329:
1.169 brouard 9330: #if defined(__GNUC__)
9331: # if defined(__GNUC_PATCHLEVEL__)
9332: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9333: + __GNUC_MINOR__ * 100 \
9334: + __GNUC_PATCHLEVEL__)
9335: # else
9336: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9337: + __GNUC_MINOR__ * 100)
9338: # endif
1.174 brouard 9339: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9340: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9341:
9342: if (uname(&sysInfo) != -1) {
9343: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9344: 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 9345: }
9346: else
9347: perror("uname() error");
1.179 brouard 9348: //#ifndef __INTEL_COMPILER
9349: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9350: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9351: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9352: #endif
1.169 brouard 9353: #endif
1.172 brouard 9354:
9355: // void main()
9356: // {
1.169 brouard 9357: #if defined(_MSC_VER)
1.174 brouard 9358: if (IsWow64()){
1.191 brouard 9359: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9360: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9361: }
9362: else{
1.191 brouard 9363: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9364: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9365: }
1.172 brouard 9366: // printf("\nPress Enter to continue...");
9367: // getchar();
9368: // }
9369:
1.169 brouard 9370: #endif
9371:
1.167 brouard 9372:
1.219 brouard 9373: }
1.136 brouard 9374:
1.219 brouard 9375: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9376: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9377: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9378: /* double ftolpl = 1.e-10; */
1.180 brouard 9379: double age, agebase, agelim;
1.203 brouard 9380: double tot;
1.180 brouard 9381:
1.202 brouard 9382: strcpy(filerespl,"PL_");
9383: strcat(filerespl,fileresu);
9384: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9385: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9386: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9387: }
1.227 brouard 9388: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9389: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9390: pstamp(ficrespl);
1.203 brouard 9391: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9392: fprintf(ficrespl,"#Age ");
9393: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9394: fprintf(ficrespl,"\n");
1.180 brouard 9395:
1.219 brouard 9396: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9397:
1.219 brouard 9398: agebase=ageminpar;
9399: agelim=agemaxpar;
1.180 brouard 9400:
1.227 brouard 9401: /* i1=pow(2,ncoveff); */
1.234 brouard 9402: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9403: if (cptcovn < 1){i1=1;}
1.180 brouard 9404:
1.238 brouard 9405: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9406: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 ! brouard 9407: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9408: continue;
1.235 brouard 9409:
1.238 brouard 9410: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9411: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9412: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9413: /* k=k+1; */
9414: /* to clean */
9415: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9416: fprintf(ficrespl,"#******");
9417: printf("#******");
9418: fprintf(ficlog,"#******");
9419: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9420: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9421: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9422: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9423: }
9424: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9425: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9426: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9427: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9428: }
9429: fprintf(ficrespl,"******\n");
9430: printf("******\n");
9431: fprintf(ficlog,"******\n");
9432: if(invalidvarcomb[k]){
9433: printf("\nCombination (%d) ignored because no case \n",k);
9434: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9435: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9436: continue;
9437: }
1.219 brouard 9438:
1.238 brouard 9439: fprintf(ficrespl,"#Age ");
9440: for(j=1;j<=cptcoveff;j++) {
9441: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9442: }
9443: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9444: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9445:
1.238 brouard 9446: for (age=agebase; age<=agelim; age++){
9447: /* for (age=agebase; age<=agebase; age++){ */
9448: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9449: fprintf(ficrespl,"%.0f ",age );
9450: for(j=1;j<=cptcoveff;j++)
9451: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9452: tot=0.;
9453: for(i=1; i<=nlstate;i++){
9454: tot += prlim[i][i];
9455: fprintf(ficrespl," %.5f", prlim[i][i]);
9456: }
9457: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9458: } /* Age */
9459: /* was end of cptcod */
9460: } /* cptcov */
9461: } /* nres */
1.219 brouard 9462: return 0;
1.180 brouard 9463: }
9464:
1.218 brouard 9465: 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){
9466: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9467:
9468: /* Computes the back prevalence limit for any combination of covariate values
9469: * at any age between ageminpar and agemaxpar
9470: */
1.235 brouard 9471: int i, j, k, i1, nres=0 ;
1.217 brouard 9472: /* double ftolpl = 1.e-10; */
9473: double age, agebase, agelim;
9474: double tot;
1.218 brouard 9475: /* double ***mobaverage; */
9476: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9477:
9478: strcpy(fileresplb,"PLB_");
9479: strcat(fileresplb,fileresu);
9480: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9481: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9482: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9483: }
9484: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9485: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9486: pstamp(ficresplb);
9487: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9488: fprintf(ficresplb,"#Age ");
9489: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9490: fprintf(ficresplb,"\n");
9491:
1.218 brouard 9492:
9493: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9494:
9495: agebase=ageminpar;
9496: agelim=agemaxpar;
9497:
9498:
1.227 brouard 9499: i1=pow(2,cptcoveff);
1.218 brouard 9500: if (cptcovn < 1){i1=1;}
1.227 brouard 9501:
1.238 brouard 9502: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9503: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 ! brouard 9504: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9505: continue;
9506: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9507: fprintf(ficresplb,"#******");
9508: printf("#******");
9509: fprintf(ficlog,"#******");
9510: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9511: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9512: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9513: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9514: }
9515: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9516: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9517: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9518: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9519: }
9520: fprintf(ficresplb,"******\n");
9521: printf("******\n");
9522: fprintf(ficlog,"******\n");
9523: if(invalidvarcomb[k]){
9524: printf("\nCombination (%d) ignored because no cases \n",k);
9525: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9526: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9527: continue;
9528: }
1.218 brouard 9529:
1.238 brouard 9530: fprintf(ficresplb,"#Age ");
9531: for(j=1;j<=cptcoveff;j++) {
9532: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9533: }
9534: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9535: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9536:
9537:
1.238 brouard 9538: for (age=agebase; age<=agelim; age++){
9539: /* for (age=agebase; age<=agebase; age++){ */
9540: if(mobilavproj > 0){
9541: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9542: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9543: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9544: }else if (mobilavproj == 0){
9545: 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);
9546: 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);
9547: exit(1);
9548: }else{
9549: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9550: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9551: }
9552: fprintf(ficresplb,"%.0f ",age );
9553: for(j=1;j<=cptcoveff;j++)
9554: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9555: tot=0.;
9556: for(i=1; i<=nlstate;i++){
9557: tot += bprlim[i][i];
9558: fprintf(ficresplb," %.5f", bprlim[i][i]);
9559: }
9560: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9561: } /* Age */
9562: /* was end of cptcod */
9563: } /* end of any combination */
9564: } /* end of nres */
1.218 brouard 9565: /* hBijx(p, bage, fage); */
9566: /* fclose(ficrespijb); */
9567:
9568: return 0;
1.217 brouard 9569: }
1.218 brouard 9570:
1.180 brouard 9571: int hPijx(double *p, int bage, int fage){
9572: /*------------- h Pij x at various ages ------------*/
9573:
9574: int stepsize;
9575: int agelim;
9576: int hstepm;
9577: int nhstepm;
1.235 brouard 9578: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9579:
9580: double agedeb;
9581: double ***p3mat;
9582:
1.201 brouard 9583: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9584: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9585: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9586: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9587: }
9588: printf("Computing pij: result on file '%s' \n", filerespij);
9589: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9590:
9591: stepsize=(int) (stepm+YEARM-1)/YEARM;
9592: /*if (stepm<=24) stepsize=2;*/
9593:
9594: agelim=AGESUP;
9595: hstepm=stepsize*YEARM; /* Every year of age */
9596: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9597:
1.180 brouard 9598: /* hstepm=1; aff par mois*/
9599: pstamp(ficrespij);
9600: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9601: i1= pow(2,cptcoveff);
1.218 brouard 9602: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9603: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9604: /* k=k+1; */
1.235 brouard 9605: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9606: for(k=1; k<=i1;k++){
1.253 ! brouard 9607: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9608: continue;
1.183 brouard 9609: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9610: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9611: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9612: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9613: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9614: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9615: }
1.183 brouard 9616: fprintf(ficrespij,"******\n");
9617:
9618: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9619: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9620: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9621:
9622: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9623:
1.183 brouard 9624: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9625: oldm=oldms;savm=savms;
1.235 brouard 9626: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9627: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9628: for(i=1; i<=nlstate;i++)
9629: for(j=1; j<=nlstate+ndeath;j++)
9630: fprintf(ficrespij," %1d-%1d",i,j);
9631: fprintf(ficrespij,"\n");
