Annotation of imach/src/imach.c, revision 1.250
1.250 ! brouard 1: /* $Id: imach.c,v 1.249 2016/09/07 17:14:18 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.250 ! brouard 4: Revision 1.249 2016/09/07 17:14:18 brouard
! 5: Summary: Starting values from frequencies
! 6:
1.249 brouard 7: Revision 1.248 2016/09/07 14:10:18 brouard
8: *** empty log message ***
9:
1.248 brouard 10: Revision 1.247 2016/09/02 11:11:21 brouard
11: *** empty log message ***
12:
1.247 brouard 13: Revision 1.246 2016/09/02 08:49:22 brouard
14: *** empty log message ***
15:
1.246 brouard 16: Revision 1.245 2016/09/02 07:25:01 brouard
17: *** empty log message ***
18:
1.245 brouard 19: Revision 1.244 2016/09/02 07:17:34 brouard
20: *** empty log message ***
21:
1.244 brouard 22: Revision 1.243 2016/09/02 06:45:35 brouard
23: *** empty log message ***
24:
1.243 brouard 25: Revision 1.242 2016/08/30 15:01:20 brouard
26: Summary: Fixing a lots
27:
1.242 brouard 28: Revision 1.241 2016/08/29 17:17:25 brouard
29: Summary: gnuplot problem in Back projection to fix
30:
1.241 brouard 31: Revision 1.240 2016/08/29 07:53:18 brouard
32: Summary: Better
33:
1.240 brouard 34: Revision 1.239 2016/08/26 15:51:03 brouard
35: Summary: Improvement in Powell output in order to copy and paste
36:
37: Author:
38:
1.239 brouard 39: Revision 1.238 2016/08/26 14:23:35 brouard
40: Summary: Starting tests of 0.99
41:
1.238 brouard 42: Revision 1.237 2016/08/26 09:20:19 brouard
43: Summary: to valgrind
44:
1.237 brouard 45: Revision 1.236 2016/08/25 10:50:18 brouard
46: *** empty log message ***
47:
1.236 brouard 48: Revision 1.235 2016/08/25 06:59:23 brouard
49: *** empty log message ***
50:
1.235 brouard 51: Revision 1.234 2016/08/23 16:51:20 brouard
52: *** empty log message ***
53:
1.234 brouard 54: Revision 1.233 2016/08/23 07:40:50 brouard
55: Summary: not working
56:
1.233 brouard 57: Revision 1.232 2016/08/22 14:20:21 brouard
58: Summary: not working
59:
1.232 brouard 60: Revision 1.231 2016/08/22 07:17:15 brouard
61: Summary: not working
62:
1.231 brouard 63: Revision 1.230 2016/08/22 06:55:53 brouard
64: Summary: Not working
65:
1.230 brouard 66: Revision 1.229 2016/07/23 09:45:53 brouard
67: Summary: Completing for func too
68:
1.229 brouard 69: Revision 1.228 2016/07/22 17:45:30 brouard
70: Summary: Fixing some arrays, still debugging
71:
1.227 brouard 72: Revision 1.226 2016/07/12 18:42:34 brouard
73: Summary: temp
74:
1.226 brouard 75: Revision 1.225 2016/07/12 08:40:03 brouard
76: Summary: saving but not running
77:
1.225 brouard 78: Revision 1.224 2016/07/01 13:16:01 brouard
79: Summary: Fixes
80:
1.224 brouard 81: Revision 1.223 2016/02/19 09:23:35 brouard
82: Summary: temporary
83:
1.223 brouard 84: Revision 1.222 2016/02/17 08:14:50 brouard
85: Summary: Probably last 0.98 stable version 0.98r6
86:
1.222 brouard 87: Revision 1.221 2016/02/15 23:35:36 brouard
88: Summary: minor bug
89:
1.220 brouard 90: Revision 1.219 2016/02/15 00:48:12 brouard
91: *** empty log message ***
92:
1.219 brouard 93: Revision 1.218 2016/02/12 11:29:23 brouard
94: Summary: 0.99 Back projections
95:
1.218 brouard 96: Revision 1.217 2015/12/23 17:18:31 brouard
97: Summary: Experimental backcast
98:
1.217 brouard 99: Revision 1.216 2015/12/18 17:32:11 brouard
100: Summary: 0.98r4 Warning and status=-2
101:
102: Version 0.98r4 is now:
103: - displaying an error when status is -1, date of interview unknown and date of death known;
104: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
105: Older changes concerning s=-2, dating from 2005 have been supersed.
106:
1.216 brouard 107: Revision 1.215 2015/12/16 08:52:24 brouard
108: Summary: 0.98r4 working
109:
1.215 brouard 110: Revision 1.214 2015/12/16 06:57:54 brouard
111: Summary: temporary not working
112:
1.214 brouard 113: Revision 1.213 2015/12/11 18:22:17 brouard
114: Summary: 0.98r4
115:
1.213 brouard 116: Revision 1.212 2015/11/21 12:47:24 brouard
117: Summary: minor typo
118:
1.212 brouard 119: Revision 1.211 2015/11/21 12:41:11 brouard
120: Summary: 0.98r3 with some graph of projected cross-sectional
121:
122: Author: Nicolas Brouard
123:
1.211 brouard 124: Revision 1.210 2015/11/18 17:41:20 brouard
125: Summary: Start working on projected prevalences
126:
1.210 brouard 127: Revision 1.209 2015/11/17 22:12:03 brouard
128: Summary: Adding ftolpl parameter
129: Author: N Brouard
130:
131: We had difficulties to get smoothed confidence intervals. It was due
132: to the period prevalence which wasn't computed accurately. The inner
133: parameter ftolpl is now an outer parameter of the .imach parameter
134: file after estepm. If ftolpl is small 1.e-4 and estepm too,
135: computation are long.
136:
1.209 brouard 137: Revision 1.208 2015/11/17 14:31:57 brouard
138: Summary: temporary
139:
1.208 brouard 140: Revision 1.207 2015/10/27 17:36:57 brouard
141: *** empty log message ***
142:
1.207 brouard 143: Revision 1.206 2015/10/24 07:14:11 brouard
144: *** empty log message ***
145:
1.206 brouard 146: Revision 1.205 2015/10/23 15:50:53 brouard
147: Summary: 0.98r3 some clarification for graphs on likelihood contributions
148:
1.205 brouard 149: Revision 1.204 2015/10/01 16:20:26 brouard
150: Summary: Some new graphs of contribution to likelihood
151:
1.204 brouard 152: Revision 1.203 2015/09/30 17:45:14 brouard
153: Summary: looking at better estimation of the hessian
154:
155: Also a better criteria for convergence to the period prevalence And
156: therefore adding the number of years needed to converge. (The
157: prevalence in any alive state shold sum to one
158:
1.203 brouard 159: Revision 1.202 2015/09/22 19:45:16 brouard
160: Summary: Adding some overall graph on contribution to likelihood. Might change
161:
1.202 brouard 162: Revision 1.201 2015/09/15 17:34:58 brouard
163: Summary: 0.98r0
164:
165: - Some new graphs like suvival functions
166: - Some bugs fixed like model=1+age+V2.
167:
1.201 brouard 168: Revision 1.200 2015/09/09 16:53:55 brouard
169: Summary: Big bug thanks to Flavia
170:
171: Even model=1+age+V2. did not work anymore
172:
1.200 brouard 173: Revision 1.199 2015/09/07 14:09:23 brouard
174: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
175:
1.199 brouard 176: Revision 1.198 2015/09/03 07:14:39 brouard
177: Summary: 0.98q5 Flavia
178:
1.198 brouard 179: Revision 1.197 2015/09/01 18:24:39 brouard
180: *** empty log message ***
181:
1.197 brouard 182: Revision 1.196 2015/08/18 23:17:52 brouard
183: Summary: 0.98q5
184:
1.196 brouard 185: Revision 1.195 2015/08/18 16:28:39 brouard
186: Summary: Adding a hack for testing purpose
187:
188: After reading the title, ftol and model lines, if the comment line has
189: a q, starting with #q, the answer at the end of the run is quit. It
190: permits to run test files in batch with ctest. The former workaround was
191: $ echo q | imach foo.imach
192:
1.195 brouard 193: Revision 1.194 2015/08/18 13:32:00 brouard
194: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
195:
1.194 brouard 196: Revision 1.193 2015/08/04 07:17:42 brouard
197: Summary: 0.98q4
198:
1.193 brouard 199: Revision 1.192 2015/07/16 16:49:02 brouard
200: Summary: Fixing some outputs
201:
1.192 brouard 202: Revision 1.191 2015/07/14 10:00:33 brouard
203: Summary: Some fixes
204:
1.191 brouard 205: Revision 1.190 2015/05/05 08:51:13 brouard
206: Summary: Adding digits in output parameters (7 digits instead of 6)
207:
208: Fix 1+age+.
209:
1.190 brouard 210: Revision 1.189 2015/04/30 14:45:16 brouard
211: Summary: 0.98q2
212:
1.189 brouard 213: Revision 1.188 2015/04/30 08:27:53 brouard
214: *** empty log message ***
215:
1.188 brouard 216: Revision 1.187 2015/04/29 09:11:15 brouard
217: *** empty log message ***
218:
1.187 brouard 219: Revision 1.186 2015/04/23 12:01:52 brouard
220: Summary: V1*age is working now, version 0.98q1
221:
222: Some codes had been disabled in order to simplify and Vn*age was
223: working in the optimization phase, ie, giving correct MLE parameters,
224: but, as usual, outputs were not correct and program core dumped.
225:
1.186 brouard 226: Revision 1.185 2015/03/11 13:26:42 brouard
227: Summary: Inclusion of compile and links command line for Intel Compiler
228:
1.185 brouard 229: Revision 1.184 2015/03/11 11:52:39 brouard
230: Summary: Back from Windows 8. Intel Compiler
231:
1.184 brouard 232: Revision 1.183 2015/03/10 20:34:32 brouard
233: Summary: 0.98q0, trying with directest, mnbrak fixed
234:
235: We use directest instead of original Powell test; probably no
236: incidence on the results, but better justifications;
237: We fixed Numerical Recipes mnbrak routine which was wrong and gave
238: wrong results.
239:
1.183 brouard 240: Revision 1.182 2015/02/12 08:19:57 brouard
241: Summary: Trying to keep directest which seems simpler and more general
242: Author: Nicolas Brouard
243:
1.182 brouard 244: Revision 1.181 2015/02/11 23:22:24 brouard
245: Summary: Comments on Powell added
246:
247: Author:
248:
1.181 brouard 249: Revision 1.180 2015/02/11 17:33:45 brouard
250: Summary: Finishing move from main to function (hpijx and prevalence_limit)
251:
1.180 brouard 252: Revision 1.179 2015/01/04 09:57:06 brouard
253: Summary: back to OS/X
254:
1.179 brouard 255: Revision 1.178 2015/01/04 09:35:48 brouard
256: *** empty log message ***
257:
1.178 brouard 258: Revision 1.177 2015/01/03 18:40:56 brouard
259: Summary: Still testing ilc32 on OSX
260:
1.177 brouard 261: Revision 1.176 2015/01/03 16:45:04 brouard
262: *** empty log message ***
263:
1.176 brouard 264: Revision 1.175 2015/01/03 16:33:42 brouard
265: *** empty log message ***
266:
1.175 brouard 267: Revision 1.174 2015/01/03 16:15:49 brouard
268: Summary: Still in cross-compilation
269:
1.174 brouard 270: Revision 1.173 2015/01/03 12:06:26 brouard
271: Summary: trying to detect cross-compilation
272:
1.173 brouard 273: Revision 1.172 2014/12/27 12:07:47 brouard
274: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
275:
1.172 brouard 276: Revision 1.171 2014/12/23 13:26:59 brouard
277: Summary: Back from Visual C
278:
279: Still problem with utsname.h on Windows
280:
1.171 brouard 281: Revision 1.170 2014/12/23 11:17:12 brouard
282: Summary: Cleaning some \%% back to %%
283:
284: The escape was mandatory for a specific compiler (which one?), but too many warnings.
285:
1.170 brouard 286: Revision 1.169 2014/12/22 23:08:31 brouard
287: Summary: 0.98p
288:
289: Outputs some informations on compiler used, OS etc. Testing on different platforms.
290:
1.169 brouard 291: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 292: Summary: update
1.169 brouard 293:
1.168 brouard 294: Revision 1.167 2014/12/22 13:50:56 brouard
295: Summary: Testing uname and compiler version and if compiled 32 or 64
296:
297: Testing on Linux 64
298:
1.167 brouard 299: Revision 1.166 2014/12/22 11:40:47 brouard
300: *** empty log message ***
301:
1.166 brouard 302: Revision 1.165 2014/12/16 11:20:36 brouard
303: Summary: After compiling on Visual C
304:
305: * imach.c (Module): Merging 1.61 to 1.162
306:
1.165 brouard 307: Revision 1.164 2014/12/16 10:52:11 brouard
308: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
309:
310: * imach.c (Module): Merging 1.61 to 1.162
311:
1.164 brouard 312: Revision 1.163 2014/12/16 10:30:11 brouard
313: * imach.c (Module): Merging 1.61 to 1.162
314:
1.163 brouard 315: Revision 1.162 2014/09/25 11:43:39 brouard
316: Summary: temporary backup 0.99!
317:
1.162 brouard 318: Revision 1.1 2014/09/16 11:06:58 brouard
319: Summary: With some code (wrong) for nlopt
320:
321: Author:
322:
323: Revision 1.161 2014/09/15 20:41:41 brouard
324: Summary: Problem with macro SQR on Intel compiler
325:
1.161 brouard 326: Revision 1.160 2014/09/02 09:24:05 brouard
327: *** empty log message ***
328:
1.160 brouard 329: Revision 1.159 2014/09/01 10:34:10 brouard
330: Summary: WIN32
331: Author: Brouard
332:
1.159 brouard 333: Revision 1.158 2014/08/27 17:11:51 brouard
334: *** empty log message ***
335:
1.158 brouard 336: Revision 1.157 2014/08/27 16:26:55 brouard
337: Summary: Preparing windows Visual studio version
338: Author: Brouard
339:
340: In order to compile on Visual studio, time.h is now correct and time_t
341: and tm struct should be used. difftime should be used but sometimes I
342: just make the differences in raw time format (time(&now).
343: Trying to suppress #ifdef LINUX
344: Add xdg-open for __linux in order to open default browser.
345:
1.157 brouard 346: Revision 1.156 2014/08/25 20:10:10 brouard
347: *** empty log message ***
348:
1.156 brouard 349: Revision 1.155 2014/08/25 18:32:34 brouard
350: Summary: New compile, minor changes
351: Author: Brouard
352:
1.155 brouard 353: Revision 1.154 2014/06/20 17:32:08 brouard
354: Summary: Outputs now all graphs of convergence to period prevalence
355:
1.154 brouard 356: Revision 1.153 2014/06/20 16:45:46 brouard
357: Summary: If 3 live state, convergence to period prevalence on same graph
358: Author: Brouard
359:
1.153 brouard 360: Revision 1.152 2014/06/18 17:54:09 brouard
361: Summary: open browser, use gnuplot on same dir than imach if not found in the path
362:
1.152 brouard 363: Revision 1.151 2014/06/18 16:43:30 brouard
364: *** empty log message ***
365:
1.151 brouard 366: Revision 1.150 2014/06/18 16:42:35 brouard
367: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
368: Author: brouard
369:
1.150 brouard 370: Revision 1.149 2014/06/18 15:51:14 brouard
371: Summary: Some fixes in parameter files errors
372: Author: Nicolas Brouard
373:
1.149 brouard 374: Revision 1.148 2014/06/17 17:38:48 brouard
375: Summary: Nothing new
376: Author: Brouard
377:
378: Just a new packaging for OS/X version 0.98nS
379:
1.148 brouard 380: Revision 1.147 2014/06/16 10:33:11 brouard
381: *** empty log message ***
382:
1.147 brouard 383: Revision 1.146 2014/06/16 10:20:28 brouard
384: Summary: Merge
385: Author: Brouard
386:
387: Merge, before building revised version.
388:
1.146 brouard 389: Revision 1.145 2014/06/10 21:23:15 brouard
390: Summary: Debugging with valgrind
391: Author: Nicolas Brouard
392:
393: Lot of changes in order to output the results with some covariates
394: After the Edimburgh REVES conference 2014, it seems mandatory to
395: improve the code.
396: No more memory valgrind error but a lot has to be done in order to
397: continue the work of splitting the code into subroutines.
398: Also, decodemodel has been improved. Tricode is still not
399: optimal. nbcode should be improved. Documentation has been added in
400: the source code.
401:
1.144 brouard 402: Revision 1.143 2014/01/26 09:45:38 brouard
403: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
404:
405: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
406: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
407:
1.143 brouard 408: Revision 1.142 2014/01/26 03:57:36 brouard
409: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
410:
411: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
412:
1.142 brouard 413: Revision 1.141 2014/01/26 02:42:01 brouard
414: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
415:
1.141 brouard 416: Revision 1.140 2011/09/02 10:37:54 brouard
417: Summary: times.h is ok with mingw32 now.
418:
1.140 brouard 419: Revision 1.139 2010/06/14 07:50:17 brouard
420: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
421: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
422:
1.139 brouard 423: Revision 1.138 2010/04/30 18:19:40 brouard
424: *** empty log message ***
425:
1.138 brouard 426: Revision 1.137 2010/04/29 18:11:38 brouard
427: (Module): Checking covariates for more complex models
428: than V1+V2. A lot of change to be done. Unstable.
429:
1.137 brouard 430: Revision 1.136 2010/04/26 20:30:53 brouard
431: (Module): merging some libgsl code. Fixing computation
432: of likelione (using inter/intrapolation if mle = 0) in order to
433: get same likelihood as if mle=1.
434: Some cleaning of code and comments added.
435:
1.136 brouard 436: Revision 1.135 2009/10/29 15:33:14 brouard
437: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
438:
1.135 brouard 439: Revision 1.134 2009/10/29 13:18:53 brouard
440: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
441:
1.134 brouard 442: Revision 1.133 2009/07/06 10:21:25 brouard
443: just nforces
444:
1.133 brouard 445: Revision 1.132 2009/07/06 08:22:05 brouard
446: Many tings
447:
1.132 brouard 448: Revision 1.131 2009/06/20 16:22:47 brouard
449: Some dimensions resccaled
450:
1.131 brouard 451: Revision 1.130 2009/05/26 06:44:34 brouard
452: (Module): Max Covariate is now set to 20 instead of 8. A
453: lot of cleaning with variables initialized to 0. Trying to make
454: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
455:
1.130 brouard 456: Revision 1.129 2007/08/31 13:49:27 lievre
457: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
458:
1.129 lievre 459: Revision 1.128 2006/06/30 13:02:05 brouard
460: (Module): Clarifications on computing e.j
461:
1.128 brouard 462: Revision 1.127 2006/04/28 18:11:50 brouard
463: (Module): Yes the sum of survivors was wrong since
464: imach-114 because nhstepm was no more computed in the age
465: loop. Now we define nhstepma in the age loop.
466: (Module): In order to speed up (in case of numerous covariates) we
467: compute health expectancies (without variances) in a first step
468: and then all the health expectancies with variances or standard
469: deviation (needs data from the Hessian matrices) which slows the
470: computation.
471: In the future we should be able to stop the program is only health
472: expectancies and graph are needed without standard deviations.
473:
1.127 brouard 474: Revision 1.126 2006/04/28 17:23:28 brouard
475: (Module): Yes the sum of survivors was wrong since
476: imach-114 because nhstepm was no more computed in the age
477: loop. Now we define nhstepma in the age loop.
478: Version 0.98h
479:
1.126 brouard 480: Revision 1.125 2006/04/04 15:20:31 lievre
481: Errors in calculation of health expectancies. Age was not initialized.
482: Forecasting file added.
483:
484: Revision 1.124 2006/03/22 17:13:53 lievre
485: Parameters are printed with %lf instead of %f (more numbers after the comma).
486: The log-likelihood is printed in the log file
487:
488: Revision 1.123 2006/03/20 10:52:43 brouard
489: * imach.c (Module): <title> changed, corresponds to .htm file
490: name. <head> headers where missing.
491:
492: * imach.c (Module): Weights can have a decimal point as for
493: English (a comma might work with a correct LC_NUMERIC environment,
494: otherwise the weight is truncated).
495: Modification of warning when the covariates values are not 0 or
496: 1.
497: Version 0.98g
498:
499: Revision 1.122 2006/03/20 09:45:41 brouard
500: (Module): Weights can have a decimal point as for
501: English (a comma might work with a correct LC_NUMERIC environment,
502: otherwise the weight is truncated).
503: Modification of warning when the covariates values are not 0 or
504: 1.
505: Version 0.98g
506:
507: Revision 1.121 2006/03/16 17:45:01 lievre
508: * imach.c (Module): Comments concerning covariates added
509:
510: * imach.c (Module): refinements in the computation of lli if
511: status=-2 in order to have more reliable computation if stepm is
512: not 1 month. Version 0.98f
513:
514: Revision 1.120 2006/03/16 15:10:38 lievre
515: (Module): refinements in the computation of lli if
516: status=-2 in order to have more reliable computation if stepm is
517: not 1 month. Version 0.98f
518:
519: Revision 1.119 2006/03/15 17:42:26 brouard
520: (Module): Bug if status = -2, the loglikelihood was
521: computed as likelihood omitting the logarithm. Version O.98e
522:
523: Revision 1.118 2006/03/14 18:20:07 brouard
524: (Module): varevsij Comments added explaining the second
525: table of variances if popbased=1 .
526: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
527: (Module): Function pstamp added
528: (Module): Version 0.98d
529:
530: Revision 1.117 2006/03/14 17:16:22 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 1.116 2006/03/06 10:29:27 brouard
538: (Module): Variance-covariance wrong links and
539: varian-covariance of ej. is needed (Saito).
540:
541: Revision 1.115 2006/02/27 12:17:45 brouard
542: (Module): One freematrix added in mlikeli! 0.98c
543:
544: Revision 1.114 2006/02/26 12:57:58 brouard
545: (Module): Some improvements in processing parameter
546: filename with strsep.
547:
548: Revision 1.113 2006/02/24 14:20:24 brouard
549: (Module): Memory leaks checks with valgrind and:
550: datafile was not closed, some imatrix were not freed and on matrix
551: allocation too.
552:
553: Revision 1.112 2006/01/30 09:55:26 brouard
554: (Module): Back to gnuplot.exe instead of wgnuplot.exe
555:
556: Revision 1.111 2006/01/25 20:38:18 brouard
557: (Module): Lots of cleaning and bugs added (Gompertz)
558: (Module): Comments can be added in data file. Missing date values
559: can be a simple dot '.'.
560:
561: Revision 1.110 2006/01/25 00:51:50 brouard
562: (Module): Lots of cleaning and bugs added (Gompertz)
563:
564: Revision 1.109 2006/01/24 19:37:15 brouard
565: (Module): Comments (lines starting with a #) are allowed in data.
566:
567: Revision 1.108 2006/01/19 18:05:42 lievre
568: Gnuplot problem appeared...
569: To be fixed
570:
571: Revision 1.107 2006/01/19 16:20:37 brouard
572: Test existence of gnuplot in imach path
573:
574: Revision 1.106 2006/01/19 13:24:36 brouard
575: Some cleaning and links added in html output
576:
577: Revision 1.105 2006/01/05 20:23:19 lievre
578: *** empty log message ***
579:
580: Revision 1.104 2005/09/30 16:11:43 lievre
581: (Module): sump fixed, loop imx fixed, and simplifications.
582: (Module): If the status is missing at the last wave but we know
583: that the person is alive, then we can code his/her status as -2
584: (instead of missing=-1 in earlier versions) and his/her
585: contributions to the likelihood is 1 - Prob of dying from last
586: health status (= 1-p13= p11+p12 in the easiest case of somebody in
587: the healthy state at last known wave). Version is 0.98
588:
589: Revision 1.103 2005/09/30 15:54:49 lievre
590: (Module): sump fixed, loop imx fixed, and simplifications.
591:
592: Revision 1.102 2004/09/15 17:31:30 brouard
593: Add the possibility to read data file including tab characters.
594:
595: Revision 1.101 2004/09/15 10:38:38 brouard
596: Fix on curr_time
597:
598: Revision 1.100 2004/07/12 18:29:06 brouard
599: Add version for Mac OS X. Just define UNIX in Makefile
600:
601: Revision 1.99 2004/06/05 08:57:40 brouard
602: *** empty log message ***
603:
604: Revision 1.98 2004/05/16 15:05:56 brouard
605: New version 0.97 . First attempt to estimate force of mortality
606: directly from the data i.e. without the need of knowing the health
607: state at each age, but using a Gompertz model: log u =a + b*age .
608: This is the basic analysis of mortality and should be done before any
609: other analysis, in order to test if the mortality estimated from the
610: cross-longitudinal survey is different from the mortality estimated
611: from other sources like vital statistic data.
612:
613: The same imach parameter file can be used but the option for mle should be -3.
614:
1.133 brouard 615: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 616: former routines in order to include the new code within the former code.
617:
618: The output is very simple: only an estimate of the intercept and of
619: the slope with 95% confident intervals.
620:
621: Current limitations:
622: A) Even if you enter covariates, i.e. with the
623: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
624: B) There is no computation of Life Expectancy nor Life Table.
625:
626: Revision 1.97 2004/02/20 13:25:42 lievre
627: Version 0.96d. Population forecasting command line is (temporarily)
628: suppressed.
629:
630: Revision 1.96 2003/07/15 15:38:55 brouard
631: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
632: rewritten within the same printf. Workaround: many printfs.
633:
634: Revision 1.95 2003/07/08 07:54:34 brouard
635: * imach.c (Repository):
636: (Repository): Using imachwizard code to output a more meaningful covariance
637: matrix (cov(a12,c31) instead of numbers.
638:
639: Revision 1.94 2003/06/27 13:00:02 brouard
640: Just cleaning
641:
642: Revision 1.93 2003/06/25 16:33:55 brouard
643: (Module): On windows (cygwin) function asctime_r doesn't
644: exist so I changed back to asctime which exists.
645: (Module): Version 0.96b
646:
647: Revision 1.92 2003/06/25 16:30:45 brouard
648: (Module): On windows (cygwin) function asctime_r doesn't
649: exist so I changed back to asctime which exists.
650:
651: Revision 1.91 2003/06/25 15:30:29 brouard
652: * imach.c (Repository): Duplicated warning errors corrected.
653: (Repository): Elapsed time after each iteration is now output. It
654: helps to forecast when convergence will be reached. Elapsed time
655: is stamped in powell. We created a new html file for the graphs
656: concerning matrix of covariance. It has extension -cov.htm.
657:
658: Revision 1.90 2003/06/24 12:34:15 brouard
659: (Module): Some bugs corrected for windows. Also, when
660: mle=-1 a template is output in file "or"mypar.txt with the design
661: of the covariance matrix to be input.
662:
663: Revision 1.89 2003/06/24 12:30:52 brouard
664: (Module): Some bugs corrected for windows. Also, when
665: mle=-1 a template is output in file "or"mypar.txt with the design
666: of the covariance matrix to be input.
667:
668: Revision 1.88 2003/06/23 17:54:56 brouard
669: * 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.
670:
671: Revision 1.87 2003/06/18 12:26:01 brouard
672: Version 0.96
673:
674: Revision 1.86 2003/06/17 20:04:08 brouard
675: (Module): Change position of html and gnuplot routines and added
676: routine fileappend.
677:
678: Revision 1.85 2003/06/17 13:12:43 brouard
679: * imach.c (Repository): Check when date of death was earlier that
680: current date of interview. It may happen when the death was just
681: prior to the death. In this case, dh was negative and likelihood
682: was wrong (infinity). We still send an "Error" but patch by
683: assuming that the date of death was just one stepm after the
684: interview.
685: (Repository): Because some people have very long ID (first column)
686: we changed int to long in num[] and we added a new lvector for
687: memory allocation. But we also truncated to 8 characters (left
688: truncation)
689: (Repository): No more line truncation errors.
690:
691: Revision 1.84 2003/06/13 21:44:43 brouard
692: * imach.c (Repository): Replace "freqsummary" at a correct
693: place. It differs from routine "prevalence" which may be called
694: many times. Probs is memory consuming and must be used with
695: parcimony.
696: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
697:
698: Revision 1.83 2003/06/10 13:39:11 lievre
699: *** empty log message ***
700:
701: Revision 1.82 2003/06/05 15:57:20 brouard
702: Add log in imach.c and fullversion number is now printed.
703:
704: */
705: /*
706: Interpolated Markov Chain
707:
708: Short summary of the programme:
709:
1.227 brouard 710: This program computes Healthy Life Expectancies or State-specific
711: (if states aren't health statuses) Expectancies from
712: cross-longitudinal data. Cross-longitudinal data consist in:
713:
714: -1- a first survey ("cross") where individuals from different ages
715: are interviewed on their health status or degree of disability (in
716: the case of a health survey which is our main interest)
717:
718: -2- at least a second wave of interviews ("longitudinal") which
719: measure each change (if any) in individual health status. Health
720: expectancies are computed from the time spent in each health state
721: according to a model. More health states you consider, more time is
722: necessary to reach the Maximum Likelihood of the parameters involved
723: in the model. The simplest model is the multinomial logistic model
724: where pij is the probability to be observed in state j at the second
725: wave conditional to be observed in state i at the first
726: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
727: etc , where 'age' is age and 'sex' is a covariate. If you want to
728: have a more complex model than "constant and age", you should modify
729: the program where the markup *Covariates have to be included here
730: again* invites you to do it. More covariates you add, slower the
1.126 brouard 731: convergence.
732:
733: The advantage of this computer programme, compared to a simple
734: multinomial logistic model, is clear when the delay between waves is not
735: identical for each individual. Also, if a individual missed an
736: intermediate interview, the information is lost, but taken into
737: account using an interpolation or extrapolation.
738:
739: hPijx is the probability to be observed in state i at age x+h
740: conditional to the observed state i at age x. The delay 'h' can be
741: split into an exact number (nh*stepm) of unobserved intermediate
742: states. This elementary transition (by month, quarter,
743: semester or year) is modelled as a multinomial logistic. The hPx
744: matrix is simply the matrix product of nh*stepm elementary matrices
745: and the contribution of each individual to the likelihood is simply
746: hPijx.
747:
748: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 749: of the life expectancies. It also computes the period (stable) prevalence.
750:
751: Back prevalence and projections:
1.227 brouard 752:
753: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
754: double agemaxpar, double ftolpl, int *ncvyearp, double
755: dateprev1,double dateprev2, int firstpass, int lastpass, int
756: mobilavproj)
757:
758: Computes the back prevalence limit for any combination of
759: covariate values k at any age between ageminpar and agemaxpar and
760: returns it in **bprlim. In the loops,
761:
762: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
763: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
764:
765: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 766: Computes for any combination of covariates k and any age between bage and fage
767: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
768: oldm=oldms;savm=savms;
1.227 brouard 769:
770: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k);
1.218 brouard 771: Computes the transition matrix starting at age 'age' over
772: 'nhstepm*hstepm*stepm' months (i.e. until
773: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 774: nhstepm*hstepm matrices.
775:
776: Returns p3mat[i][j][h] after calling
777: p3mat[i][j][h]=matprod2(newm,
778: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
779: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
780: oldm);
1.226 brouard 781:
782: Important routines
783:
784: - func (or funcone), computes logit (pij) distinguishing
785: o fixed variables (single or product dummies or quantitative);
786: o varying variables by:
787: (1) wave (single, product dummies, quantitative),
788: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
789: % fixed dummy (treated) or quantitative (not done because time-consuming);
790: % varying dummy (not done) or quantitative (not done);
791: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
792: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
793: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
794: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
795: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 796:
1.226 brouard 797:
798:
1.133 brouard 799: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
800: Institut national d'études démographiques, Paris.
1.126 brouard 801: This software have been partly granted by Euro-REVES, a concerted action
802: from the European Union.
803: It is copyrighted identically to a GNU software product, ie programme and
804: software can be distributed freely for non commercial use. Latest version
805: can be accessed at http://euroreves.ined.fr/imach .
806:
807: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
808: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
809:
810: **********************************************************************/
811: /*
812: main
813: read parameterfile
814: read datafile
815: concatwav
816: freqsummary
817: if (mle >= 1)
818: mlikeli
819: print results files
820: if mle==1
821: computes hessian
822: read end of parameter file: agemin, agemax, bage, fage, estepm
823: begin-prev-date,...
824: open gnuplot file
825: open html file
1.145 brouard 826: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
827: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
828: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
829: freexexit2 possible for memory heap.
830:
831: h Pij x | pij_nom ficrestpij
832: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
833: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
834: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
835:
836: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
837: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
838: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
839: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
840: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
841:
1.126 brouard 842: forecasting if prevfcast==1 prevforecast call prevalence()
843: health expectancies
844: Variance-covariance of DFLE
845: prevalence()
846: movingaverage()
847: varevsij()
848: if popbased==1 varevsij(,popbased)
849: total life expectancies
850: Variance of period (stable) prevalence
851: end
852: */
853:
1.187 brouard 854: /* #define DEBUG */
855: /* #define DEBUGBRENT */
1.203 brouard 856: /* #define DEBUGLINMIN */
857: /* #define DEBUGHESS */
858: #define DEBUGHESSIJ
1.224 brouard 859: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 860: #define POWELL /* Instead of NLOPT */
1.224 brouard 861: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 862: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
863: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 864:
865: #include <math.h>
866: #include <stdio.h>
867: #include <stdlib.h>
868: #include <string.h>
1.226 brouard 869: #include <ctype.h>
1.159 brouard 870:
871: #ifdef _WIN32
872: #include <io.h>
1.172 brouard 873: #include <windows.h>
874: #include <tchar.h>
1.159 brouard 875: #else
1.126 brouard 876: #include <unistd.h>
1.159 brouard 877: #endif
1.126 brouard 878:
879: #include <limits.h>
880: #include <sys/types.h>
1.171 brouard 881:
882: #if defined(__GNUC__)
883: #include <sys/utsname.h> /* Doesn't work on Windows */
884: #endif
885:
1.126 brouard 886: #include <sys/stat.h>
887: #include <errno.h>
1.159 brouard 888: /* extern int errno; */
1.126 brouard 889:
1.157 brouard 890: /* #ifdef LINUX */
891: /* #include <time.h> */
892: /* #include "timeval.h" */
893: /* #else */
894: /* #include <sys/time.h> */
895: /* #endif */
896:
1.126 brouard 897: #include <time.h>
898:
1.136 brouard 899: #ifdef GSL
900: #include <gsl/gsl_errno.h>
901: #include <gsl/gsl_multimin.h>
902: #endif
903:
1.167 brouard 904:
1.162 brouard 905: #ifdef NLOPT
906: #include <nlopt.h>
907: typedef struct {
908: double (* function)(double [] );
909: } myfunc_data ;
910: #endif
911:
1.126 brouard 912: /* #include <libintl.h> */
913: /* #define _(String) gettext (String) */
914:
1.141 brouard 915: #define MAXLINE 1024 /* Was 256. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 916:
917: #define GNUPLOTPROGRAM "gnuplot"
918: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
919: #define FILENAMELENGTH 132
920:
921: #define GLOCK_ERROR_NOPATH -1 /* empty path */
922: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
923:
1.144 brouard 924: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
925: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 926:
927: #define NINTERVMAX 8
1.144 brouard 928: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
929: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
930: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 931: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 932: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
933: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 934: #define MAXN 20000
1.144 brouard 935: #define YEARM 12. /**< Number of months per year */
1.218 brouard 936: /* #define AGESUP 130 */
937: #define AGESUP 150
938: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 939: #define AGEBASE 40
1.194 brouard 940: #define AGEOVERFLOW 1.e20
1.164 brouard 941: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 942: #ifdef _WIN32
943: #define DIRSEPARATOR '\\'
944: #define CHARSEPARATOR "\\"
945: #define ODIRSEPARATOR '/'
946: #else
1.126 brouard 947: #define DIRSEPARATOR '/'
948: #define CHARSEPARATOR "/"
949: #define ODIRSEPARATOR '\\'
950: #endif
951:
1.250 ! brouard 952: /* $Id: imach.c,v 1.249 2016/09/07 17:14:18 brouard Exp $ */
1.126 brouard 953: /* $State: Exp $ */
1.196 brouard 954: #include "version.h"
955: char version[]=__IMACH_VERSION__;
1.224 brouard 956: 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";
1.250 ! brouard 957: char fullversion[]="$Revision: 1.249 $ $Date: 2016/09/07 17:14:18 $";
1.126 brouard 958: char strstart[80];
959: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 960: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 961: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 962: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
963: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
964: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 965: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
966: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 967: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
968: int cptcovprodnoage=0; /**< Number of covariate products without age */
969: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 970: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
971: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 972: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 973: int nsd=0; /**< Total number of single dummy variables (output) */
974: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 975: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 976: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 977: int ntveff=0; /**< ntveff number of effective time varying variables */
978: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 979: int cptcov=0; /* Working variable */
1.218 brouard 980: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 981: int npar=NPARMAX;
982: int nlstate=2; /* Number of live states */
983: int ndeath=1; /* Number of dead states */
1.130 brouard 984: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 985: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 986: int popbased=0;
987:
988: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 989: int maxwav=0; /* Maxim number of waves */
990: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
991: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
992: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 993: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 994: int mle=1, weightopt=0;
1.126 brouard 995: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
996: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
997: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
998: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 999: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1000: int selected(int kvar); /* Is covariate kvar selected for printing results */
1001:
1.130 brouard 1002: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1003: double **matprod2(); /* test */
1.126 brouard 1004: double **oldm, **newm, **savm; /* Working pointers to matrices */
1005: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1006: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1007:
1.136 brouard 1008: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1009: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1010: FILE *ficlog, *ficrespow;
1.130 brouard 1011: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1012: double fretone; /* Only one call to likelihood */
1.130 brouard 1013: long ipmx=0; /* Number of contributions */
1.126 brouard 1014: double sw; /* Sum of weights */
1015: char filerespow[FILENAMELENGTH];
1016: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1017: FILE *ficresilk;
1018: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1019: FILE *ficresprobmorprev;
1020: FILE *fichtm, *fichtmcov; /* Html File */
1021: FILE *ficreseij;
1022: char filerese[FILENAMELENGTH];
1023: FILE *ficresstdeij;
1024: char fileresstde[FILENAMELENGTH];
1025: FILE *ficrescveij;
1026: char filerescve[FILENAMELENGTH];
1027: FILE *ficresvij;
1028: char fileresv[FILENAMELENGTH];
1029: FILE *ficresvpl;
1030: char fileresvpl[FILENAMELENGTH];
1031: char title[MAXLINE];
1.234 brouard 1032: char model[MAXLINE]; /**< The model line */
1.217 brouard 1033: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1034: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1035: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1036: char command[FILENAMELENGTH];
1037: int outcmd=0;
1038:
1.217 brouard 1039: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1040: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1041: char filelog[FILENAMELENGTH]; /* Log file */
1042: char filerest[FILENAMELENGTH];
1043: char fileregp[FILENAMELENGTH];
1044: char popfile[FILENAMELENGTH];
1045:
1046: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1047:
1.157 brouard 1048: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1049: /* struct timezone tzp; */
1050: /* extern int gettimeofday(); */
1051: struct tm tml, *gmtime(), *localtime();
1052:
1053: extern time_t time();
1054:
1055: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1056: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1057: struct tm tm;
1058:
1.126 brouard 1059: char strcurr[80], strfor[80];
1060:
1061: char *endptr;
1062: long lval;
1063: double dval;
1064:
1065: #define NR_END 1
1066: #define FREE_ARG char*
1067: #define FTOL 1.0e-10
1068:
1069: #define NRANSI
1.240 brouard 1070: #define ITMAX 200
1071: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1072:
1073: #define TOL 2.0e-4
1074:
1075: #define CGOLD 0.3819660
1076: #define ZEPS 1.0e-10
1077: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1078:
1079: #define GOLD 1.618034
1080: #define GLIMIT 100.0
1081: #define TINY 1.0e-20
1082:
1083: static double maxarg1,maxarg2;
1084: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1085: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1086:
1087: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1088: #define rint(a) floor(a+0.5)
1.166 brouard 1089: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1090: #define mytinydouble 1.0e-16
1.166 brouard 1091: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1092: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1093: /* static double dsqrarg; */
1094: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1095: static double sqrarg;
1096: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1097: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1098: int agegomp= AGEGOMP;
1099:
1100: int imx;
1101: int stepm=1;
1102: /* Stepm, step in month: minimum step interpolation*/
1103:
1104: int estepm;
1105: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1106:
1107: int m,nb;
1108: long *num;
1.197 brouard 1109: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1110: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1111: covariate for which somebody answered excluding
1112: undefined. Usually 2: 0 and 1. */
1113: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1114: covariate for which somebody answered including
1115: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1116: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1117: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1118: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1119: double *ageexmed,*agecens;
1120: double dateintmean=0;
1121:
1122: double *weight;
1123: int **s; /* Status */
1.141 brouard 1124: double *agedc;
1.145 brouard 1125: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1126: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1127: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1128: double **coqvar; /* Fixed quantitative covariate iqv */
1129: double ***cotvar; /* Time varying covariate itv */
1130: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1131: double idx;
1132: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1133: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1134: /*k 1 2 3 4 5 6 7 8 9 */
1135: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1136: /* Tndvar[k] 1 2 3 4 5 */
1137: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1138: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1139: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1140: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1141: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1142: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1143: /* Tprod[i]=k 4 7 */
1144: /* Tage[i]=k 5 8 */
1145: /* */
1146: /* Type */
1147: /* V 1 2 3 4 5 */
1148: /* F F V V V */
1149: /* D Q D D Q */
1150: /* */
1151: int *TvarsD;
1152: int *TvarsDind;
1153: int *TvarsQ;
1154: int *TvarsQind;
1155:
1.235 brouard 1156: #define MAXRESULTLINES 10
1157: int nresult=0;
1158: int TKresult[MAXRESULTLINES];
1.237 brouard 1159: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1160: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1161: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1162: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1163: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1164: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1165:
1.234 brouard 1166: /* 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 1167: 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 */
1168: 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 */
1169: 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 */
1170: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1171: 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 */
1172: 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 1173: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1174: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1175: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1176: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1177: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1178: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1179: 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 */
1180: 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 */
1181:
1.230 brouard 1182: int *Tvarsel; /**< Selected covariates for output */
1183: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1184: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1185: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1186: 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 1187: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1188: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1189: int *Tage;
1.227 brouard 1190: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1191: 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 1192: 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*/
1193: 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 1194: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1195: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1196: int **Tvard;
1197: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1198: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1199: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1200: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1201: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1202: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1203: double *lsurv, *lpop, *tpop;
1204:
1.231 brouard 1205: #define FD 1; /* Fixed dummy covariate */
1206: #define FQ 2; /* Fixed quantitative covariate */
1207: #define FP 3; /* Fixed product covariate */
1208: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1209: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1210: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1211: #define VD 10; /* Varying dummy covariate */
1212: #define VQ 11; /* Varying quantitative covariate */
1213: #define VP 12; /* Varying product covariate */
1214: #define VPDD 13; /* Varying product dummy*dummy covariate */
1215: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1216: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1217: #define APFD 16; /* Age product * fixed dummy covariate */
1218: #define APFQ 17; /* Age product * fixed quantitative covariate */
1219: #define APVD 18; /* Age product * varying dummy covariate */
1220: #define APVQ 19; /* Age product * varying quantitative covariate */
1221:
1222: #define FTYPE 1; /* Fixed covariate */
1223: #define VTYPE 2; /* Varying covariate (loop in wave) */
1224: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1225:
1226: struct kmodel{
1227: int maintype; /* main type */
1228: int subtype; /* subtype */
1229: };
1230: struct kmodel modell[NCOVMAX];
1231:
1.143 brouard 1232: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1233: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1234:
1235: /**************** split *************************/
1236: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1237: {
1238: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1239: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1240: */
1241: char *ss; /* pointer */
1.186 brouard 1242: int l1=0, l2=0; /* length counters */
1.126 brouard 1243:
1244: l1 = strlen(path ); /* length of path */
1245: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1246: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1247: if ( ss == NULL ) { /* no directory, so determine current directory */
1248: strcpy( name, path ); /* we got the fullname name because no directory */
1249: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1250: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1251: /* get current working directory */
1252: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1253: #ifdef WIN32
1254: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1255: #else
1256: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1257: #endif
1.126 brouard 1258: return( GLOCK_ERROR_GETCWD );
1259: }
1260: /* got dirc from getcwd*/
1261: printf(" DIRC = %s \n",dirc);
1.205 brouard 1262: } else { /* strip directory from path */
1.126 brouard 1263: ss++; /* after this, the filename */
1264: l2 = strlen( ss ); /* length of filename */
1265: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1266: strcpy( name, ss ); /* save file name */
1267: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1268: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1269: printf(" DIRC2 = %s \n",dirc);
1270: }
1271: /* We add a separator at the end of dirc if not exists */
1272: l1 = strlen( dirc ); /* length of directory */
1273: if( dirc[l1-1] != DIRSEPARATOR ){
1274: dirc[l1] = DIRSEPARATOR;
1275: dirc[l1+1] = 0;
1276: printf(" DIRC3 = %s \n",dirc);
1277: }
1278: ss = strrchr( name, '.' ); /* find last / */
1279: if (ss >0){
1280: ss++;
1281: strcpy(ext,ss); /* save extension */
1282: l1= strlen( name);
1283: l2= strlen(ss)+1;
1284: strncpy( finame, name, l1-l2);
1285: finame[l1-l2]= 0;
1286: }
1287:
1288: return( 0 ); /* we're done */
1289: }
1290:
1291:
1292: /******************************************/
1293:
1294: void replace_back_to_slash(char *s, char*t)
1295: {
1296: int i;
1297: int lg=0;
1298: i=0;
1299: lg=strlen(t);
1300: for(i=0; i<= lg; i++) {
1301: (s[i] = t[i]);
1302: if (t[i]== '\\') s[i]='/';
1303: }
1304: }
1305:
1.132 brouard 1306: char *trimbb(char *out, char *in)
1.137 brouard 1307: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1308: char *s;
1309: s=out;
1310: while (*in != '\0'){
1.137 brouard 1311: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1312: in++;
1313: }
1314: *out++ = *in++;
1315: }
1316: *out='\0';
1317: return s;
1318: }
1319:
1.187 brouard 1320: /* char *substrchaine(char *out, char *in, char *chain) */
1321: /* { */
1322: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1323: /* char *s, *t; */
1324: /* t=in;s=out; */
1325: /* while ((*in != *chain) && (*in != '\0')){ */
1326: /* *out++ = *in++; */
1327: /* } */
1328:
1329: /* /\* *in matches *chain *\/ */
1330: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1331: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1332: /* } */
1333: /* in--; chain--; */
1334: /* while ( (*in != '\0')){ */
1335: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1336: /* *out++ = *in++; */
1337: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1338: /* } */
1339: /* *out='\0'; */
1340: /* out=s; */
1341: /* return out; */
1342: /* } */
1343: char *substrchaine(char *out, char *in, char *chain)
1344: {
1345: /* Substract chain 'chain' from 'in', return and output 'out' */
1346: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1347:
1348: char *strloc;
1349:
1350: strcpy (out, in);
1351: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1352: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1353: if(strloc != NULL){
1354: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1355: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1356: /* strcpy (strloc, strloc +strlen(chain));*/
1357: }
1358: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1359: return out;
1360: }
1361:
1362:
1.145 brouard 1363: char *cutl(char *blocc, char *alocc, char *in, char occ)
1364: {
1.187 brouard 1365: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1366: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1367: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1368: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1369: */
1.160 brouard 1370: char *s, *t;
1.145 brouard 1371: t=in;s=in;
1372: while ((*in != occ) && (*in != '\0')){
1373: *alocc++ = *in++;
1374: }
1375: if( *in == occ){
1376: *(alocc)='\0';
1377: s=++in;
1378: }
1379:
1380: if (s == t) {/* occ not found */
1381: *(alocc-(in-s))='\0';
1382: in=s;
1383: }
1384: while ( *in != '\0'){
1385: *blocc++ = *in++;
1386: }
1387:
1388: *blocc='\0';
1389: return t;
1390: }
1.137 brouard 1391: char *cutv(char *blocc, char *alocc, char *in, char occ)
1392: {
1.187 brouard 1393: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1394: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1395: gives blocc="abcdef2ghi" and alocc="j".
1396: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1397: */
1398: char *s, *t;
1399: t=in;s=in;
1400: while (*in != '\0'){
1401: while( *in == occ){
1402: *blocc++ = *in++;
1403: s=in;
1404: }
1405: *blocc++ = *in++;
1406: }
1407: if (s == t) /* occ not found */
1408: *(blocc-(in-s))='\0';
1409: else
1410: *(blocc-(in-s)-1)='\0';
1411: in=s;
1412: while ( *in != '\0'){
1413: *alocc++ = *in++;
1414: }
1415:
1416: *alocc='\0';
1417: return s;
1418: }
1419:
1.126 brouard 1420: int nbocc(char *s, char occ)
1421: {
1422: int i,j=0;
1423: int lg=20;
1424: i=0;
1425: lg=strlen(s);
1426: for(i=0; i<= lg; i++) {
1.234 brouard 1427: if (s[i] == occ ) j++;
1.126 brouard 1428: }
1429: return j;
1430: }
1431:
1.137 brouard 1432: /* void cutv(char *u,char *v, char*t, char occ) */
1433: /* { */
1434: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1435: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1436: /* gives u="abcdef2ghi" and v="j" *\/ */
1437: /* int i,lg,j,p=0; */
1438: /* i=0; */
1439: /* lg=strlen(t); */
1440: /* for(j=0; j<=lg-1; j++) { */
1441: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1442: /* } */
1.126 brouard 1443:
1.137 brouard 1444: /* for(j=0; j<p; j++) { */
1445: /* (u[j] = t[j]); */
1446: /* } */
1447: /* u[p]='\0'; */
1.126 brouard 1448:
1.137 brouard 1449: /* for(j=0; j<= lg; j++) { */
1450: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1451: /* } */
1452: /* } */
1.126 brouard 1453:
1.160 brouard 1454: #ifdef _WIN32
1455: char * strsep(char **pp, const char *delim)
1456: {
1457: char *p, *q;
1458:
1459: if ((p = *pp) == NULL)
1460: return 0;
1461: if ((q = strpbrk (p, delim)) != NULL)
1462: {
1463: *pp = q + 1;
1464: *q = '\0';
1465: }
1466: else
1467: *pp = 0;
1468: return p;
1469: }
1470: #endif
1471:
1.126 brouard 1472: /********************** nrerror ********************/
1473:
1474: void nrerror(char error_text[])
1475: {
1476: fprintf(stderr,"ERREUR ...\n");
1477: fprintf(stderr,"%s\n",error_text);
1478: exit(EXIT_FAILURE);
1479: }
1480: /*********************** vector *******************/
1481: double *vector(int nl, int nh)
1482: {
1483: double *v;
1484: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1485: if (!v) nrerror("allocation failure in vector");
1486: return v-nl+NR_END;
1487: }
1488:
1489: /************************ free vector ******************/
1490: void free_vector(double*v, int nl, int nh)
1491: {
1492: free((FREE_ARG)(v+nl-NR_END));
1493: }
1494:
1495: /************************ivector *******************************/
1496: int *ivector(long nl,long nh)
1497: {
1498: int *v;
1499: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1500: if (!v) nrerror("allocation failure in ivector");
1501: return v-nl+NR_END;
1502: }
1503:
1504: /******************free ivector **************************/
1505: void free_ivector(int *v, long nl, long nh)
1506: {
1507: free((FREE_ARG)(v+nl-NR_END));
1508: }
1509:
1510: /************************lvector *******************************/
1511: long *lvector(long nl,long nh)
1512: {
1513: long *v;
1514: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1515: if (!v) nrerror("allocation failure in ivector");
1516: return v-nl+NR_END;
1517: }
1518:
1519: /******************free lvector **************************/
1520: void free_lvector(long *v, long nl, long nh)
1521: {
1522: free((FREE_ARG)(v+nl-NR_END));
1523: }
1524:
1525: /******************* imatrix *******************************/
1526: int **imatrix(long nrl, long nrh, long ncl, long nch)
1527: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1528: {
1529: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1530: int **m;
1531:
1532: /* allocate pointers to rows */
1533: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1534: if (!m) nrerror("allocation failure 1 in matrix()");
1535: m += NR_END;
1536: m -= nrl;
1537:
1538:
1539: /* allocate rows and set pointers to them */
1540: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1541: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1542: m[nrl] += NR_END;
1543: m[nrl] -= ncl;
1544:
1545: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1546:
1547: /* return pointer to array of pointers to rows */
1548: return m;
1549: }
1550:
1551: /****************** free_imatrix *************************/
1552: void free_imatrix(m,nrl,nrh,ncl,nch)
1553: int **m;
1554: long nch,ncl,nrh,nrl;
1555: /* free an int matrix allocated by imatrix() */
1556: {
1557: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1558: free((FREE_ARG) (m+nrl-NR_END));
1559: }
1560:
1561: /******************* matrix *******************************/
1562: double **matrix(long nrl, long nrh, long ncl, long nch)
1563: {
1564: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1565: double **m;
1566:
1567: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1568: if (!m) nrerror("allocation failure 1 in matrix()");
1569: m += NR_END;
1570: m -= nrl;
1571:
1572: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1573: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1574: m[nrl] += NR_END;
1575: m[nrl] -= ncl;
1576:
1577: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1578: return m;
1.145 brouard 1579: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1580: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1581: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1582: */
1583: }
1584:
1585: /*************************free matrix ************************/
1586: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1587: {
1588: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1589: free((FREE_ARG)(m+nrl-NR_END));
1590: }
1591:
1592: /******************* ma3x *******************************/
1593: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1594: {
1595: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1596: double ***m;
1597:
1598: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1599: if (!m) nrerror("allocation failure 1 in matrix()");
1600: m += NR_END;
1601: m -= nrl;
1602:
1603: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1604: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1605: m[nrl] += NR_END;
1606: m[nrl] -= ncl;
1607:
1608: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1609:
1610: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1611: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1612: m[nrl][ncl] += NR_END;
1613: m[nrl][ncl] -= nll;
1614: for (j=ncl+1; j<=nch; j++)
1615: m[nrl][j]=m[nrl][j-1]+nlay;
1616:
1617: for (i=nrl+1; i<=nrh; i++) {
1618: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1619: for (j=ncl+1; j<=nch; j++)
1620: m[i][j]=m[i][j-1]+nlay;
1621: }
1622: return m;
1623: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1624: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1625: */
1626: }
1627:
1628: /*************************free ma3x ************************/
1629: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1630: {
1631: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1632: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1633: free((FREE_ARG)(m+nrl-NR_END));
1634: }
1635:
1636: /*************** function subdirf ***********/
1637: char *subdirf(char fileres[])
1638: {
1639: /* Caution optionfilefiname is hidden */
1640: strcpy(tmpout,optionfilefiname);
1641: strcat(tmpout,"/"); /* Add to the right */
1642: strcat(tmpout,fileres);
1643: return tmpout;
1644: }
1645:
1646: /*************** function subdirf2 ***********/
1647: char *subdirf2(char fileres[], char *preop)
1648: {
1649:
1650: /* Caution optionfilefiname is hidden */
1651: strcpy(tmpout,optionfilefiname);
1652: strcat(tmpout,"/");
1653: strcat(tmpout,preop);
1654: strcat(tmpout,fileres);
1655: return tmpout;
1656: }
1657:
1658: /*************** function subdirf3 ***********/
1659: char *subdirf3(char fileres[], char *preop, char *preop2)
1660: {
1661:
1662: /* Caution optionfilefiname is hidden */
1663: strcpy(tmpout,optionfilefiname);
1664: strcat(tmpout,"/");
1665: strcat(tmpout,preop);
1666: strcat(tmpout,preop2);
1667: strcat(tmpout,fileres);
1668: return tmpout;
1669: }
1.213 brouard 1670:
1671: /*************** function subdirfext ***********/
1672: char *subdirfext(char fileres[], char *preop, char *postop)
1673: {
1674:
1675: strcpy(tmpout,preop);
1676: strcat(tmpout,fileres);
1677: strcat(tmpout,postop);
1678: return tmpout;
1679: }
1.126 brouard 1680:
1.213 brouard 1681: /*************** function subdirfext3 ***********/
1682: char *subdirfext3(char fileres[], char *preop, char *postop)
1683: {
1684:
1685: /* Caution optionfilefiname is hidden */
1686: strcpy(tmpout,optionfilefiname);
1687: strcat(tmpout,"/");
1688: strcat(tmpout,preop);
1689: strcat(tmpout,fileres);
1690: strcat(tmpout,postop);
1691: return tmpout;
1692: }
1693:
1.162 brouard 1694: char *asc_diff_time(long time_sec, char ascdiff[])
1695: {
1696: long sec_left, days, hours, minutes;
1697: days = (time_sec) / (60*60*24);
1698: sec_left = (time_sec) % (60*60*24);
1699: hours = (sec_left) / (60*60) ;
1700: sec_left = (sec_left) %(60*60);
1701: minutes = (sec_left) /60;
1702: sec_left = (sec_left) % (60);
1703: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1704: return ascdiff;
1705: }
1706:
1.126 brouard 1707: /***************** f1dim *************************/
1708: extern int ncom;
1709: extern double *pcom,*xicom;
1710: extern double (*nrfunc)(double []);
1711:
1712: double f1dim(double x)
1713: {
1714: int j;
1715: double f;
1716: double *xt;
1717:
1718: xt=vector(1,ncom);
1719: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1720: f=(*nrfunc)(xt);
1721: free_vector(xt,1,ncom);
1722: return f;
1723: }
1724:
1725: /*****************brent *************************/
1726: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1727: {
1728: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1729: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1730: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1731: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1732: * returned function value.