9632: for (h=0; h<=nhstepm; h++){
9633: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9634: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9635: for(i=1; i<=nlstate;i++)
9636: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9637: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9638: fprintf(ficrespij,"\n");
9639: }
1.183 brouard 9640: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9641: fprintf(ficrespij,"\n");
9642: }
1.180 brouard 9643: /*}*/
9644: }
1.218 brouard 9645: return 0;
1.180 brouard 9646: }
1.218 brouard 9647:
9648: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9649: /*------------- h Bij x at various ages ------------*/
9650:
9651: int stepsize;
1.218 brouard 9652: /* int agelim; */
9653: int ageminl;
1.217 brouard 9654: int hstepm;
9655: int nhstepm;
1.238 brouard 9656: int h, i, i1, j, k, nres;
1.218 brouard 9657:
1.217 brouard 9658: double agedeb;
9659: double ***p3mat;
1.218 brouard 9660:
9661: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9662: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9663: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9664: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9665: }
9666: printf("Computing pij back: result on file '%s' \n", filerespijb);
9667: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9668:
9669: stepsize=(int) (stepm+YEARM-1)/YEARM;
9670: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9671:
1.218 brouard 9672: /* agelim=AGESUP; */
9673: ageminl=30;
9674: hstepm=stepsize*YEARM; /* Every year of age */
9675: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9676:
9677: /* hstepm=1; aff par mois*/
9678: pstamp(ficrespijb);
9679: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9680: i1= pow(2,cptcoveff);
1.218 brouard 9681: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9682: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9683: /* k=k+1; */
1.238 brouard 9684: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9685: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 ! brouard 9686: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9687: continue;
9688: fprintf(ficrespijb,"\n#****** ");
9689: for(j=1;j<=cptcoveff;j++)
9690: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9691: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9692: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9693: }
9694: fprintf(ficrespijb,"******\n");
9695: if(invalidvarcomb[k]){
9696: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9697: continue;
9698: }
9699:
9700: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9701: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9702: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9703: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9704: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9705:
9706: /* nhstepm=nhstepm*YEARM; aff par mois*/
9707:
9708: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9709: /* oldm=oldms;savm=savms; */
9710: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9711: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9712: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9713: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9714: for(i=1; i<=nlstate;i++)
9715: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9716: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9717: fprintf(ficrespijb,"\n");
1.238 brouard 9718: for (h=0; h<=nhstepm; h++){
9719: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9720: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9721: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9722: for(i=1; i<=nlstate;i++)
9723: for(j=1; j<=nlstate+ndeath;j++)
9724: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9725: fprintf(ficrespijb,"\n");
9726: }
9727: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9728: fprintf(ficrespijb,"\n");
9729: } /* end age deb */
9730: } /* end combination */
9731: } /* end nres */
1.218 brouard 9732: return 0;
9733: } /* hBijx */
1.217 brouard 9734:
1.180 brouard 9735:
1.136 brouard 9736: /***********************************************/
9737: /**************** Main Program *****************/
9738: /***********************************************/
9739:
9740: int main(int argc, char *argv[])
9741: {
9742: #ifdef GSL
9743: const gsl_multimin_fminimizer_type *T;
9744: size_t iteri = 0, it;
9745: int rval = GSL_CONTINUE;
9746: int status = GSL_SUCCESS;
9747: double ssval;
9748: #endif
9749: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9750: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9751: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9752: int jj, ll, li, lj, lk;
1.136 brouard 9753: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9754: int num_filled;
1.136 brouard 9755: int itimes;
9756: int NDIM=2;
9757: int vpopbased=0;
1.235 brouard 9758: int nres=0;
1.136 brouard 9759:
1.164 brouard 9760: char ca[32], cb[32];
1.136 brouard 9761: /* FILE *fichtm; *//* Html File */
9762: /* FILE *ficgp;*/ /*Gnuplot File */
9763: struct stat info;
1.191 brouard 9764: double agedeb=0.;
1.194 brouard 9765:
9766: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9767: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9768:
1.165 brouard 9769: double fret;
1.191 brouard 9770: double dum=0.; /* Dummy variable */
1.136 brouard 9771: double ***p3mat;
1.218 brouard 9772: /* double ***mobaverage; */
1.164 brouard 9773:
9774: char line[MAXLINE];
1.197 brouard 9775: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9776:
1.234 brouard 9777: char modeltemp[MAXLINE];
1.230 brouard 9778: char resultline[MAXLINE];
9779:
1.136 brouard 9780: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9781: char *tok, *val; /* pathtot */
1.136 brouard 9782: int firstobs=1, lastobs=10;
1.195 brouard 9783: int c, h , cpt, c2;
1.191 brouard 9784: int jl=0;
9785: int i1, j1, jk, stepsize=0;
1.194 brouard 9786: int count=0;
9787:
1.164 brouard 9788: int *tab;
1.136 brouard 9789: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9790: int backcast=0;
1.136 brouard 9791: int mobilav=0,popforecast=0;
1.191 brouard 9792: int hstepm=0, nhstepm=0;
1.136 brouard 9793: int agemortsup;
9794: float sumlpop=0.;
9795: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9796: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9797:
1.191 brouard 9798: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9799: double ftolpl=FTOL;
9800: double **prlim;
1.217 brouard 9801: double **bprlim;
1.136 brouard 9802: double ***param; /* Matrix of parameters */
1.251 brouard 9803: double ***paramstart; /* Matrix of starting parameter values */
9804: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 9805: double **matcov; /* Matrix of covariance */
1.203 brouard 9806: double **hess; /* Hessian matrix */
1.136 brouard 9807: double ***delti3; /* Scale */
9808: double *delti; /* Scale */
9809: double ***eij, ***vareij;
9810: double **varpl; /* Variances of prevalence limits by age */
9811: double *epj, vepp;
1.164 brouard 9812:
1.136 brouard 9813: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9814: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9815:
1.136 brouard 9816: double **ximort;
1.145 brouard 9817: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9818: int *dcwave;
9819:
1.164 brouard 9820: char z[1]="c";
1.136 brouard 9821:
9822: /*char *strt;*/
9823: char strtend[80];
1.126 brouard 9824:
1.164 brouard 9825:
1.126 brouard 9826: /* setlocale (LC_ALL, ""); */
9827: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9828: /* textdomain (PACKAGE); */
9829: /* setlocale (LC_CTYPE, ""); */
9830: /* setlocale (LC_MESSAGES, ""); */
9831:
9832: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9833: rstart_time = time(NULL);
9834: /* (void) gettimeofday(&start_time,&tzp);*/
9835: start_time = *localtime(&rstart_time);
1.126 brouard 9836: curr_time=start_time;
1.157 brouard 9837: /*tml = *localtime(&start_time.tm_sec);*/
9838: /* strcpy(strstart,asctime(&tml)); */
9839: strcpy(strstart,asctime(&start_time));
1.126 brouard 9840:
9841: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9842: /* tp.tm_sec = tp.tm_sec +86400; */
9843: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9844: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9845: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9846: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9847: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9848: /* strt=asctime(&tmg); */
9849: /* printf("Time(after) =%s",strstart); */
9850: /* (void) time (&time_value);
9851: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9852: * tm = *localtime(&time_value);
9853: * strstart=asctime(&tm);
9854: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9855: */
9856:
9857: nberr=0; /* Number of errors and warnings */
9858: nbwarn=0;
1.184 brouard 9859: #ifdef WIN32
9860: _getcwd(pathcd, size);
9861: #else
1.126 brouard 9862: getcwd(pathcd, size);
1.184 brouard 9863: #endif
1.191 brouard 9864: syscompilerinfo(0);
1.196 brouard 9865: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9866: if(argc <=1){
9867: printf("\nEnter the parameter file name: ");
1.205 brouard 9868: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9869: printf("ERROR Empty parameter file name\n");
9870: goto end;
9871: }
1.126 brouard 9872: i=strlen(pathr);
9873: if(pathr[i-1]=='\n')
9874: pathr[i-1]='\0';
1.156 brouard 9875: i=strlen(pathr);
1.205 brouard 9876: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9877: pathr[i-1]='\0';
1.205 brouard 9878: }
9879: i=strlen(pathr);
9880: if( i==0 ){
9881: printf("ERROR Empty parameter file name\n");
9882: goto end;
9883: }
9884: for (tok = pathr; tok != NULL; ){
1.126 brouard 9885: printf("Pathr |%s|\n",pathr);
9886: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9887: printf("val= |%s| pathr=%s\n",val,pathr);
9888: strcpy (pathtot, val);
9889: if(pathr[0] == '\0') break; /* Dirty */
9890: }
9891: }
9892: else{
9893: strcpy(pathtot,argv[1]);
9894: }
9895: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9896: /*cygwin_split_path(pathtot,path,optionfile);
9897: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9898: /* cutv(path,optionfile,pathtot,'\\');*/
9899:
9900: /* Split argv[0], imach program to get pathimach */
9901: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9902: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9903: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9904: /* strcpy(pathimach,argv[0]); */
9905: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9906: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9907: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9908: #ifdef WIN32
9909: _chdir(path); /* Can be a relative path */
9910: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9911: #else
1.126 brouard 9912: chdir(path); /* Can be a relative path */
1.184 brouard 9913: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9914: #endif
9915: printf("Current directory %s!\n",pathcd);
1.126 brouard 9916: strcpy(command,"mkdir ");
9917: strcat(command,optionfilefiname);
9918: if((outcmd=system(command)) != 0){
1.169 brouard 9919: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9920: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9921: /* fclose(ficlog); */
9922: /* exit(1); */
9923: }
9924: /* if((imk=mkdir(optionfilefiname))<0){ */
9925: /* perror("mkdir"); */
9926: /* } */
9927:
9928: /*-------- arguments in the command line --------*/
9929:
1.186 brouard 9930: /* Main Log file */
1.126 brouard 9931: strcat(filelog, optionfilefiname);
9932: strcat(filelog,".log"); /* */
9933: if((ficlog=fopen(filelog,"w"))==NULL) {
9934: printf("Problem with logfile %s\n",filelog);
9935: goto end;