1733: */
1.126 brouard 1734: int iter;
1735: double a,b,d,etemp;
1.159 brouard 1736: double fu=0,fv,fw,fx;
1.164 brouard 1737: double ftemp=0.;
1.126 brouard 1738: double p,q,r,tol1,tol2,u,v,w,x,xm;
1739: double e=0.0;
1740:
1741: a=(ax < cx ? ax : cx);
1742: b=(ax > cx ? ax : cx);
1743: x=w=v=bx;
1744: fw=fv=fx=(*f)(x);
1745: for (iter=1;iter<=ITMAX;iter++) {
1746: xm=0.5*(a+b);
1747: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1748: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1749: printf(".");fflush(stdout);
1750: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1751: #ifdef DEBUGBRENT
1.126 brouard 1752: 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);
1753: 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);
1754: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1755: #endif
1756: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1757: *xmin=x;
1758: return fx;
1759: }
1760: ftemp=fu;
1761: if (fabs(e) > tol1) {
1762: r=(x-w)*(fx-fv);
1763: q=(x-v)*(fx-fw);
1764: p=(x-v)*q-(x-w)*r;
1765: q=2.0*(q-r);
1766: if (q > 0.0) p = -p;
1767: q=fabs(q);
1768: etemp=e;
1769: e=d;
1770: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1771: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1772: else {
1.224 brouard 1773: d=p/q;
1774: u=x+d;
1775: if (u-a < tol2 || b-u < tol2)
1776: d=SIGN(tol1,xm-x);
1.126 brouard 1777: }
1778: } else {
1779: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1780: }
1781: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1782: fu=(*f)(u);
1783: if (fu <= fx) {
1784: if (u >= x) a=x; else b=x;
1785: SHFT(v,w,x,u)
1.183 brouard 1786: SHFT(fv,fw,fx,fu)
1787: } else {
1788: if (u < x) a=u; else b=u;
1789: if (fu <= fw || w == x) {
1.224 brouard 1790: v=w;
1791: w=u;
1792: fv=fw;
1793: fw=fu;
1.183 brouard 1794: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1795: v=u;
1796: fv=fu;
1.183 brouard 1797: }
1798: }
1.126 brouard 1799: }
1800: nrerror("Too many iterations in brent");
1801: *xmin=x;
1802: return fx;
1803: }
1804:
1805: /****************** mnbrak ***********************/
1806:
1807: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1808: double (*func)(double))
1.183 brouard 1809: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1810: the downhill direction (defined by the function as evaluated at the initial points) and returns
1811: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1812: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1813: */
1.126 brouard 1814: double ulim,u,r,q, dum;
1815: double fu;
1.187 brouard 1816:
1817: double scale=10.;
1818: int iterscale=0;
1819:
1820: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1821: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1822:
1823:
1824: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1825: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1826: /* *bx = *ax - (*ax - *bx)/scale; */
1827: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1828: /* } */
1829:
1.126 brouard 1830: if (*fb > *fa) {
1831: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1832: SHFT(dum,*fb,*fa,dum)
1833: }
1.126 brouard 1834: *cx=(*bx)+GOLD*(*bx-*ax);
1835: *fc=(*func)(*cx);
1.183 brouard 1836: #ifdef DEBUG
1.224 brouard 1837: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1838: 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 1839: #endif
1.224 brouard 1840: 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 1841: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1842: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1843: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1844: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1845: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1846: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1847: fu=(*func)(u);
1.163 brouard 1848: #ifdef DEBUG
1849: /* f(x)=A(x-u)**2+f(u) */
1850: double A, fparabu;
1851: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1852: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1853: 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);
1854: 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 1855: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1856: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1857: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1858: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1859: #endif
1.184 brouard 1860: #ifdef MNBRAKORIGINAL
1.183 brouard 1861: #else
1.191 brouard 1862: /* if (fu > *fc) { */
1863: /* #ifdef DEBUG */
1864: /* printf("mnbrak4 fu > fc \n"); */
1865: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1866: /* #endif */
1867: /* /\* 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 *\\/ *\/ */
1868: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1869: /* dum=u; /\* Shifting c and u *\/ */
1870: /* u = *cx; */
1871: /* *cx = dum; */
1872: /* dum = fu; */
1873: /* fu = *fc; */
1874: /* *fc =dum; */
1875: /* } else { /\* end *\/ */
1876: /* #ifdef DEBUG */
1877: /* printf("mnbrak3 fu < fc \n"); */
1878: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1879: /* #endif */
1880: /* dum=u; /\* Shifting c and u *\/ */
1881: /* u = *cx; */
1882: /* *cx = dum; */
1883: /* dum = fu; */
1884: /* fu = *fc; */
1885: /* *fc =dum; */
1886: /* } */
1.224 brouard 1887: #ifdef DEBUGMNBRAK
1888: double A, fparabu;
1889: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1890: fparabu= *fa - A*(*ax-u)*(*ax-u);
1891: 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);
1892: 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 1893: #endif
1.191 brouard 1894: dum=u; /* Shifting c and u */
1895: u = *cx;
1896: *cx = dum;
1897: dum = fu;
1898: fu = *fc;
1899: *fc =dum;
1.183 brouard 1900: #endif
1.162 brouard 1901: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1902: #ifdef DEBUG
1.224 brouard 1903: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1904: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1905: #endif
1.126 brouard 1906: fu=(*func)(u);
1907: if (fu < *fc) {
1.183 brouard 1908: #ifdef DEBUG
1.224 brouard 1909: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1910: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1911: #endif
1912: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1913: SHFT(*fb,*fc,fu,(*func)(u))
1914: #ifdef DEBUG
1915: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1916: #endif
1917: }
1.162 brouard 1918: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1919: #ifdef DEBUG
1.224 brouard 1920: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1921: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1922: #endif
1.126 brouard 1923: u=ulim;
1924: fu=(*func)(u);
1.183 brouard 1925: } else { /* u could be left to b (if r > q parabola has a maximum) */
1926: #ifdef DEBUG
1.224 brouard 1927: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1928: 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 1929: #endif
1.126 brouard 1930: u=(*cx)+GOLD*(*cx-*bx);
1931: fu=(*func)(u);
1.224 brouard 1932: #ifdef DEBUG
1933: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1934: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1935: #endif
1.183 brouard 1936: } /* end tests */
1.126 brouard 1937: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1938: SHFT(*fa,*fb,*fc,fu)
1939: #ifdef DEBUG
1.224 brouard 1940: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1941: 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 1942: #endif
1943: } /* 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 1944: }
1945:
1946: /*************** linmin ************************/
1.162 brouard 1947: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
1948: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
1949: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
1950: the value of func at the returned location p . This is actually all accomplished by calling the
1951: routines mnbrak and brent .*/
1.126 brouard 1952: int ncom;
1953: double *pcom,*xicom;
1954: double (*nrfunc)(double []);
1955:
1.224 brouard 1956: #ifdef LINMINORIGINAL
1.126 brouard 1957: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 1958: #else
1959: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
1960: #endif
1.126 brouard 1961: {
1962: double brent(double ax, double bx, double cx,
1963: double (*f)(double), double tol, double *xmin);
1964: double f1dim(double x);
1965: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
1966: double *fc, double (*func)(double));
1967: int j;
1968: double xx,xmin,bx,ax;
1969: double fx,fb,fa;
1.187 brouard 1970:
1.203 brouard 1971: #ifdef LINMINORIGINAL
1972: #else
1973: double scale=10., axs, xxs; /* Scale added for infinity */
1974: #endif
1975:
1.126 brouard 1976: ncom=n;
1977: pcom=vector(1,n);
1978: xicom=vector(1,n);
1979: nrfunc=func;
1980: for (j=1;j<=n;j++) {
1981: pcom[j]=p[j];
1.202 brouard 1982: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 1983: }
1.187 brouard 1984:
1.203 brouard 1985: #ifdef LINMINORIGINAL
1986: xx=1.;
1987: #else
1988: axs=0.0;
1989: xxs=1.;
1990: do{
1991: xx= xxs;
1992: #endif
1.187 brouard 1993: ax=0.;
1994: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
1995: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
1996: /* 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)) */
1997: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
1998: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
1999: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2000: /* 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 2001: #ifdef LINMINORIGINAL
2002: #else
2003: if (fx != fx){
1.224 brouard 2004: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2005: printf("|");
2006: fprintf(ficlog,"|");
1.203 brouard 2007: #ifdef DEBUGLINMIN
1.224 brouard 2008: 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 2009: #endif
2010: }
1.224 brouard 2011: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2012: #endif
2013:
1.191 brouard 2014: #ifdef DEBUGLINMIN
2015: 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 2016: 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 2017: #endif
1.224 brouard 2018: #ifdef LINMINORIGINAL
2019: #else
2020: if(fb == fx){ /* Flat function in the direction */
2021: xmin=xx;
2022: *flat=1;
2023: }else{
2024: *flat=0;
2025: #endif
2026: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2027: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2028: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2029: /* fmin = f(p[j] + xmin * xi[j]) */
2030: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2031: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2032: #ifdef DEBUG
1.224 brouard 2033: 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);
2034: 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);
2035: #endif
2036: #ifdef LINMINORIGINAL
2037: #else
2038: }
1.126 brouard 2039: #endif
1.191 brouard 2040: #ifdef DEBUGLINMIN
2041: printf("linmin end ");
1.202 brouard 2042: fprintf(ficlog,"linmin end ");
1.191 brouard 2043: #endif
1.126 brouard 2044: for (j=1;j<=n;j++) {
1.203 brouard 2045: #ifdef LINMINORIGINAL
2046: xi[j] *= xmin;
2047: #else
2048: #ifdef DEBUGLINMIN
2049: if(xxs <1.0)
2050: printf(" before xi[%d]=%12.8f", j,xi[j]);
2051: #endif
2052: 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) */
2053: #ifdef DEBUGLINMIN
2054: if(xxs <1.0)
2055: 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 );
2056: #endif
2057: #endif
1.187 brouard 2058: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2059: }
1.191 brouard 2060: #ifdef DEBUGLINMIN
1.203 brouard 2061: printf("\n");
1.191 brouard 2062: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2063: 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 2064: for (j=1;j<=n;j++) {
1.202 brouard 2065: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2066: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2067: if(j % ncovmodel == 0){
1.191 brouard 2068: printf("\n");
1.202 brouard 2069: fprintf(ficlog,"\n");
2070: }
1.191 brouard 2071: }
1.203 brouard 2072: #else
1.191 brouard 2073: #endif
1.126 brouard 2074: free_vector(xicom,1,n);
2075: free_vector(pcom,1,n);
2076: }
2077:
2078:
2079: /*************** powell ************************/
1.162 brouard 2080: /*
2081: Minimization of a function func of n variables. Input consists of an initial starting point
2082: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2083: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2084: such that failure to decrease by more than this amount on one iteration signals doneness. On
2085: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2086: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2087: */
1.224 brouard 2088: #ifdef LINMINORIGINAL
2089: #else
2090: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2091: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2092: #endif
1.126 brouard 2093: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2094: double (*func)(double []))
2095: {
1.224 brouard 2096: #ifdef LINMINORIGINAL
2097: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2098: double (*func)(double []));
1.224 brouard 2099: #else
1.241 brouard 2100: void linmin(double p[], double xi[], int n, double *fret,
2101: double (*func)(double []),int *flat);
1.224 brouard 2102: #endif
1.239 brouard 2103: int i,ibig,j,jk,k;
1.126 brouard 2104: double del,t,*pt,*ptt,*xit;
1.181 brouard 2105: double directest;
1.126 brouard 2106: double fp,fptt;
2107: double *xits;
2108: int niterf, itmp;
1.224 brouard 2109: #ifdef LINMINORIGINAL
2110: #else
2111:
2112: flatdir=ivector(1,n);
2113: for (j=1;j<=n;j++) flatdir[j]=0;
2114: #endif
1.126 brouard 2115:
2116: pt=vector(1,n);
2117: ptt=vector(1,n);
2118: xit=vector(1,n);
2119: xits=vector(1,n);
2120: *fret=(*func)(p);
2121: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2122: rcurr_time = time(NULL);
1.126 brouard 2123: for (*iter=1;;++(*iter)) {
1.187 brouard 2124: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2125: ibig=0;
2126: del=0.0;
1.157 brouard 2127: rlast_time=rcurr_time;
2128: /* (void) gettimeofday(&curr_time,&tzp); */
2129: rcurr_time = time(NULL);
2130: curr_time = *localtime(&rcurr_time);
2131: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2132: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2133: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2134: for (i=1;i<=n;i++) {
1.126 brouard 2135: fprintf(ficrespow," %.12lf", p[i]);
2136: }
1.239 brouard 2137: fprintf(ficrespow,"\n");fflush(ficrespow);
2138: printf("\n#model= 1 + age ");
2139: fprintf(ficlog,"\n#model= 1 + age ");
2140: if(nagesqr==1){
1.241 brouard 2141: printf(" + age*age ");
2142: fprintf(ficlog," + age*age ");
1.239 brouard 2143: }
2144: for(j=1;j <=ncovmodel-2;j++){
2145: if(Typevar[j]==0) {
2146: printf(" + V%d ",Tvar[j]);
2147: fprintf(ficlog," + V%d ",Tvar[j]);
2148: }else if(Typevar[j]==1) {
2149: printf(" + V%d*age ",Tvar[j]);
2150: fprintf(ficlog," + V%d*age ",Tvar[j]);
2151: }else if(Typevar[j]==2) {
2152: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2153: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2154: }
2155: }
1.126 brouard 2156: printf("\n");
1.239 brouard 2157: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2158: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2159: fprintf(ficlog,"\n");
1.239 brouard 2160: for(i=1,jk=1; i <=nlstate; i++){
2161: for(k=1; k <=(nlstate+ndeath); k++){
2162: if (k != i) {
2163: printf("%d%d ",i,k);
2164: fprintf(ficlog,"%d%d ",i,k);
2165: for(j=1; j <=ncovmodel; j++){
2166: printf("%12.7f ",p[jk]);
2167: fprintf(ficlog,"%12.7f ",p[jk]);
2168: jk++;
2169: }
2170: printf("\n");
2171: fprintf(ficlog,"\n");
2172: }
2173: }
2174: }
1.241 brouard 2175: if(*iter <=3 && *iter >1){
1.157 brouard 2176: tml = *localtime(&rcurr_time);
2177: strcpy(strcurr,asctime(&tml));
2178: rforecast_time=rcurr_time;
1.126 brouard 2179: itmp = strlen(strcurr);
2180: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2181: strcurr[itmp-1]='\0';
1.162 brouard 2182: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2183: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2184: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2185: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2186: forecast_time = *localtime(&rforecast_time);
2187: strcpy(strfor,asctime(&forecast_time));
2188: itmp = strlen(strfor);
2189: if(strfor[itmp-1]=='\n')
2190: strfor[itmp-1]='\0';
2191: 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);
2192: 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 2193: }
2194: }
1.187 brouard 2195: for (i=1;i<=n;i++) { /* For each direction i */
2196: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2197: fptt=(*fret);
2198: #ifdef DEBUG
1.203 brouard 2199: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2200: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2201: #endif
1.203 brouard 2202: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2203: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2204: #ifdef LINMINORIGINAL
1.188 brouard 2205: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2206: #else
2207: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2208: flatdir[i]=flat; /* Function is vanishing in that direction i */
2209: #endif
2210: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2211: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2212: /* because that direction will be replaced unless the gain del is small */
2213: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2214: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2215: /* with the new direction. */
2216: del=fabs(fptt-(*fret));
2217: ibig=i;
1.126 brouard 2218: }
2219: #ifdef DEBUG
2220: printf("%d %.12e",i,(*fret));
2221: fprintf(ficlog,"%d %.12e",i,(*fret));
2222: for (j=1;j<=n;j++) {
1.224 brouard 2223: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2224: printf(" x(%d)=%.12e",j,xit[j]);
2225: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2226: }
2227: for(j=1;j<=n;j++) {
1.225 brouard 2228: printf(" p(%d)=%.12e",j,p[j]);
2229: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2230: }
2231: printf("\n");
2232: fprintf(ficlog,"\n");
2233: #endif
1.187 brouard 2234: } /* end loop on each direction i */
2235: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2236: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2237: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2238: for(j=1;j<=n;j++) {
1.225 brouard 2239: if(flatdir[j] >0){
2240: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2241: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2242: }
2243: /* printf("\n"); */
2244: /* fprintf(ficlog,"\n"); */
2245: }
1.243 brouard 2246: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2247: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2248: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2249: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2250: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2251: /* decreased of more than 3.84 */
2252: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2253: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2254: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2255:
1.188 brouard 2256: /* Starting the program with initial values given by a former maximization will simply change */
2257: /* the scales of the directions and the directions, because the are reset to canonical directions */
2258: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2259: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2260: #ifdef DEBUG
2261: int k[2],l;
2262: k[0]=1;
2263: k[1]=-1;
2264: printf("Max: %.12e",(*func)(p));
2265: fprintf(ficlog,"Max: %.12e",(*func)(p));
2266: for (j=1;j<=n;j++) {
2267: printf(" %.12e",p[j]);
2268: fprintf(ficlog," %.12e",p[j]);
2269: }
2270: printf("\n");
2271: fprintf(ficlog,"\n");
2272: for(l=0;l<=1;l++) {
2273: for (j=1;j<=n;j++) {
2274: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2275: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2276: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2277: }
2278: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2279: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2280: }
2281: #endif
2282:
1.224 brouard 2283: #ifdef LINMINORIGINAL
2284: #else
2285: free_ivector(flatdir,1,n);
2286: #endif
1.126 brouard 2287: free_vector(xit,1,n);
2288: free_vector(xits,1,n);
2289: free_vector(ptt,1,n);
2290: free_vector(pt,1,n);
2291: return;
1.192 brouard 2292: } /* enough precision */
1.240 brouard 2293: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2294: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2295: ptt[j]=2.0*p[j]-pt[j];
2296: xit[j]=p[j]-pt[j];
2297: pt[j]=p[j];
2298: }
1.181 brouard 2299: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2300: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2301: if (*iter <=4) {
1.225 brouard 2302: #else
2303: #endif
1.224 brouard 2304: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2305: #else
1.161 brouard 2306: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2307: #endif
1.162 brouard 2308: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2309: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2310: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2311: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2312: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2313: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2314: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2315: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2316: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2317: /* Even if f3 <f1, directest can be negative and t >0 */
2318: /* mu² and del² are equal when f3=f1 */
2319: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2320: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2321: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2322: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2323: #ifdef NRCORIGINAL
2324: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2325: #else
2326: 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 2327: t= t- del*SQR(fp-fptt);
1.183 brouard 2328: #endif
1.202 brouard 2329: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2330: #ifdef DEBUG
1.181 brouard 2331: 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);
2332: 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 2333: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2334: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2335: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2336: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2337: 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);
2338: 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);
2339: #endif
1.183 brouard 2340: #ifdef POWELLORIGINAL
2341: if (t < 0.0) { /* Then we use it for new direction */
2342: #else
1.182 brouard 2343: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2344: 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 2345: 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 2346: 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 2347: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2348: }
1.181 brouard 2349: if (directest < 0.0) { /* Then we use it for new direction */
2350: #endif
1.191 brouard 2351: #ifdef DEBUGLINMIN
1.234 brouard 2352: printf("Before linmin in direction P%d-P0\n",n);
2353: for (j=1;j<=n;j++) {
2354: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2355: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2356: if(j % ncovmodel == 0){
2357: printf("\n");
2358: fprintf(ficlog,"\n");
2359: }
2360: }
1.224 brouard 2361: #endif
2362: #ifdef LINMINORIGINAL
1.234 brouard 2363: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2364: #else
1.234 brouard 2365: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2366: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2367: #endif
1.234 brouard 2368:
1.191 brouard 2369: #ifdef DEBUGLINMIN
1.234 brouard 2370: for (j=1;j<=n;j++) {
2371: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2372: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2373: if(j % ncovmodel == 0){
2374: printf("\n");
2375: fprintf(ficlog,"\n");
2376: }
2377: }
1.224 brouard 2378: #endif
1.234 brouard 2379: for (j=1;j<=n;j++) {
2380: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2381: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2382: }
1.224 brouard 2383: #ifdef LINMINORIGINAL
2384: #else
1.234 brouard 2385: for (j=1, flatd=0;j<=n;j++) {
2386: if(flatdir[j]>0)
2387: flatd++;
2388: }
2389: if(flatd >0){
2390: printf("%d flat directions\n",flatd);
2391: fprintf(ficlog,"%d flat directions\n",flatd);
2392: for (j=1;j<=n;j++) {
2393: if(flatdir[j]>0){
2394: printf("%d ",j);
2395: fprintf(ficlog,"%d ",j);
2396: }
2397: }
2398: printf("\n");
2399: fprintf(ficlog,"\n");
2400: }
1.191 brouard 2401: #endif
1.234 brouard 2402: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2403: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2404:
1.126 brouard 2405: #ifdef DEBUG
1.234 brouard 2406: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2407: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2408: for(j=1;j<=n;j++){
2409: printf(" %lf",xit[j]);
2410: fprintf(ficlog," %lf",xit[j]);
2411: }
2412: printf("\n");
2413: fprintf(ficlog,"\n");
1.126 brouard 2414: #endif
1.192 brouard 2415: } /* end of t or directest negative */
1.224 brouard 2416: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2417: #else
1.234 brouard 2418: } /* end if (fptt < fp) */
1.192 brouard 2419: #endif
1.225 brouard 2420: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2421: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2422: #else
1.224 brouard 2423: #endif
1.234 brouard 2424: } /* loop iteration */
1.126 brouard 2425: }
1.234 brouard 2426:
1.126 brouard 2427: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2428:
1.235 brouard 2429: 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 2430: {
1.235 brouard 2431: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2432: (and selected quantitative values in nres)
2433: by left multiplying the unit
1.234 brouard 2434: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2435: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2436: /* Wx is row vector: population in state 1, population in state 2, population dead */
2437: /* or prevalence in state 1, prevalence in state 2, 0 */
2438: /* newm is the matrix after multiplications, its rows are identical at a factor */
2439: /* Initial matrix pimij */
2440: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2441: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2442: /* 0, 0 , 1} */
2443: /*
2444: * and after some iteration: */
2445: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2446: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2447: /* 0, 0 , 1} */
2448: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2449: /* {0.51571254859325999, 0.4842874514067399, */
2450: /* 0.51326036147820708, 0.48673963852179264} */
2451: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2452:
1.126 brouard 2453: int i, ii,j,k;
1.209 brouard 2454: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2455: /* double **matprod2(); */ /* test */
1.218 brouard 2456: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2457: double **newm;
1.209 brouard 2458: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2459: int ncvloop=0;
1.169 brouard 2460:
1.209 brouard 2461: min=vector(1,nlstate);
2462: max=vector(1,nlstate);
2463: meandiff=vector(1,nlstate);
2464:
1.218 brouard 2465: /* Starting with matrix unity */
1.126 brouard 2466: for (ii=1;ii<=nlstate+ndeath;ii++)
2467: for (j=1;j<=nlstate+ndeath;j++){
2468: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2469: }
1.169 brouard 2470:
2471: cov[1]=1.;
2472:
2473: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2474: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2475: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2476: ncvloop++;
1.126 brouard 2477: newm=savm;
2478: /* Covariates have to be included here again */
1.138 brouard 2479: cov[2]=agefin;
1.187 brouard 2480: if(nagesqr==1)
2481: cov[3]= agefin*agefin;;
1.234 brouard 2482: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2483: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2484: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2485: /* 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 2486: }
2487: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2488: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2489: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2490: /* 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 2491: }
1.237 brouard 2492: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2493: if(Dummy[Tvar[Tage[k]]]){
2494: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2495: } else{
1.235 brouard 2496: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2497: }
1.235 brouard 2498: /* 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 2499: }
1.237 brouard 2500: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2501: /* 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 2502: if(Dummy[Tvard[k][1]==0]){
2503: if(Dummy[Tvard[k][2]==0]){
2504: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2505: }else{
2506: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2507: }
2508: }else{
2509: if(Dummy[Tvard[k][2]==0]){
2510: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2511: }else{
2512: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2513: }
2514: }
1.234 brouard 2515: }
1.138 brouard 2516: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2517: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2518: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2519: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2520: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2521: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2522: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2523:
1.126 brouard 2524: savm=oldm;
2525: oldm=newm;
1.209 brouard 2526:
2527: for(j=1; j<=nlstate; j++){
2528: max[j]=0.;
2529: min[j]=1.;
2530: }
2531: for(i=1;i<=nlstate;i++){
2532: sumnew=0;
2533: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2534: for(j=1; j<=nlstate; j++){
2535: prlim[i][j]= newm[i][j]/(1-sumnew);
2536: max[j]=FMAX(max[j],prlim[i][j]);
2537: min[j]=FMIN(min[j],prlim[i][j]);
2538: }
2539: }
2540:
1.126 brouard 2541: maxmax=0.;
1.209 brouard 2542: for(j=1; j<=nlstate; j++){
2543: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2544: maxmax=FMAX(maxmax,meandiff[j]);
2545: /* 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 2546: } /* j loop */
1.203 brouard 2547: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2548: /* 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 2549: if(maxmax < ftolpl){
1.209 brouard 2550: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2551: free_vector(min,1,nlstate);
2552: free_vector(max,1,nlstate);
2553: free_vector(meandiff,1,nlstate);
1.126 brouard 2554: return prlim;
2555: }
1.169 brouard 2556: } /* age loop */
1.208 brouard 2557: /* After some age loop it doesn't converge */
1.209 brouard 2558: 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 2559: 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 2560: /* 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); */
2561: free_vector(min,1,nlstate);
2562: free_vector(max,1,nlstate);
2563: free_vector(meandiff,1,nlstate);
1.208 brouard 2564:
1.169 brouard 2565: return prlim; /* should not reach here */
1.126 brouard 2566: }
2567:
1.217 brouard 2568:
2569: /**** Back Prevalence limit (stable or period prevalence) ****************/
2570:
1.218 brouard 2571: /* 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) */
2572: /* 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 2573: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2574: {
1.218 brouard 2575: /* Computes the prevalence limit in each live state at age x and covariate ij by left multiplying the unit
1.217 brouard 2576: matrix by transitions matrix until convergence is reached with precision ftolpl */
2577: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2578: /* Wx is row vector: population in state 1, population in state 2, population dead */
2579: /* or prevalence in state 1, prevalence in state 2, 0 */
2580: /* newm is the matrix after multiplications, its rows are identical at a factor */
2581: /* Initial matrix pimij */
2582: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2583: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2584: /* 0, 0 , 1} */
2585: /*
2586: * and after some iteration: */
2587: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2588: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2589: /* 0, 0 , 1} */
2590: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2591: /* {0.51571254859325999, 0.4842874514067399, */
2592: /* 0.51326036147820708, 0.48673963852179264} */
2593: /* If we start from prlim again, prlim tends to a constant matrix */
2594:
2595: int i, ii,j,k;
1.247 brouard 2596: int first=0;
1.217 brouard 2597: double *min, *max, *meandiff, maxmax,sumnew=0.;
2598: /* double **matprod2(); */ /* test */
2599: double **out, cov[NCOVMAX+1], **bmij();
2600: double **newm;
1.218 brouard 2601: double **dnewm, **doldm, **dsavm; /* for use */
2602: double **oldm, **savm; /* for use */
2603:
1.217 brouard 2604: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2605: int ncvloop=0;
2606:
2607: min=vector(1,nlstate);
2608: max=vector(1,nlstate);
2609: meandiff=vector(1,nlstate);
2610:
1.218 brouard 2611: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2612: oldm=oldms; savm=savms;
2613:
2614: /* Starting with matrix unity */
2615: for (ii=1;ii<=nlstate+ndeath;ii++)
2616: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2617: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2618: }
2619:
2620: cov[1]=1.;
2621:
2622: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2623: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2624: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2625: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2626: ncvloop++;
1.218 brouard 2627: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2628: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2629: /* Covariates have to be included here again */
2630: cov[2]=agefin;
2631: if(nagesqr==1)
2632: cov[3]= agefin*agefin;;
1.242 brouard 2633: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2634: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2635: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
2636: /* 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)); */
2637: }
2638: /* for (k=1; k<=cptcovn;k++) { */
2639: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2640: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2641: /* /\* 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])]); *\/ */
2642: /* } */
2643: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2644: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2645: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2646: /* 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]); */
2647: }
2648: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2649: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2650: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2651: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2652: for (k=1; k<=cptcovage;k++){ /* For product with age */
2653: if(Dummy[Tvar[Tage[k]]]){
2654: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2655: } else{
2656: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2657: }
2658: /* 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]); */
2659: }
2660: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2661: /* 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]); */
2662: if(Dummy[Tvard[k][1]==0]){
2663: if(Dummy[Tvard[k][2]==0]){
2664: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2665: }else{
2666: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2667: }
2668: }else{
2669: if(Dummy[Tvard[k][2]==0]){
2670: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2671: }else{
2672: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2673: }
2674: }
1.217 brouard 2675: }
2676:
2677: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2678: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2679: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2680: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2681: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2682: /* ij should be linked to the correct index of cov */
2683: /* age and covariate values ij are in 'cov', but we need to pass
2684: * ij for the observed prevalence at age and status and covariate
2685: * number: prevacurrent[(int)agefin][ii][ij]
2686: */
2687: /* 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 *\/ */
2688: /* 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 *\/ */
2689: 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 2690: savm=oldm;
2691: oldm=newm;
2692: for(j=1; j<=nlstate; j++){
2693: max[j]=0.;
2694: min[j]=1.;
2695: }
2696: for(j=1; j<=nlstate; j++){
2697: for(i=1;i<=nlstate;i++){
1.234 brouard 2698: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2699: bprlim[i][j]= newm[i][j];
2700: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2701: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2702: }
2703: }
1.218 brouard 2704:
1.217 brouard 2705: maxmax=0.;
2706: for(i=1; i<=nlstate; i++){
2707: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2708: maxmax=FMAX(maxmax,meandiff[i]);
2709: /* 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); */
2710: } /* j loop */
2711: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2712: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2713: if(maxmax < ftolpl){
1.220 brouard 2714: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2715: free_vector(min,1,nlstate);
2716: free_vector(max,1,nlstate);
2717: free_vector(meandiff,1,nlstate);
2718: return bprlim;
2719: }
2720: } /* age loop */
2721: /* After some age loop it doesn't converge */
1.247 brouard 2722: if(first){
2723: first=1;
2724: 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\
2725: 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);
2726: }
2727: 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 2728: 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);
2729: /* 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); */
2730: free_vector(min,1,nlstate);
2731: free_vector(max,1,nlstate);
2732: free_vector(meandiff,1,nlstate);
2733:
2734: return bprlim; /* should not reach here */
2735: }
2736:
1.126 brouard 2737: /*************** transition probabilities ***************/
2738:
2739: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2740: {
1.138 brouard 2741: /* According to parameters values stored in x and the covariate's values stored in cov,
2742: computes the probability to be observed in state j being in state i by appying the
2743: model to the ncovmodel covariates (including constant and age).
2744: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2745: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2746: ncth covariate in the global vector x is given by the formula:
2747: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2748: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2749: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2750: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2751: Outputs ps[i][j] the probability to be observed in j being in j according to
2752: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2753: */
2754: double s1, lnpijopii;
1.126 brouard 2755: /*double t34;*/
1.164 brouard 2756: int i,j, nc, ii, jj;
1.126 brouard 2757:
1.223 brouard 2758: for(i=1; i<= nlstate; i++){
2759: for(j=1; j<i;j++){
2760: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2761: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2762: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2763: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2764: }
2765: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2766: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2767: }
2768: for(j=i+1; j<=nlstate+ndeath;j++){
2769: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2770: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2771: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2772: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2773: }
2774: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2775: }
2776: }
1.218 brouard 2777:
1.223 brouard 2778: for(i=1; i<= nlstate; i++){
2779: s1=0;
2780: for(j=1; j<i; j++){
2781: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2782: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2783: }
2784: for(j=i+1; j<=nlstate+ndeath; j++){
2785: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2786: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2787: }
2788: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2789: ps[i][i]=1./(s1+1.);
2790: /* Computing other pijs */
2791: for(j=1; j<i; j++)
2792: ps[i][j]= exp(ps[i][j])*ps[i][i];
2793: for(j=i+1; j<=nlstate+ndeath; j++)
2794: ps[i][j]= exp(ps[i][j])*ps[i][i];
2795: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2796: } /* end i */
1.218 brouard 2797:
1.223 brouard 2798: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2799: for(jj=1; jj<= nlstate+ndeath; jj++){
2800: ps[ii][jj]=0;
2801: ps[ii][ii]=1;
2802: }
2803: }
1.218 brouard 2804:
2805:
1.223 brouard 2806: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2807: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2808: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2809: /* } */
2810: /* printf("\n "); */
2811: /* } */
2812: /* printf("\n ");printf("%lf ",cov[2]);*/
2813: /*
2814: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2815: goto end;*/
1.223 brouard 2816: return ps;
1.126 brouard 2817: }
2818:
1.218 brouard 2819: /*************** backward transition probabilities ***************/
2820:
2821: /* 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 ) */
2822: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2823: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2824: {
1.222 brouard 2825: /* Computes the backward probability at age agefin and covariate ij
2826: * and returns in **ps as well as **bmij.
2827: */
1.218 brouard 2828: int i, ii, j,k;
1.222 brouard 2829:
2830: double **out, **pmij();
2831: double sumnew=0.;
1.218 brouard 2832: double agefin;
1.222 brouard 2833:
2834: double **dnewm, **dsavm, **doldm;
2835: double **bbmij;
2836:
1.218 brouard 2837: doldm=ddoldms; /* global pointers */
1.222 brouard 2838: dnewm=ddnewms;
2839: dsavm=ddsavms;
2840:
2841: agefin=cov[2];
2842: /* bmij *//* age is cov[2], ij is included in cov, but we need for
2843: the observed prevalence (with this covariate ij) */
2844: dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate);
2845: /* We do have the matrix Px in savm and we need pij */
2846: for (j=1;j<=nlstate+ndeath;j++){
2847: sumnew=0.; /* w1 p11 + w2 p21 only on live states */
2848: for (ii=1;ii<=nlstate;ii++){
2849: sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij];
2850: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
2851: for (ii=1;ii<=nlstate+ndeath;ii++){
2852: if(sumnew >= 1.e-10){
2853: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2854: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2855: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2856: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2857: /* }else */
2858: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
2859: }else{
1.242 brouard 2860: ;
2861: /* 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 2862: }
2863: } /*End ii */
2864: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] */
2865: /* left Product of this diag matrix by dsavm=Px (newm=dsavm*doldm) */
2866: bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
2867: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2868: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2869: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2870: /* left Product of this matrix by diag matrix of prevalences (savm) */
2871: for (j=1;j<=nlstate+ndeath;j++){
2872: for (ii=1;ii<=nlstate+ndeath;ii++){
2873: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2874: }
2875: } /* End j, At the end oldm is diag[1/(w_1p1i+w_2 p2i)] */
2876: ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
2877: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2878: /* end bmij */
2879: return ps;
1.218 brouard 2880: }
1.217 brouard 2881: /*************** transition probabilities ***************/
2882:
1.218 brouard 2883: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2884: {
2885: /* According to parameters values stored in x and the covariate's values stored in cov,
2886: computes the probability to be observed in state j being in state i by appying the
2887: model to the ncovmodel covariates (including constant and age).
2888: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2889: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2890: ncth covariate in the global vector x is given by the formula:
2891: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2892: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2893: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2894: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2895: Outputs ps[i][j] the probability to be observed in j being in j according to
2896: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2897: */
2898: double s1, lnpijopii;
2899: /*double t34;*/
2900: int i,j, nc, ii, jj;
2901:
1.234 brouard 2902: for(i=1; i<= nlstate; i++){
2903: for(j=1; j<i;j++){
2904: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2905: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2906: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2907: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2908: }
2909: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2910: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2911: }
2912: for(j=i+1; j<=nlstate+ndeath;j++){
2913: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2914: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2915: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2916: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2917: }
2918: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2919: }
2920: }
2921:
2922: for(i=1; i<= nlstate; i++){
2923: s1=0;
2924: for(j=1; j<i; j++){
2925: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2926: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2927: }
2928: for(j=i+1; j<=nlstate+ndeath; j++){
2929: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2930: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2931: }
2932: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2933: ps[i][i]=1./(s1+1.);
2934: /* Computing other pijs */
2935: for(j=1; j<i; j++)
2936: ps[i][j]= exp(ps[i][j])*ps[i][i];
2937: for(j=i+1; j<=nlstate+ndeath; j++)
2938: ps[i][j]= exp(ps[i][j])*ps[i][i];
2939: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2940: } /* end i */
2941:
2942: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2943: for(jj=1; jj<= nlstate+ndeath; jj++){
2944: ps[ii][jj]=0;
2945: ps[ii][ii]=1;
2946: }
2947: }
2948: /* Added for backcast */ /* Transposed matrix too */
2949: for(jj=1; jj<= nlstate+ndeath; jj++){
2950: s1=0.;
2951: for(ii=1; ii<= nlstate+ndeath; ii++){
2952: s1+=ps[ii][jj];
2953: }
2954: for(ii=1; ii<= nlstate; ii++){
2955: ps[ii][jj]=ps[ii][jj]/s1;
2956: }
2957: }
2958: /* Transposition */
2959: for(jj=1; jj<= nlstate+ndeath; jj++){
2960: for(ii=jj; ii<= nlstate+ndeath; ii++){
2961: s1=ps[ii][jj];
2962: ps[ii][jj]=ps[jj][ii];
2963: ps[jj][ii]=s1;
2964: }
2965: }
2966: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2967: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2968: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2969: /* } */
2970: /* printf("\n "); */
2971: /* } */
2972: /* printf("\n ");printf("%lf ",cov[2]);*/
2973: /*
2974: for(i=1; i<= npar; i++) printf("%f ",x[i]);
2975: goto end;*/
2976: return ps;
1.217 brouard 2977: }
2978:
2979:
1.126 brouard 2980: /**************** Product of 2 matrices ******************/
2981:
1.145 brouard 2982: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 2983: {
2984: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
2985: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
2986: /* in, b, out are matrice of pointers which should have been initialized
2987: before: only the contents of out is modified. The function returns
2988: a pointer to pointers identical to out */
1.145 brouard 2989: int i, j, k;
1.126 brouard 2990: for(i=nrl; i<= nrh; i++)
1.145 brouard 2991: for(k=ncolol; k<=ncoloh; k++){
2992: out[i][k]=0.;
2993: for(j=ncl; j<=nch; j++)
2994: out[i][k] +=in[i][j]*b[j][k];
2995: }
1.126 brouard 2996: return out;
2997: }
2998:
2999:
3000: /************* Higher Matrix Product ***************/
3001:
1.235 brouard 3002: 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 3003: {
1.218 brouard 3004: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3005: 'nhstepm*hstepm*stepm' months (i.e. until
3006: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3007: nhstepm*hstepm matrices.
3008: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3009: (typically every 2 years instead of every month which is too big
3010: for the memory).
3011: Model is determined by parameters x and covariates have to be
3012: included manually here.
3013:
3014: */
3015:
3016: int i, j, d, h, k;
1.131 brouard 3017: double **out, cov[NCOVMAX+1];
1.126 brouard 3018: double **newm;
1.187 brouard 3019: double agexact;
1.214 brouard 3020: double agebegin, ageend;
1.126 brouard 3021:
3022: /* Hstepm could be zero and should return the unit matrix */
3023: for (i=1;i<=nlstate+ndeath;i++)
3024: for (j=1;j<=nlstate+ndeath;j++){
3025: oldm[i][j]=(i==j ? 1.0 : 0.0);
3026: po[i][j][0]=(i==j ? 1.0 : 0.0);
3027: }
3028: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3029: for(h=1; h <=nhstepm; h++){
3030: for(d=1; d <=hstepm; d++){
3031: newm=savm;
3032: /* Covariates have to be included here again */
3033: cov[1]=1.;
1.214 brouard 3034: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3035: cov[2]=agexact;
3036: if(nagesqr==1)
1.227 brouard 3037: cov[3]= agexact*agexact;
1.235 brouard 3038: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3039: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3040: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3041: /* 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)); */
3042: }
3043: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3044: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3045: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3046: /* 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]); */
3047: }
3048: for (k=1; k<=cptcovage;k++){
3049: if(Dummy[Tvar[Tage[k]]]){
3050: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3051: } else{
3052: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3053: }
3054: /* 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]); */
3055: }
3056: for (k=1; k<=cptcovprod;k++){ /* */
3057: /* 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]); */
3058: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3059: }
3060: /* for (k=1; k<=cptcovn;k++) */
3061: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3062: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3063: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3064: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3065: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3066:
3067:
1.126 brouard 3068: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3069: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3070: /* right multiplication of oldm by the current matrix */
1.126 brouard 3071: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3072: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3073: /* if((int)age == 70){ */
3074: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3075: /* for(i=1; i<=nlstate+ndeath; i++) { */
3076: /* printf("%d pmmij ",i); */
3077: /* for(j=1;j<=nlstate+ndeath;j++) { */
3078: /* printf("%f ",pmmij[i][j]); */
3079: /* } */
3080: /* printf(" oldm "); */
3081: /* for(j=1;j<=nlstate+ndeath;j++) { */
3082: /* printf("%f ",oldm[i][j]); */
3083: /* } */
3084: /* printf("\n"); */
3085: /* } */
3086: /* } */
1.126 brouard 3087: savm=oldm;
3088: oldm=newm;
3089: }
3090: for(i=1; i<=nlstate+ndeath; i++)
3091: for(j=1;j<=nlstate+ndeath;j++) {
1.218 brouard 3092: po[i][j][h]=newm[i][j];
3093: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3094: }
1.128 brouard 3095: /*printf("h=%d ",h);*/
1.126 brouard 3096: } /* end h */
1.218 brouard 3097: /* printf("\n H=%d \n",h); */
1.126 brouard 3098: return po;
3099: }
3100:
1.217 brouard 3101: /************* Higher Back Matrix Product ***************/
1.218 brouard 3102: /* 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 3103: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij )
1.217 brouard 3104: {
1.218 brouard 3105: /* Computes the transition matrix starting at age 'age' over
1.217 brouard 3106: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3107: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3108: nhstepm*hstepm matrices.
3109: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3110: (typically every 2 years instead of every month which is too big
1.217 brouard 3111: for the memory).
1.218 brouard 3112: Model is determined by parameters x and covariates have to be
3113: included manually here.