9936: }
9937: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9938: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9939: fprintf(ficlog,"\nEnter the parameter file name: \n");
9940: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9941: path=%s \n\
9942: optionfile=%s\n\
9943: optionfilext=%s\n\
1.156 brouard 9944: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9945:
1.197 brouard 9946: syscompilerinfo(1);
1.167 brouard 9947:
1.126 brouard 9948: printf("Local time (at start):%s",strstart);
9949: fprintf(ficlog,"Local time (at start): %s",strstart);
9950: fflush(ficlog);
9951: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9952: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9953:
9954: /* */
9955: strcpy(fileres,"r");
9956: strcat(fileres, optionfilefiname);
1.201 brouard 9957: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9958: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9959: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9960:
1.186 brouard 9961: /* Main ---------arguments file --------*/
1.126 brouard 9962:
9963: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9964: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9965: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9966: fflush(ficlog);
1.149 brouard 9967: /* goto end; */
9968: exit(70);
1.126 brouard 9969: }
9970:
9971:
9972:
9973: strcpy(filereso,"o");
1.201 brouard 9974: strcat(filereso,fileresu);
1.126 brouard 9975: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9976: printf("Problem with Output resultfile: %s\n", filereso);
9977: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9978: fflush(ficlog);
9979: goto end;
9980: }
9981:
9982: /* Reads comments: lines beginning with '#' */
9983: numlinepar=0;
1.197 brouard 9984:
9985: /* First parameter line */
9986: while(fgets(line, MAXLINE, ficpar)) {
9987: /* If line starts with a # it is a comment */
9988: if (line[0] == '#') {
9989: numlinepar++;
9990: fputs(line,stdout);
9991: fputs(line,ficparo);
9992: fputs(line,ficlog);
9993: continue;
9994: }else
9995: break;
9996: }
9997: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9998: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9999: if (num_filled != 5) {
10000: printf("Should be 5 parameters\n");
10001: }
1.126 brouard 10002: numlinepar++;
1.197 brouard 10003: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10004: }
10005: /* Second parameter line */
10006: while(fgets(line, MAXLINE, ficpar)) {
10007: /* If line starts with a # it is a comment */
10008: if (line[0] == '#') {
10009: numlinepar++;
10010: fputs(line,stdout);
10011: fputs(line,ficparo);
10012: fputs(line,ficlog);
10013: continue;
10014: }else
10015: break;
10016: }
1.223 brouard 10017: 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", \
10018: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10019: if (num_filled != 11) {
10020: 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 10021: printf("but line=%s\n",line);
1.197 brouard 10022: }
1.223 brouard 10023: 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 10024: }
1.203 brouard 10025: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10026: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10027: /* Third parameter line */
10028: while(fgets(line, MAXLINE, ficpar)) {
10029: /* If line starts with a # it is a comment */
10030: if (line[0] == '#') {
10031: numlinepar++;
10032: fputs(line,stdout);
10033: fputs(line,ficparo);
10034: fputs(line,ficlog);
10035: continue;
10036: }else
10037: break;
10038: }
1.201 brouard 10039: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
10040: if (num_filled == 0)
10041: model[0]='\0';
10042: else if (num_filled != 1){
1.197 brouard 10043: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10044: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10045: model[0]='\0';
10046: goto end;
10047: }
10048: else{
10049: if (model[0]=='+'){
10050: for(i=1; i<=strlen(model);i++)
10051: modeltemp[i-1]=model[i];
1.201 brouard 10052: strcpy(model,modeltemp);
1.197 brouard 10053: }
10054: }
1.199 brouard 10055: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10056: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10057: }
10058: /* 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); */
10059: /* numlinepar=numlinepar+3; /\* In general *\/ */
10060: /* 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 10061: 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);
10062: 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 10063: fflush(ficlog);
1.190 brouard 10064: /* if(model[0]=='#'|| model[0]== '\0'){ */
10065: if(model[0]=='#'){
1.187 brouard 10066: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10067: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10068: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10069: if(mle != -1){
10070: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10071: exit(1);
10072: }
10073: }
1.126 brouard 10074: while((c=getc(ficpar))=='#' && c!= EOF){
10075: ungetc(c,ficpar);
10076: fgets(line, MAXLINE, ficpar);
10077: numlinepar++;
1.195 brouard 10078: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10079: z[0]=line[1];
10080: }
10081: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10082: fputs(line, stdout);
10083: //puts(line);
1.126 brouard 10084: fputs(line,ficparo);
10085: fputs(line,ficlog);
10086: }
10087: ungetc(c,ficpar);
10088:
10089:
1.145 brouard 10090: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10091: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10092: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10093: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10094: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10095: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10096: v1+v2*age+v2*v3 makes cptcovn = 3
10097: */
10098: if (strlen(model)>1)
1.187 brouard 10099: 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 10100: else
1.187 brouard 10101: ncovmodel=2; /* Constant and age */
1.133 brouard 10102: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10103: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10104: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10105: 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);
10106: 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);
10107: fflush(stdout);
10108: fclose (ficlog);
10109: goto end;
10110: }
1.126 brouard 10111: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10112: delti=delti3[1][1];
10113: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10114: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10115: /* We could also provide initial parameters values giving by simple logistic regression
10116: * only one way, that is without matrix product. We will have nlstate maximizations */
10117: /* for(i=1;i<nlstate;i++){ */
10118: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10119: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10120: /* } */
1.126 brouard 10121: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10122: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10123: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10124: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10125: fclose (ficparo);
10126: fclose (ficlog);
10127: goto end;
10128: exit(0);
1.220 brouard 10129: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10130: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10131: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10132: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10133: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10134: matcov=matrix(1,npar,1,npar);
1.203 brouard 10135: hess=matrix(1,npar,1,npar);
1.220 brouard 10136: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10137: /* Read guessed parameters */
1.126 brouard 10138: /* Reads comments: lines beginning with '#' */
10139: while((c=getc(ficpar))=='#' && c!= EOF){
10140: ungetc(c,ficpar);
10141: fgets(line, MAXLINE, ficpar);
10142: numlinepar++;
1.141 brouard 10143: fputs(line,stdout);
1.126 brouard 10144: fputs(line,ficparo);
10145: fputs(line,ficlog);
10146: }
10147: ungetc(c,ficpar);
10148:
10149: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10150: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10151: for(i=1; i <=nlstate; i++){
1.234 brouard 10152: j=0;
1.126 brouard 10153: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10154: if(jj==i) continue;
10155: j++;
10156: fscanf(ficpar,"%1d%1d",&i1,&j1);
10157: if ((i1 != i) || (j1 != jj)){
10158: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10159: It might be a problem of design; if ncovcol and the model are correct\n \
10160: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10161: exit(1);
10162: }
10163: fprintf(ficparo,"%1d%1d",i1,j1);
10164: if(mle==1)
10165: printf("%1d%1d",i,jj);
10166: fprintf(ficlog,"%1d%1d",i,jj);
10167: for(k=1; k<=ncovmodel;k++){
10168: fscanf(ficpar," %lf",¶m[i][j][k]);
10169: if(mle==1){
10170: printf(" %lf",param[i][j][k]);
10171: fprintf(ficlog," %lf",param[i][j][k]);
10172: }
10173: else
10174: fprintf(ficlog," %lf",param[i][j][k]);
10175: fprintf(ficparo," %lf",param[i][j][k]);
10176: }
10177: fscanf(ficpar,"\n");
10178: numlinepar++;
10179: if(mle==1)
10180: printf("\n");
10181: fprintf(ficlog,"\n");
10182: fprintf(ficparo,"\n");
1.126 brouard 10183: }
10184: }
10185: fflush(ficlog);
1.234 brouard 10186:
1.251 brouard 10187: /* Reads parameters values */
1.126 brouard 10188: p=param[1][1];
1.251 brouard 10189: pstart=paramstart[1][1];
1.126 brouard 10190:
10191: /* Reads comments: lines beginning with '#' */
10192: while((c=getc(ficpar))=='#' && c!= EOF){
10193: ungetc(c,ficpar);
10194: fgets(line, MAXLINE, ficpar);
10195: numlinepar++;
1.141 brouard 10196: fputs(line,stdout);
1.126 brouard 10197: fputs(line,ficparo);
10198: fputs(line,ficlog);
10199: }
10200: ungetc(c,ficpar);
10201:
10202: for(i=1; i <=nlstate; i++){
10203: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10204: fscanf(ficpar,"%1d%1d",&i1,&j1);
10205: if ( (i1-i) * (j1-j) != 0){
10206: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10207: exit(1);
10208: }
10209: printf("%1d%1d",i,j);
10210: fprintf(ficparo,"%1d%1d",i1,j1);
10211: fprintf(ficlog,"%1d%1d",i1,j1);
10212: for(k=1; k<=ncovmodel;k++){
10213: fscanf(ficpar,"%le",&delti3[i][j][k]);
10214: printf(" %le",delti3[i][j][k]);
10215: fprintf(ficparo," %le",delti3[i][j][k]);
10216: fprintf(ficlog," %le",delti3[i][j][k]);
10217: }
10218: fscanf(ficpar,"\n");
10219: numlinepar++;
10220: printf("\n");
10221: fprintf(ficparo,"\n");
10222: fprintf(ficlog,"\n");
1.126 brouard 10223: }
10224: }
10225: fflush(ficlog);
1.234 brouard 10226:
1.145 brouard 10227: /* Reads covariance matrix */
1.126 brouard 10228: delti=delti3[1][1];
1.220 brouard 10229:
10230:
1.126 brouard 10231: /* 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 10232:
1.126 brouard 10233: /* Reads comments: lines beginning with '#' */
10234: while((c=getc(ficpar))=='#' && c!= EOF){
10235: ungetc(c,ficpar);
10236: fgets(line, MAXLINE, ficpar);
10237: numlinepar++;
1.141 brouard 10238: fputs(line,stdout);
1.126 brouard 10239: fputs(line,ficparo);
10240: fputs(line,ficlog);
10241: }
10242: ungetc(c,ficpar);
1.220 brouard 10243:
1.126 brouard 10244: matcov=matrix(1,npar,1,npar);
1.203 brouard 10245: hess=matrix(1,npar,1,npar);
1.131 brouard 10246: for(i=1; i <=npar; i++)
10247: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10248:
1.194 brouard 10249: /* Scans npar lines */
1.126 brouard 10250: for(i=1; i <=npar; i++){
1.226 brouard 10251: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10252: if(count != 3){
1.226 brouard 10253: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10254: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10255: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10256: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10257: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10258: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10259: exit(1);
1.220 brouard 10260: }else{
1.226 brouard 10261: if(mle==1)
10262: printf("%1d%1d%d",i1,j1,jk);
10263: }
10264: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10265: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10266: for(j=1; j <=i; j++){
1.226 brouard 10267: fscanf(ficpar," %le",&matcov[i][j]);
10268: if(mle==1){
10269: printf(" %.5le",matcov[i][j]);
10270: }
10271: fprintf(ficlog," %.5le",matcov[i][j]);
10272: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10273: }
10274: fscanf(ficpar,"\n");
10275: numlinepar++;
10276: if(mle==1)
1.220 brouard 10277: printf("\n");
1.126 brouard 10278: fprintf(ficlog,"\n");
10279: fprintf(ficparo,"\n");
10280: }
1.194 brouard 10281: /* End of read covariance matrix npar lines */
1.126 brouard 10282: for(i=1; i <=npar; i++)
10283: for(j=i+1;j<=npar;j++)
1.226 brouard 10284: matcov[i][j]=matcov[j][i];
1.126 brouard 10285:
10286: if(mle==1)
10287: printf("\n");
10288: fprintf(ficlog,"\n");
10289:
10290: fflush(ficlog);
10291:
10292: /*-------- Rewriting parameter file ----------*/
10293: strcpy(rfileres,"r"); /* "Rparameterfile */
10294: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10295: strcat(rfileres,"."); /* */
10296: strcat(rfileres,optionfilext); /* Other files have txt extension */
10297: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10298: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10299: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10300: }
10301: fprintf(ficres,"#%s\n",version);
10302: } /* End of mle != -3 */
1.218 brouard 10303:
1.186 brouard 10304: /* Main data
10305: */
1.126 brouard 10306: n= lastobs;
10307: num=lvector(1,n);
10308: moisnais=vector(1,n);
10309: annais=vector(1,n);
10310: moisdc=vector(1,n);
10311: andc=vector(1,n);
1.220 brouard 10312: weight=vector(1,n);
1.126 brouard 10313: agedc=vector(1,n);
10314: cod=ivector(1,n);
1.220 brouard 10315: for(i=1;i<=n;i++){
1.234 brouard 10316: num[i]=0;
10317: moisnais[i]=0;
10318: annais[i]=0;
10319: moisdc[i]=0;
10320: andc[i]=0;
10321: agedc[i]=0;
10322: cod[i]=0;
10323: weight[i]=1.0; /* Equal weights, 1 by default */
10324: }
1.126 brouard 10325: mint=matrix(1,maxwav,1,n);
10326: anint=matrix(1,maxwav,1,n);
1.131 brouard 10327: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10328: tab=ivector(1,NCOVMAX);
1.144 brouard 10329: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10330: 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 10331:
1.136 brouard 10332: /* Reads data from file datafile */
10333: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10334: goto end;
10335:
10336: /* Calculation of the number of parameters from char model */
1.234 brouard 10337: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10338: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10339: k=3 V4 Tvar[k=3]= 4 (from V4)
10340: k=2 V1 Tvar[k=2]= 1 (from V1)
10341: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10342: */
10343:
10344: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10345: TvarsDind=ivector(1,NCOVMAX); /* */
10346: TvarsD=ivector(1,NCOVMAX); /* */
10347: TvarsQind=ivector(1,NCOVMAX); /* */
10348: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10349: TvarF=ivector(1,NCOVMAX); /* */
10350: TvarFind=ivector(1,NCOVMAX); /* */
10351: TvarV=ivector(1,NCOVMAX); /* */
10352: TvarVind=ivector(1,NCOVMAX); /* */
10353: TvarA=ivector(1,NCOVMAX); /* */
10354: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10355: TvarFD=ivector(1,NCOVMAX); /* */
10356: TvarFDind=ivector(1,NCOVMAX); /* */
10357: TvarFQ=ivector(1,NCOVMAX); /* */
10358: TvarFQind=ivector(1,NCOVMAX); /* */
10359: TvarVD=ivector(1,NCOVMAX); /* */
10360: TvarVDind=ivector(1,NCOVMAX); /* */
10361: TvarVQ=ivector(1,NCOVMAX); /* */
10362: TvarVQind=ivector(1,NCOVMAX); /* */
10363:
1.230 brouard 10364: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10365: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10366: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10367: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10368: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10369: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10370: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10371: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10372: */
10373: /* For model-covariate k tells which data-covariate to use but
10374: because this model-covariate is a construction we invent a new column
10375: ncovcol + k1
10376: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10377: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10378: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10379: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10380: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10381: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10382: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10383: */
1.145 brouard 10384: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10385: 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 10386: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10387: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10388: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10389: 4 covariates (3 plus signs)
10390: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10391: */
1.230 brouard 10392: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10393: * individual dummy, fixed or varying:
10394: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10395: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10396: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10397: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10398: * Tmodelind[1]@9={9,0,3,2,}*/
10399: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10400: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10401: * individual quantitative, fixed or varying:
10402: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10403: * 3, 1, 0, 0, 0, 0, 0, 0},
10404: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10405: /* Main decodemodel */
10406:
1.187 brouard 10407:
1.223 brouard 10408: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10409: goto end;
10410:
1.137 brouard 10411: if((double)(lastobs-imx)/(double)imx > 1.10){
10412: nbwarn++;
10413: 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);
10414: 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);
10415: }
1.136 brouard 10416: /* if(mle==1){*/
1.137 brouard 10417: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10418: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10419: }
10420:
10421: /*-calculation of age at interview from date of interview and age at death -*/
10422: agev=matrix(1,maxwav,1,imx);
10423:
10424: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10425: goto end;
10426:
1.126 brouard 10427:
1.136 brouard 10428: agegomp=(int)agemin;
10429: free_vector(moisnais,1,n);
10430: free_vector(annais,1,n);
1.126 brouard 10431: /* free_matrix(mint,1,maxwav,1,n);
10432: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10433: /* free_vector(moisdc,1,n); */
10434: /* free_vector(andc,1,n); */
1.145 brouard 10435: /* */
10436:
1.126 brouard 10437: wav=ivector(1,imx);
1.214 brouard 10438: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10439: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10440: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10441: 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.*/
10442: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10443: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10444:
10445: /* Concatenates waves */
1.214 brouard 10446: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10447: Death is a valid wave (if date is known).
10448: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10449: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10450: and mw[mi+1][i]. dh depends on stepm.
10451: */
10452:
1.126 brouard 10453: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10454: /* Concatenates waves */
1.145 brouard 10455:
1.215 brouard 10456: free_vector(moisdc,1,n);
10457: free_vector(andc,1,n);
10458:
1.126 brouard 10459: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10460: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10461: ncodemax[1]=1;
1.145 brouard 10462: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10463: cptcoveff=0;
1.220 brouard 10464: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10465: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10466: }
10467:
10468: ncovcombmax=pow(2,cptcoveff);
10469: invalidvarcomb=ivector(1, ncovcombmax);
10470: for(i=1;i<ncovcombmax;i++)
10471: invalidvarcomb[i]=0;
10472:
1.211 brouard 10473: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10474: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10475: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10476:
1.200 brouard 10477: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10478: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10479: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10480: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10481: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10482: * (currently 0 or 1) in the data.
10483: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10484: * corresponding modality (h,j).
10485: */
10486:
1.145 brouard 10487: h=0;
10488: /*if (cptcovn > 0) */
1.126 brouard 10489: m=pow(2,cptcoveff);
10490:
1.144 brouard 10491: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10492: * For k=4 covariates, h goes from 1 to m=2**k
10493: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10494: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10495: * h\k 1 2 3 4
1.143 brouard 10496: *______________________________
10497: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10498: * 2 2 1 1 1
10499: * 3 i=2 1 2 1 1
10500: * 4 2 2 1 1
10501: * 5 i=3 1 i=2 1 2 1
10502: * 6 2 1 2 1
10503: * 7 i=4 1 2 2 1
10504: * 8 2 2 2 1
1.197 brouard 10505: * 9 i=5 1 i=3 1 i=2 1 2
10506: * 10 2 1 1 2
10507: * 11 i=6 1 2 1 2
10508: * 12 2 2 1 2
10509: * 13 i=7 1 i=4 1 2 2
10510: * 14 2 1 2 2
10511: * 15 i=8 1 2 2 2
10512: * 16 2 2 2 2
1.143 brouard 10513: */
1.212 brouard 10514: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10515: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10516: * and the value of each covariate?
10517: * V1=1, V2=1, V3=2, V4=1 ?
10518: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10519: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10520: * In order to get the real value in the data, we use nbcode
10521: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10522: * We are keeping this crazy system in order to be able (in the future?)
10523: * to have more than 2 values (0 or 1) for a covariate.
10524: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10525: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10526: * bbbbbbbb
10527: * 76543210
10528: * h-1 00000101 (6-1=5)
1.219 brouard 10529: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10530: * &
10531: * 1 00000001 (1)
1.219 brouard 10532: * 00000000 = 1 & ((h-1) >> (k-1))
10533: * +1= 00000001 =1
1.211 brouard 10534: *
10535: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10536: * h' 1101 =2^3+2^2+0x2^1+2^0
10537: * >>k' 11
10538: * & 00000001
10539: * = 00000001
10540: * +1 = 00000010=2 = codtabm(14,3)
10541: * Reverse h=6 and m=16?