1.217 brouard 3114:
1.222 brouard 3115: */
1.217 brouard 3116:
3117: int i, j, d, h, k;
3118: double **out, cov[NCOVMAX+1];
3119: double **newm;
3120: double agexact;
3121: double agebegin, ageend;
1.222 brouard 3122: double **oldm, **savm;
1.217 brouard 3123:
1.222 brouard 3124: oldm=oldms;savm=savms;
1.217 brouard 3125: /* Hstepm could be zero and should return the unit matrix */
3126: for (i=1;i<=nlstate+ndeath;i++)
3127: for (j=1;j<=nlstate+ndeath;j++){
3128: oldm[i][j]=(i==j ? 1.0 : 0.0);
3129: po[i][j][0]=(i==j ? 1.0 : 0.0);
3130: }
3131: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3132: for(h=1; h <=nhstepm; h++){
3133: for(d=1; d <=hstepm; d++){
3134: newm=savm;
3135: /* Covariates have to be included here again */
3136: cov[1]=1.;
3137: agexact=age-((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
3138: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3139: cov[2]=agexact;
3140: if(nagesqr==1)
1.222 brouard 3141: cov[3]= agexact*agexact;
1.218 brouard 3142: for (k=1; k<=cptcovn;k++)
1.222 brouard 3143: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)];
3144: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.217 brouard 3145: for (k=1; k<=cptcovage;k++) /* Should start at cptcovn+1 */
1.222 brouard 3146: /* cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
3147: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3148: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
1.217 brouard 3149: for (k=1; k<=cptcovprod;k++) /* Useless because included in cptcovn */
1.222 brouard 3150: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
3151: /* 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 3152:
3153:
1.217 brouard 3154: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3155: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3156: /* Careful transposed matrix */
1.222 brouard 3157: /* age is in cov[2] */
1.218 brouard 3158: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3159: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3160: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3161: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3162: /* if((int)age == 70){ */
3163: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3164: /* for(i=1; i<=nlstate+ndeath; i++) { */
3165: /* printf("%d pmmij ",i); */
3166: /* for(j=1;j<=nlstate+ndeath;j++) { */
3167: /* printf("%f ",pmmij[i][j]); */
3168: /* } */
3169: /* printf(" oldm "); */
3170: /* for(j=1;j<=nlstate+ndeath;j++) { */
3171: /* printf("%f ",oldm[i][j]); */
3172: /* } */
3173: /* printf("\n"); */
3174: /* } */
3175: /* } */
3176: savm=oldm;
3177: oldm=newm;
3178: }
3179: for(i=1; i<=nlstate+ndeath; i++)
3180: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3181: po[i][j][h]=newm[i][j];
3182: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3183: }
3184: /*printf("h=%d ",h);*/
3185: } /* end h */
1.222 brouard 3186: /* printf("\n H=%d \n",h); */
1.217 brouard 3187: return po;
3188: }
3189:
3190:
1.162 brouard 3191: #ifdef NLOPT
3192: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3193: double fret;
3194: double *xt;
3195: int j;
3196: myfunc_data *d2 = (myfunc_data *) pd;
3197: /* xt = (p1-1); */
3198: xt=vector(1,n);
3199: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3200:
3201: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3202: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3203: printf("Function = %.12lf ",fret);
3204: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3205: printf("\n");
3206: free_vector(xt,1,n);
3207: return fret;
3208: }
3209: #endif
1.126 brouard 3210:
3211: /*************** log-likelihood *************/
3212: double func( double *x)
3213: {
1.226 brouard 3214: int i, ii, j, k, mi, d, kk;
3215: int ioffset=0;
3216: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3217: double **out;
3218: double lli; /* Individual log likelihood */
3219: int s1, s2;
1.228 brouard 3220: 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 3221: double bbh, survp;
3222: long ipmx;
3223: double agexact;
3224: /*extern weight */
3225: /* We are differentiating ll according to initial status */
3226: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3227: /*for(i=1;i<imx;i++)
3228: printf(" %d\n",s[4][i]);
3229: */
1.162 brouard 3230:
1.226 brouard 3231: ++countcallfunc;
1.162 brouard 3232:
1.226 brouard 3233: cov[1]=1.;
1.126 brouard 3234:
1.226 brouard 3235: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3236: ioffset=0;
1.226 brouard 3237: if(mle==1){
3238: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3239: /* Computes the values of the ncovmodel covariates of the model
3240: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3241: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3242: to be observed in j being in i according to the model.
3243: */
1.243 brouard 3244: ioffset=2+nagesqr ;
1.233 brouard 3245: /* Fixed */
1.234 brouard 3246: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3247: 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)*/
3248: }
1.226 brouard 3249: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3250: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3251: has been calculated etc */
3252: /* For an individual i, wav[i] gives the number of effective waves */
3253: /* We compute the contribution to Likelihood of each effective transition
3254: mw[mi][i] is real wave of the mi th effectve wave */
3255: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3256: s2=s[mw[mi+1][i]][i];
3257: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3258: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3259: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3260: */
3261: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3262: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3263: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3264: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3265: }
3266: for (ii=1;ii<=nlstate+ndeath;ii++)
3267: for (j=1;j<=nlstate+ndeath;j++){
3268: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3269: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3270: }
3271: for(d=0; d<dh[mi][i]; d++){
3272: newm=savm;
3273: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3274: cov[2]=agexact;
3275: if(nagesqr==1)
3276: cov[3]= agexact*agexact; /* Should be changed here */
3277: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3278: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3279: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3280: else
3281: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3282: }
3283: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3284: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3285: savm=oldm;
3286: oldm=newm;
3287: } /* end mult */
3288:
3289: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3290: /* But now since version 0.9 we anticipate for bias at large stepm.
3291: * If stepm is larger than one month (smallest stepm) and if the exact delay
3292: * (in months) between two waves is not a multiple of stepm, we rounded to
3293: * the nearest (and in case of equal distance, to the lowest) interval but now
3294: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3295: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3296: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3297: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3298: * -stepm/2 to stepm/2 .
3299: * For stepm=1 the results are the same as for previous versions of Imach.
3300: * For stepm > 1 the results are less biased than in previous versions.
3301: */
1.234 brouard 3302: s1=s[mw[mi][i]][i];
3303: s2=s[mw[mi+1][i]][i];
3304: bbh=(double)bh[mi][i]/(double)stepm;
3305: /* bias bh is positive if real duration
3306: * is higher than the multiple of stepm and negative otherwise.
3307: */
3308: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3309: if( s2 > nlstate){
3310: /* i.e. if s2 is a death state and if the date of death is known
3311: then the contribution to the likelihood is the probability to
3312: die between last step unit time and current step unit time,
3313: which is also equal to probability to die before dh
3314: minus probability to die before dh-stepm .
3315: In version up to 0.92 likelihood was computed
3316: as if date of death was unknown. Death was treated as any other
3317: health state: the date of the interview describes the actual state
3318: and not the date of a change in health state. The former idea was
3319: to consider that at each interview the state was recorded
3320: (healthy, disable or death) and IMaCh was corrected; but when we
3321: introduced the exact date of death then we should have modified
3322: the contribution of an exact death to the likelihood. This new
3323: contribution is smaller and very dependent of the step unit
3324: stepm. It is no more the probability to die between last interview
3325: and month of death but the probability to survive from last
3326: interview up to one month before death multiplied by the
3327: probability to die within a month. Thanks to Chris
3328: Jackson for correcting this bug. Former versions increased
3329: mortality artificially. The bad side is that we add another loop
3330: which slows down the processing. The difference can be up to 10%
3331: lower mortality.
3332: */
3333: /* If, at the beginning of the maximization mostly, the
3334: cumulative probability or probability to be dead is
3335: constant (ie = 1) over time d, the difference is equal to
3336: 0. out[s1][3] = savm[s1][3]: probability, being at state
3337: s1 at precedent wave, to be dead a month before current
3338: wave is equal to probability, being at state s1 at
3339: precedent wave, to be dead at mont of the current
3340: wave. Then the observed probability (that this person died)
3341: is null according to current estimated parameter. In fact,
3342: it should be very low but not zero otherwise the log go to
3343: infinity.
3344: */
1.183 brouard 3345: /* #ifdef INFINITYORIGINAL */
3346: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3347: /* #else */
3348: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3349: /* lli=log(mytinydouble); */
3350: /* else */
3351: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3352: /* #endif */
1.226 brouard 3353: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3354:
1.226 brouard 3355: } else if ( s2==-1 ) { /* alive */
3356: for (j=1,survp=0. ; j<=nlstate; j++)
3357: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3358: /*survp += out[s1][j]; */
3359: lli= log(survp);
3360: }
3361: else if (s2==-4) {
3362: for (j=3,survp=0. ; j<=nlstate; j++)
3363: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3364: lli= log(survp);
3365: }
3366: else if (s2==-5) {
3367: for (j=1,survp=0. ; j<=2; j++)
3368: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3369: lli= log(survp);
3370: }
3371: else{
3372: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3373: /* 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 */
3374: }
3375: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3376: /*if(lli ==000.0)*/
3377: /*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); */
3378: ipmx +=1;
3379: sw += weight[i];
3380: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3381: /* if (lli < log(mytinydouble)){ */
3382: /* 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); */
3383: /* 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]); */
3384: /* } */
3385: } /* end of wave */
3386: } /* end of individual */
3387: } else if(mle==2){
3388: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3389: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3390: for(mi=1; mi<= wav[i]-1; mi++){
3391: for (ii=1;ii<=nlstate+ndeath;ii++)
3392: for (j=1;j<=nlstate+ndeath;j++){
3393: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3394: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3395: }
3396: for(d=0; d<=dh[mi][i]; d++){
3397: newm=savm;
3398: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3399: cov[2]=agexact;
3400: if(nagesqr==1)
3401: cov[3]= agexact*agexact;
3402: for (kk=1; kk<=cptcovage;kk++) {
3403: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3404: }
3405: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3406: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3407: savm=oldm;
3408: oldm=newm;
3409: } /* end mult */
3410:
3411: s1=s[mw[mi][i]][i];
3412: s2=s[mw[mi+1][i]][i];
3413: bbh=(double)bh[mi][i]/(double)stepm;
3414: 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 */
3415: ipmx +=1;
3416: sw += weight[i];
3417: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3418: } /* end of wave */
3419: } /* end of individual */
3420: } else if(mle==3){ /* exponential inter-extrapolation */
3421: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3422: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3423: for(mi=1; mi<= wav[i]-1; mi++){
3424: for (ii=1;ii<=nlstate+ndeath;ii++)
3425: for (j=1;j<=nlstate+ndeath;j++){
3426: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3427: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3428: }
3429: for(d=0; d<dh[mi][i]; d++){
3430: newm=savm;
3431: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3432: cov[2]=agexact;
3433: if(nagesqr==1)
3434: cov[3]= agexact*agexact;
3435: for (kk=1; kk<=cptcovage;kk++) {
3436: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3437: }
3438: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3439: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3440: savm=oldm;
3441: oldm=newm;
3442: } /* end mult */
3443:
3444: s1=s[mw[mi][i]][i];
3445: s2=s[mw[mi+1][i]][i];
3446: bbh=(double)bh[mi][i]/(double)stepm;
3447: 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 */
3448: ipmx +=1;
3449: sw += weight[i];
3450: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3451: } /* end of wave */
3452: } /* end of individual */
3453: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3454: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3455: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3456: for(mi=1; mi<= wav[i]-1; mi++){
3457: for (ii=1;ii<=nlstate+ndeath;ii++)
3458: for (j=1;j<=nlstate+ndeath;j++){
3459: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3460: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3461: }
3462: for(d=0; d<dh[mi][i]; d++){
3463: newm=savm;
3464: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3465: cov[2]=agexact;
3466: if(nagesqr==1)
3467: cov[3]= agexact*agexact;
3468: for (kk=1; kk<=cptcovage;kk++) {
3469: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3470: }
1.126 brouard 3471:
1.226 brouard 3472: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3473: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3474: savm=oldm;
3475: oldm=newm;
3476: } /* end mult */
3477:
3478: s1=s[mw[mi][i]][i];
3479: s2=s[mw[mi+1][i]][i];
3480: if( s2 > nlstate){
3481: lli=log(out[s1][s2] - savm[s1][s2]);
3482: } else if ( s2==-1 ) { /* alive */
3483: for (j=1,survp=0. ; j<=nlstate; j++)
3484: survp += out[s1][j];
3485: lli= log(survp);
3486: }else{
3487: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3488: }
3489: ipmx +=1;
3490: sw += weight[i];
3491: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3492: /* 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 3493: } /* end of wave */
3494: } /* end of individual */
3495: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3496: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3497: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3498: for(mi=1; mi<= wav[i]-1; mi++){
3499: for (ii=1;ii<=nlstate+ndeath;ii++)
3500: for (j=1;j<=nlstate+ndeath;j++){
3501: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3502: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3503: }
3504: for(d=0; d<dh[mi][i]; d++){
3505: newm=savm;
3506: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3507: cov[2]=agexact;
3508: if(nagesqr==1)
3509: cov[3]= agexact*agexact;
3510: for (kk=1; kk<=cptcovage;kk++) {
3511: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3512: }
1.126 brouard 3513:
1.226 brouard 3514: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3515: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3516: savm=oldm;
3517: oldm=newm;
3518: } /* end mult */
3519:
3520: s1=s[mw[mi][i]][i];
3521: s2=s[mw[mi+1][i]][i];
3522: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3523: ipmx +=1;
3524: sw += weight[i];
3525: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3526: /*printf("i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],out[s1][s2],savm[s1][s2]);*/
3527: } /* end of wave */
3528: } /* end of individual */
3529: } /* End of if */
3530: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3531: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3532: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3533: return -l;
1.126 brouard 3534: }
3535:
3536: /*************** log-likelihood *************/
3537: double funcone( double *x)
3538: {
1.228 brouard 3539: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3540: int i, ii, j, k, mi, d, kk;
1.228 brouard 3541: int ioffset=0;
1.131 brouard 3542: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3543: double **out;
3544: double lli; /* Individual log likelihood */
3545: double llt;
3546: int s1, s2;
1.228 brouard 3547: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3548:
1.126 brouard 3549: double bbh, survp;
1.187 brouard 3550: double agexact;
1.214 brouard 3551: double agebegin, ageend;
1.126 brouard 3552: /*extern weight */
3553: /* We are differentiating ll according to initial status */
3554: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3555: /*for(i=1;i<imx;i++)
3556: printf(" %d\n",s[4][i]);
3557: */
3558: cov[1]=1.;
3559:
3560: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3561: ioffset=0;
3562: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3563: /* ioffset=2+nagesqr+cptcovage; */
3564: ioffset=2+nagesqr;
1.232 brouard 3565: /* Fixed */
1.224 brouard 3566: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3567: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3568: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3569: 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)*/
3570: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3571: /* cov[2+6]=covar[Tvar[6]][i]; */
3572: /* cov[2+6]=covar[2][i]; V2 */
3573: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3574: /* cov[2+7]=covar[Tvar[7]][i]; */
3575: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3576: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3577: /* cov[2+9]=covar[Tvar[9]][i]; */
3578: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3579: }
1.232 brouard 3580: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3581: /* 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?)*\/ */
3582: /* } */
1.231 brouard 3583: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3584: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3585: /* } */
1.225 brouard 3586:
1.233 brouard 3587:
3588: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3589: /* Wave varying (but not age varying) */
3590: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3591: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3592: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3593: }
1.232 brouard 3594: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3595: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3596: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3597: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3598: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3599: /* 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 3600: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3601: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3602: /* /\* 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]); *\/ */
3603: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3604: /* } */
1.126 brouard 3605: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3606: for (j=1;j<=nlstate+ndeath;j++){
3607: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3608: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3609: }
1.214 brouard 3610:
3611: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3612: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3613: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3614: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3615: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3616: and mw[mi+1][i]. dh depends on stepm.*/
3617: newm=savm;
1.247 brouard 3618: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3619: cov[2]=agexact;
3620: if(nagesqr==1)
3621: cov[3]= agexact*agexact;
3622: for (kk=1; kk<=cptcovage;kk++) {
3623: if(!FixedV[Tvar[Tage[kk]]])
3624: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3625: else
3626: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3627: }
3628: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3629: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3630: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3631: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3632: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3633: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3634: savm=oldm;
3635: oldm=newm;
1.126 brouard 3636: } /* end mult */
3637:
3638: s1=s[mw[mi][i]][i];
3639: s2=s[mw[mi+1][i]][i];
1.217 brouard 3640: /* if(s2==-1){ */
3641: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3642: /* /\* exit(1); *\/ */
3643: /* } */
1.126 brouard 3644: bbh=(double)bh[mi][i]/(double)stepm;
3645: /* bias is positive if real duration
3646: * is higher than the multiple of stepm and negative otherwise.
3647: */
3648: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3649: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3650: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3651: for (j=1,survp=0. ; j<=nlstate; j++)
3652: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3653: lli= log(survp);
1.126 brouard 3654: }else if (mle==1){
1.242 brouard 3655: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3656: } else if(mle==2){
1.242 brouard 3657: 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 3658: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3659: 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 3660: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3661: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3662: } else{ /* mle=0 back to 1 */
1.242 brouard 3663: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3664: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3665: } /* End of if */
3666: ipmx +=1;
3667: sw += weight[i];
3668: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3669: /*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 3670: if(globpr){
1.246 brouard 3671: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3672: %11.6f %11.6f %11.6f ", \
1.242 brouard 3673: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3674: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3675: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3676: llt +=ll[k]*gipmx/gsw;
3677: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3678: }
3679: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3680: }
1.232 brouard 3681: } /* end of wave */
3682: } /* end of individual */
3683: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3684: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3685: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3686: if(globpr==0){ /* First time we count the contributions and weights */
3687: gipmx=ipmx;
3688: gsw=sw;
3689: }
3690: return -l;
1.126 brouard 3691: }
3692:
3693:
3694: /*************** function likelione ***********/
3695: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3696: {
3697: /* This routine should help understanding what is done with
3698: the selection of individuals/waves and
3699: to check the exact contribution to the likelihood.
3700: Plotting could be done.
3701: */
3702: int k;
3703:
3704: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3705: strcpy(fileresilk,"ILK_");
1.202 brouard 3706: strcat(fileresilk,fileresu);
1.126 brouard 3707: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3708: printf("Problem with resultfile: %s\n", fileresilk);
3709: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3710: }
1.214 brouard 3711: 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");
3712: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3713: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3714: for(k=1; k<=nlstate; k++)
3715: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3716: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3717: }
3718:
3719: *fretone=(*funcone)(p);
3720: if(*globpri !=0){
3721: fclose(ficresilk);
1.205 brouard 3722: if (mle ==0)
3723: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3724: else if(mle >=1)
3725: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3726: 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 3727:
1.208 brouard 3728:
3729: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3730: 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 3731: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3732: }
1.207 brouard 3733: 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 3734: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3735: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3736: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3737: fflush(fichtm);
1.205 brouard 3738: }
1.126 brouard 3739: return;
3740: }
3741:
3742:
3743: /*********** Maximum Likelihood Estimation ***************/
3744:
3745: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3746: {
1.165 brouard 3747: int i,j, iter=0;
1.126 brouard 3748: double **xi;
3749: double fret;
3750: double fretone; /* Only one call to likelihood */
3751: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3752:
3753: #ifdef NLOPT
3754: int creturn;
3755: nlopt_opt opt;
3756: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3757: double *lb;
3758: double minf; /* the minimum objective value, upon return */
3759: double * p1; /* Shifted parameters from 0 instead of 1 */
3760: myfunc_data dinst, *d = &dinst;
3761: #endif
3762:
3763:
1.126 brouard 3764: xi=matrix(1,npar,1,npar);
3765: for (i=1;i<=npar;i++)
3766: for (j=1;j<=npar;j++)
3767: xi[i][j]=(i==j ? 1.0 : 0.0);
3768: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3769: strcpy(filerespow,"POW_");
1.126 brouard 3770: strcat(filerespow,fileres);
3771: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3772: printf("Problem with resultfile: %s\n", filerespow);
3773: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3774: }
3775: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3776: for (i=1;i<=nlstate;i++)
3777: for(j=1;j<=nlstate+ndeath;j++)
3778: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3779: fprintf(ficrespow,"\n");
1.162 brouard 3780: #ifdef POWELL
1.126 brouard 3781: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3782: #endif
1.126 brouard 3783:
1.162 brouard 3784: #ifdef NLOPT
3785: #ifdef NEWUOA
3786: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3787: #else
3788: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3789: #endif
3790: lb=vector(0,npar-1);
3791: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3792: nlopt_set_lower_bounds(opt, lb);
3793: nlopt_set_initial_step1(opt, 0.1);
3794:
3795: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3796: d->function = func;
3797: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3798: nlopt_set_min_objective(opt, myfunc, d);
3799: nlopt_set_xtol_rel(opt, ftol);
3800: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3801: printf("nlopt failed! %d\n",creturn);
3802: }
3803: else {
3804: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3805: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3806: iter=1; /* not equal */
3807: }
3808: nlopt_destroy(opt);
3809: #endif
1.126 brouard 3810: free_matrix(xi,1,npar,1,npar);
3811: fclose(ficrespow);
1.203 brouard 3812: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3813: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3814: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3815:
3816: }
3817:
3818: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3819: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3820: {
3821: double **a,**y,*x,pd;
1.203 brouard 3822: /* double **hess; */
1.164 brouard 3823: int i, j;
1.126 brouard 3824: int *indx;
3825:
3826: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3827: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3828: void lubksb(double **a, int npar, int *indx, double b[]) ;
3829: void ludcmp(double **a, int npar, int *indx, double *d) ;
3830: double gompertz(double p[]);
1.203 brouard 3831: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3832:
3833: printf("\nCalculation of the hessian matrix. Wait...\n");
3834: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3835: for (i=1;i<=npar;i++){
1.203 brouard 3836: printf("%d-",i);fflush(stdout);
3837: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3838:
3839: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3840:
3841: /* printf(" %f ",p[i]);
3842: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3843: }
3844:
3845: for (i=1;i<=npar;i++) {
3846: for (j=1;j<=npar;j++) {
3847: if (j>i) {
1.203 brouard 3848: printf(".%d-%d",i,j);fflush(stdout);
3849: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3850: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3851:
3852: hess[j][i]=hess[i][j];
3853: /*printf(" %lf ",hess[i][j]);*/
3854: }
3855: }
3856: }
3857: printf("\n");
3858: fprintf(ficlog,"\n");
3859:
3860: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3861: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3862:
3863: a=matrix(1,npar,1,npar);
3864: y=matrix(1,npar,1,npar);
3865: x=vector(1,npar);
3866: indx=ivector(1,npar);
3867: for (i=1;i<=npar;i++)
3868: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3869: ludcmp(a,npar,indx,&pd);
3870:
3871: for (j=1;j<=npar;j++) {
3872: for (i=1;i<=npar;i++) x[i]=0;
3873: x[j]=1;
3874: lubksb(a,npar,indx,x);
3875: for (i=1;i<=npar;i++){
3876: matcov[i][j]=x[i];
3877: }
3878: }
3879:
3880: printf("\n#Hessian matrix#\n");
3881: fprintf(ficlog,"\n#Hessian matrix#\n");
3882: for (i=1;i<=npar;i++) {
3883: for (j=1;j<=npar;j++) {
1.203 brouard 3884: printf("%.6e ",hess[i][j]);
3885: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3886: }
3887: printf("\n");
3888: fprintf(ficlog,"\n");
3889: }
3890:
1.203 brouard 3891: /* printf("\n#Covariance matrix#\n"); */
3892: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3893: /* for (i=1;i<=npar;i++) { */
3894: /* for (j=1;j<=npar;j++) { */
3895: /* printf("%.6e ",matcov[i][j]); */
3896: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3897: /* } */
3898: /* printf("\n"); */
3899: /* fprintf(ficlog,"\n"); */
3900: /* } */
3901:
1.126 brouard 3902: /* Recompute Inverse */
1.203 brouard 3903: /* for (i=1;i<=npar;i++) */
3904: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
3905: /* ludcmp(a,npar,indx,&pd); */
3906:
3907: /* printf("\n#Hessian matrix recomputed#\n"); */
3908:
3909: /* for (j=1;j<=npar;j++) { */
3910: /* for (i=1;i<=npar;i++) x[i]=0; */
3911: /* x[j]=1; */
3912: /* lubksb(a,npar,indx,x); */
3913: /* for (i=1;i<=npar;i++){ */
3914: /* y[i][j]=x[i]; */
3915: /* printf("%.3e ",y[i][j]); */
3916: /* fprintf(ficlog,"%.3e ",y[i][j]); */
3917: /* } */
3918: /* printf("\n"); */
3919: /* fprintf(ficlog,"\n"); */
3920: /* } */
3921:
3922: /* Verifying the inverse matrix */
3923: #ifdef DEBUGHESS
3924: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 3925:
1.203 brouard 3926: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
3927: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 3928:
3929: for (j=1;j<=npar;j++) {
3930: for (i=1;i<=npar;i++){
1.203 brouard 3931: printf("%.2f ",y[i][j]);
3932: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 3933: }
3934: printf("\n");
3935: fprintf(ficlog,"\n");
3936: }
1.203 brouard 3937: #endif
1.126 brouard 3938:
3939: free_matrix(a,1,npar,1,npar);
3940: free_matrix(y,1,npar,1,npar);
3941: free_vector(x,1,npar);
3942: free_ivector(indx,1,npar);
1.203 brouard 3943: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 3944:
3945:
3946: }
3947:
3948: /*************** hessian matrix ****************/
3949: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 3950: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 3951: int i;
3952: int l=1, lmax=20;
1.203 brouard 3953: double k1,k2, res, fx;
1.132 brouard 3954: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 3955: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
3956: int k=0,kmax=10;
3957: double l1;
3958:
3959: fx=func(x);
3960: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 3961: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 3962: l1=pow(10,l);
3963: delts=delt;
3964: for(k=1 ; k <kmax; k=k+1){
3965: delt = delta*(l1*k);
3966: p2[theta]=x[theta] +delt;
1.145 brouard 3967: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 3968: p2[theta]=x[theta]-delt;
3969: k2=func(p2)-fx;
3970: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 3971: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 3972:
1.203 brouard 3973: #ifdef DEBUGHESSII
1.126 brouard 3974: 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);
3975: 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);
3976: #endif
3977: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
3978: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
3979: k=kmax;
3980: }
3981: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 3982: k=kmax; l=lmax*10;
1.126 brouard 3983: }
3984: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
3985: delts=delt;
3986: }
1.203 brouard 3987: } /* End loop k */
1.126 brouard 3988: }
3989: delti[theta]=delts;
3990: return res;
3991:
3992: }
3993:
1.203 brouard 3994: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 3995: {
3996: int i;
1.164 brouard 3997: int l=1, lmax=20;
1.126 brouard 3998: double k1,k2,k3,k4,res,fx;
1.132 brouard 3999: double p2[MAXPARM+1];
1.203 brouard 4000: int k, kmax=1;
4001: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4002:
4003: int firstime=0;
1.203 brouard 4004:
1.126 brouard 4005: fx=func(x);
1.203 brouard 4006: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4007: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4008: p2[thetai]=x[thetai]+delti[thetai]*k;
4009: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4010: k1=func(p2)-fx;
4011:
1.203 brouard 4012: p2[thetai]=x[thetai]+delti[thetai]*k;
4013: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4014: k2=func(p2)-fx;
4015:
1.203 brouard 4016: p2[thetai]=x[thetai]-delti[thetai]*k;
4017: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4018: k3=func(p2)-fx;
4019:
1.203 brouard 4020: p2[thetai]=x[thetai]-delti[thetai]*k;
4021: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4022: k4=func(p2)-fx;
1.203 brouard 4023: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4024: if(k1*k2*k3*k4 <0.){
1.208 brouard 4025: firstime=1;
1.203 brouard 4026: kmax=kmax+10;
1.208 brouard 4027: }
4028: if(kmax >=10 || firstime ==1){
1.246 brouard 4029: 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);
4030: 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 4031: 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);
4032: 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);
4033: }
4034: #ifdef DEBUGHESSIJ
4035: v1=hess[thetai][thetai];
4036: v2=hess[thetaj][thetaj];
4037: cv12=res;
4038: /* Computing eigen value of Hessian matrix */
4039: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4040: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4041: if ((lc2 <0) || (lc1 <0) ){
4042: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4043: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4044: 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);
4045: 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);
4046: }
1.126 brouard 4047: #endif
4048: }
4049: return res;
4050: }
4051:
1.203 brouard 4052: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4053: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4054: /* { */
4055: /* int i; */
4056: /* int l=1, lmax=20; */
4057: /* double k1,k2,k3,k4,res,fx; */
4058: /* double p2[MAXPARM+1]; */
4059: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4060: /* int k=0,kmax=10; */
4061: /* double l1; */
4062:
4063: /* fx=func(x); */
4064: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4065: /* l1=pow(10,l); */
4066: /* delts=delt; */
4067: /* for(k=1 ; k <kmax; k=k+1){ */
4068: /* delt = delti*(l1*k); */
4069: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4070: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4071: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4072: /* k1=func(p2)-fx; */
4073:
4074: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4075: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4076: /* k2=func(p2)-fx; */
4077:
4078: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4079: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4080: /* k3=func(p2)-fx; */
4081:
4082: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4083: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4084: /* k4=func(p2)-fx; */
4085: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4086: /* #ifdef DEBUGHESSIJ */
4087: /* 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); */
4088: /* 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); */
4089: /* #endif */
4090: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4091: /* k=kmax; */
4092: /* } */
4093: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4094: /* k=kmax; l=lmax*10; */
4095: /* } */
4096: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4097: /* delts=delt; */
4098: /* } */
4099: /* } /\* End loop k *\/ */
4100: /* } */
4101: /* delti[theta]=delts; */
4102: /* return res; */
4103: /* } */
4104:
4105:
1.126 brouard 4106: /************** Inverse of matrix **************/
4107: void ludcmp(double **a, int n, int *indx, double *d)
4108: {
4109: int i,imax,j,k;
4110: double big,dum,sum,temp;
4111: double *vv;
4112:
4113: vv=vector(1,n);
4114: *d=1.0;
4115: for (i=1;i<=n;i++) {
4116: big=0.0;
4117: for (j=1;j<=n;j++)
4118: if ((temp=fabs(a[i][j])) > big) big=temp;
4119: if (big == 0.0) nrerror("Singular matrix in routine ludcmp");
4120: vv[i]=1.0/big;
4121: }
4122: for (j=1;j<=n;j++) {
4123: for (i=1;i<j;i++) {
4124: sum=a[i][j];
4125: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4126: a[i][j]=sum;
4127: }
4128: big=0.0;
4129: for (i=j;i<=n;i++) {
4130: sum=a[i][j];
4131: for (k=1;k<j;k++)
4132: sum -= a[i][k]*a[k][j];
4133: a[i][j]=sum;
4134: if ( (dum=vv[i]*fabs(sum)) >= big) {
4135: big=dum;
4136: imax=i;
4137: }
4138: }
4139: if (j != imax) {
4140: for (k=1;k<=n;k++) {
4141: dum=a[imax][k];
4142: a[imax][k]=a[j][k];
4143: a[j][k]=dum;
4144: }
4145: *d = -(*d);
4146: vv[imax]=vv[j];
4147: }
4148: indx[j]=imax;
4149: if (a[j][j] == 0.0) a[j][j]=TINY;
4150: if (j != n) {
4151: dum=1.0/(a[j][j]);
4152: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4153: }
4154: }
4155: free_vector(vv,1,n); /* Doesn't work */
4156: ;
4157: }
4158:
4159: void lubksb(double **a, int n, int *indx, double b[])
4160: {
4161: int i,ii=0,ip,j;
4162: double sum;
4163:
4164: for (i=1;i<=n;i++) {
4165: ip=indx[i];
4166: sum=b[ip];
4167: b[ip]=b[i];
4168: if (ii)
4169: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4170: else if (sum) ii=i;
4171: b[i]=sum;
4172: }
4173: for (i=n;i>=1;i--) {
4174: sum=b[i];
4175: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4176: b[i]=sum/a[i][i];
4177: }
4178: }
4179:
4180: void pstamp(FILE *fichier)
4181: {
1.196 brouard 4182: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4183: }
4184:
4185: /************ Frequencies ********************/
1.250 ! brouard 4186: void freqsummary(char fileres[], double p[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4187: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4188: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 ! brouard 4189: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4190:
1.250 ! brouard 4191: int i, m, jk, j1, bool, z1,j, k, iv, jj=0;
1.226 brouard 4192: int iind=0, iage=0;
4193: int mi; /* Effective wave */
4194: int first;
4195: double ***freq; /* Frequencies */
4196: double *meanq;
4197: double **meanqt;
4198: double *pp, **prop, *posprop, *pospropt;
4199: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4200: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4201: double agebegin, ageend;
4202:
4203: pp=vector(1,nlstate);
4204: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4205: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4206: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4207: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4208: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4209: meanqt=matrix(1,lastpass,1,nqtveff);
4210: strcpy(fileresp,"P_");
4211: strcat(fileresp,fileresu);
4212: /*strcat(fileresphtm,fileresu);*/
4213: if((ficresp=fopen(fileresp,"w"))==NULL) {
4214: printf("Problem with prevalence resultfile: %s\n", fileresp);
4215: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4216: exit(0);
4217: }
1.240 brouard 4218:
1.226 brouard 4219: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4220: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4221: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4222: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4223: fflush(ficlog);
4224: exit(70);
4225: }
4226: else{
4227: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4228: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4229: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4230: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4231: }
1.237 brouard 4232: 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 4233:
1.226 brouard 4234: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4235: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4236: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4237: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4238: fflush(ficlog);