10542: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10543: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10544: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10545: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10546: * V3=decodtabm(14,3,2**4)=2
10547: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10548: *(h-1) >> (j-1) 0011 =13 >> 2
10549: * &1 000000001
10550: * = 000000001
10551: * +1= 000000010 =2
10552: * 2211
10553: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10554: * V3=2
1.220 brouard 10555: * codtabm and decodtabm are identical
1.211 brouard 10556: */
10557:
1.145 brouard 10558:
10559: free_ivector(Ndum,-1,NCOVMAX);
10560:
10561:
1.126 brouard 10562:
1.186 brouard 10563: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10564: strcpy(optionfilegnuplot,optionfilefiname);
10565: if(mle==-3)
1.201 brouard 10566: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10567: strcat(optionfilegnuplot,".gp");
10568:
10569: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10570: printf("Problem with file %s",optionfilegnuplot);
10571: }
10572: else{
1.204 brouard 10573: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10574: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10575: //fprintf(ficgp,"set missing 'NaNq'\n");
10576: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10577: }
10578: /* fclose(ficgp);*/
1.186 brouard 10579:
10580:
10581: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10582:
10583: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10584: if(mle==-3)
1.201 brouard 10585: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10586: strcat(optionfilehtm,".htm");
10587: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10588: printf("Problem with %s \n",optionfilehtm);
10589: exit(0);
1.126 brouard 10590: }
10591:
10592: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10593: strcat(optionfilehtmcov,"-cov.htm");
10594: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10595: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10596: }
10597: else{
10598: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10599: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10600: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10601: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10602: }
10603:
1.213 brouard 10604: 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 10605: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10606: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10607: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10608: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10609: \n\
10610: <hr size=\"2\" color=\"#EC5E5E\">\
10611: <ul><li><h4>Parameter files</h4>\n\
10612: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10613: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10614: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10615: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10616: - Date and time at start: %s</ul>\n",\
10617: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10618: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10619: fileres,fileres,\
10620: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10621: fflush(fichtm);
10622:
10623: strcpy(pathr,path);
10624: strcat(pathr,optionfilefiname);
1.184 brouard 10625: #ifdef WIN32
10626: _chdir(optionfilefiname); /* Move to directory named optionfile */
10627: #else
1.126 brouard 10628: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10629: #endif
10630:
1.126 brouard 10631:
1.220 brouard 10632: /* Calculates basic frequencies. Computes observed prevalence at single age
10633: and for any valid combination of covariates
1.126 brouard 10634: and prints on file fileres'p'. */
1.251 brouard 10635: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10636: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10637:
10638: fprintf(fichtm,"\n");
10639: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10640: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10641: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10642: imx,agemin,agemax,jmin,jmax,jmean);
10643: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10644: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10645: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10646: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10647: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10648:
1.126 brouard 10649: /* For Powell, parameters are in a vector p[] starting at p[1]
10650: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10651: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10652:
10653: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10654: /* For mortality only */
1.126 brouard 10655: if (mle==-3){
1.136 brouard 10656: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10657: for(i=1;i<=NDIM;i++)
10658: for(j=1;j<=NDIM;j++)
10659: ximort[i][j]=0.;
1.186 brouard 10660: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10661: cens=ivector(1,n);
10662: ageexmed=vector(1,n);
10663: agecens=vector(1,n);
10664: dcwave=ivector(1,n);
1.223 brouard 10665:
1.126 brouard 10666: for (i=1; i<=imx; i++){
10667: dcwave[i]=-1;
10668: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10669: if (s[m][i]>nlstate) {
10670: dcwave[i]=m;
10671: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10672: break;
10673: }
1.126 brouard 10674: }
1.226 brouard 10675:
1.126 brouard 10676: for (i=1; i<=imx; i++) {
10677: if (wav[i]>0){
1.226 brouard 10678: ageexmed[i]=agev[mw[1][i]][i];
10679: j=wav[i];
10680: agecens[i]=1.;
10681:
10682: if (ageexmed[i]> 1 && wav[i] > 0){
10683: agecens[i]=agev[mw[j][i]][i];
10684: cens[i]= 1;
10685: }else if (ageexmed[i]< 1)
10686: cens[i]= -1;
10687: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10688: cens[i]=0 ;
1.126 brouard 10689: }
10690: else cens[i]=-1;
10691: }
10692:
10693: for (i=1;i<=NDIM;i++) {
10694: for (j=1;j<=NDIM;j++)
1.226 brouard 10695: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10696: }
10697:
1.145 brouard 10698: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10699: /*printf("%lf %lf", p[1], p[2]);*/
10700:
10701:
1.136 brouard 10702: #ifdef GSL
10703: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10704: #else
1.126 brouard 10705: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10706: #endif
1.201 brouard 10707: strcpy(filerespow,"POW-MORT_");
10708: strcat(filerespow,fileresu);
1.126 brouard 10709: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10710: printf("Problem with resultfile: %s\n", filerespow);
10711: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10712: }
1.136 brouard 10713: #ifdef GSL
10714: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10715: #else
1.126 brouard 10716: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10717: #endif
1.126 brouard 10718: /* for (i=1;i<=nlstate;i++)
10719: for(j=1;j<=nlstate+ndeath;j++)
10720: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10721: */
10722: fprintf(ficrespow,"\n");
1.136 brouard 10723: #ifdef GSL
10724: /* gsl starts here */
10725: T = gsl_multimin_fminimizer_nmsimplex;
10726: gsl_multimin_fminimizer *sfm = NULL;
10727: gsl_vector *ss, *x;
10728: gsl_multimin_function minex_func;
10729:
10730: /* Initial vertex size vector */
10731: ss = gsl_vector_alloc (NDIM);
10732:
10733: if (ss == NULL){
10734: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10735: }
10736: /* Set all step sizes to 1 */
10737: gsl_vector_set_all (ss, 0.001);
10738:
10739: /* Starting point */
1.126 brouard 10740:
1.136 brouard 10741: x = gsl_vector_alloc (NDIM);
10742:
10743: if (x == NULL){
10744: gsl_vector_free(ss);
10745: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10746: }
10747:
10748: /* Initialize method and iterate */
10749: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10750: /* gsl_vector_set(x, 0, 0.0268); */
10751: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10752: gsl_vector_set(x, 0, p[1]);
10753: gsl_vector_set(x, 1, p[2]);
10754:
10755: minex_func.f = &gompertz_f;
10756: minex_func.n = NDIM;
10757: minex_func.params = (void *)&p; /* ??? */
10758:
10759: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10760: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10761:
10762: printf("Iterations beginning .....\n\n");
10763: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10764:
10765: iteri=0;
10766: while (rval == GSL_CONTINUE){
10767: iteri++;
10768: status = gsl_multimin_fminimizer_iterate(sfm);
10769:
10770: if (status) printf("error: %s\n", gsl_strerror (status));
10771: fflush(0);
10772:
10773: if (status)
10774: break;
10775:
10776: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10777: ssval = gsl_multimin_fminimizer_size (sfm);
10778:
10779: if (rval == GSL_SUCCESS)
10780: printf ("converged to a local maximum at\n");
10781:
10782: printf("%5d ", iteri);
10783: for (it = 0; it < NDIM; it++){
10784: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10785: }
10786: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10787: }
10788:
10789: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10790:
10791: gsl_vector_free(x); /* initial values */
10792: gsl_vector_free(ss); /* inital step size */
10793: for (it=0; it<NDIM; it++){
10794: p[it+1]=gsl_vector_get(sfm->x,it);
10795: fprintf(ficrespow," %.12lf", p[it]);
10796: }
10797: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10798: #endif
10799: #ifdef POWELL
10800: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10801: #endif
1.126 brouard 10802: fclose(ficrespow);
10803:
1.203 brouard 10804: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10805:
10806: for(i=1; i <=NDIM; i++)
10807: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10808: matcov[i][j]=matcov[j][i];
1.126 brouard 10809:
10810: printf("\nCovariance matrix\n ");
1.203 brouard 10811: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10812: for(i=1; i <=NDIM; i++) {
10813: for(j=1;j<=NDIM;j++){
1.220 brouard 10814: printf("%f ",matcov[i][j]);
10815: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10816: }
1.203 brouard 10817: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10818: }
10819:
10820: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10821: for (i=1;i<=NDIM;i++) {
1.126 brouard 10822: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10823: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10824: }
1.126 brouard 10825: lsurv=vector(1,AGESUP);
10826: lpop=vector(1,AGESUP);
10827: tpop=vector(1,AGESUP);
10828: lsurv[agegomp]=100000;
10829:
10830: for (k=agegomp;k<=AGESUP;k++) {
10831: agemortsup=k;
10832: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10833: }
10834:
10835: for (k=agegomp;k<agemortsup;k++)
10836: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10837:
10838: for (k=agegomp;k<agemortsup;k++){
10839: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10840: sumlpop=sumlpop+lpop[k];
10841: }
10842:
10843: tpop[agegomp]=sumlpop;
10844: for (k=agegomp;k<(agemortsup-3);k++){
10845: /* tpop[k+1]=2;*/
10846: tpop[k+1]=tpop[k]-lpop[k];
10847: }
10848:
10849:
10850: printf("\nAge lx qx dx Lx Tx e(x)\n");
10851: for (k=agegomp;k<(agemortsup-2);k++)
10852: 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]);
10853:
10854:
10855: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10856: ageminpar=50;
10857: agemaxpar=100;
1.194 brouard 10858: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10859: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10860: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10861: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10862: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10863: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10864: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10865: }else{
10866: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10867: 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 10868: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10869: }
1.201 brouard 10870: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10871: stepm, weightopt,\
10872: model,imx,p,matcov,agemortsup);
10873:
10874: free_vector(lsurv,1,AGESUP);
10875: free_vector(lpop,1,AGESUP);
10876: free_vector(tpop,1,AGESUP);