4239: exit(70);
1.240 brouard 4240: } else{
1.226 brouard 4241: 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 4242: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4243: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4244: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4245: }
1.240 brouard 4246: 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);
4247:
1.226 brouard 4248: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4249: j1=0;
1.126 brouard 4250:
1.227 brouard 4251: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4252: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4253: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4254:
4255:
1.226 brouard 4256: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4257: reference=low_education V1=0,V2=0
4258: med_educ V1=1 V2=0,
4259: high_educ V1=0 V2=1
4260: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4261: */
1.249 brouard 4262: dateintsum=0;
4263: k2cpt=0;
4264:
4265: for (j = 0; j <= cptcoveff; j+=cptcoveff){
4266: first=1;
1.227 brouard 4267: 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 */
1.226 brouard 4268: posproptt=0.;
4269: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4270: scanf("%d", i);*/
4271: for (i=-5; i<=nlstate+ndeath; i++)
4272: for (jk=-5; jk<=nlstate+ndeath; jk++)
1.240 brouard 4273: for(m=iagemin; m <= iagemax+3; m++)
4274: freq[i][jk][m]=0;
4275:
1.226 brouard 4276: for (i=1; i<=nlstate; i++) {
4277: for(m=iagemin; m <= iagemax+3; m++)
1.240 brouard 4278: prop[i][m]=0;
1.226 brouard 4279: posprop[i]=0;
4280: pospropt[i]=0;
4281: }
1.227 brouard 4282: /* for (z1=1; z1<= nqfveff; z1++) { */
4283: /* meanq[z1]+=0.; */
4284: /* for(m=1;m<=lastpass;m++){ */
4285: /* meanqt[m][z1]=0.; */
4286: /* } */
4287: /* } */
1.240 brouard 4288:
1.249 brouard 4289: /* dateintsum=0; */
4290: /* k2cpt=0; */
4291:
1.227 brouard 4292: /* For that combination of covariate j1, we count and print the frequencies in one pass */
1.226 brouard 4293: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4294: bool=1;
1.249 brouard 4295: if(j !=0){
1.227 brouard 4296: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.234 brouard 4297: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
1.227 brouard 4298: /* for (z1=1; z1<= nqfveff; z1++) { */
4299: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4300: /* } */
1.250 ! brouard 4301: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.234 brouard 4302: /* if(Tvaraff[z1] ==-20){ */
4303: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4304: /* }else if(Tvaraff[z1] ==-10){ */
4305: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4306: /* }else */
1.250 ! brouard 4307: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
! 4308: /* Tests if this individual iind responded to combination j1 (V4=1 V3=0) */
! 4309: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.234 brouard 4310: /* 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",
4311: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4312: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4313: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4314: } /* Onlyf fixed */
4315: } /* end z1 */
4316: } /* cptcovn > 0 */
1.227 brouard 4317: } /* end any */
1.249 brouard 4318: }/* end j==0 */
1.227 brouard 4319: if (bool==1){ /* We selected an individual iind satisfying combination j1 or all fixed */
1.234 brouard 4320: /* for(m=firstpass; m<=lastpass; m++){ */
4321: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4322: m=mw[mi][iind];
1.249 brouard 4323: if(j!=0){
1.234 brouard 4324: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4325: for (z1=1; z1<=cptcoveff; z1++) {
4326: if( Fixed[Tmodelind[z1]]==1){
4327: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.250 ! brouard 4328: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
! 4329: value is -1, we don't select. It differs from the
! 4330: constant and age model which counts them. */
1.249 brouard 4331: bool=0; /* not selected */
1.234 brouard 4332: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4333: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4334: bool=0;
4335: }
4336: }
4337: }
4338: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
1.249 brouard 4339: } /* end j==0 */
1.234 brouard 4340: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4341: if(bool==1){
4342: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4343: and mw[mi+1][iind]. dh depends on stepm. */
4344: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4345: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4346: if(m >=firstpass && m <=lastpass){
4347: k2=anint[m][iind]+(mint[m][iind]/12.);
4348: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4349: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4350: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4351: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4352: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4353: if (m<lastpass) {
4354: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4355: /* 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]); */
4356: if(s[m][iind]==-1)
4357: 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.));
4358: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.250 ! brouard 4359: /* if((int)agev[m][iind] == 55) */
! 4360: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
1.234 brouard 4361: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4362: 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 */
4363: }
4364: } /* end if between passes */
1.249 brouard 4365: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4366: dateintsum=dateintsum+k2; /* on all covariates ?*/
1.234 brouard 4367: k2cpt++;
4368: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
4369: }
1.250 ! brouard 4370: }else{
! 4371: bool=1;
! 4372: }/* end bool 2 */
1.234 brouard 4373: } /* end m */
1.226 brouard 4374: } /* end bool */
4375: } /* end iind = 1 to imx */
4376: /* prop[s][age] is feeded for any initial and valid live state as well as
4377: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
1.240 brouard 4378:
4379:
1.226 brouard 4380: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
4381: pstamp(ficresp);
1.249 brouard 4382: if (cptcoveff>0 && j!=0){
1.226 brouard 4383: fprintf(ficresp, "\n#********** Variable ");
4384: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4385: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
1.240 brouard 4386: fprintf(ficlog, "\n#********** Variable ");
1.227 brouard 4387: for (z1=1; z1<=cptcoveff; z1++){
1.240 brouard 4388: if(DummyV[z1]){
4389: fprintf(ficresp, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4390: fprintf(ficresphtm, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4391: fprintf(ficresphtmfr, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4392: fprintf(ficlog, "V%d (fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4393: }else{
4394: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4395: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4396: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4397: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4398: }
1.226 brouard 4399: }
4400: fprintf(ficresp, "**********\n#");
4401: fprintf(ficresphtm, "**********</h3>\n");
4402: fprintf(ficresphtmfr, "**********</h3>\n");
4403: fprintf(ficlog, "**********\n");
4404: }
4405: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
4406: for(i=1; i<=nlstate;i++) {
1.240 brouard 4407: fprintf(ficresp, " Age Prev(%d) N(%d) N ",i,i);
1.226 brouard 4408: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4409: }
4410: fprintf(ficresp, "\n");
4411: fprintf(ficresphtm, "\n");
1.240 brouard 4412:
1.226 brouard 4413: /* Header of frequency table by age */
4414: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4415: fprintf(ficresphtmfr,"<th>Age</th> ");
4416: for(jk=-1; jk <=nlstate+ndeath; jk++){
4417: for(m=-1; m <=nlstate+ndeath; m++){
1.234 brouard 4418: if(jk!=0 && m!=0)
4419: fprintf(ficresphtmfr,"<th>%d%d</th> ",jk,m);
1.226 brouard 4420: }
4421: }
4422: fprintf(ficresphtmfr, "\n");
1.240 brouard 4423:
1.226 brouard 4424: /* For each age */
4425: for(iage=iagemin; iage <= iagemax+3; iage++){
4426: fprintf(ficresphtm,"<tr>");
4427: if(iage==iagemax+1){
1.240 brouard 4428: fprintf(ficlog,"1");
4429: fprintf(ficresphtmfr,"<tr><th>0</th> ");
1.226 brouard 4430: }else if(iage==iagemax+2){
1.240 brouard 4431: fprintf(ficlog,"0");
4432: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
1.226 brouard 4433: }else if(iage==iagemax+3){
1.240 brouard 4434: fprintf(ficlog,"Total");
4435: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
1.226 brouard 4436: }else{
1.240 brouard 4437: if(first==1){
4438: first=0;
4439: printf("See log file for details...\n");
4440: }
4441: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4442: fprintf(ficlog,"Age %d", iage);
1.226 brouard 4443: }
4444: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4445: for(m=-1, pp[jk]=0; m <=nlstate+ndeath ; m++)
4446: pp[jk] += freq[jk][m][iage];
1.226 brouard 4447: }
4448: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4449: for(m=-1, pos=0; m <=0 ; m++)
4450: pos += freq[jk][m][iage];
4451: if(pp[jk]>=1.e-10){
4452: if(first==1){
4453: printf(" %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4454: }
4455: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",jk,pp[jk],jk,100*pos/pp[jk]);
4456: }else{
4457: if(first==1)
4458: printf(" %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4459: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",jk,pp[jk],jk);
4460: }
1.226 brouard 4461: }
1.240 brouard 4462:
1.226 brouard 4463: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4464: /* posprop[jk]=0; */
4465: for(m=0, pp[jk]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4466: pp[jk] += freq[jk][m][iage];
1.226 brouard 4467: } /* pp[jk] is the total number of transitions starting from state jk and any ending status until this age */
1.240 brouard 4468:
1.226 brouard 4469: for(jk=1,pos=0, pospropta=0.; jk <=nlstate ; jk++){
1.240 brouard 4470: pos += pp[jk]; /* pos is the total number of transitions until this age */
4471: posprop[jk] += prop[jk][iage]; /* prop is the number of transitions from a live state
4472: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4473: pospropta += prop[jk][iage]; /* prop is the number of transitions from a live state
4474: from jk at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
1.226 brouard 4475: }
4476: for(jk=1; jk <=nlstate ; jk++){
1.240 brouard 4477: if(pos>=1.e-5){
4478: if(first==1)
4479: printf(" %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4480: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",jk,pp[jk],jk,100*pp[jk]/pos);
4481: }else{
4482: if(first==1)
4483: printf(" %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4484: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",jk,pp[jk],jk);
4485: }
4486: if( iage <= iagemax){
4487: if(pos>=1.e-5){
4488: fprintf(ficresp," %d %.5f %.0f %.0f",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4489: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[jk][iage]/pospropta, prop[jk][iage],pospropta);
4490: /*probs[iage][jk][j1]= pp[jk]/pos;*/
4491: /*printf("\niage=%d jk=%d j1=%d %.5f %.0f %.0f %f",iage,jk,j1,pp[jk]/pos, pp[jk],pos,probs[iage][jk][j1]);*/
4492: }
4493: else{
4494: fprintf(ficresp," %d NaNq %.0f %.0f",iage,prop[jk][iage],pospropta);
4495: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[jk][iage],pospropta);
4496: }
4497: }
4498: pospropt[jk] +=posprop[jk];
1.226 brouard 4499: } /* end loop jk */
4500: /* pospropt=0.; */
4501: for(jk=-1; jk <=nlstate+ndeath; jk++){
1.240 brouard 4502: for(m=-1; m <=nlstate+ndeath; m++){
4503: if(freq[jk][m][iage] !=0 ) { /* minimizing output */
4504: if(first==1){
4505: printf(" %d%d=%.0f",jk,m,freq[jk][m][iage]);
4506: }
4507: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iage]);
4508: }
4509: if(jk!=0 && m!=0)
4510: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[jk][m][iage]);
4511: }
1.226 brouard 4512: } /* end loop jk */
4513: posproptt=0.;
4514: for(jk=1; jk <=nlstate; jk++){
1.240 brouard 4515: posproptt += pospropt[jk];
1.226 brouard 4516: }
4517: fprintf(ficresphtmfr,"</tr>\n ");
4518: if(iage <= iagemax){
1.240 brouard 4519: fprintf(ficresp,"\n");
4520: fprintf(ficresphtm,"</tr>\n");
1.226 brouard 4521: }
4522: if(first==1)
1.240 brouard 4523: printf("Others in log...\n");
1.226 brouard 4524: fprintf(ficlog,"\n");
4525: } /* end loop age iage */
4526: fprintf(ficresphtm,"<tr><th>Tot</th>");
4527: for(jk=1; jk <=nlstate ; jk++){
4528: if(posproptt < 1.e-5){
1.240 brouard 4529: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[jk],posproptt);
1.226 brouard 4530: }else{
1.240 brouard 4531: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[jk]/posproptt,pospropt[jk],posproptt);
1.226 brouard 4532: }
4533: }
4534: fprintf(ficresphtm,"</tr>\n");
4535: fprintf(ficresphtm,"</table>\n");
4536: fprintf(ficresphtmfr,"</table>\n");
4537: if(posproptt < 1.e-5){
4538: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4539: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4540: fprintf(ficres,"\n This combination (%d) is not valid and no result will be produced\n\n",j1);
4541: invalidvarcomb[j1]=1;
4542: }else{
4543: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4544: invalidvarcomb[j1]=0;
4545: }
4546: fprintf(ficresphtmfr,"</table>\n");
4547: } /* end selected combination of covariate j1 */
1.249 brouard 4548: if(j==0){ /* We can estimate starting values from the occurences in each case */
1.250 ! brouard 4549: printf("#Freqsummary\n");
! 4550: fprintf(ficlog,"\n");
! 4551: for(i=1,jk=1; i <=nlstate; i++){
! 4552: for(k=1; k <=(nlstate+ndeath); k++){
! 4553: if (k != i) {
! 4554: printf("%d%d ",i,k);
! 4555: fprintf(ficlog,"%d%d ",i,k);
! 4556: for(jj=1; jj <=ncovmodel; jj++){
! 4557: if(jj==1){
! 4558: printf("%12.7f ln(%12.1f/%12.1f)= %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]));
! 4559: fprintf(ficlog,"%12.7f ln(%12.1f/%12.1f)= %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]));
! 4560: }
! 4561: /* printf("%12.7f )", param[i][jj][k]); */
! 4562: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
! 4563: jk++;
! 4564: }
! 4565: printf("\n");
! 4566: fprintf(ficlog,"\n");
! 4567: }
! 4568: }
! 4569: }
! 4570: printf("#Freqsummary\n");
! 4571: fprintf(ficlog,"\n");
1.249 brouard 4572: for(jk=-1; jk <=nlstate+ndeath; jk++){
4573: for(m=-1; m <=nlstate+ndeath; m++){
4574: /* param[i]|j][k]= freq[jk][m][iagemax+3] */
1.250 ! brouard 4575: printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
! 4576: fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]);
! 4577: /* if(freq[jk][m][iage] !=0 ) { /\* minimizing output *\/ */
! 4578: /* printf(" %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
! 4579: /* fprintf(ficlog," %d%d=%.0f",jk,m,freq[jk][m][iagemax+3]); */
! 4580: /* } */
1.249 brouard 4581: }
4582: } /* end loop jk */
1.250 ! brouard 4583: printf("\n");
! 4584: fprintf(ficlog,"\n");
! 4585: } /* if j=0 */
1.249 brouard 4586: } /* end j */
1.226 brouard 4587: dateintmean=dateintsum/k2cpt;
1.240 brouard 4588:
1.226 brouard 4589: fclose(ficresp);
4590: fclose(ficresphtm);
4591: fclose(ficresphtmfr);
4592: free_vector(meanq,1,nqfveff);
4593: free_matrix(meanqt,1,lastpass,1,nqtveff);
4594: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4595: free_vector(pospropt,1,nlstate);
4596: free_vector(posprop,1,nlstate);
4597: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+3+AGEMARGE);
4598: free_vector(pp,1,nlstate);
4599: /* End of freqsummary */
4600: }
1.126 brouard 4601:
4602: /************ Prevalence ********************/
1.227 brouard 4603: 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)
4604: {
4605: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4606: in each health status at the date of interview (if between dateprev1 and dateprev2).
4607: We still use firstpass and lastpass as another selection.
4608: */
1.126 brouard 4609:
1.227 brouard 4610: int i, m, jk, j1, bool, z1,j, iv;
4611: int mi; /* Effective wave */
4612: int iage;
4613: double agebegin, ageend;
4614:
4615: double **prop;
4616: double posprop;
4617: double y2; /* in fractional years */
4618: int iagemin, iagemax;
4619: int first; /** to stop verbosity which is redirected to log file */
4620:
4621: iagemin= (int) agemin;
4622: iagemax= (int) agemax;
4623: /*pp=vector(1,nlstate);*/
4624: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4625: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4626: j1=0;
1.222 brouard 4627:
1.227 brouard 4628: /*j=cptcoveff;*/
4629: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4630:
1.227 brouard 4631: first=1;
4632: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4633: for (i=1; i<=nlstate; i++)
4634: for(iage=iagemin-AGEMARGE; iage <= iagemax+3+AGEMARGE; iage++)
4635: prop[i][iage]=0.0;
4636: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4637: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4638: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4639:
4640: for (i=1; i<=imx; i++) { /* Each individual */
4641: bool=1;
4642: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4643: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4644: m=mw[mi][i];
4645: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4646: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4647: for (z1=1; z1<=cptcoveff; z1++){
4648: if( Fixed[Tmodelind[z1]]==1){
4649: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4650: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4651: bool=0;
4652: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4653: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4654: bool=0;
4655: }
4656: }
4657: if(bool==1){ /* Otherwise we skip that wave/person */
4658: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4659: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4660: if(m >=firstpass && m <=lastpass){
4661: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4662: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4663: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4664: if(agev[m][i]==1) agev[m][i]=iagemax+2;
4665: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+3+AGEMARGE){
4666: 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);
4667: exit(1);
4668: }
4669: if (s[m][i]>0 && s[m][i]<=nlstate) {
4670: /*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]]);*/
4671: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4672: prop[s[m][i]][iagemax+3] += weight[i];
4673: } /* end valid statuses */
4674: } /* end selection of dates */
4675: } /* end selection of waves */
4676: } /* end bool */
4677: } /* end wave */
4678: } /* end individual */
4679: for(i=iagemin; i <= iagemax+3; i++){
4680: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4681: posprop += prop[jk][i];
4682: }
4683:
4684: for(jk=1; jk <=nlstate ; jk++){
4685: if( i <= iagemax){
4686: if(posprop>=1.e-5){
4687: probs[i][jk][j1]= prop[jk][i]/posprop;
4688: } else{
4689: if(first==1){
4690: first=0;
4691: 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]);
4692: }
4693: }
4694: }
4695: }/* end jk */
4696: }/* end i */
1.222 brouard 4697: /*} *//* end i1 */
1.227 brouard 4698: } /* end j1 */
1.222 brouard 4699:
1.227 brouard 4700: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
4701: /*free_vector(pp,1,nlstate);*/
4702: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+3+AGEMARGE);
4703: } /* End of prevalence */
1.126 brouard 4704:
4705: /************* Waves Concatenation ***************/
4706:
4707: 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)
4708: {
4709: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
4710: Death is a valid wave (if date is known).
4711: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
4712: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4713: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 4714: */
1.126 brouard 4715:
1.224 brouard 4716: int i=0, mi=0, m=0, mli=0;
1.126 brouard 4717: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
4718: double sum=0., jmean=0.;*/
1.224 brouard 4719: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 4720: int j, k=0,jk, ju, jl;
4721: double sum=0.;
4722: first=0;
1.214 brouard 4723: firstwo=0;
1.217 brouard 4724: firsthree=0;
1.218 brouard 4725: firstfour=0;
1.164 brouard 4726: jmin=100000;
1.126 brouard 4727: jmax=-1;
4728: jmean=0.;
1.224 brouard 4729:
4730: /* Treating live states */
1.214 brouard 4731: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 4732: mi=0; /* First valid wave */
1.227 brouard 4733: mli=0; /* Last valid wave */
1.126 brouard 4734: m=firstpass;
1.214 brouard 4735: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 4736: 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 */
4737: mli=m-1;/* mw[++mi][i]=m-1; */
4738: }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 */
4739: mw[++mi][i]=m;
4740: mli=m;
1.224 brouard 4741: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
4742: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 4743: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 4744: }
1.227 brouard 4745: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 4746: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 4747: break;
1.224 brouard 4748: #else
1.227 brouard 4749: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
4750: if(firsthree == 0){
4751: 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);
4752: firsthree=1;
4753: }
4754: 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);
4755: mw[++mi][i]=m;
4756: mli=m;
4757: }
4758: if(s[m][i]==-2){ /* Vital status is really unknown */
4759: nbwarn++;
4760: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
4761: 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);
4762: 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);
4763: }
4764: break;
4765: }
4766: break;
1.224 brouard 4767: #endif
1.227 brouard 4768: }/* End m >= lastpass */
1.126 brouard 4769: }/* end while */
1.224 brouard 4770:
1.227 brouard 4771: /* 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 4772: /* After last pass */
1.224 brouard 4773: /* Treating death states */
1.214 brouard 4774: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 4775: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
4776: /* } */
1.126 brouard 4777: mi++; /* Death is another wave */
4778: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 4779: /* Only death is a correct wave */
1.126 brouard 4780: mw[mi][i]=m;
1.224 brouard 4781: }
4782: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.227 brouard 4783: 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 4784: /* m++; */
4785: /* mi++; */
4786: /* s[m][i]=nlstate+1; /\* We are setting the status to the last of non live state *\/ */
4787: /* mw[mi][i]=m; */
1.218 brouard 4788: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 4789: 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 */
4790: nbwarn++;
4791: if(firstfiv==0){
4792: 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 );
4793: firstfiv=1;
4794: }else{
4795: 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 );
4796: }
4797: }else{ /* Death occured afer last wave potential bias */
4798: nberr++;
4799: if(firstwo==0){
4800: 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 );
4801: firstwo=1;
4802: }
4803: 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 );
4804: }
1.218 brouard 4805: }else{ /* end date of interview is known */
1.227 brouard 4806: /* death is known but not confirmed by death status at any wave */
4807: if(firstfour==0){
4808: 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 );
4809: firstfour=1;
4810: }
4811: 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 4812: }
1.224 brouard 4813: } /* end if date of death is known */
4814: #endif
4815: wav[i]=mi; /* mi should be the last effective wave (or mli) */
4816: /* wav[i]=mw[mi][i]; */
1.126 brouard 4817: if(mi==0){
4818: nbwarn++;
4819: if(first==0){
1.227 brouard 4820: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
4821: first=1;
1.126 brouard 4822: }
4823: if(first==1){
1.227 brouard 4824: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 4825: }
4826: } /* end mi==0 */
4827: } /* End individuals */
1.214 brouard 4828: /* wav and mw are no more changed */
1.223 brouard 4829:
1.214 brouard 4830:
1.126 brouard 4831: for(i=1; i<=imx; i++){
4832: for(mi=1; mi<wav[i];mi++){
4833: if (stepm <=0)
1.227 brouard 4834: dh[mi][i]=1;
1.126 brouard 4835: else{
1.227 brouard 4836: if (s[mw[mi+1][i]][i] > nlstate) { /* A death */
4837: if (agedc[i] < 2*AGESUP) {
4838: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
4839: if(j==0) j=1; /* Survives at least one month after exam */
4840: else if(j<0){
4841: nberr++;
4842: 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]);
4843: j=1; /* Temporary Dangerous patch */
4844: 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);
4845: 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]);
4846: 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);
4847: }
4848: k=k+1;
4849: if (j >= jmax){
4850: jmax=j;
4851: ijmax=i;
4852: }
4853: if (j <= jmin){
4854: jmin=j;
4855: ijmin=i;
4856: }
4857: sum=sum+j;
4858: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
4859: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
4860: }
4861: }
4862: else{
4863: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 4864: /* 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 4865:
1.227 brouard 4866: k=k+1;
4867: if (j >= jmax) {
4868: jmax=j;
4869: ijmax=i;
4870: }
4871: else if (j <= jmin){
4872: jmin=j;
4873: ijmin=i;
4874: }
4875: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
4876: /*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]);*/
4877: if(j<0){
4878: nberr++;
4879: 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]);
4880: 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]);
4881: }
4882: sum=sum+j;
4883: }
4884: jk= j/stepm;
4885: jl= j -jk*stepm;
4886: ju= j -(jk+1)*stepm;
4887: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
4888: if(jl==0){
4889: dh[mi][i]=jk;
4890: bh[mi][i]=0;
4891: }else{ /* We want a negative bias in order to only have interpolation ie
4892: * to avoid the price of an extra matrix product in likelihood */
4893: dh[mi][i]=jk+1;
4894: bh[mi][i]=ju;
4895: }
4896: }else{
4897: if(jl <= -ju){
4898: dh[mi][i]=jk;
4899: bh[mi][i]=jl; /* bias is positive if real duration
4900: * is higher than the multiple of stepm and negative otherwise.
4901: */
4902: }
4903: else{
4904: dh[mi][i]=jk+1;
4905: bh[mi][i]=ju;
4906: }
4907: if(dh[mi][i]==0){
4908: dh[mi][i]=1; /* At least one step */
4909: bh[mi][i]=ju; /* At least one step */
4910: /* 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);*/
4911: }
4912: } /* end if mle */
1.126 brouard 4913: }
4914: } /* end wave */
4915: }
4916: jmean=sum/k;
4917: 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 4918: 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 4919: }
1.126 brouard 4920:
4921: /*********** Tricode ****************************/
1.220 brouard 4922: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 4923: {
4924: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
4925: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
4926: * Boring subroutine which should only output nbcode[Tvar[j]][k]
4927: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
4928: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
4929: */
1.130 brouard 4930:
1.242 brouard 4931: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
4932: int modmaxcovj=0; /* Modality max of covariates j */
4933: int cptcode=0; /* Modality max of covariates j */
4934: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 4935:
4936:
1.242 brouard 4937: /* cptcoveff=0; */
4938: /* *cptcov=0; */
1.126 brouard 4939:
1.242 brouard 4940: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 4941:
1.242 brouard 4942: /* Loop on covariates without age and products and no quantitative variable */
4943: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
4944: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
4945: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
4946: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
4947: switch(Fixed[k]) {
4948: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
4949: 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*/
4950: ij=(int)(covar[Tvar[k]][i]);
4951: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
4952: * If product of Vn*Vm, still boolean *:
4953: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
4954: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
4955: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
4956: modality of the nth covariate of individual i. */
4957: if (ij > modmaxcovj)
4958: modmaxcovj=ij;
4959: else if (ij < modmincovj)
4960: modmincovj=ij;
4961: if ((ij < -1) && (ij > NCOVMAX)){
4962: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
4963: exit(1);
4964: }else
4965: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
4966: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
4967: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
4968: /* getting the maximum value of the modality of the covariate
4969: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
4970: female ies 1, then modmaxcovj=1.
4971: */
4972: } /* end for loop on individuals i */
4973: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4974: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
4975: cptcode=modmaxcovj;
4976: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
4977: /*for (i=0; i<=cptcode; i++) {*/
4978: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
4979: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4980: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
4981: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
4982: if( j != -1){
4983: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
4984: covariate for which somebody answered excluding
4985: undefined. Usually 2: 0 and 1. */
4986: }
4987: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
4988: covariate for which somebody answered including
4989: undefined. Usually 3: -1, 0 and 1. */
4990: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
4991: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
4992: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 4993:
1.242 brouard 4994: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
4995: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
4996: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
4997: /* modmincovj=3; modmaxcovj = 7; */
4998: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
4999: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5000: /* defining two dummy variables: variables V1_1 and V1_2.*/
5001: /* nbcode[Tvar[j]][ij]=k; */
5002: /* nbcode[Tvar[j]][1]=0; */
5003: /* nbcode[Tvar[j]][2]=1; */
5004: /* nbcode[Tvar[j]][3]=2; */
5005: /* To be continued (not working yet). */
5006: ij=0; /* ij is similar to i but can jump over null modalities */
5007: 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*/
5008: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5009: break;
5010: }
5011: ij++;
5012: 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*/
5013: cptcode = ij; /* New max modality for covar j */
5014: } /* end of loop on modality i=-1 to 1 or more */
5015: break;
5016: case 1: /* Testing on varying covariate, could be simple and
5017: * should look at waves or product of fixed *
5018: * varying. No time to test -1, assuming 0 and 1 only */
5019: ij=0;
5020: for(i=0; i<=1;i++){
5021: nbcode[Tvar[k]][++ij]=i;
5022: }
5023: break;
5024: default:
5025: break;
5026: } /* end switch */
5027: } /* end dummy test */
5028:
5029: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5030: /* /\*recode from 0 *\/ */
5031: /* k is a modality. If we have model=V1+V1*sex */
5032: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5033: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5034: /* } */
5035: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5036: /* if (ij > ncodemax[j]) { */
5037: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5038: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5039: /* break; */
5040: /* } */
5041: /* } /\* end of loop on modality k *\/ */
5042: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5043:
5044: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5045: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5046: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5047: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5048: 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 */
5049: 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 */
5050: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5051: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5052:
5053: ij=0;
5054: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5055: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5056: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5057: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5058: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5059: /* If product not in single variable we don't print results */
5060: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5061: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5062: 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*/
5063: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5064: 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 */
5065: if(Fixed[k]!=0)
5066: anyvaryingduminmodel=1;
5067: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5068: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5069: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5070: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5071: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5072: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5073: }
5074: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5075: /* ij--; */
5076: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5077: *cptcov=ij; /*Number of total real effective covariates: effective
5078: * because they can be excluded from the model and real
5079: * if in the model but excluded because missing values, but how to get k from ij?*/
5080: for(j=ij+1; j<= cptcovt; j++){
5081: Tvaraff[j]=0;
5082: Tmodelind[j]=0;
5083: }
5084: for(j=ntveff+1; j<= cptcovt; j++){
5085: TmodelInvind[j]=0;
5086: }
5087: /* To be sorted */
5088: ;
5089: }
1.126 brouard 5090:
1.145 brouard 5091:
1.126 brouard 5092: /*********** Health Expectancies ****************/
5093:
1.235 brouard 5094: 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 5095:
5096: {
5097: /* Health expectancies, no variances */
1.164 brouard 5098: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5099: int nhstepma, nstepma; /* Decreasing with age */
5100: double age, agelim, hf;
5101: double ***p3mat;
5102: double eip;
5103:
1.238 brouard 5104: /* pstamp(ficreseij); */
1.126 brouard 5105: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5106: fprintf(ficreseij,"# Age");
5107: for(i=1; i<=nlstate;i++){
5108: for(j=1; j<=nlstate;j++){
5109: fprintf(ficreseij," e%1d%1d ",i,j);
5110: }
5111: fprintf(ficreseij," e%1d. ",i);
5112: }
5113: fprintf(ficreseij,"\n");
5114:
5115:
5116: if(estepm < stepm){
5117: printf ("Problem %d lower than %d\n",estepm, stepm);
5118: }
5119: else hstepm=estepm;
5120: /* We compute the life expectancy from trapezoids spaced every estepm months
5121: * This is mainly to measure the difference between two models: for example
5122: * if stepm=24 months pijx are given only every 2 years and by summing them
5123: * we are calculating an estimate of the Life Expectancy assuming a linear
5124: * progression in between and thus overestimating or underestimating according
5125: * to the curvature of the survival function. If, for the same date, we
5126: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5127: * to compare the new estimate of Life expectancy with the same linear
5128: * hypothesis. A more precise result, taking into account a more precise
5129: * curvature will be obtained if estepm is as small as stepm. */
5130:
5131: /* For example we decided to compute the life expectancy with the smallest unit */
5132: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5133: nhstepm is the number of hstepm from age to agelim
5134: nstepm is the number of stepm from age to agelin.
5135: Look at hpijx to understand the reason of that which relies in memory size
5136: and note for a fixed period like estepm months */
5137: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5138: survival function given by stepm (the optimization length). Unfortunately it
5139: means that if the survival funtion is printed only each two years of age and if
5140: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5141: results. So we changed our mind and took the option of the best precision.
5142: */
5143: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5144:
5145: agelim=AGESUP;
5146: /* If stepm=6 months */
5147: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5148: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5149:
5150: /* nhstepm age range expressed in number of stepm */
5151: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5152: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5153: /* if (stepm >= YEARM) hstepm=1;*/
5154: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5155: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5156:
5157: for (age=bage; age<=fage; age ++){
5158: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5159: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5160: /* if (stepm >= YEARM) hstepm=1;*/
5161: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5162:
5163: /* If stepm=6 months */
5164: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5165: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5166:
1.235 brouard 5167: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5168:
5169: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5170:
5171: printf("%d|",(int)age);fflush(stdout);
5172: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5173:
5174: /* Computing expectancies */
5175: for(i=1; i<=nlstate;i++)
5176: for(j=1; j<=nlstate;j++)
5177: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5178: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5179:
5180: /* 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]);*/
5181:
5182: }
5183:
5184: fprintf(ficreseij,"%3.0f",age );
5185: for(i=1; i<=nlstate;i++){
5186: eip=0;
5187: for(j=1; j<=nlstate;j++){
5188: eip +=eij[i][j][(int)age];
5189: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5190: }
5191: fprintf(ficreseij,"%9.4f", eip );
5192: }
5193: fprintf(ficreseij,"\n");
5194:
5195: }
5196: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5197: printf("\n");
5198: fprintf(ficlog,"\n");
5199:
5200: }
5201:
1.235 brouard 5202: 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 5203:
5204: {
5205: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5206: to initial status i, ei. .
1.126 brouard 5207: */
5208: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5209: int nhstepma, nstepma; /* Decreasing with age */
5210: double age, agelim, hf;
5211: double ***p3matp, ***p3matm, ***varhe;
5212: double **dnewm,**doldm;
5213: double *xp, *xm;
5214: double **gp, **gm;
5215: double ***gradg, ***trgradg;
5216: int theta;
5217:
5218: double eip, vip;
5219:
5220: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5221: xp=vector(1,npar);
5222: xm=vector(1,npar);
5223: dnewm=matrix(1,nlstate*nlstate,1,npar);
5224: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5225:
5226: pstamp(ficresstdeij);
5227: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5228: fprintf(ficresstdeij,"# Age");
5229: for(i=1; i<=nlstate;i++){
5230: for(j=1; j<=nlstate;j++)
5231: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5232: fprintf(ficresstdeij," e%1d. ",i);
5233: }
5234: fprintf(ficresstdeij,"\n");
5235:
5236: pstamp(ficrescveij);
5237: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5238: fprintf(ficrescveij,"# Age");
5239: for(i=1; i<=nlstate;i++)
5240: for(j=1; j<=nlstate;j++){
5241: cptj= (j-1)*nlstate+i;
5242: for(i2=1; i2<=nlstate;i2++)
5243: for(j2=1; j2<=nlstate;j2++){
5244: cptj2= (j2-1)*nlstate+i2;
5245: if(cptj2 <= cptj)
5246: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5247: }
5248: }
5249: fprintf(ficrescveij,"\n");
5250:
5251: if(estepm < stepm){
5252: printf ("Problem %d lower than %d\n",estepm, stepm);
5253: }
5254: else hstepm=estepm;
5255: /* We compute the life expectancy from trapezoids spaced every estepm months
5256: * This is mainly to measure the difference between two models: for example
5257: * if stepm=24 months pijx are given only every 2 years and by summing them
5258: * we are calculating an estimate of the Life Expectancy assuming a linear
5259: * progression in between and thus overestimating or underestimating according
5260: * to the curvature of the survival function. If, for the same date, we
5261: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5262: * to compare the new estimate of Life expectancy with the same linear
5263: * hypothesis. A more precise result, taking into account a more precise
5264: * curvature will be obtained if estepm is as small as stepm. */
5265:
5266: /* For example we decided to compute the life expectancy with the smallest unit */
5267: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5268: nhstepm is the number of hstepm from age to agelim
5269: nstepm is the number of stepm from age to agelin.
5270: Look at hpijx to understand the reason of that which relies in memory size
5271: and note for a fixed period like estepm months */
5272: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5273: survival function given by stepm (the optimization length). Unfortunately it
5274: means that if the survival funtion is printed only each two years of age and if
5275: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5276: results. So we changed our mind and took the option of the best precision.