1.220 brouard 10877: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10878: free_ivector(cens,1,n);
10879: free_vector(agecens,1,n);
10880: free_ivector(dcwave,1,n);
1.220 brouard 10881: #ifdef GSL
1.136 brouard 10882: #endif
1.186 brouard 10883: } /* Endof if mle==-3 mortality only */
1.205 brouard 10884: /* Standard */
10885: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10886: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10887: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10888: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10889: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10890: for (k=1; k<=npar;k++)
10891: printf(" %d %8.5f",k,p[k]);
10892: printf("\n");
1.205 brouard 10893: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10894: /* mlikeli uses func not funcone */
1.247 brouard 10895: /* for(i=1;i<nlstate;i++){ */
10896: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10897: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10898: /* } */
1.205 brouard 10899: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10900: }
10901: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10902: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10903: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10904: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10905: }
10906: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10907: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10908: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10909: for (k=1; k<=npar;k++)
10910: printf(" %d %8.5f",k,p[k]);
10911: printf("\n");
10912:
10913: /*--------- results files --------------*/
1.224 brouard 10914: 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 10915:
10916:
10917: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10918: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10919: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10920: for(i=1,jk=1; i <=nlstate; i++){
10921: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10922: if (k != i) {
10923: printf("%d%d ",i,k);
10924: fprintf(ficlog,"%d%d ",i,k);
10925: fprintf(ficres,"%1d%1d ",i,k);
10926: for(j=1; j <=ncovmodel; j++){
10927: printf("%12.7f ",p[jk]);
10928: fprintf(ficlog,"%12.7f ",p[jk]);
10929: fprintf(ficres,"%12.7f ",p[jk]);
10930: jk++;
10931: }
10932: printf("\n");
10933: fprintf(ficlog,"\n");
10934: fprintf(ficres,"\n");
10935: }
1.126 brouard 10936: }
10937: }
1.203 brouard 10938: if(mle != 0){
10939: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10940: ftolhess=ftol; /* Usually correct */
1.203 brouard 10941: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10942: 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");
10943: 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");
10944: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10945: for(k=1; k <=(nlstate+ndeath); k++){
10946: if (k != i) {
10947: printf("%d%d ",i,k);
10948: fprintf(ficlog,"%d%d ",i,k);
10949: for(j=1; j <=ncovmodel; j++){
10950: 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]));
10951: 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]));
10952: jk++;
10953: }
10954: printf("\n");
10955: fprintf(ficlog,"\n");
10956: }
10957: }
1.193 brouard 10958: }
1.203 brouard 10959: } /* end of hesscov and Wald tests */
1.225 brouard 10960:
1.203 brouard 10961: /* */
1.126 brouard 10962: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10963: printf("# Scales (for hessian or gradient estimation)\n");
10964: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10965: for(i=1,jk=1; i <=nlstate; i++){
10966: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10967: if (j!=i) {
10968: fprintf(ficres,"%1d%1d",i,j);
10969: printf("%1d%1d",i,j);
10970: fprintf(ficlog,"%1d%1d",i,j);
10971: for(k=1; k<=ncovmodel;k++){
10972: printf(" %.5e",delti[jk]);
10973: fprintf(ficlog," %.5e",delti[jk]);
10974: fprintf(ficres," %.5e",delti[jk]);
10975: jk++;
10976: }
10977: printf("\n");
10978: fprintf(ficlog,"\n");
10979: fprintf(ficres,"\n");
10980: }
1.126 brouard 10981: }
10982: }
10983:
10984: 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 10985: if(mle >= 1) /* To big for the screen */
1.126 brouard 10986: 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");
10987: 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");
10988: /* # 121 Var(a12)\n\ */
10989: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10990: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10991: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10992: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10993: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10994: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10995: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10996:
10997:
10998: /* Just to have a covariance matrix which will be more understandable
10999: even is we still don't want to manage dictionary of variables
11000: */
11001: for(itimes=1;itimes<=2;itimes++){
11002: jj=0;
11003: for(i=1; i <=nlstate; i++){
1.225 brouard 11004: for(j=1; j <=nlstate+ndeath; j++){
11005: if(j==i) continue;
11006: for(k=1; k<=ncovmodel;k++){
11007: jj++;
11008: ca[0]= k+'a'-1;ca[1]='\0';
11009: if(itimes==1){
11010: if(mle>=1)
11011: printf("#%1d%1d%d",i,j,k);
11012: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11013: fprintf(ficres,"#%1d%1d%d",i,j,k);
11014: }else{
11015: if(mle>=1)
11016: printf("%1d%1d%d",i,j,k);
11017: fprintf(ficlog,"%1d%1d%d",i,j,k);
11018: fprintf(ficres,"%1d%1d%d",i,j,k);
11019: }
11020: ll=0;
11021: for(li=1;li <=nlstate; li++){
11022: for(lj=1;lj <=nlstate+ndeath; lj++){
11023: if(lj==li) continue;
11024: for(lk=1;lk<=ncovmodel;lk++){
11025: ll++;
11026: if(ll<=jj){
11027: cb[0]= lk +'a'-1;cb[1]='\0';
11028: if(ll<jj){
11029: if(itimes==1){
11030: if(mle>=1)
11031: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11032: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11033: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11034: }else{
11035: if(mle>=1)
11036: printf(" %.5e",matcov[jj][ll]);
11037: fprintf(ficlog," %.5e",matcov[jj][ll]);
11038: fprintf(ficres," %.5e",matcov[jj][ll]);
11039: }
11040: }else{
11041: if(itimes==1){
11042: if(mle>=1)
11043: printf(" Var(%s%1d%1d)",ca,i,j);
11044: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11045: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11046: }else{
11047: if(mle>=1)
11048: printf(" %.7e",matcov[jj][ll]);
11049: fprintf(ficlog," %.7e",matcov[jj][ll]);
11050: fprintf(ficres," %.7e",matcov[jj][ll]);
11051: }
11052: }
11053: }
11054: } /* end lk */
11055: } /* end lj */
11056: } /* end li */
11057: if(mle>=1)
11058: printf("\n");
11059: fprintf(ficlog,"\n");
11060: fprintf(ficres,"\n");
11061: numlinepar++;
11062: } /* end k*/
11063: } /*end j */
1.126 brouard 11064: } /* end i */
11065: } /* end itimes */
11066:
11067: fflush(ficlog);
11068: fflush(ficres);
1.225 brouard 11069: while(fgets(line, MAXLINE, ficpar)) {
11070: /* If line starts with a # it is a comment */
11071: if (line[0] == '#') {
11072: numlinepar++;
11073: fputs(line,stdout);
11074: fputs(line,ficparo);
11075: fputs(line,ficlog);
11076: continue;
11077: }else
11078: break;
11079: }
11080:
1.209 brouard 11081: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11082: /* ungetc(c,ficpar); */
11083: /* fgets(line, MAXLINE, ficpar); */
11084: /* fputs(line,stdout); */
11085: /* fputs(line,ficparo); */
11086: /* } */
11087: /* ungetc(c,ficpar); */
1.126 brouard 11088:
11089: estepm=0;
1.209 brouard 11090: 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 11091:
11092: if (num_filled != 6) {
11093: 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);
11094: 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);
11095: goto end;
11096: }
11097: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11098: }
11099: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11100: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11101:
1.209 brouard 11102: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11103: if (estepm==0 || estepm < stepm) estepm=stepm;
11104: if (fage <= 2) {
11105: bage = ageminpar;
11106: fage = agemaxpar;
11107: }
11108:
11109: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11110: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11111: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11112:
1.186 brouard 11113: /* Other stuffs, more or less useful */
1.126 brouard 11114: while((c=getc(ficpar))=='#' && c!= EOF){
11115: ungetc(c,ficpar);
11116: fgets(line, MAXLINE, ficpar);
1.141 brouard 11117: fputs(line,stdout);
1.126 brouard 11118: fputs(line,ficparo);
11119: }
11120: ungetc(c,ficpar);
11121:
11122: fscanf(ficpar,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav);
11123: 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);
11124: 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);
11125: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11126: 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);
11127:
11128: while((c=getc(ficpar))=='#' && c!= EOF){
11129: ungetc(c,ficpar);
11130: fgets(line, MAXLINE, ficpar);
1.141 brouard 11131: fputs(line,stdout);
1.126 brouard 11132: fputs(line,ficparo);
11133: }
11134: ungetc(c,ficpar);
11135:
11136:
11137: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11138: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11139:
11140: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 11141: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 11142: fprintf(ficparo,"pop_based=%d\n",popbased);
11143: fprintf(ficres,"pop_based=%d\n",popbased);
11144:
11145: while((c=getc(ficpar))=='#' && c!= EOF){
11146: ungetc(c,ficpar);
11147: fgets(line, MAXLINE, ficpar);
1.141 brouard 11148: fputs(line,stdout);
1.238 brouard 11149: fputs(line,ficres);
1.126 brouard 11150: fputs(line,ficparo);
11151: }
11152: ungetc(c,ficpar);
11153:
11154: fscanf(ficpar,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj);
11155: 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);
11156: 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);
11157: 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);
11158: 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);
11159: /* day and month of proj2 are not used but only year anproj2.*/
11160:
1.217 brouard 11161: while((c=getc(ficpar))=='#' && c!= EOF){
11162: ungetc(c,ficpar);
11163: fgets(line, MAXLINE, ficpar);
11164: fputs(line,stdout);
11165: fputs(line,ficparo);
1.238 brouard 11166: fputs(line,ficres);
1.217 brouard 11167: }
11168: ungetc(c,ficpar);
11169:
11170: fscanf(ficpar,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj);
1.223 brouard 11171: 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);
11172: 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);
11173: fprintf(ficres,"backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
1.217 brouard 11174: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 11175:
1.230 brouard 11176: /* Results */
1.235 brouard 11177: nresult=0;
1.230 brouard 11178: while(fgets(line, MAXLINE, ficpar)) {
11179: /* If line starts with a # it is a comment */
11180: if (line[0] == '#') {
11181: numlinepar++;
11182: fputs(line,stdout);
11183: fputs(line,ficparo);
11184: fputs(line,ficlog);
1.238 brouard 11185: fputs(line,ficres);
1.230 brouard 11186: continue;
11187: }else
11188: break;
11189: }
1.240 brouard 11190: if (!feof(ficpar))
1.230 brouard 11191: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11192: if (num_filled == 0){
1.230 brouard 11193: resultline[0]='\0';
1.253 ! brouard 11194: printf("Warning %d: no result line should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
1.240 brouard 11195: break;
11196: } else if (num_filled != 1){
1.253 ! brouard 11197: printf("ERROR %d: result line should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
1.230 brouard 11198: }
1.235 brouard 11199: nresult++; /* Sum of resultlines */
11200: printf("Result %d: result=%s\n",nresult, resultline);
11201: if(nresult > MAXRESULTLINES){
11202: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11203: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11204: goto end;
11205: }
11206: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11207: fprintf(ficparo,"result: %s\n",resultline);
11208: fprintf(ficres,"result: %s\n",resultline);