5277: */
5278: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5279:
5280: /* If stepm=6 months */
5281: /* nhstepm age range expressed in number of stepm */
5282: agelim=AGESUP;
5283: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5284: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5285: /* if (stepm >= YEARM) hstepm=1;*/
5286: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5287:
5288: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5289: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5290: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5291: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5292: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5293: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5294:
5295: for (age=bage; age<=fage; age ++){
5296: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5297: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5298: /* if (stepm >= YEARM) hstepm=1;*/
5299: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5300:
1.126 brouard 5301: /* If stepm=6 months */
5302: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5303: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5304:
5305: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5306:
1.126 brouard 5307: /* Computing Variances of health expectancies */
5308: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5309: decrease memory allocation */
5310: for(theta=1; theta <=npar; theta++){
5311: for(i=1; i<=npar; i++){
1.222 brouard 5312: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5313: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5314: }
1.235 brouard 5315: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5316: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5317:
1.126 brouard 5318: for(j=1; j<= nlstate; j++){
1.222 brouard 5319: for(i=1; i<=nlstate; i++){
5320: for(h=0; h<=nhstepm-1; h++){
5321: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5322: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5323: }
5324: }
1.126 brouard 5325: }
1.218 brouard 5326:
1.126 brouard 5327: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5328: for(h=0; h<=nhstepm-1; h++){
5329: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5330: }
1.126 brouard 5331: }/* End theta */
5332:
5333:
5334: for(h=0; h<=nhstepm-1; h++)
5335: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5336: for(theta=1; theta <=npar; theta++)
5337: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5338:
1.218 brouard 5339:
1.222 brouard 5340: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5341: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5342: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5343:
1.222 brouard 5344: printf("%d|",(int)age);fflush(stdout);
5345: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5346: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5347: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5348: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5349: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5350: for(ij=1;ij<=nlstate*nlstate;ij++)
5351: for(ji=1;ji<=nlstate*nlstate;ji++)
5352: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5353: }
5354: }
1.218 brouard 5355:
1.126 brouard 5356: /* Computing expectancies */
1.235 brouard 5357: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5358: for(i=1; i<=nlstate;i++)
5359: for(j=1; j<=nlstate;j++)
1.222 brouard 5360: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5361: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5362:
1.222 brouard 5363: /* 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 5364:
1.222 brouard 5365: }
1.218 brouard 5366:
1.126 brouard 5367: fprintf(ficresstdeij,"%3.0f",age );
5368: for(i=1; i<=nlstate;i++){
5369: eip=0.;
5370: vip=0.;
5371: for(j=1; j<=nlstate;j++){
1.222 brouard 5372: eip += eij[i][j][(int)age];
5373: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5374: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5375: 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 5376: }
5377: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5378: }
5379: fprintf(ficresstdeij,"\n");
1.218 brouard 5380:
1.126 brouard 5381: fprintf(ficrescveij,"%3.0f",age );
5382: for(i=1; i<=nlstate;i++)
5383: for(j=1; j<=nlstate;j++){
1.222 brouard 5384: cptj= (j-1)*nlstate+i;
5385: for(i2=1; i2<=nlstate;i2++)
5386: for(j2=1; j2<=nlstate;j2++){
5387: cptj2= (j2-1)*nlstate+i2;
5388: if(cptj2 <= cptj)
5389: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5390: }
1.126 brouard 5391: }
5392: fprintf(ficrescveij,"\n");
1.218 brouard 5393:
1.126 brouard 5394: }
5395: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5396: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5397: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5398: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5399: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5400: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5401: printf("\n");
5402: fprintf(ficlog,"\n");
1.218 brouard 5403:
1.126 brouard 5404: free_vector(xm,1,npar);
5405: free_vector(xp,1,npar);
5406: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5407: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5408: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5409: }
1.218 brouard 5410:
1.126 brouard 5411: /************ Variance ******************/
1.235 brouard 5412: 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 5413: {
5414: /* Variance of health expectancies */
5415: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5416: /* double **newm;*/
5417: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5418:
5419: /* int movingaverage(); */
5420: double **dnewm,**doldm;
5421: double **dnewmp,**doldmp;
5422: int i, j, nhstepm, hstepm, h, nstepm ;
5423: int k;
5424: double *xp;
5425: double **gp, **gm; /* for var eij */
5426: double ***gradg, ***trgradg; /*for var eij */
5427: double **gradgp, **trgradgp; /* for var p point j */
5428: double *gpp, *gmp; /* for var p point j */
5429: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5430: double ***p3mat;
5431: double age,agelim, hf;
5432: /* double ***mobaverage; */
5433: int theta;
5434: char digit[4];
5435: char digitp[25];
5436:
5437: char fileresprobmorprev[FILENAMELENGTH];
5438:
5439: if(popbased==1){
5440: if(mobilav!=0)
5441: strcpy(digitp,"-POPULBASED-MOBILAV_");
5442: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5443: }
5444: else
5445: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5446:
1.218 brouard 5447: /* if (mobilav!=0) { */
5448: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5449: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5450: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5451: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5452: /* } */
5453: /* } */
5454:
5455: strcpy(fileresprobmorprev,"PRMORPREV-");
5456: sprintf(digit,"%-d",ij);
5457: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5458: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5459: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5460: strcat(fileresprobmorprev,fileresu);
5461: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5462: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5463: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5464: }
5465: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5466: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5467: pstamp(ficresprobmorprev);
5468: 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 5469: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5470: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5471: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5472: }
5473: for(j=1;j<=cptcoveff;j++)
5474: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5475: fprintf(ficresprobmorprev,"\n");
5476:
1.218 brouard 5477: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5478: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5479: fprintf(ficresprobmorprev," p.%-d SE",j);
5480: for(i=1; i<=nlstate;i++)
5481: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5482: }
5483: fprintf(ficresprobmorprev,"\n");
5484:
5485: fprintf(ficgp,"\n# Routine varevsij");
5486: fprintf(ficgp,"\nunset title \n");
5487: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5488: 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");
5489: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5490: /* } */
5491: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5492: pstamp(ficresvij);
5493: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5494: if(popbased==1)
5495: 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);
5496: else
5497: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5498: fprintf(ficresvij,"# Age");
5499: for(i=1; i<=nlstate;i++)
5500: for(j=1; j<=nlstate;j++)
5501: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5502: fprintf(ficresvij,"\n");
5503:
5504: xp=vector(1,npar);
5505: dnewm=matrix(1,nlstate,1,npar);
5506: doldm=matrix(1,nlstate,1,nlstate);
5507: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5508: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5509:
5510: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5511: gpp=vector(nlstate+1,nlstate+ndeath);
5512: gmp=vector(nlstate+1,nlstate+ndeath);
5513: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5514:
1.218 brouard 5515: if(estepm < stepm){
5516: printf ("Problem %d lower than %d\n",estepm, stepm);
5517: }
5518: else hstepm=estepm;
5519: /* For example we decided to compute the life expectancy with the smallest unit */
5520: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5521: nhstepm is the number of hstepm from age to agelim
5522: nstepm is the number of stepm from age to agelim.
5523: Look at function hpijx to understand why because of memory size limitations,
5524: we decided (b) to get a life expectancy respecting the most precise curvature of the
5525: survival function given by stepm (the optimization length). Unfortunately it
5526: means that if the survival funtion is printed every two years of age and if
5527: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5528: results. So we changed our mind and took the option of the best precision.
5529: */
5530: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5531: agelim = AGESUP;
5532: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5533: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5534: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5535: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5536: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5537: gp=matrix(0,nhstepm,1,nlstate);
5538: gm=matrix(0,nhstepm,1,nlstate);
5539:
5540:
5541: for(theta=1; theta <=npar; theta++){
5542: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5543: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5544: }
5545:
1.242 brouard 5546: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5547:
5548: if (popbased==1) {
5549: if(mobilav ==0){
5550: for(i=1; i<=nlstate;i++)
5551: prlim[i][i]=probs[(int)age][i][ij];
5552: }else{ /* mobilav */
5553: for(i=1; i<=nlstate;i++)
5554: prlim[i][i]=mobaverage[(int)age][i][ij];
5555: }
5556: }
5557:
1.235 brouard 5558: 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 5559: for(j=1; j<= nlstate; j++){
5560: for(h=0; h<=nhstepm; h++){
5561: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5562: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5563: }
5564: }
5565: /* Next for computing probability of death (h=1 means
5566: computed over hstepm matrices product = hstepm*stepm months)
5567: as a weighted average of prlim.
5568: */
5569: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5570: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5571: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5572: }
5573: /* end probability of death */
5574:
5575: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5576: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5577:
1.242 brouard 5578: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5579:
5580: if (popbased==1) {
5581: if(mobilav ==0){
5582: for(i=1; i<=nlstate;i++)
5583: prlim[i][i]=probs[(int)age][i][ij];
5584: }else{ /* mobilav */
5585: for(i=1; i<=nlstate;i++)
5586: prlim[i][i]=mobaverage[(int)age][i][ij];
5587: }
5588: }
5589:
1.235 brouard 5590: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5591:
5592: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5593: for(h=0; h<=nhstepm; h++){
5594: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5595: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5596: }
5597: }
5598: /* This for computing probability of death (h=1 means
5599: computed over hstepm matrices product = hstepm*stepm months)
5600: as a weighted average of prlim.
5601: */
5602: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5603: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5604: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5605: }
5606: /* end probability of death */
5607:
5608: for(j=1; j<= nlstate; j++) /* vareij */
5609: for(h=0; h<=nhstepm; h++){
5610: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5611: }
5612:
5613: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5614: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5615: }
5616:
5617: } /* End theta */
5618:
5619: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5620:
5621: for(h=0; h<=nhstepm; h++) /* veij */
5622: for(j=1; j<=nlstate;j++)
5623: for(theta=1; theta <=npar; theta++)
5624: trgradg[h][j][theta]=gradg[h][theta][j];
5625:
5626: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5627: for(theta=1; theta <=npar; theta++)
5628: trgradgp[j][theta]=gradgp[theta][j];
5629:
5630:
5631: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5632: for(i=1;i<=nlstate;i++)
5633: for(j=1;j<=nlstate;j++)
5634: vareij[i][j][(int)age] =0.;
5635:
5636: for(h=0;h<=nhstepm;h++){
5637: for(k=0;k<=nhstepm;k++){
5638: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5639: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5640: for(i=1;i<=nlstate;i++)
5641: for(j=1;j<=nlstate;j++)
5642: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5643: }
5644: }
5645:
5646: /* pptj */
5647: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5648: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5649: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5650: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5651: varppt[j][i]=doldmp[j][i];
5652: /* end ppptj */
5653: /* x centered again */
5654:
1.242 brouard 5655: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5656:
5657: if (popbased==1) {
5658: if(mobilav ==0){
5659: for(i=1; i<=nlstate;i++)
5660: prlim[i][i]=probs[(int)age][i][ij];
5661: }else{ /* mobilav */
5662: for(i=1; i<=nlstate;i++)
5663: prlim[i][i]=mobaverage[(int)age][i][ij];
5664: }
5665: }
5666:
5667: /* This for computing probability of death (h=1 means
5668: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5669: as a weighted average of prlim.
5670: */
1.235 brouard 5671: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5672: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5673: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5674: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5675: }
5676: /* end probability of death */
5677:
5678: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5679: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5680: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5681: for(i=1; i<=nlstate;i++){
5682: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5683: }
5684: }
5685: fprintf(ficresprobmorprev,"\n");
5686:
5687: fprintf(ficresvij,"%.0f ",age );
5688: for(i=1; i<=nlstate;i++)
5689: for(j=1; j<=nlstate;j++){
5690: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5691: }
5692: fprintf(ficresvij,"\n");
5693: free_matrix(gp,0,nhstepm,1,nlstate);
5694: free_matrix(gm,0,nhstepm,1,nlstate);
5695: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
5696: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
5697: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5698: } /* End age */
5699: free_vector(gpp,nlstate+1,nlstate+ndeath);
5700: free_vector(gmp,nlstate+1,nlstate+ndeath);
5701: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
5702: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
5703: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
5704: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
5705: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
5706: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
5707: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
5708: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
5709: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
5710: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
5711: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
5712: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
5713: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
5714: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
5715: 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);
5716: /* 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 5717: */
1.218 brouard 5718: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
5719: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 5720:
1.218 brouard 5721: free_vector(xp,1,npar);
5722: free_matrix(doldm,1,nlstate,1,nlstate);
5723: free_matrix(dnewm,1,nlstate,1,npar);
5724: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5725: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
5726: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5727: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5728: fclose(ficresprobmorprev);
5729: fflush(ficgp);
5730: fflush(fichtm);
5731: } /* end varevsij */
1.126 brouard 5732:
5733: /************ Variance of prevlim ******************/
1.235 brouard 5734: 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 5735: {
1.205 brouard 5736: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 5737: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 5738:
1.126 brouard 5739: double **dnewm,**doldm;
5740: int i, j, nhstepm, hstepm;
5741: double *xp;
5742: double *gp, *gm;
5743: double **gradg, **trgradg;
1.208 brouard 5744: double **mgm, **mgp;
1.126 brouard 5745: double age,agelim;
5746: int theta;
5747:
5748: pstamp(ficresvpl);
5749: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 5750: fprintf(ficresvpl,"# Age ");
5751: if(nresult >=1)
5752: fprintf(ficresvpl," Result# ");
1.126 brouard 5753: for(i=1; i<=nlstate;i++)
5754: fprintf(ficresvpl," %1d-%1d",i,i);
5755: fprintf(ficresvpl,"\n");
5756:
5757: xp=vector(1,npar);
5758: dnewm=matrix(1,nlstate,1,npar);
5759: doldm=matrix(1,nlstate,1,nlstate);
5760:
5761: hstepm=1*YEARM; /* Every year of age */
5762: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
5763: agelim = AGESUP;
5764: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5765: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5766: if (stepm >= YEARM) hstepm=1;
5767: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
5768: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 5769: mgp=matrix(1,npar,1,nlstate);
5770: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 5771: gp=vector(1,nlstate);
5772: gm=vector(1,nlstate);
5773:
5774: for(theta=1; theta <=npar; theta++){
5775: for(i=1; i<=npar; i++){ /* Computes gradient */
5776: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5777: }
1.209 brouard 5778: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5779: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5780: else
1.235 brouard 5781: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5782: for(i=1;i<=nlstate;i++){
1.126 brouard 5783: gp[i] = prlim[i][i];
1.208 brouard 5784: mgp[theta][i] = prlim[i][i];
5785: }
1.126 brouard 5786: for(i=1; i<=npar; i++) /* Computes gradient */
5787: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 5788: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 5789: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 5790: else
1.235 brouard 5791: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 5792: for(i=1;i<=nlstate;i++){
1.126 brouard 5793: gm[i] = prlim[i][i];
1.208 brouard 5794: mgm[theta][i] = prlim[i][i];
5795: }
1.126 brouard 5796: for(i=1;i<=nlstate;i++)
5797: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 5798: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 5799: } /* End theta */
5800:
5801: trgradg =matrix(1,nlstate,1,npar);
5802:
5803: for(j=1; j<=nlstate;j++)
5804: for(theta=1; theta <=npar; theta++)
5805: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 5806: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5807: /* printf("\nmgm mgp %d ",(int)age); */
5808: /* for(j=1; j<=nlstate;j++){ */
5809: /* printf(" %d ",j); */
5810: /* for(theta=1; theta <=npar; theta++) */
5811: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
5812: /* printf("\n "); */
5813: /* } */
5814: /* } */
5815: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
5816: /* printf("\n gradg %d ",(int)age); */
5817: /* for(j=1; j<=nlstate;j++){ */
5818: /* printf("%d ",j); */
5819: /* for(theta=1; theta <=npar; theta++) */
5820: /* printf("%d %lf ",theta,gradg[theta][j]); */
5821: /* printf("\n "); */
5822: /* } */
5823: /* } */
1.126 brouard 5824:
5825: for(i=1;i<=nlstate;i++)
5826: varpl[i][(int)age] =0.;
1.209 brouard 5827: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 5828: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5829: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
5830: }else{
1.126 brouard 5831: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
5832: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 5833: }
1.126 brouard 5834: for(i=1;i<=nlstate;i++)
5835: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
5836:
5837: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 5838: if(nresult >=1)
5839: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 5840: for(i=1; i<=nlstate;i++)
5841: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
5842: fprintf(ficresvpl,"\n");
5843: free_vector(gp,1,nlstate);
5844: free_vector(gm,1,nlstate);
1.208 brouard 5845: free_matrix(mgm,1,npar,1,nlstate);
5846: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 5847: free_matrix(gradg,1,npar,1,nlstate);
5848: free_matrix(trgradg,1,nlstate,1,npar);
5849: } /* End age */
5850:
5851: free_vector(xp,1,npar);
5852: free_matrix(doldm,1,nlstate,1,npar);
5853: free_matrix(dnewm,1,nlstate,1,nlstate);
5854:
5855: }
5856:
5857: /************ Variance of one-step probabilities ******************/
5858: 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 5859: {
5860: int i, j=0, k1, l1, tj;
5861: int k2, l2, j1, z1;
5862: int k=0, l;
5863: int first=1, first1, first2;
5864: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
5865: double **dnewm,**doldm;
5866: double *xp;
5867: double *gp, *gm;
5868: double **gradg, **trgradg;
5869: double **mu;
5870: double age, cov[NCOVMAX+1];
5871: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
5872: int theta;
5873: char fileresprob[FILENAMELENGTH];
5874: char fileresprobcov[FILENAMELENGTH];
5875: char fileresprobcor[FILENAMELENGTH];
5876: double ***varpij;
5877:
5878: strcpy(fileresprob,"PROB_");
5879: strcat(fileresprob,fileres);
5880: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
5881: printf("Problem with resultfile: %s\n", fileresprob);
5882: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
5883: }
5884: strcpy(fileresprobcov,"PROBCOV_");
5885: strcat(fileresprobcov,fileresu);
5886: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
5887: printf("Problem with resultfile: %s\n", fileresprobcov);
5888: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
5889: }
5890: strcpy(fileresprobcor,"PROBCOR_");
5891: strcat(fileresprobcor,fileresu);
5892: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
5893: printf("Problem with resultfile: %s\n", fileresprobcor);
5894: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
5895: }
5896: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5897: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
5898: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5899: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
5900: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5901: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
5902: pstamp(ficresprob);
5903: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
5904: fprintf(ficresprob,"# Age");
5905: pstamp(ficresprobcov);
5906: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
5907: fprintf(ficresprobcov,"# Age");
5908: pstamp(ficresprobcor);
5909: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
5910: fprintf(ficresprobcor,"# Age");
1.126 brouard 5911:
5912:
1.222 brouard 5913: for(i=1; i<=nlstate;i++)
5914: for(j=1; j<=(nlstate+ndeath);j++){
5915: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
5916: fprintf(ficresprobcov," p%1d-%1d ",i,j);
5917: fprintf(ficresprobcor," p%1d-%1d ",i,j);
5918: }
5919: /* fprintf(ficresprob,"\n");
5920: fprintf(ficresprobcov,"\n");
5921: fprintf(ficresprobcor,"\n");
5922: */
5923: xp=vector(1,npar);
5924: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5925: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
5926: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
5927: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
5928: first=1;
5929: fprintf(ficgp,"\n# Routine varprob");
5930: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
5931: fprintf(fichtm,"\n");
5932:
5933: 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);
5934: 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);
5935: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 5936: and drawn. It helps understanding how is the covariance between two incidences.\
5937: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 5938: 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 5939: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
5940: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
5941: standard deviations wide on each axis. <br>\
5942: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
5943: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
5944: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
5945:
1.222 brouard 5946: cov[1]=1;
5947: /* tj=cptcoveff; */
1.225 brouard 5948: tj = (int) pow(2,cptcoveff);
1.222 brouard 5949: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
5950: j1=0;
1.224 brouard 5951: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 5952: if (cptcovn>0) {
5953: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 5954: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5955: fprintf(ficresprob, "**********\n#\n");
5956: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 5957: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5958: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 5959:
1.222 brouard 5960: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 5961: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5962: fprintf(ficgp, "**********\n#\n");
1.220 brouard 5963:
5964:
1.222 brouard 5965: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 5966: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5967: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 5968:
1.222 brouard 5969: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 5970: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 5971: fprintf(ficresprobcor, "**********\n#");
5972: if(invalidvarcomb[j1]){
5973: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
5974: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
5975: continue;
5976: }
5977: }
5978: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
5979: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
5980: gp=vector(1,(nlstate)*(nlstate+ndeath));
5981: gm=vector(1,(nlstate)*(nlstate+ndeath));
5982: for (age=bage; age<=fage; age ++){
5983: cov[2]=age;
5984: if(nagesqr==1)
5985: cov[3]= age*age;
5986: for (k=1; k<=cptcovn;k++) {
5987: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
5988: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
5989: * 1 1 1 1 1
5990: * 2 2 1 1 1
5991: * 3 1 2 1 1
5992: */
5993: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
5994: }
5995: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
5996: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
5997: for (k=1; k<=cptcovprod;k++)
5998: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 5999:
6000:
1.222 brouard 6001: for(theta=1; theta <=npar; theta++){
6002: for(i=1; i<=npar; i++)
6003: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6004:
1.222 brouard 6005: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6006:
1.222 brouard 6007: k=0;
6008: for(i=1; i<= (nlstate); i++){
6009: for(j=1; j<=(nlstate+ndeath);j++){
6010: k=k+1;
6011: gp[k]=pmmij[i][j];
6012: }
6013: }
1.220 brouard 6014:
1.222 brouard 6015: for(i=1; i<=npar; i++)
6016: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6017:
1.222 brouard 6018: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6019: k=0;
6020: for(i=1; i<=(nlstate); i++){
6021: for(j=1; j<=(nlstate+ndeath);j++){
6022: k=k+1;
6023: gm[k]=pmmij[i][j];
6024: }
6025: }
1.220 brouard 6026:
1.222 brouard 6027: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6028: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6029: }
1.126 brouard 6030:
1.222 brouard 6031: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6032: for(theta=1; theta <=npar; theta++)
6033: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6034:
1.222 brouard 6035: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6036: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6037:
1.222 brouard 6038: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6039:
1.222 brouard 6040: k=0;
6041: for(i=1; i<=(nlstate); i++){
6042: for(j=1; j<=(nlstate+ndeath);j++){
6043: k=k+1;
6044: mu[k][(int) age]=pmmij[i][j];
6045: }
6046: }
6047: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6048: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6049: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6050:
1.222 brouard 6051: /*printf("\n%d ",(int)age);
6052: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6053: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6054: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6055: }*/
1.220 brouard 6056:
1.222 brouard 6057: fprintf(ficresprob,"\n%d ",(int)age);
6058: fprintf(ficresprobcov,"\n%d ",(int)age);
6059: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6060:
1.222 brouard 6061: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6062: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6063: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6064: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6065: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6066: }
6067: i=0;
6068: for (k=1; k<=(nlstate);k++){
6069: for (l=1; l<=(nlstate+ndeath);l++){
6070: i++;
6071: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6072: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6073: for (j=1; j<=i;j++){
6074: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6075: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6076: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6077: }
6078: }
6079: }/* end of loop for state */
6080: } /* end of loop for age */
6081: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6082: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6083: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6084: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6085:
6086: /* Confidence intervalle of pij */
6087: /*
6088: fprintf(ficgp,"\nunset parametric;unset label");
6089: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6090: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6091: 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);
6092: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6093: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6094: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6095: */
6096:
6097: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6098: first1=1;first2=2;
6099: for (k2=1; k2<=(nlstate);k2++){
6100: for (l2=1; l2<=(nlstate+ndeath);l2++){
6101: if(l2==k2) continue;
6102: j=(k2-1)*(nlstate+ndeath)+l2;
6103: for (k1=1; k1<=(nlstate);k1++){
6104: for (l1=1; l1<=(nlstate+ndeath);l1++){
6105: if(l1==k1) continue;
6106: i=(k1-1)*(nlstate+ndeath)+l1;
6107: if(i<=j) continue;
6108: for (age=bage; age<=fage; age ++){
6109: if ((int)age %5==0){
6110: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6111: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6112: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6113: mu1=mu[i][(int) age]/stepm*YEARM ;
6114: mu2=mu[j][(int) age]/stepm*YEARM;
6115: c12=cv12/sqrt(v1*v2);
6116: /* Computing eigen value of matrix of covariance */
6117: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6118: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6119: if ((lc2 <0) || (lc1 <0) ){
6120: if(first2==1){
6121: first1=0;
6122: 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);
6123: }
6124: 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);
6125: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6126: /* lc2=fabs(lc2); */
6127: }
1.220 brouard 6128:
1.222 brouard 6129: /* Eigen vectors */
6130: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6131: /*v21=sqrt(1.-v11*v11); *//* error */
6132: v21=(lc1-v1)/cv12*v11;
6133: v12=-v21;
6134: v22=v11;
6135: tnalp=v21/v11;
6136: if(first1==1){
6137: first1=0;
6138: 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);
6139: }
6140: 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);
6141: /*printf(fignu*/
6142: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6143: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6144: if(first==1){
6145: first=0;
6146: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6147: fprintf(ficgp,"\nset parametric;unset label");
6148: 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);
6149: fprintf(ficgp,"\nset ter svg size 640, 480");
6150: fprintf(fichtmcov,"\n<br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6151: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6152: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6153: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6154: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6155: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6156: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6157: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6158: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6159: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6160: 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", \
6161: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6162: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6163: }else{
6164: first=0;
6165: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6166: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6167: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6168: 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", \
6169: mu1,std,v11,sqrt(lc1),v12,sqrt(lc2), \
6170: mu2,std,v21,sqrt(lc1),v22,sqrt(lc2));
6171: }/* if first */
6172: } /* age mod 5 */
6173: } /* end loop age */
6174: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6175: first=1;
6176: } /*l12 */
6177: } /* k12 */
6178: } /*l1 */
6179: }/* k1 */
6180: } /* loop on combination of covariates j1 */
6181: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6182: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6183: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6184: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6185: free_vector(xp,1,npar);
6186: fclose(ficresprob);
6187: fclose(ficresprobcov);
6188: fclose(ficresprobcor);
6189: fflush(ficgp);
6190: fflush(fichtmcov);
6191: }
1.126 brouard 6192:
6193:
6194: /******************* Printing html file ***********/
1.201 brouard 6195: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6196: int lastpass, int stepm, int weightopt, char model[],\
6197: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.217 brouard 6198: int popforecast, int prevfcast, int backcast, int estepm , \
1.213 brouard 6199: double jprev1, double mprev1,double anprev1, double dateprev1, \
6200: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6201: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6202:
6203: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6204: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6205: </ul>");
1.237 brouard 6206: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6207: </ul>", model);
1.214 brouard 6208: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6209: 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",
6210: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6211: 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 6212: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6213: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6214: fprintf(fichtm,"\
6215: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6216: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6217: fprintf(fichtm,"\
1.217 brouard 6218: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6219: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6220: fprintf(fichtm,"\
1.126 brouard 6221: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6222: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6223: fprintf(fichtm,"\
1.217 brouard 6224: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6225: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6226: fprintf(fichtm,"\
1.211 brouard 6227: - (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 6228: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6229: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6230: if(prevfcast==1){
6231: fprintf(fichtm,"\
6232: - Prevalence projections by age and states: \
1.201 brouard 6233: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6234: }
1.126 brouard 6235:
1.222 brouard 6236: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
1.126 brouard 6237:
1.225 brouard 6238: m=pow(2,cptcoveff);
1.222 brouard 6239: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6240:
1.222 brouard 6241: jj1=0;
1.237 brouard 6242:
6243: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6244: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.237 brouard 6245: if(TKresult[nres]!= k1)
6246: continue;
1.220 brouard 6247:
1.222 brouard 6248: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6249: jj1++;
6250: if (cptcovn > 0) {
6251: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6252: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6253: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6254: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6255: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6256: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6257: }
1.237 brouard 6258: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6259: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6260: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6261: }
6262:
1.230 brouard 6263: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6264: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6265: if(invalidvarcomb[k1]){
6266: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6267: printf("\nCombination (%d) ignored because no cases \n",k1);
6268: continue;
6269: }
6270: }
6271: /* aij, bij */
1.241 brouard 6272: 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> \
6273: <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 6274: /* Pij */
1.241 brouard 6275: 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> \
6276: <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 6277: /* Quasi-incidences */
6278: 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 6279: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6280: 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 6281: 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> \
6282: <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 6283: /* Survival functions (period) in state j */
6284: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6285: 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> \
6286: <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 6287: }
6288: /* State specific survival functions (period) */
6289: for(cpt=1; cpt<=nlstate;cpt++){
6290: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6291: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6292: <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 6293: }
6294: /* Period (stable) prevalence in each health state */
6295: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6296: 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> \
6297: <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 6298: }
6299: if(backcast==1){
6300: /* Period (stable) back prevalence in each health state */
6301: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6302: 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> \
6303: <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 6304: }
1.217 brouard 6305: }
1.222 brouard 6306: if(prevfcast==1){
6307: /* Projection of prevalence up to period (stable) prevalence in each health state */
6308: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6309: 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> \
6310: <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 6311: }
6312: }
1.220 brouard 6313:
1.222 brouard 6314: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6315: 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> \
6316: <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 6317: }
6318: /* } /\* end i1 *\/ */
6319: }/* End k1 */
6320: fprintf(fichtm,"</ul>");
1.126 brouard 6321:
1.222 brouard 6322: fprintf(fichtm,"\
1.126 brouard 6323: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6324: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6325: - 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 6326: But because parameters are usually highly correlated (a higher incidence of disability \
6327: and a higher incidence of recovery can give very close observed transition) it might \
6328: be very useful to look not only at linear confidence intervals estimated from the \
6329: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6330: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6331: covariance matrix of the one-step probabilities. \
6332: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6333:
1.222 brouard 6334: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6335: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6336: fprintf(fichtm,"\
1.126 brouard 6337: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6338: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6339:
1.222 brouard 6340: fprintf(fichtm,"\
1.126 brouard 6341: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6342: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6343: fprintf(fichtm,"\
1.126 brouard 6344: - 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): \
6345: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6346: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6347: fprintf(fichtm,"\
1.126 brouard 6348: - (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): \
6349: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6350: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6351: fprintf(fichtm,"\
1.128 brouard 6352: - 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 6353: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6354: fprintf(fichtm,"\
1.128 brouard 6355: - 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 6356: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6357: fprintf(fichtm,"\
1.126 brouard 6358: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6359: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6360:
6361: /* if(popforecast==1) fprintf(fichtm,"\n */
6362: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6363: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6364: /* <br>",fileres,fileres,fileres,fileres); */
6365: /* else */
6366: /* 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 6367: fflush(fichtm);
6368: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6369:
1.225 brouard 6370: m=pow(2,cptcoveff);
1.222 brouard 6371: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6372:
1.222 brouard 6373: jj1=0;
1.237 brouard 6374:
1.241 brouard 6375: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6376: for(k1=1; k1<=m;k1++){
1.237 brouard 6377: if(TKresult[nres]!= k1)
6378: continue;
1.222 brouard 6379: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6380: jj1++;
1.126 brouard 6381: if (cptcovn > 0) {
6382: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6383: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6384: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6385: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6386: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6387: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6388: }
6389:
1.126 brouard 6390: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6391:
1.222 brouard 6392: if(invalidvarcomb[k1]){
6393: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6394: continue;
6395: }
1.126 brouard 6396: }
6397: for(cpt=1; cpt<=nlstate;cpt++) {
1.218 brouard 6398: fprintf(fichtm,"\n<br>- Observed (cross-sectional) and period (incidence based) \
1.241 brouard 6399: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
6400: <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 6401: }
6402: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6403: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6404: true period expectancies (those weighted with period prevalences are also\
6405: drawn in addition to the population based expectancies computed using\
1.241 brouard 6406: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6407: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6408: /* } /\* end i1 *\/ */
6409: }/* End k1 */
1.241 brouard 6410: }/* End nres */
1.222 brouard 6411: fprintf(fichtm,"</ul>");
6412: fflush(fichtm);
1.126 brouard 6413: }
6414:
6415: /******************* Gnuplot file **************/
1.223 brouard 6416: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[]){
1.126 brouard 6417:
6418: char dirfileres[132],optfileres[132];
1.223 brouard 6419: char gplotcondition[132];
1.237 brouard 6420: 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 6421: int lv=0, vlv=0, kl=0;
1.130 brouard 6422: int ng=0;
1.201 brouard 6423: int vpopbased;
1.223 brouard 6424: int ioffset; /* variable offset for columns */