11209: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11210: while(fgets(line, MAXLINE, ficpar)) {
11211: /* If line starts with a # it is a comment */
11212: if (line[0] == '#') {
11213: numlinepar++;
11214: fputs(line,stdout);
11215: fputs(line,ficparo);
1.238 brouard 11216: fputs(line,ficres);
1.230 brouard 11217: fputs(line,ficlog);
11218: continue;
11219: }else
11220: break;
11221: }
11222: if (feof(ficpar))
11223: break;
11224: else{ /* Processess output results for this combination of covariate values */
11225: }
1.240 brouard 11226: } /* end while */
1.230 brouard 11227:
11228:
1.126 brouard 11229:
1.230 brouard 11230: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11231: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11232:
11233: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11234: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11235: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11236: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11237: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11238: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11239: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11240: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11241: }else{
1.218 brouard 11242: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11243: }
11244: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11245: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11246: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11247:
1.225 brouard 11248: /*------------ free_vector -------------*/
11249: /* chdir(path); */
1.220 brouard 11250:
1.215 brouard 11251: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11252: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11253: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11254: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11255: free_lvector(num,1,n);
11256: free_vector(agedc,1,n);
11257: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11258: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11259: fclose(ficparo);
11260: fclose(ficres);
1.220 brouard 11261:
11262:
1.186 brouard 11263: /* Other results (useful)*/
1.220 brouard 11264:
11265:
1.126 brouard 11266: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11267: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11268: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11269: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11270: fclose(ficrespl);
11271:
11272: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11273: /*#include "hpijx.h"*/
11274: hPijx(p, bage, fage);
1.145 brouard 11275: fclose(ficrespij);
1.227 brouard 11276:
1.220 brouard 11277: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11278: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11279: k=1;
1.126 brouard 11280: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11281:
1.219 brouard 11282: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11283: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11284: for(i=1;i<=AGESUP;i++)
1.219 brouard 11285: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11286: for(k=1;k<=ncovcombmax;k++)
11287: probs[i][j][k]=0.;
1.219 brouard 11288: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11289: if (mobilav!=0 ||mobilavproj !=0 ) {
11290: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11291: for(i=1;i<=AGESUP;i++)
11292: for(j=1;j<=nlstate;j++)
11293: for(k=1;k<=ncovcombmax;k++)
11294: mobaverages[i][j][k]=0.;
1.219 brouard 11295: mobaverage=mobaverages;
11296: if (mobilav!=0) {
1.235 brouard 11297: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11298: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11299: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11300: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11301: }
1.219 brouard 11302: }
11303: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11304: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11305: else if (mobilavproj !=0) {
1.235 brouard 11306: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11307: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11308: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11309: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11310: }
1.219 brouard 11311: }
11312: }/* end if moving average */
1.227 brouard 11313:
1.126 brouard 11314: /*---------- Forecasting ------------------*/
11315: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11316: if(prevfcast==1){
11317: /* if(stepm ==1){*/
1.225 brouard 11318: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11319: }
1.217 brouard 11320: if(backcast==1){
1.219 brouard 11321: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11322: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11323: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11324:
11325: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11326:
11327: bprlim=matrix(1,nlstate,1,nlstate);
11328: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11329: fclose(ficresplb);
11330:
1.222 brouard 11331: hBijx(p, bage, fage, mobaverage);
11332: fclose(ficrespijb);
1.219 brouard 11333: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11334:
11335: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11336: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11337: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11338: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11339: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11340: }
1.217 brouard 11341:
1.186 brouard 11342:
11343: /* ------ Other prevalence ratios------------ */
1.126 brouard 11344:
1.215 brouard 11345: free_ivector(wav,1,imx);
11346: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11347: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11348: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11349:
11350:
1.127 brouard 11351: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11352:
1.201 brouard 11353: strcpy(filerese,"E_");
11354: strcat(filerese,fileresu);
1.126 brouard 11355: if((ficreseij=fopen(filerese,"w"))==NULL) {
11356: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11357: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11358: }
1.208 brouard 11359: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11360: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11361:
11362: pstamp(ficreseij);
1.219 brouard 11363:
1.235 brouard 11364: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11365: if (cptcovn < 1){i1=1;}
11366:
11367: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11368: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 ! brouard 11369: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11370: continue;
1.219 brouard 11371: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11372: printf("\n#****** ");
1.225 brouard 11373: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11374: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11375: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11376: }
11377: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11378: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11379: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11380: }
11381: fprintf(ficreseij,"******\n");
1.235 brouard 11382: printf("******\n");
1.219 brouard 11383:
11384: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11385: oldm=oldms;savm=savms;
1.235 brouard 11386: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11387:
1.219 brouard 11388: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11389: }
11390: fclose(ficreseij);
1.208 brouard 11391: printf("done evsij\n");fflush(stdout);
11392: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11393:
1.227 brouard 11394: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11395:
11396:
1.201 brouard 11397: strcpy(filerest,"T_");
11398: strcat(filerest,fileresu);
1.127 brouard 11399: if((ficrest=fopen(filerest,"w"))==NULL) {
11400: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11401: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11402: }
1.208 brouard 11403: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11404: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11405:
1.126 brouard 11406:
1.201 brouard 11407: strcpy(fileresstde,"STDE_");
11408: strcat(fileresstde,fileresu);
1.126 brouard 11409: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11410: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11411: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11412: }
1.227 brouard 11413: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11414: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11415:
1.201 brouard 11416: strcpy(filerescve,"CVE_");
11417: strcat(filerescve,fileresu);
1.126 brouard 11418: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11419: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11420: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11421: }
1.227 brouard 11422: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11423: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11424:
1.201 brouard 11425: strcpy(fileresv,"V_");
11426: strcat(fileresv,fileresu);
1.126 brouard 11427: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11428: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11429: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11430: }
1.227 brouard 11431: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11432: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11433:
1.145 brouard 11434: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11435: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11436:
1.235 brouard 11437: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11438: if (cptcovn < 1){i1=1;}
11439:
11440: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11441: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 ! brouard 11442: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11443: continue;
1.242 brouard 11444: printf("\n#****** Result for:");
11445: fprintf(ficrest,"\n#****** Result for:");
11446: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11447: for(j=1;j<=cptcoveff;j++){
11448: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11449: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11450: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11451: }
1.235 brouard 11452: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11453: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11454: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11455: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11456: }
1.208 brouard 11457: fprintf(ficrest,"******\n");
1.227 brouard 11458: fprintf(ficlog,"******\n");
11459: printf("******\n");
1.208 brouard 11460:
11461: fprintf(ficresstdeij,"\n#****** ");
11462: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11463: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11464: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11465: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11466: }
1.235 brouard 11467: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11468: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11469: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11470: }
1.208 brouard 11471: fprintf(ficresstdeij,"******\n");
11472: fprintf(ficrescveij,"******\n");
11473:
11474: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11475: /* pstamp(ficresvij); */
1.225 brouard 11476: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11477: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11478: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11479: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11480: }
1.208 brouard 11481: fprintf(ficresvij,"******\n");
11482:
11483: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11484: oldm=oldms;savm=savms;
1.235 brouard 11485: printf(" cvevsij ");
11486: fprintf(ficlog, " cvevsij ");
11487: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11488: printf(" end cvevsij \n ");
11489: fprintf(ficlog, " end cvevsij \n ");
11490:
11491: /*
11492: */
11493: /* goto endfree; */
11494:
11495: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11496: pstamp(ficrest);
11497:
11498:
11499: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11500: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11501: cptcod= 0; /* To be deleted */
11502: printf("varevsij vpopbased=%d \n",vpopbased);
11503: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11504: 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 11505: 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 ");
11506: if(vpopbased==1)
11507: 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);
11508: else
11509: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11510: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11511: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11512: fprintf(ficrest,"\n");
11513: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11514: epj=vector(1,nlstate+1);
11515: printf("Computing age specific period (stable) prevalences in each health state \n");
11516: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11517: for(age=bage; age <=fage ;age++){
1.235 brouard 11518: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11519: if (vpopbased==1) {
11520: if(mobilav ==0){
11521: for(i=1; i<=nlstate;i++)
11522: prlim[i][i]=probs[(int)age][i][k];
11523: }else{ /* mobilav */
11524: for(i=1; i<=nlstate;i++)
11525: prlim[i][i]=mobaverage[(int)age][i][k];
11526: }
11527: }
1.219 brouard 11528:
1.227 brouard 11529: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11530: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11531: /* printf(" age %4.0f ",age); */
11532: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11533: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11534: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11535: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11536: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11537: }
11538: epj[nlstate+1] +=epj[j];
11539: }
11540: /* printf(" age %4.0f \n",age); */
1.219 brouard 11541:
1.227 brouard 11542: for(i=1, vepp=0.;i <=nlstate;i++)
11543: for(j=1;j <=nlstate;j++)
11544: vepp += vareij[i][j][(int)age];
11545: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11546: for(j=1;j <=nlstate;j++){
11547: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11548: }
11549: fprintf(ficrest,"\n");
11550: }
1.208 brouard 11551: } /* End vpopbased */
11552: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11553: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11554: free_vector(epj,1,nlstate+1);
1.235 brouard 11555: printf("done selection\n");fflush(stdout);
11556: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11557:
1.145 brouard 11558: /*}*/
1.235 brouard 11559: } /* End k selection */
1.227 brouard 11560:
11561: printf("done State-specific expectancies\n");fflush(stdout);
11562: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11563:
1.126 brouard 11564: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11565:
1.201 brouard 11566: strcpy(fileresvpl,"VPL_");
11567: strcat(fileresvpl,fileresu);
1.126 brouard 11568: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11569: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11570: exit(0);
11571: }
1.208 brouard 11572: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11573: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11574:
1.145 brouard 11575: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11576: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11577:
1.235 brouard 11578: i1=pow(2,cptcoveff);
11579: if (cptcovn < 1){i1=1;}
11580:
11581: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11582: for(k=1; k<=i1;k++){
1.253 ! brouard 11583: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11584: continue;
1.227 brouard 11585: fprintf(ficresvpl,"\n#****** ");
11586: printf("\n#****** ");
11587: fprintf(ficlog,"\n#****** ");
11588: for(j=1;j<=cptcoveff;j++) {
11589: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11590: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11591: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11592: }
1.235 brouard 11593: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11594: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11595: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11596: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11597: }
1.227 brouard 11598: fprintf(ficresvpl,"******\n");
11599: printf("******\n");
11600: fprintf(ficlog,"******\n");
11601:
11602: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11603: oldm=oldms;savm=savms;
1.235 brouard 11604: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11605: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11606: /*}*/
1.126 brouard 11607: }
1.227 brouard 11608:
1.126 brouard 11609: fclose(ficresvpl);
1.208 brouard 11610: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11611: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11612:
11613: free_vector(weight,1,n);
11614: free_imatrix(Tvard,1,NCOVMAX,1,2);
11615: free_imatrix(s,1,maxwav+1,1,n);
11616: free_matrix(anint,1,maxwav,1,n);
11617: free_matrix(mint,1,maxwav,1,n);
11618: free_ivector(cod,1,n);
11619: free_ivector(tab,1,NCOVMAX);
11620: fclose(ficresstdeij);
11621: fclose(ficrescveij);
11622: fclose(ficresvij);
11623: fclose(ficrest);
11624: fclose(ficpar);
11625:
11626:
1.126 brouard 11627: /*---------- End : free ----------------*/
1.219 brouard 11628: if (mobilav!=0 ||mobilavproj !=0)
11629: 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 11630: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11631: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11632: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11633: } /* mle==-3 arrives here for freeing */
1.227 brouard 11634: /* endfree:*/
11635: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11636: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11637: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11638: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11639: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11640: free_matrix(coqvar,1,maxwav,1,n);
11641: free_matrix(covar,0,NCOVMAX,1,n);
11642: free_matrix(matcov,1,npar,1,npar);
11643: free_matrix(hess,1,npar,1,npar);
11644: /*free_vector(delti,1,npar);*/
11645: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11646: free_matrix(agev,1,maxwav,1,imx);
11647: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11648:
11649: free_ivector(ncodemax,1,NCOVMAX);
11650: free_ivector(ncodemaxwundef,1,NCOVMAX);
11651: free_ivector(Dummy,-1,NCOVMAX);
11652: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11653: free_ivector(DummyV,1,NCOVMAX);
11654: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11655: free_ivector(Typevar,-1,NCOVMAX);
11656: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11657: free_ivector(TvarsQ,1,NCOVMAX);
11658: free_ivector(TvarsQind,1,NCOVMAX);
11659: free_ivector(TvarsD,1,NCOVMAX);
11660: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11661: free_ivector(TvarFD,1,NCOVMAX);
11662: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11663: free_ivector(TvarF,1,NCOVMAX);
11664: free_ivector(TvarFind,1,NCOVMAX);
11665: free_ivector(TvarV,1,NCOVMAX);
11666: free_ivector(TvarVind,1,NCOVMAX);
11667: free_ivector(TvarA,1,NCOVMAX);
11668: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11669: free_ivector(TvarFQ,1,NCOVMAX);
11670: free_ivector(TvarFQind,1,NCOVMAX);
11671: free_ivector(TvarVD,1,NCOVMAX);
11672: free_ivector(TvarVDind,1,NCOVMAX);
11673: free_ivector(TvarVQ,1,NCOVMAX);
11674: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11675: free_ivector(Tvarsel,1,NCOVMAX);
11676: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11677: free_ivector(Tposprod,1,NCOVMAX);
11678: free_ivector(Tprod,1,NCOVMAX);
11679: free_ivector(Tvaraff,1,NCOVMAX);
11680: free_ivector(invalidvarcomb,1,ncovcombmax);
11681: free_ivector(Tage,1,NCOVMAX);
11682: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11683: free_ivector(TmodelInvind,1,NCOVMAX);
11684: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11685:
11686: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11687: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11688: fflush(fichtm);
11689: fflush(ficgp);
11690:
1.227 brouard 11691:
1.126 brouard 11692: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11693: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11694: 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 11695: }else{
11696: printf("End of Imach\n");
11697: fprintf(ficlog,"End of Imach\n");
11698: }
11699: printf("See log file on %s\n",filelog);
11700: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11701: /*(void) gettimeofday(&end_time,&tzp);*/
11702: rend_time = time(NULL);
11703: end_time = *localtime(&rend_time);
11704: /* tml = *localtime(&end_time.tm_sec); */
11705: strcpy(strtend,asctime(&end_time));
1.126 brouard 11706: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11707: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11708: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11709:
1.157 brouard 11710: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11711: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11712: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11713: /* printf("Total time was %d uSec.\n", total_usecs);*/
11714: /* if(fileappend(fichtm,optionfilehtm)){ */
11715: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11716: fclose(fichtm);
11717: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11718: fclose(fichtmcov);
11719: fclose(ficgp);
11720: fclose(ficlog);
11721: /*------ End -----------*/
1.227 brouard 11722:
11723:
11724: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11725: #ifdef WIN32
1.227 brouard 11726: if (_chdir(pathcd) != 0)
11727: printf("Can't move to directory %s!\n",path);
11728: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11729: #else
1.227 brouard 11730: if(chdir(pathcd) != 0)
11731: printf("Can't move to directory %s!\n", path);
11732: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11733: #endif
1.126 brouard 11734: printf("Current directory %s!\n",pathcd);
11735: /*strcat(plotcmd,CHARSEPARATOR);*/
11736: sprintf(plotcmd,"gnuplot");
1.157 brouard 11737: #ifdef _WIN32
1.126 brouard 11738: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11739: #endif
11740: if(!stat(plotcmd,&info)){
1.158 brouard 11741: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11742: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11743: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11744: }else
11745: strcpy(pplotcmd,plotcmd);
1.157 brouard 11746: #ifdef __unix
1.126 brouard 11747: strcpy(plotcmd,GNUPLOTPROGRAM);
11748: if(!stat(plotcmd,&info)){
1.158 brouard 11749: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11750: }else
11751: strcpy(pplotcmd,plotcmd);
11752: #endif
11753: }else
11754: strcpy(pplotcmd,plotcmd);
11755:
11756: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11757: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11758:
1.126 brouard 11759: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11760: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11761: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11762: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11763: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11764: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11765: }
1.158 brouard 11766: printf(" Successful, please wait...");
1.126 brouard 11767: while (z[0] != 'q') {
11768: /* chdir(path); */
1.154 brouard 11769: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11770: scanf("%s",z);
11771: /* if (z[0] == 'c') system("./imach"); */
11772: if (z[0] == 'e') {
1.158 brouard 11773: #ifdef __APPLE__
1.152 brouard 11774: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11775: #elif __linux
11776: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11777: #else
1.152 brouard 11778: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11779: #endif
11780: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11781: system(pplotcmd);
1.126 brouard 11782: }
11783: else if (z[0] == 'g') system(plotcmd);
11784: else if (z[0] == 'q') exit(0);
11785: }
1.227 brouard 11786: end:
1.126 brouard 11787: while (z[0] != 'q') {
1.195 brouard 11788: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11789: scanf("%s",z);
11790: }
11791: }
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