1.235 brouard 6425: int nres=0; /* Index of resultline */
1.219 brouard 6426:
1.126 brouard 6427: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6428: /* printf("Problem with file %s",optionfilegnuplot); */
6429: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6430: /* } */
6431:
6432: /*#ifdef windows */
6433: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6434: /*#endif */
1.225 brouard 6435: m=pow(2,cptcoveff);
1.126 brouard 6436:
1.202 brouard 6437: /* Contribution to likelihood */
6438: /* Plot the probability implied in the likelihood */
1.223 brouard 6439: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6440: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6441: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6442: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6443: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6444: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6445: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6446: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6447: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6448: 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));
6449: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6450: 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));
6451: for (i=1; i<= nlstate ; i ++) {
6452: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6453: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6454: 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);
6455: for (j=2; j<= nlstate+ndeath ; j ++) {
6456: 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);
6457: }
6458: fprintf(ficgp,";\nset out; unset ylabel;\n");
6459: }
6460: /* 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 */
6461: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6462: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6463: fprintf(ficgp,"\nset out;unset log\n");
6464: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6465:
1.126 brouard 6466: strcpy(dirfileres,optionfilefiname);
6467: strcpy(optfileres,"vpl");
1.223 brouard 6468: /* 1eme*/
1.238 brouard 6469: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6470: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6471: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6472: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
6473: if(TKresult[nres]!= k1)
6474: continue;
6475: /* We are interested in selected combination by the resultline */
1.246 brouard 6476: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6477: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
6478: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6479: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6480: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6481: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6482: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6483: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6484: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6485: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6486: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6487: }
6488: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6489: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6490: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6491: }
1.246 brouard 6492: /* printf("\n#\n"); */
1.238 brouard 6493: fprintf(ficgp,"\n#\n");
6494: if(invalidvarcomb[k1]){
6495: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6496: continue;
6497: }
1.235 brouard 6498:
1.241 brouard 6499: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6500: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
6501: 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 6502:
1.238 brouard 6503: for (i=1; i<= nlstate ; i ++) {
6504: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6505: else fprintf(ficgp," %%*lf (%%*lf)");
6506: }
1.242 brouard 6507: 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 6508: for (i=1; i<= nlstate ; i ++) {
6509: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6510: else fprintf(ficgp," %%*lf (%%*lf)");
6511: }
1.242 brouard 6512: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres);
1.238 brouard 6513: for (i=1; i<= nlstate ; i ++) {
6514: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6515: else fprintf(ficgp," %%*lf (%%*lf)");
6516: }
6517: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1));
6518: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6519: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6520: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6521: if(cptcoveff ==0){
1.245 brouard 6522: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6523: }else{
6524: kl=0;
6525: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6526: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6527: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6528: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6529: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6530: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6531: kl++;
1.238 brouard 6532: /* 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 *\/ */
6533: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6534: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6535: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6536: if(k==cptcoveff){
1.245 brouard 6537: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6538: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6539: }else{
6540: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6541: kl++;
6542: }
6543: } /* end covariate */
6544: } /* end if no covariate */
6545: } /* end if backcast */
6546: fprintf(ficgp,"\nset out \n");
6547: } /* nres */
1.201 brouard 6548: } /* k1 */
6549: } /* cpt */
1.235 brouard 6550:
6551:
1.126 brouard 6552: /*2 eme*/
1.238 brouard 6553: for (k1=1; k1<= m ; k1 ++){
6554: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6555: if(TKresult[nres]!= k1)
6556: continue;
6557: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
6558: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6559: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6560: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6561: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6562: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6563: vlv= nbcode[Tvaraff[k]][lv];
6564: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6565: }
1.237 brouard 6566: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6567: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6568: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6569: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6570: }
1.211 brouard 6571: fprintf(ficgp,"\n#\n");
1.223 brouard 6572: if(invalidvarcomb[k1]){
6573: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6574: continue;
6575: }
1.219 brouard 6576:
1.241 brouard 6577: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6578: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
6579: if(vpopbased==0)
6580: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
6581: else
6582: fprintf(ficgp,"\nreplot ");
6583: for (i=1; i<= nlstate+1 ; i ++) {
6584: k=2*i;
6585: 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);
6586: for (j=1; j<= nlstate+1 ; j ++) {
6587: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6588: else fprintf(ficgp," %%*lf (%%*lf)");
6589: }
6590: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6591: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
6592: 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);
6593: for (j=1; j<= nlstate+1 ; j ++) {
6594: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6595: else fprintf(ficgp," %%*lf (%%*lf)");
6596: }
6597: fprintf(ficgp,"\" t\"\" w l lt 0,");
6598: 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);
6599: for (j=1; j<= nlstate+1 ; j ++) {
6600: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6601: else fprintf(ficgp," %%*lf (%%*lf)");
6602: }
6603: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6604: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6605: } /* state */
6606: } /* vpopbased */
1.244 brouard 6607: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; \n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6608: } /* end nres */
6609: } /* k1 end 2 eme*/
6610:
6611:
6612: /*3eme*/
6613: for (k1=1; k1<= m ; k1 ++){
6614: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.240 brouard 6615: if(TKresult[nres]!= k1)
1.238 brouard 6616: continue;
6617:
6618: for (cpt=1; cpt<= nlstate ; cpt ++) {
6619: fprintf(ficgp,"\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
6620: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6621: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6622: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6623: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6624: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6625: vlv= nbcode[Tvaraff[k]][lv];
6626: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6627: }
6628: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6629: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6630: }
6631: fprintf(ficgp,"\n#\n");
6632: if(invalidvarcomb[k1]){
6633: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6634: continue;
6635: }
6636:
6637: /* k=2+nlstate*(2*cpt-2); */
6638: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 6639: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.238 brouard 6640: fprintf(ficgp,"set ter svg size 640, 480\n\
1.201 brouard 6641: 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 6642: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6643: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6644: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
6645: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
6646: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
6647: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 6648:
1.238 brouard 6649: */
6650: for (i=1; i< nlstate ; i ++) {
6651: 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);
6652: /* 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 6653:
1.238 brouard 6654: }
6655: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),k1-1,k1-1,k+nlstate,cpt);
6656: }
6657: } /* end nres */
6658: } /* end kl 3eme */
1.126 brouard 6659:
1.223 brouard 6660: /* 4eme */
1.201 brouard 6661: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 6662: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
6663: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6664: if(TKresult[nres]!= k1)
1.223 brouard 6665: continue;
1.238 brouard 6666: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
6667: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
6668: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6669: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6670: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6671: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6672: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6673: vlv= nbcode[Tvaraff[k]][lv];
6674: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6675: }
6676: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6677: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6678: }
6679: fprintf(ficgp,"\n#\n");
6680: if(invalidvarcomb[k1]){
6681: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6682: continue;
1.223 brouard 6683: }
1.238 brouard 6684:
1.241 brouard 6685: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.238 brouard 6686: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6687: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6688: k=3;
6689: for (i=1; i<= nlstate ; i ++){
6690: if(i==1){
6691: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6692: }else{
6693: fprintf(ficgp,", '' ");
6694: }
6695: l=(nlstate+ndeath)*(i-1)+1;
6696: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6697: for (j=2; j<= nlstate+ndeath ; j ++)
6698: fprintf(ficgp,"+$%d",k+l+j-1);
6699: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
6700: } /* nlstate */
6701: fprintf(ficgp,"\nset out\n");
6702: } /* end cpt state*/
6703: } /* end nres */
6704: } /* end covariate k1 */
6705:
1.220 brouard 6706: /* 5eme */
1.201 brouard 6707: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 6708: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
6709: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6710: if(TKresult[nres]!= k1)
1.227 brouard 6711: continue;
1.238 brouard 6712: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
6713: 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);
6714: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6715: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6716: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6717: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6718: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6719: vlv= nbcode[Tvaraff[k]][lv];
6720: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6721: }
6722: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6723: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6724: }
6725: fprintf(ficgp,"\n#\n");
6726: if(invalidvarcomb[k1]){
6727: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6728: continue;
6729: }
1.227 brouard 6730:
1.241 brouard 6731: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.238 brouard 6732: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
6733: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
6734: k=3;
6735: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6736: if(j==1)
6737: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6738: else
6739: fprintf(ficgp,", '' ");
6740: l=(nlstate+ndeath)*(cpt-1) +j;
6741: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
6742: /* for (i=2; i<= nlstate+ndeath ; i ++) */
6743: /* fprintf(ficgp,"+$%d",k+l+i-1); */
6744: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
6745: } /* nlstate */
6746: fprintf(ficgp,", '' ");
6747: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
6748: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
6749: l=(nlstate+ndeath)*(cpt-1) +j;
6750: if(j < nlstate)
6751: fprintf(ficgp,"$%d +",k+l);
6752: else
6753: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
6754: }
6755: fprintf(ficgp,"\nset out\n");
6756: } /* end cpt state*/
6757: } /* end covariate */
6758: } /* end nres */
1.227 brouard 6759:
1.220 brouard 6760: /* 6eme */
1.202 brouard 6761: /* CV preval stable (period) for each covariate */
1.237 brouard 6762: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6763: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6764: if(TKresult[nres]!= k1)
6765: continue;
1.153 brouard 6766: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6767:
1.211 brouard 6768: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 6769: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 6770: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6771: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6772: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6773: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6774: vlv= nbcode[Tvaraff[k]][lv];
6775: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6776: }
1.237 brouard 6777: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6778: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6779: }
1.211 brouard 6780: fprintf(ficgp,"\n#\n");
1.223 brouard 6781: if(invalidvarcomb[k1]){
1.227 brouard 6782: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6783: continue;
1.223 brouard 6784: }
1.227 brouard 6785:
1.241 brouard 6786: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.126 brouard 6787: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6788: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 6789: k=3; /* Offset */
1.153 brouard 6790: for (i=1; i<= nlstate ; i ++){
1.227 brouard 6791: if(i==1)
6792: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
6793: else
6794: fprintf(ficgp,", '' ");
6795: l=(nlstate+ndeath)*(i-1)+1;
6796: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
6797: for (j=2; j<= nlstate ; j ++)
6798: fprintf(ficgp,"+$%d",k+l+j-1);
6799: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 6800: } /* nlstate */
1.201 brouard 6801: fprintf(ficgp,"\nset out\n");
1.153 brouard 6802: } /* end cpt state*/
6803: } /* end covariate */
1.227 brouard 6804:
6805:
1.220 brouard 6806: /* 7eme */
1.218 brouard 6807: if(backcast == 1){
1.217 brouard 6808: /* CV back preval stable (period) for each covariate */
1.237 brouard 6809: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6810: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6811: if(TKresult[nres]!= k1)
6812: continue;
1.218 brouard 6813: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6814: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
6815: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
6816: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
6817: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6818: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 6819: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 6820: vlv= nbcode[Tvaraff[k]][lv];
6821: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6822: }
1.237 brouard 6823: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6824: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6825: }
1.227 brouard 6826: fprintf(ficgp,"\n#\n");
6827: if(invalidvarcomb[k1]){
6828: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6829: continue;
6830: }
6831:
1.241 brouard 6832: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.227 brouard 6833: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 6834: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6835: k=3; /* Offset */
6836: for (i=1; i<= nlstate ; i ++){
6837: if(i==1)
6838: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
6839: else
6840: fprintf(ficgp,", '' ");
6841: /* l=(nlstate+ndeath)*(i-1)+1; */
6842: l=(nlstate+ndeath)*(cpt-1)+1;
6843: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
6844: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
6845: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+(cpt-1)+i-1); /* a vérifier */
6846: /* for (j=2; j<= nlstate ; j ++) */
6847: /* fprintf(ficgp,"+$%d",k+l+j-1); */
6848: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
6849: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
6850: } /* nlstate */
6851: fprintf(ficgp,"\nset out\n");
1.218 brouard 6852: } /* end cpt state*/
6853: } /* end covariate */
6854: } /* End if backcast */
6855:
1.223 brouard 6856: /* 8eme */
1.218 brouard 6857: if(prevfcast==1){
6858: /* Projection from cross-sectional to stable (period) for each covariate */
6859:
1.237 brouard 6860: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
6861: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6862: if(TKresult[nres]!= k1)
6863: continue;
1.211 brouard 6864: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.227 brouard 6865: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
6866: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
6867: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6868: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6869: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6870: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6871: vlv= nbcode[Tvaraff[k]][lv];
6872: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
6873: }
1.237 brouard 6874: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6875: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6876: }
1.227 brouard 6877: fprintf(ficgp,"\n#\n");
6878: if(invalidvarcomb[k1]){
6879: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6880: continue;
6881: }
6882:
6883: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 6884: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.227 brouard 6885: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 6886: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 6887: for (i=1; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
6888: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6889: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6890: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6891: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6892: if(i==1){
6893: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
6894: }else{
6895: fprintf(ficgp,",\\\n '' ");
6896: }
6897: if(cptcoveff ==0){ /* No covariate */
6898: ioffset=2; /* Age is in 2 */
6899: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6900: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6901: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
6902: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
6903: fprintf(ficgp," u %d:(", ioffset);
6904: if(i==nlstate+1)
6905: fprintf(ficgp," $%d/(1.-$%d)) t 'pw.%d' with line ", \
6906: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6907: else
6908: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
6909: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6910: }else{ /* more than 2 covariates */
6911: if(cptcoveff ==1){
6912: ioffset=4; /* Age is in 4 */
6913: }else{
6914: ioffset=6; /* Age is in 6 */
6915: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
6916: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
6917: }
6918: fprintf(ficgp," u %d:(",ioffset);
6919: kl=0;
6920: strcpy(gplotcondition,"(");
6921: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
6922: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
6923: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6924: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6925: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6926: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
6927: kl++;
6928: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
6929: kl++;
6930: if(k <cptcoveff && cptcoveff>1)
6931: sprintf(gplotcondition+strlen(gplotcondition)," && ");
6932: }
6933: strcpy(gplotcondition+strlen(gplotcondition),")");
6934: /* 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 *\/ */
6935: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6936: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6937: /* '' 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*/
6938: if(i==nlstate+1){
6939: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p.%d' with line ", gplotcondition, \
6940: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
6941: }else{
6942: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
6943: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
6944: }
6945: } /* end if covariate */
6946: } /* nlstate */
6947: fprintf(ficgp,"\nset out\n");
1.223 brouard 6948: } /* end cpt state*/
6949: } /* end covariate */
6950: } /* End if prevfcast */
1.227 brouard 6951:
6952:
1.238 brouard 6953: /* 9eme writing MLE parameters */
6954: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 6955: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 6956: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 6957: for(k=1; k <=(nlstate+ndeath); k++){
6958: if (k != i) {
1.227 brouard 6959: fprintf(ficgp,"# current state %d\n",k);
6960: for(j=1; j <=ncovmodel; j++){
6961: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
6962: jk++;
6963: }
6964: fprintf(ficgp,"\n");
1.126 brouard 6965: }
6966: }
1.223 brouard 6967: }
1.187 brouard 6968: fprintf(ficgp,"##############\n#\n");
1.227 brouard 6969:
1.145 brouard 6970: /*goto avoid;*/
1.238 brouard 6971: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
6972: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 6973: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
6974: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
6975: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
6976: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
6977: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6978: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6979: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6980: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
6981: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
6982: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
6983: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
6984: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
6985: fprintf(ficgp,"#\n");
1.223 brouard 6986: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 6987: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 6988: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 6989: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.237 brouard 6990: fprintf(ficgp,"# jk=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
6991: for(jk=1; jk <=m; jk++) /* For each combination of covariate */
6992: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
6993: if(TKresult[nres]!= jk)
6994: continue;
6995: fprintf(ficgp,"# Combination of dummy jk=%d and ",jk);
6996: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6997: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6998: }
6999: fprintf(ficgp,"\n#\n");
1.241 brouard 7000: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),jk,ng,nres);
1.223 brouard 7001: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7002: if (ng==1){
7003: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7004: fprintf(ficgp,"\nunset log y");
7005: }else if (ng==2){
7006: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7007: fprintf(ficgp,"\nset log y");
7008: }else if (ng==3){
7009: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7010: fprintf(ficgp,"\nset log y");
7011: }else
7012: fprintf(ficgp,"\nunset title ");
7013: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7014: i=1;
7015: for(k2=1; k2<=nlstate; k2++) {
7016: k3=i;
7017: for(k=1; k<=(nlstate+ndeath); k++) {
7018: if (k != k2){
7019: switch( ng) {
7020: case 1:
7021: if(nagesqr==0)
7022: fprintf(ficgp," p%d+p%d*x",i,i+1);
7023: else /* nagesqr =1 */
7024: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7025: break;
7026: case 2: /* ng=2 */
7027: if(nagesqr==0)
7028: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7029: else /* nagesqr =1 */
7030: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7031: break;
7032: case 3:
7033: if(nagesqr==0)
7034: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7035: else /* nagesqr =1 */
7036: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7037: break;
7038: }
7039: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7040: ijp=1; /* product no age */
7041: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7042: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7043: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7044: if(j==Tage[ij]) { /* Product by age */
7045: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7046: if(DummyV[j]==0){
1.237 brouard 7047: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7048: }else{ /* quantitative */
7049: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
7050: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7051: }
7052: ij++;
7053: }
7054: }else if(j==Tprod[ijp]) { /* */
7055: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7056: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7057: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7058: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.237 brouard 7059: /* 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)]); */
7060: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7061: }else{ /* Vn is dummy and Vm is quanti */
7062: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(jk,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
7063: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7064: }
7065: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7066: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7067: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7068: }else{ /* Both quanti */
7069: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7070: }
7071: }
1.238 brouard 7072: ijp++;
1.237 brouard 7073: }
7074: } else{ /* simple covariate */
7075: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(jk,j)]); /\* Valgrind bug nbcode *\/ */
7076: if(Dummy[j]==0){
7077: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7078: }else{ /* quantitative */
7079: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.223 brouard 7080: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7081: }
1.237 brouard 7082: } /* end simple */
7083: } /* end j */
1.223 brouard 7084: }else{
7085: i=i-ncovmodel;
7086: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7087: fprintf(ficgp," (1.");
7088: }
1.227 brouard 7089:
1.223 brouard 7090: if(ng != 1){
7091: fprintf(ficgp,")/(1");
1.227 brouard 7092:
1.223 brouard 7093: for(k1=1; k1 <=nlstate; k1++){
7094: if(nagesqr==0)
7095: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(k1-1)*ncovmodel,k3+(k1-1)*ncovmodel+1);
7096: else /* nagesqr =1 */
7097: 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 7098:
1.223 brouard 7099: ij=1;
7100: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7101: if((j-2)==Tage[ij]) { /* Bug valgrind */
7102: if(ij <=cptcovage) { /* Bug valgrind */
1.223 brouard 7103: fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,j-2)]);
7104: /* fprintf(ficgp,"+p%d*%d*x",k3+(k1-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(jk,Tvar[j-2])]); */
7105: ij++;
7106: }
7107: }
7108: else
1.225 brouard 7109: 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 7110: }
7111: fprintf(ficgp,")");
7112: }
7113: fprintf(ficgp,")");
7114: if(ng ==2)
7115: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7116: else /* ng= 3 */
7117: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7118: }else{ /* end ng <> 1 */
7119: if( k !=k2) /* logit p11 is hard to draw */
7120: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7121: }
7122: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7123: fprintf(ficgp,",");
7124: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7125: fprintf(ficgp,",");
7126: i=i+ncovmodel;
7127: } /* end k */
7128: } /* end k2 */
7129: fprintf(ficgp,"\n set out\n");
7130: } /* end jk */
7131: } /* end ng */
7132: /* avoid: */
7133: fflush(ficgp);
1.126 brouard 7134: } /* end gnuplot */
7135:
7136:
7137: /*************** Moving average **************/
1.219 brouard 7138: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7139: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7140:
1.222 brouard 7141: int i, cpt, cptcod;
7142: int modcovmax =1;
7143: int mobilavrange, mob;
7144: int iage=0;
7145:
7146: double sum=0.;
7147: double age;
7148: double *sumnewp, *sumnewm;
7149: double *agemingood, *agemaxgood; /* Currently identical for all covariates */
7150:
7151:
1.225 brouard 7152: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7153: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7154:
7155: sumnewp = vector(1,ncovcombmax);
7156: sumnewm = vector(1,ncovcombmax);
7157: agemingood = vector(1,ncovcombmax);
7158: agemaxgood = vector(1,ncovcombmax);
7159:
7160: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7161: sumnewm[cptcod]=0.;
7162: sumnewp[cptcod]=0.;
7163: agemingood[cptcod]=0;
7164: agemaxgood[cptcod]=0;
7165: }
7166: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7167:
7168: if(mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7169: if(mobilav==1) mobilavrange=5; /* default */
7170: else mobilavrange=mobilav;
7171: for (age=bage; age<=fage; age++)
7172: for (i=1; i<=nlstate;i++)
7173: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7174: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7175: /* We keep the original values on the extreme ages bage, fage and for
7176: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7177: we use a 5 terms etc. until the borders are no more concerned.
7178: */
7179: for (mob=3;mob <=mobilavrange;mob=mob+2){
7180: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
7181: for (i=1; i<=nlstate;i++){
7182: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7183: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7184: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7185: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7186: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7187: }
7188: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
7189: }
7190: }
7191: }/* end age */
7192: }/* end mob */
7193: }else
7194: return -1;
7195: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7196: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7197: if(invalidvarcomb[cptcod]){
7198: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7199: continue;
7200: }
1.219 brouard 7201:
1.222 brouard 7202: agemingood[cptcod]=fage-(mob-1)/2;
7203: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, finding the youngest wrong */
7204: sumnewm[cptcod]=0.;
7205: for (i=1; i<=nlstate;i++){
7206: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7207: }
7208: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7209: agemingood[cptcod]=age;
7210: }else{ /* bad */
7211: for (i=1; i<=nlstate;i++){
7212: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7213: } /* i */
7214: } /* end bad */
7215: }/* age */
7216: sum=0.;
7217: for (i=1; i<=nlstate;i++){
7218: sum+=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7219: }
7220: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7221: 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);
7222: /* for (i=1; i<=nlstate;i++){ */
7223: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7224: /* } /\* i *\/ */
7225: } /* end bad */
7226: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7227: /* From youngest, finding the oldest wrong */
7228: agemaxgood[cptcod]=bage+(mob-1)/2;
7229: for (age=bage+(mob-1)/2; age<=fage; age++){
7230: sumnewm[cptcod]=0.;
7231: for (i=1; i<=nlstate;i++){
7232: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7233: }
7234: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7235: agemaxgood[cptcod]=age;
7236: }else{ /* bad */
7237: for (i=1; i<=nlstate;i++){
7238: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7239: } /* i */
7240: } /* end bad */
7241: }/* age */
7242: sum=0.;
7243: for (i=1; i<=nlstate;i++){
7244: sum+=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7245: }
7246: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7247: 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);
7248: /* for (i=1; i<=nlstate;i++){ */
7249: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7250: /* } /\* i *\/ */
7251: } /* end bad */
7252:
7253: for (age=bage; age<=fage; age++){
1.235 brouard 7254: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7255: sumnewp[cptcod]=0.;
7256: sumnewm[cptcod]=0.;
7257: for (i=1; i<=nlstate;i++){
7258: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7259: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7260: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7261: }
7262: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7263: }
7264: /* printf("\n"); */
7265: /* } */
7266: /* brutal averaging */
7267: for (i=1; i<=nlstate;i++){
7268: for (age=1; age<=bage; age++){
7269: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7270: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7271: }
7272: for (age=fage; age<=AGESUP; age++){
7273: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7274: /* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); */
7275: }
7276: } /* end i status */
7277: for (i=nlstate+1; i<=nlstate+ndeath;i++){
7278: for (age=1; age<=AGESUP; age++){
7279: /*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*/
7280: mobaverage[(int)age][i][cptcod]=0.;
7281: }
7282: }
7283: }/* end cptcod */
7284: free_vector(sumnewm,1, ncovcombmax);
7285: free_vector(sumnewp,1, ncovcombmax);
7286: free_vector(agemaxgood,1, ncovcombmax);
7287: free_vector(agemingood,1, ncovcombmax);
7288: return 0;
7289: }/* End movingaverage */
1.218 brouard 7290:
1.126 brouard 7291:
7292: /************** Forecasting ******************/
1.235 brouard 7293: 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 7294: /* proj1, year, month, day of starting projection
7295: agemin, agemax range of age
7296: dateprev1 dateprev2 range of dates during which prevalence is computed
7297: anproj2 year of en of projection (same day and month as proj1).
7298: */
1.235 brouard 7299: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7300: double agec; /* generic age */
7301: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7302: double *popeffectif,*popcount;
7303: double ***p3mat;
1.218 brouard 7304: /* double ***mobaverage; */
1.126 brouard 7305: char fileresf[FILENAMELENGTH];
7306:
7307: agelim=AGESUP;
1.211 brouard 7308: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7309: in each health status at the date of interview (if between dateprev1 and dateprev2).
7310: We still use firstpass and lastpass as another selection.
7311: */
1.214 brouard 7312: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7313: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7314:
1.201 brouard 7315: strcpy(fileresf,"F_");
7316: strcat(fileresf,fileresu);
1.126 brouard 7317: if((ficresf=fopen(fileresf,"w"))==NULL) {
7318: printf("Problem with forecast resultfile: %s\n", fileresf);
7319: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7320: }
1.235 brouard 7321: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7322: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7323:
1.225 brouard 7324: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7325:
7326:
7327: stepsize=(int) (stepm+YEARM-1)/YEARM;
7328: if (stepm<=12) stepsize=1;
7329: if(estepm < stepm){
7330: printf ("Problem %d lower than %d\n",estepm, stepm);
7331: }
7332: else hstepm=estepm;
7333:
7334: hstepm=hstepm/stepm;
7335: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7336: fractional in yp1 */
7337: anprojmean=yp;
7338: yp2=modf((yp1*12),&yp);
7339: mprojmean=yp;
7340: yp1=modf((yp2*30.5),&yp);
7341: jprojmean=yp;
7342: if(jprojmean==0) jprojmean=1;
7343: if(mprojmean==0) jprojmean=1;
7344:
1.227 brouard 7345: i1=pow(2,cptcoveff);
1.126 brouard 7346: if (cptcovn < 1){i1=1;}
7347:
7348: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7349:
7350: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7351:
1.126 brouard 7352: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7353: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7354: for(k=1; k<=i1;k++){
7355: if(TKresult[nres]!= k)
7356: continue;
1.227 brouard 7357: if(invalidvarcomb[k]){
7358: printf("\nCombination (%d) projection ignored because no cases \n",k);
7359: continue;
7360: }
7361: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7362: for(j=1;j<=cptcoveff;j++) {
7363: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7364: }
1.235 brouard 7365: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7366: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7367: }
1.227 brouard 7368: fprintf(ficresf," yearproj age");
7369: for(j=1; j<=nlstate+ndeath;j++){
7370: for(i=1; i<=nlstate;i++)
7371: fprintf(ficresf," p%d%d",i,j);
7372: fprintf(ficresf," wp.%d",j);
7373: }
7374: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7375: fprintf(ficresf,"\n");
7376: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7377: for (agec=fage; agec>=(ageminpar-1); agec--){
7378: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7379: nhstepm = nhstepm/hstepm;
7380: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7381: oldm=oldms;savm=savms;
1.235 brouard 7382: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7383:
7384: for (h=0; h<=nhstepm; h++){
7385: if (h*hstepm/YEARM*stepm ==yearp) {
7386: fprintf(ficresf,"\n");
7387: for(j=1;j<=cptcoveff;j++)
7388: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7389: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7390: }
7391: for(j=1; j<=nlstate+ndeath;j++) {
7392: ppij=0.;
7393: for(i=1; i<=nlstate;i++) {
7394: if (mobilav==1)
7395: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
7396: else {
7397: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
7398: }
7399: if (h*hstepm/YEARM*stepm== yearp) {
7400: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7401: }
7402: } /* end i */
7403: if (h*hstepm/YEARM*stepm==yearp) {
7404: fprintf(ficresf," %.3f", ppij);
7405: }
7406: }/* end j */
7407: } /* end h */
7408: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7409: } /* end agec */
7410: } /* end yearp */
7411: } /* end k */
1.219 brouard 7412:
1.126 brouard 7413: fclose(ficresf);
1.215 brouard 7414: printf("End of Computing forecasting \n");
7415: fprintf(ficlog,"End of Computing forecasting\n");
7416:
1.126 brouard 7417: }
7418:
1.218 brouard 7419: /* /\************** Back Forecasting ******************\/ */
1.225 brouard 7420: /* 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 7421: /* /\* back1, year, month, day of starting backection */
7422: /* agemin, agemax range of age */
7423: /* dateprev1 dateprev2 range of dates during which prevalence is computed */
7424: /* anback2 year of en of backection (same day and month as back1). */
7425: /* *\/ */
7426: /* int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1; */
7427: /* double agec; /\* generic age *\/ */
7428: /* double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean; */
7429: /* double *popeffectif,*popcount; */
7430: /* double ***p3mat; */
7431: /* /\* double ***mobaverage; *\/ */
7432: /* char fileresfb[FILENAMELENGTH]; */
7433:
7434: /* agelim=AGESUP; */
7435: /* /\* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people */
7436: /* in each health status at the date of interview (if between dateprev1 and dateprev2). */
7437: /* We still use firstpass and lastpass as another selection. */
7438: /* *\/ */
7439: /* /\* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ *\/ */
7440: /* /\* firstpass, lastpass, stepm, weightopt, model); *\/ */
7441: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
7442:
7443: /* strcpy(fileresfb,"FB_"); */
7444: /* strcat(fileresfb,fileresu); */
7445: /* if((ficresfb=fopen(fileresfb,"w"))==NULL) { */
7446: /* printf("Problem with back forecast resultfile: %s\n", fileresfb); */
7447: /* fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb); */
7448: /* } */
7449: /* printf("Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7450: /* fprintf(ficlog,"Computing back forecasting: result on file '%s', please wait... \n", fileresfb); */
7451:
1.225 brouard 7452: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.218 brouard 7453:
7454: /* /\* if (mobilav!=0) { *\/ */
7455: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7456: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7457: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7458: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7459: /* /\* } *\/ */
7460: /* /\* } *\/ */
7461:
7462: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7463: /* if (stepm<=12) stepsize=1; */
7464: /* if(estepm < stepm){ */
7465: /* printf ("Problem %d lower than %d\n",estepm, stepm); */
7466: /* } */
7467: /* else hstepm=estepm; */
7468:
7469: /* hstepm=hstepm/stepm; */
7470: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
7471: /* fractional in yp1 *\/ */
7472: /* anprojmean=yp; */
7473: /* yp2=modf((yp1*12),&yp); */
7474: /* mprojmean=yp; */
7475: /* yp1=modf((yp2*30.5),&yp); */
7476: /* jprojmean=yp; */
7477: /* if(jprojmean==0) jprojmean=1; */
7478: /* if(mprojmean==0) jprojmean=1; */
7479:
1.225 brouard 7480: /* i1=cptcoveff; */
1.218 brouard 7481: /* if (cptcovn < 1){i1=1;} */
1.217 brouard 7482:
1.218 brouard 7483: /* fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2); */
1.217 brouard 7484:
1.218 brouard 7485: /* fprintf(ficresfb,"#****** Routine prevbackforecast **\n"); */
7486:
7487: /* /\* if (h==(int)(YEARM*yearp)){ *\/ */
7488: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
1.225 brouard 7489: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
1.218 brouard 7490: /* k=k+1; */
7491: /* fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#"); */
1.225 brouard 7492: /* for(j=1;j<=cptcoveff;j++) { */
1.218 brouard 7493: /* fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7494: /* } */
7495: /* fprintf(ficresfb," yearbproj age"); */
7496: /* for(j=1; j<=nlstate+ndeath;j++){ */
7497: /* for(i=1; i<=nlstate;i++) */
7498: /* fprintf(ficresfb," p%d%d",i,j); */
7499: /* fprintf(ficresfb," p.%d",j); */
7500: /* } */
7501: /* for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) { */
7502: /* /\* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { *\/ */
7503: /* fprintf(ficresfb,"\n"); */
7504: /* fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
7505: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
7506: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
7507: /* nhstepm = nhstepm/hstepm; */
7508: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7509: /* oldm=oldms;savm=savms; */
7510: /* hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm,oldm,savm, dnewm, doldm, dsavm, k); */
7511: /* for (h=0; h<=nhstepm; h++){ */
7512: /* if (h*hstepm/YEARM*stepm ==yearp) { */
7513: /* fprintf(ficresfb,"\n"); */
1.225 brouard 7514: /* for(j=1;j<=cptcoveff;j++) */
1.218 brouard 7515: /* fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7516: /* fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm); */
7517: /* } */
7518: /* for(j=1; j<=nlstate+ndeath;j++) { */
7519: /* ppij=0.; */
7520: /* for(i=1; i<=nlstate;i++) { */
7521: /* if (mobilav==1) */
7522: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][cptcod]; */
7523: /* else { */
7524: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][cptcod]; */
7525: /* } */
7526: /* if (h*hstepm/YEARM*stepm== yearp) { */
7527: /* fprintf(ficresfb," %.3f", p3mat[i][j][h]); */
7528: /* } */
7529: /* } /\* end i *\/ */
7530: /* if (h*hstepm/YEARM*stepm==yearp) { */
7531: /* fprintf(ficresfb," %.3f", ppij); */
7532: /* } */
7533: /* }/\* end j *\/ */
7534: /* } /\* end h *\/ */
7535: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7536: /* } /\* end agec *\/ */
7537: /* } /\* end yearp *\/ */
7538: /* } /\* end cptcod *\/ */
7539: /* } /\* end cptcov *\/ */
7540:
7541: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7542:
7543: /* fclose(ficresfb); */
7544: /* printf("End of Computing Back forecasting \n"); */
7545: /* fprintf(ficlog,"End of Computing Back forecasting\n"); */
1.217 brouard 7546:
1.218 brouard 7547: /* } */
1.217 brouard 7548:
1.126 brouard 7549: /************** Forecasting *****not tested NB*************/
1.227 brouard 7550: /* 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 7551:
1.227 brouard 7552: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
7553: /* int *popage; */
7554: /* double calagedatem, agelim, kk1, kk2; */
7555: /* double *popeffectif,*popcount; */
7556: /* double ***p3mat,***tabpop,***tabpopprev; */
7557: /* /\* double ***mobaverage; *\/ */
7558: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 7559:
1.227 brouard 7560: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7561: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7562: /* agelim=AGESUP; */
7563: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 7564:
1.227 brouard 7565: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 7566:
7567:
1.227 brouard 7568: /* strcpy(filerespop,"POP_"); */
7569: /* strcat(filerespop,fileresu); */
7570: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
7571: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
7572: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
7573: /* } */
7574: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
7575: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 7576:
1.227 brouard 7577: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 7578:
1.227 brouard 7579: /* /\* if (mobilav!=0) { *\/ */
7580: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
7581: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
7582: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7583: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
7584: /* /\* } *\/ */
7585: /* /\* } *\/ */
1.126 brouard 7586:
1.227 brouard 7587: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
7588: /* if (stepm<=12) stepsize=1; */
1.126 brouard 7589:
1.227 brouard 7590: /* agelim=AGESUP; */
1.126 brouard 7591:
1.227 brouard 7592: /* hstepm=1; */
7593: /* hstepm=hstepm/stepm; */
1.218 brouard 7594:
1.227 brouard 7595: /* if (popforecast==1) { */
7596: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
7597: /* printf("Problem with population file : %s\n",popfile);exit(0); */
7598: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
7599: /* } */
7600: /* popage=ivector(0,AGESUP); */
7601: /* popeffectif=vector(0,AGESUP); */
7602: /* popcount=vector(0,AGESUP); */
1.126 brouard 7603:
1.227 brouard 7604: /* i=1; */
7605: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 7606:
1.227 brouard 7607: /* imx=i; */
7608: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
7609: /* } */
1.218 brouard 7610:
1.227 brouard 7611: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
7612: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
7613: /* k=k+1; */
7614: /* fprintf(ficrespop,"\n#******"); */
7615: /* for(j=1;j<=cptcoveff;j++) { */
7616: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
7617: /* } */
7618: /* fprintf(ficrespop,"******\n"); */
7619: /* fprintf(ficrespop,"# Age"); */
7620: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
7621: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 7622:
1.227 brouard 7623: /* for (cpt=0; cpt<=0;cpt++) { */
7624: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 7625:
1.227 brouard 7626: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7627: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7628: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7629:
1.227 brouard 7630: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7631: /* oldm=oldms;savm=savms; */
7632: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 7633:
1.227 brouard 7634: /* for (h=0; h<=nhstepm; h++){ */
7635: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7636: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7637: /* } */
7638: /* for(j=1; j<=nlstate+ndeath;j++) { */
7639: /* kk1=0.;kk2=0; */
7640: /* for(i=1; i<=nlstate;i++) { */
7641: /* if (mobilav==1) */
7642: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
7643: /* else { */
7644: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
7645: /* } */
7646: /* } */
7647: /* if (h==(int)(calagedatem+12*cpt)){ */
7648: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
7649: /* /\*fprintf(ficrespop," %.3f", kk1); */
7650: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
7651: /* } */
7652: /* } */
7653: /* for(i=1; i<=nlstate;i++){ */
7654: /* kk1=0.; */
7655: /* for(j=1; j<=nlstate;j++){ */
7656: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
7657: /* } */
7658: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
7659: /* } */
1.218 brouard 7660:
1.227 brouard 7661: /* if (h==(int)(calagedatem+12*cpt)) */
7662: /* for(j=1; j<=nlstate;j++) */
7663: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
7664: /* } */
7665: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7666: /* } */
7667: /* } */
1.218 brouard 7668:
1.227 brouard 7669: /* /\******\/ */
1.218 brouard 7670:
1.227 brouard 7671: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
7672: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
7673: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
7674: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
7675: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 7676:
1.227 brouard 7677: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7678: /* oldm=oldms;savm=savms; */
7679: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
7680: /* for (h=0; h<=nhstepm; h++){ */
7681: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
7682: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
7683: /* } */
7684: /* for(j=1; j<=nlstate+ndeath;j++) { */
7685: /* kk1=0.;kk2=0; */
7686: /* for(i=1; i<=nlstate;i++) { */
7687: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
7688: /* } */
7689: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
7690: /* } */
7691: /* } */
7692: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
7693: /* } */
7694: /* } */
7695: /* } */
7696: /* } */
1.218 brouard 7697:
1.227 brouard 7698: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 7699:
1.227 brouard 7700: /* if (popforecast==1) { */
7701: /* free_ivector(popage,0,AGESUP); */
7702: /* free_vector(popeffectif,0,AGESUP); */
7703: /* free_vector(popcount,0,AGESUP); */
7704: /* } */
7705: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7706: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7707: /* fclose(ficrespop); */
7708: /* } /\* End of popforecast *\/ */
1.218 brouard 7709:
1.126 brouard 7710: int fileappend(FILE *fichier, char *optionfich)
7711: {
7712: if((fichier=fopen(optionfich,"a"))==NULL) {
7713: printf("Problem with file: %s\n", optionfich);
7714: fprintf(ficlog,"Problem with file: %s\n", optionfich);
7715: return (0);
7716: }
7717: fflush(fichier);
7718: return (1);
7719: }
7720:
7721:
7722: /**************** function prwizard **********************/
7723: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
7724: {
7725:
7726: /* Wizard to print covariance matrix template */
7727:
1.164 brouard 7728: char ca[32], cb[32];
7729: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 7730: int numlinepar;
7731:
7732: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7733: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
7734: for(i=1; i <=nlstate; i++){
7735: jj=0;
7736: for(j=1; j <=nlstate+ndeath; j++){
7737: if(j==i) continue;
7738: jj++;
7739: /*ca[0]= k+'a'-1;ca[1]='\0';*/
7740: printf("%1d%1d",i,j);
7741: fprintf(ficparo,"%1d%1d",i,j);
7742: for(k=1; k<=ncovmodel;k++){
7743: /* printf(" %lf",param[i][j][k]); */
7744: /* fprintf(ficparo," %lf",param[i][j][k]); */
7745: printf(" 0.");
7746: fprintf(ficparo," 0.");
7747: }
7748: printf("\n");
7749: fprintf(ficparo,"\n");
7750: }
7751: }
7752: printf("# Scales (for hessian or gradient estimation)\n");
7753: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
7754: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
7755: for(i=1; i <=nlstate; i++){
7756: jj=0;
7757: for(j=1; j <=nlstate+ndeath; j++){
7758: if(j==i) continue;
7759: jj++;
7760: fprintf(ficparo,"%1d%1d",i,j);
7761: printf("%1d%1d",i,j);
7762: fflush(stdout);
7763: for(k=1; k<=ncovmodel;k++){
7764: /* printf(" %le",delti3[i][j][k]); */
7765: /* fprintf(ficparo," %le",delti3[i][j][k]); */
7766: printf(" 0.");
7767: fprintf(ficparo," 0.");
7768: }
7769: numlinepar++;
7770: printf("\n");
7771: fprintf(ficparo,"\n");
7772: }
7773: }
7774: printf("# Covariance matrix\n");
7775: /* # 121 Var(a12)\n\ */
7776: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7777: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
7778: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
7779: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
7780: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
7781: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
7782: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7783: fflush(stdout);
7784: fprintf(ficparo,"# Covariance matrix\n");
7785: /* # 121 Var(a12)\n\ */
7786: /* # 122 Cov(b12,a12) Var(b12)\n\ */
7787: /* # ...\n\ */
7788: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
7789:
7790: for(itimes=1;itimes<=2;itimes++){
7791: jj=0;
7792: for(i=1; i <=nlstate; i++){
7793: for(j=1; j <=nlstate+ndeath; j++){
7794: if(j==i) continue;
7795: for(k=1; k<=ncovmodel;k++){
7796: jj++;
7797: ca[0]= k+'a'-1;ca[1]='\0';
7798: if(itimes==1){
7799: printf("#%1d%1d%d",i,j,k);
7800: fprintf(ficparo,"#%1d%1d%d",i,j,k);
7801: }else{
7802: printf("%1d%1d%d",i,j,k);
7803: fprintf(ficparo,"%1d%1d%d",i,j,k);
7804: /* printf(" %.5le",matcov[i][j]); */
7805: }
7806: ll=0;
7807: for(li=1;li <=nlstate; li++){
7808: for(lj=1;lj <=nlstate+ndeath; lj++){
7809: if(lj==li) continue;
7810: for(lk=1;lk<=ncovmodel;lk++){
7811: ll++;
7812: if(ll<=jj){
7813: cb[0]= lk +'a'-1;cb[1]='\0';
7814: if(ll<jj){
7815: if(itimes==1){
7816: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7817: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
7818: }else{
7819: printf(" 0.");
7820: fprintf(ficparo," 0.");
7821: }
7822: }else{
7823: if(itimes==1){
7824: printf(" Var(%s%1d%1d)",ca,i,j);
7825: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
7826: }else{
7827: printf(" 0.");
7828: fprintf(ficparo," 0.");
7829: }
7830: }
7831: }
7832: } /* end lk */
7833: } /* end lj */
7834: } /* end li */
7835: printf("\n");
7836: fprintf(ficparo,"\n");
7837: numlinepar++;
7838: } /* end k*/
7839: } /*end j */
7840: } /* end i */
7841: } /* end itimes */
7842:
7843: } /* end of prwizard */
7844: /******************* Gompertz Likelihood ******************************/
7845: double gompertz(double x[])
7846: {
7847: double A,B,L=0.0,sump=0.,num=0.;
7848: int i,n=0; /* n is the size of the sample */
7849:
1.220 brouard 7850: for (i=1;i<=imx ; i++) {
1.126 brouard 7851: sump=sump+weight[i];
7852: /* sump=sump+1;*/
7853: num=num+1;
7854: }
7855:
7856:
7857: /* for (i=0; i<=imx; i++)
7858: 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]);*/
7859:
7860: for (i=1;i<=imx ; i++)
7861: {
7862: if (cens[i] == 1 && wav[i]>1)
7863: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
7864:
7865: if (cens[i] == 0 && wav[i]>1)
7866: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
7867: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
7868:
7869: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7870: if (wav[i] > 1 ) { /* ??? */
7871: L=L+A*weight[i];
7872: /* 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]);*/
7873: }
7874: }
7875:
7876: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7877:
7878: return -2*L*num/sump;
7879: }
7880:
1.136 brouard 7881: #ifdef GSL
7882: /******************* Gompertz_f Likelihood ******************************/
7883: double gompertz_f(const gsl_vector *v, void *params)
7884: {
7885: double A,B,LL=0.0,sump=0.,num=0.;
7886: double *x= (double *) v->data;
7887: int i,n=0; /* n is the size of the sample */
7888:
7889: for (i=0;i<=imx-1 ; i++) {
7890: sump=sump+weight[i];
7891: /* sump=sump+1;*/
7892: num=num+1;
7893: }
7894:
7895:
7896: /* for (i=0; i<=imx; i++)
7897: 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]);*/
7898: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
7899: for (i=1;i<=imx ; i++)
7900: {
7901: if (cens[i] == 1 && wav[i]>1)
7902: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
7903:
7904: if (cens[i] == 0 && wav[i]>1)
7905: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
7906: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
7907:
7908: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
7909: if (wav[i] > 1 ) { /* ??? */
7910: LL=LL+A*weight[i];
7911: /* 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]);*/
7912: }
7913: }
7914:
7915: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
7916: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
7917:
7918: return -2*LL*num/sump;
7919: }
7920: #endif
7921:
1.126 brouard 7922: /******************* Printing html file ***********/
1.201 brouard 7923: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7924: int lastpass, int stepm, int weightopt, char model[],\
7925: int imx, double p[],double **matcov,double agemortsup){
7926: int i,k;
7927:
7928: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
7929: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
7930: for (i=1;i<=2;i++)
7931: 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 7932: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 7933: fprintf(fichtm,"</ul>");
7934:
7935: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
7936:
7937: 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>");
7938:
7939: for (k=agegomp;k<(agemortsup-2);k++)
7940: 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]);
7941:
7942:
7943: fflush(fichtm);
7944: }
7945:
7946: /******************* Gnuplot file **************/
1.201 brouard 7947: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 7948:
7949: char dirfileres[132],optfileres[132];
1.164 brouard 7950:
1.126 brouard 7951: int ng;
7952:
7953:
7954: /*#ifdef windows */
7955: fprintf(ficgp,"cd \"%s\" \n",pathc);
7956: /*#endif */
7957:
7958:
7959: strcpy(dirfileres,optionfilefiname);
7960: strcpy(optfileres,"vpl");
1.199 brouard 7961: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 7962: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 7963: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 7964: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 7965: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
7966:
7967: }
7968:
1.136 brouard 7969: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
7970: {
1.126 brouard 7971:
1.136 brouard 7972: /*-------- data file ----------*/
7973: FILE *fic;
7974: char dummy[]=" ";
1.240 brouard 7975: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 7976: int lstra;
1.136 brouard 7977: int linei, month, year,iout;
7978: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 7979: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 7980: char *stratrunc;
1.223 brouard 7981:
1.240 brouard 7982: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
7983: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 7984:
1.240 brouard 7985: for(v=1; v <=ncovcol;v++){
7986: DummyV[v]=0;
7987: FixedV[v]=0;
7988: }
7989: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
7990: DummyV[v]=1;
7991: FixedV[v]=0;
7992: }
7993: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
7994: DummyV[v]=0;
7995: FixedV[v]=1;
7996: }
7997: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
7998: DummyV[v]=1;
7999: FixedV[v]=1;
8000: }
8001: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8002: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8003: 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]);
8004: }
1.126 brouard 8005:
1.136 brouard 8006: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8007: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8008: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8009: }
1.126 brouard 8010:
1.136 brouard 8011: i=1;
8012: linei=0;
8013: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8014: linei=linei+1;
8015: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8016: if(line[j] == '\t')
8017: line[j] = ' ';
8018: }
8019: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8020: ;
8021: };
8022: line[j+1]=0; /* Trims blanks at end of line */
8023: if(line[0]=='#'){
8024: fprintf(ficlog,"Comment line\n%s\n",line);
8025: printf("Comment line\n%s\n",line);
8026: continue;
8027: }
8028: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8029: strcpy(line, linetmp);
1.223 brouard 8030:
8031: /* Loops on waves */
8032: for (j=maxwav;j>=1;j--){
8033: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8034: cutv(stra, strb, line, ' ');
8035: if(strb[0]=='.') { /* Missing value */
8036: lval=-1;
8037: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8038: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8039: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8040: 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);
8041: 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);
8042: return 1;
8043: }
8044: }else{
8045: errno=0;
8046: /* what_kind_of_number(strb); */
8047: dval=strtod(strb,&endptr);
8048: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8049: /* if(strb != endptr && *endptr == '\0') */
8050: /* dval=dlval; */
8051: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8052: if( strb[0]=='\0' || (*endptr != '\0')){
8053: 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);
8054: 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);
8055: return 1;
8056: }
8057: cotqvar[j][iv][i]=dval;
8058: cotvar[j][ntv+iv][i]=dval;
8059: }
8060: strcpy(line,stra);
1.223 brouard 8061: }/* end loop ntqv */
1.225 brouard 8062:
1.223 brouard 8063: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8064: cutv(stra, strb, line, ' ');
8065: if(strb[0]=='.') { /* Missing value */
8066: lval=-1;
8067: }else{
8068: errno=0;
8069: lval=strtol(strb,&endptr,10);
8070: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8071: if( strb[0]=='\0' || (*endptr != '\0')){
8072: 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);
8073: 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);
8074: return 1;
8075: }
8076: }
8077: if(lval <-1 || lval >1){
8078: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8079: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8080: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8081: For example, for multinomial values like 1, 2 and 3,\n \
8082: build V1=0 V2=0 for the reference value (1),\n \
8083: V1=1 V2=0 for (2) \n \
1.223 brouard 8084: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8085: output of IMaCh is often meaningless.\n \
1.223 brouard 8086: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8087: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8088: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8089: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8090: For example, for multinomial values like 1, 2 and 3,\n \
8091: build V1=0 V2=0 for the reference value (1),\n \
8092: V1=1 V2=0 for (2) \n \
1.223 brouard 8093: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8094: output of IMaCh is often meaningless.\n \
1.223 brouard 8095: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8096: return 1;
8097: }
8098: cotvar[j][iv][i]=(double)(lval);
8099: strcpy(line,stra);
1.223 brouard 8100: }/* end loop ntv */
1.225 brouard 8101:
1.223 brouard 8102: /* Statuses at wave */
1.137 brouard 8103: cutv(stra, strb, line, ' ');
1.223 brouard 8104: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8105: lval=-1;
1.136 brouard 8106: }else{
1.238 brouard 8107: errno=0;
8108: lval=strtol(strb,&endptr,10);
8109: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8110: if( strb[0]=='\0' || (*endptr != '\0')){
8111: 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);
8112: 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);
8113: return 1;
8114: }
1.136 brouard 8115: }
1.225 brouard 8116:
1.136 brouard 8117: s[j][i]=lval;
1.225 brouard 8118:
1.223 brouard 8119: /* Date of Interview */
1.136 brouard 8120: strcpy(line,stra);
8121: cutv(stra, strb,line,' ');
1.169 brouard 8122: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8123: }
1.169 brouard 8124: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8125: month=99;
8126: year=9999;
1.136 brouard 8127: }else{
1.225 brouard 8128: 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);
8129: 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);
8130: return 1;
1.136 brouard 8131: }
8132: anint[j][i]= (double) year;
8133: mint[j][i]= (double)month;
8134: strcpy(line,stra);
1.223 brouard 8135: } /* End loop on waves */
1.225 brouard 8136:
1.223 brouard 8137: /* Date of death */
1.136 brouard 8138: cutv(stra, strb,line,' ');
1.169 brouard 8139: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8140: }
1.169 brouard 8141: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8142: month=99;
8143: year=9999;
8144: }else{
1.141 brouard 8145: 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 8146: 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);
8147: return 1;
1.136 brouard 8148: }
8149: andc[i]=(double) year;
8150: moisdc[i]=(double) month;
8151: strcpy(line,stra);
8152:
1.223 brouard 8153: /* Date of birth */
1.136 brouard 8154: cutv(stra, strb,line,' ');
1.169 brouard 8155: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8156: }
1.169 brouard 8157: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8158: month=99;
8159: year=9999;
8160: }else{
1.141 brouard 8161: 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);
8162: 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 8163: return 1;
1.136 brouard 8164: }
8165: if (year==9999) {
1.141 brouard 8166: 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);
8167: 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 8168: return 1;
8169:
1.136 brouard 8170: }
8171: annais[i]=(double)(year);
8172: moisnais[i]=(double)(month);
8173: strcpy(line,stra);
1.225 brouard 8174:
1.223 brouard 8175: /* Sample weight */
1.136 brouard 8176: cutv(stra, strb,line,' ');
8177: errno=0;
8178: dval=strtod(strb,&endptr);
8179: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8180: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8181: 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 8182: fflush(ficlog);
8183: return 1;
8184: }
8185: weight[i]=dval;
8186: strcpy(line,stra);
1.225 brouard 8187:
1.223 brouard 8188: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8189: cutv(stra, strb, line, ' ');
8190: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8191: lval=-1;
1.223 brouard 8192: }else{
1.225 brouard 8193: errno=0;
8194: /* what_kind_of_number(strb); */
8195: dval=strtod(strb,&endptr);
8196: /* if(strb != endptr && *endptr == '\0') */
8197: /* dval=dlval; */
8198: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8199: if( strb[0]=='\0' || (*endptr != '\0')){
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) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);
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) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);fflush(ficlog);
8202: return 1;
8203: }
8204: coqvar[iv][i]=dval;
1.226 brouard 8205: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8206: }
8207: strcpy(line,stra);
8208: }/* end loop nqv */
1.136 brouard 8209:
1.223 brouard 8210: /* Covariate values */
1.136 brouard 8211: for (j=ncovcol;j>=1;j--){
8212: cutv(stra, strb,line,' ');
1.223 brouard 8213: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8214: lval=-1;
1.136 brouard 8215: }else{
1.225 brouard 8216: errno=0;
8217: lval=strtol(strb,&endptr,10);
8218: if( strb[0]=='\0' || (*endptr != '\0')){
8219: 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);
8220: 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);
8221: return 1;
8222: }
1.136 brouard 8223: }
8224: if(lval <-1 || lval >1){
1.225 brouard 8225: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8226: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8227: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8228: For example, for multinomial values like 1, 2 and 3,\n \
8229: build V1=0 V2=0 for the reference value (1),\n \
8230: V1=1 V2=0 for (2) \n \
1.136 brouard 8231: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8232: output of IMaCh is often meaningless.\n \
1.136 brouard 8233: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8234: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8235: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8236: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8237: For example, for multinomial values like 1, 2 and 3,\n \
8238: build V1=0 V2=0 for the reference value (1),\n \
8239: V1=1 V2=0 for (2) \n \
1.136 brouard 8240: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8241: output of IMaCh is often meaningless.\n \
1.136 brouard 8242: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8243: return 1;
1.136 brouard 8244: }
8245: covar[j][i]=(double)(lval);
8246: strcpy(line,stra);
8247: }
8248: lstra=strlen(stra);
1.225 brouard 8249:
1.136 brouard 8250: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8251: stratrunc = &(stra[lstra-9]);
8252: num[i]=atol(stratrunc);
8253: }
8254: else
8255: num[i]=atol(stra);
8256: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8257: 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;}*/
8258:
8259: i=i+1;
8260: } /* End loop reading data */
1.225 brouard 8261:
1.136 brouard 8262: *imax=i-1; /* Number of individuals */
8263: fclose(fic);
1.225 brouard 8264:
1.136 brouard 8265: return (0);
1.164 brouard 8266: /* endread: */
1.225 brouard 8267: printf("Exiting readdata: ");
8268: fclose(fic);
8269: return (1);
1.223 brouard 8270: }
1.126 brouard 8271:
1.234 brouard 8272: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8273: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8274: while (*p2 == ' ')
1.234 brouard 8275: p2++;
8276: /* while ((*p1++ = *p2++) !=0) */
8277: /* ; */
8278: /* do */
8279: /* while (*p2 == ' ') */
8280: /* p2++; */
8281: /* while (*p1++ == *p2++); */
8282: *stri=p2;
1.145 brouard 8283: }
8284:
1.235 brouard 8285: int decoderesult ( char resultline[], int nres)
1.230 brouard 8286: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8287: {
1.235 brouard 8288: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8289: char resultsav[MAXLINE];
1.234 brouard 8290: int resultmodel[MAXLINE];
8291: int modelresult[MAXLINE];
1.230 brouard 8292: char stra[80], strb[80], strc[80], strd[80],stre[80];
8293:
1.234 brouard 8294: removefirstspace(&resultline);
1.233 brouard 8295: printf("decoderesult:%s\n",resultline);
1.230 brouard 8296:
8297: if (strstr(resultline,"v") !=0){
8298: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8299: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8300: return 1;
8301: }
8302: trimbb(resultsav, resultline);
8303: if (strlen(resultsav) >1){
8304: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8305: }
1.234 brouard 8306: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8307: 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);
8308: 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);
8309: }
8310: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8311: if(nbocc(resultsav,'=') >1){
8312: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8313: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8314: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8315: }else
8316: cutl(strc,strd,resultsav,'=');
1.230 brouard 8317: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8318:
1.230 brouard 8319: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8320: Tvarsel[k]=atoi(strc);
8321: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8322: /* cptcovsel++; */
8323: if (nbocc(stra,'=') >0)
8324: strcpy(resultsav,stra); /* and analyzes it */
8325: }
1.235 brouard 8326: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8327: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8328: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8329: match=0;
1.236 brouard 8330: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8331: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8332: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8333: match=1;
8334: break;
8335: }
8336: }
8337: if(match == 0){
8338: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8339: }
8340: }
8341: }
1.235 brouard 8342: /* Checking for missing or useless values in comparison of current model needs */
8343: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8344: match=0;
1.235 brouard 8345: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8346: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8347: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8348: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8349: ++match;
8350: }
8351: }
8352: }
8353: if(match == 0){
8354: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8355: }else if(match > 1){
8356: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8357: }
8358: }
1.235 brouard 8359:
1.234 brouard 8360: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8361: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8362: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8363: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8364: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8365: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8366: /* 1 0 0 0 */
8367: /* 2 1 0 0 */
8368: /* 3 0 1 0 */
8369: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8370: /* 5 0 0 1 */
8371: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8372: /* 7 0 1 1 */
8373: /* 8 1 1 1 */
1.237 brouard 8374: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8375: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8376: /* V5*age V5 known which value for nres? */
8377: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8378: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8379: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8380: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8381: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8382: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8383: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8384: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8385: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8386: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8387: k4++;;
8388: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8389: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8390: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8391: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8392: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8393: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8394: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8395: k4q++;;
8396: }
8397: }
1.234 brouard 8398:
1.235 brouard 8399: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8400: return (0);
8401: }
1.235 brouard 8402:
1.230 brouard 8403: int decodemodel( char model[], int lastobs)
8404: /**< This routine decodes the model and returns:
1.224 brouard 8405: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8406: * - nagesqr = 1 if age*age in the model, otherwise 0.
8407: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8408: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8409: * - cptcovage number of covariates with age*products =2
8410: * - cptcovs number of simple covariates
8411: * - 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
8412: * which is a new column after the 9 (ncovcol) variables.
8413: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8414: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8415: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8416: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8417: */
1.136 brouard 8418: {
1.238 brouard 8419: int i, j, k, ks, v;
1.227 brouard 8420: int j1, k1, k2, k3, k4;
1.136 brouard 8421: char modelsav[80];
1.145 brouard 8422: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8423: char *strpt;
1.136 brouard 8424:
1.145 brouard 8425: /*removespace(model);*/
1.136 brouard 8426: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8427: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8428: if (strstr(model,"AGE") !=0){
1.192 brouard 8429: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8430: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8431: return 1;
8432: }
1.141 brouard 8433: if (strstr(model,"v") !=0){
8434: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8435: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8436: return 1;
8437: }
1.187 brouard 8438: strcpy(modelsav,model);
8439: if ((strpt=strstr(model,"age*age")) !=0){
8440: printf(" strpt=%s, model=%s\n",strpt, model);
8441: if(strpt != model){
1.234 brouard 8442: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8443: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8444: corresponding column of parameters.\n",model);
1.234 brouard 8445: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8446: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8447: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8448: return 1;
1.225 brouard 8449: }
1.187 brouard 8450: nagesqr=1;
8451: if (strstr(model,"+age*age") !=0)
1.234 brouard 8452: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8453: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8454: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8455: else
1.234 brouard 8456: substrchaine(modelsav, model, "age*age");
1.187 brouard 8457: }else
8458: nagesqr=0;
8459: if (strlen(modelsav) >1){
8460: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
8461: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 8462: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 8463: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 8464: * cst, age and age*age
8465: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
8466: /* including age products which are counted in cptcovage.
8467: * but the covariates which are products must be treated
8468: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 8469: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
8470: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 8471:
8472:
1.187 brouard 8473: /* Design
8474: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
8475: * < ncovcol=8 >
8476: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
8477: * k= 1 2 3 4 5 6 7 8
8478: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
8479: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 8480: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
8481: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 8482: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
8483: * Tage[++cptcovage]=k
8484: * if products, new covar are created after ncovcol with k1
8485: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
8486: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
8487: * 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
8488: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
8489: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
8490: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
8491: * < ncovcol=8 >
8492: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
8493: * k= 1 2 3 4 5 6 7 8 9 10 11 12
8494: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
8495: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8496: * p Tprod[1]@2={ 6, 5}
8497: *p Tvard[1][1]@4= {7, 8, 5, 6}
8498: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
8499: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
8500: *How to reorganize?
8501: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
8502: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
8503: * {2, 1, 4, 8, 5, 6, 3, 7}
8504: * Struct []
8505: */
1.225 brouard 8506:
1.187 brouard 8507: /* This loop fills the array Tvar from the string 'model'.*/
8508: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
8509: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
8510: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
8511: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
8512: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
8513: /* k=1 Tvar[1]=2 (from V2) */
8514: /* k=5 Tvar[5] */
8515: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 8516: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 8517: /* } */
1.198 brouard 8518: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 8519: /*
8520: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 8521: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
8522: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
8523: }
1.187 brouard 8524: cptcovage=0;
8525: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 8526: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 8527: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 8528: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
8529: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
8530: /*scanf("%d",i);*/
8531: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
8532: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
8533: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
8534: /* covar is not filled and then is empty */
8535: cptcovprod--;
8536: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
8537: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
8538: Typevar[k]=1; /* 1 for age product */
8539: cptcovage++; /* Sums the number of covariates which include age as a product */
8540: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
8541: /*printf("stre=%s ", stre);*/
8542: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
8543: cptcovprod--;
8544: cutl(stre,strb,strc,'V');
8545: Tvar[k]=atoi(stre);
8546: Typevar[k]=1; /* 1 for age product */
8547: cptcovage++;
8548: Tage[cptcovage]=k;
8549: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
8550: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
8551: cptcovn++;
8552: cptcovprodnoage++;k1++;
8553: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
8554: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
8555: because this model-covariate is a construction we invent a new column
8556: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
8557: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
8558: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
8559: Typevar[k]=2; /* 2 for double fixed dummy covariates */
8560: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
8561: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
8562: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
8563: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
8564: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
8565: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
8566: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
8567: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 8568: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 8569: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
8570: for (i=1; i<=lastobs;i++){
8571: /* Computes the new covariate which is a product of
8572: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
8573: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
8574: }
8575: } /* End age is not in the model */
8576: } /* End if model includes a product */
8577: else { /* no more sum */
8578: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
8579: /* scanf("%d",i);*/
8580: cutl(strd,strc,strb,'V');
8581: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
8582: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
8583: Tvar[k]=atoi(strd);
8584: Typevar[k]=0; /* 0 for simple covariates */
8585: }
8586: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 8587: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 8588: scanf("%d",i);*/
1.187 brouard 8589: } /* end of loop + on total covariates */
8590: } /* end if strlen(modelsave == 0) age*age might exist */
8591: } /* end if strlen(model == 0) */
1.136 brouard 8592:
8593: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
8594: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 8595:
1.136 brouard 8596: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 8597: printf("cptcovprod=%d ", cptcovprod);
8598: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
8599: scanf("%d ",i);*/
8600:
8601:
1.230 brouard 8602: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
8603: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 8604: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
8605: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
8606: k = 1 2 3 4 5 6 7 8 9
8607: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
8608: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 8609: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
8610: Dummy[k] 1 0 0 0 3 1 1 2 3
8611: Tmodelind[combination of covar]=k;
1.225 brouard 8612: */
8613: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 8614: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 8615: /* 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 8616: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 8617: printf("Model=%s\n\
8618: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8619: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8620: 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);
8621: fprintf(ficlog,"Model=%s\n\
8622: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
8623: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
8624: 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 8625: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 8626: 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 */
8627: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 8628: Fixed[k]= 0;
8629: Dummy[k]= 0;
1.225 brouard 8630: ncoveff++;
1.232 brouard 8631: ncovf++;
1.234 brouard 8632: nsd++;
8633: modell[k].maintype= FTYPE;
8634: TvarsD[nsd]=Tvar[k];
8635: TvarsDind[nsd]=k;
8636: TvarF[ncovf]=Tvar[k];
8637: TvarFind[ncovf]=k;
8638: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8639: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8640: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
8641: Fixed[k]= 0;
8642: Dummy[k]= 0;
8643: ncoveff++;
8644: ncovf++;
8645: modell[k].maintype= FTYPE;
8646: TvarF[ncovf]=Tvar[k];
8647: TvarFind[ncovf]=k;
1.230 brouard 8648: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 8649: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 8650: }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 8651: Fixed[k]= 0;
8652: Dummy[k]= 1;
1.230 brouard 8653: nqfveff++;
1.234 brouard 8654: modell[k].maintype= FTYPE;
8655: modell[k].subtype= FQ;
8656: nsq++;
8657: TvarsQ[nsq]=Tvar[k];
8658: TvarsQind[nsq]=k;
1.232 brouard 8659: ncovf++;
1.234 brouard 8660: TvarF[ncovf]=Tvar[k];
8661: TvarFind[ncovf]=k;
1.231 brouard 8662: 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 8663: 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 8664: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 8665: Fixed[k]= 1;
8666: Dummy[k]= 0;
1.225 brouard 8667: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 8668: modell[k].maintype= VTYPE;
8669: modell[k].subtype= VD;
8670: nsd++;
8671: TvarsD[nsd]=Tvar[k];
8672: TvarsDind[nsd]=k;
8673: ncovv++; /* Only simple time varying variables */
8674: TvarV[ncovv]=Tvar[k];
1.242 brouard 8675: 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 8676: 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 */
8677: 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 8678: 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);
8679: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 8680: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 8681: Fixed[k]= 1;
8682: Dummy[k]= 1;
8683: nqtveff++;
8684: modell[k].maintype= VTYPE;
8685: modell[k].subtype= VQ;
8686: ncovv++; /* Only simple time varying variables */
8687: nsq++;
8688: TvarsQ[nsq]=Tvar[k];
8689: TvarsQind[nsq]=k;
8690: TvarV[ncovv]=Tvar[k];
1.242 brouard 8691: 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 8692: 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 */
8693: 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 8694: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
8695: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
8696: 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 8697: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 8698: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 8699: ncova++;
8700: TvarA[ncova]=Tvar[k];
8701: TvarAind[ncova]=k;
1.231 brouard 8702: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 8703: Fixed[k]= 2;
8704: Dummy[k]= 2;
8705: modell[k].maintype= ATYPE;
8706: modell[k].subtype= APFD;
8707: /* ncoveff++; */
1.227 brouard 8708: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 8709: Fixed[k]= 2;
8710: Dummy[k]= 3;
8711: modell[k].maintype= ATYPE;
8712: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
8713: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 8714: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 8715: Fixed[k]= 3;
8716: Dummy[k]= 2;
8717: modell[k].maintype= ATYPE;
8718: modell[k].subtype= APVD; /* Product age * varying dummy */
8719: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 8720: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8721: Fixed[k]= 3;
8722: Dummy[k]= 3;
8723: modell[k].maintype= ATYPE;
8724: modell[k].subtype= APVQ; /* Product age * varying quantitative */
8725: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 8726: }
8727: }else if (Typevar[k] == 2) { /* product without age */
8728: k1=Tposprod[k];
8729: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 8730: if(Tvard[k1][2] <=ncovcol){
8731: Fixed[k]= 1;
8732: Dummy[k]= 0;
8733: modell[k].maintype= FTYPE;
8734: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
8735: ncovf++; /* Fixed variables without age */
8736: TvarF[ncovf]=Tvar[k];
8737: TvarFind[ncovf]=k;
8738: }else if(Tvard[k1][2] <=ncovcol+nqv){
8739: Fixed[k]= 0; /* or 2 ?*/
8740: Dummy[k]= 1;
8741: modell[k].maintype= FTYPE;
8742: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
8743: ncovf++; /* Varying variables without age */
8744: TvarF[ncovf]=Tvar[k];
8745: TvarFind[ncovf]=k;
8746: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8747: Fixed[k]= 1;
8748: Dummy[k]= 0;
8749: modell[k].maintype= VTYPE;
8750: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
8751: ncovv++; /* Varying variables without age */
8752: TvarV[ncovv]=Tvar[k];
8753: TvarVind[ncovv]=k;
8754: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8755: Fixed[k]= 1;
8756: Dummy[k]= 1;
8757: modell[k].maintype= VTYPE;
8758: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
8759: ncovv++; /* Varying variables without age */
8760: TvarV[ncovv]=Tvar[k];
8761: TvarVind[ncovv]=k;
8762: }
1.227 brouard 8763: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 8764: if(Tvard[k1][2] <=ncovcol){
8765: Fixed[k]= 0; /* or 2 ?*/
8766: Dummy[k]= 1;
8767: modell[k].maintype= FTYPE;
8768: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
8769: ncovf++; /* Fixed variables without age */
8770: TvarF[ncovf]=Tvar[k];
8771: TvarFind[ncovf]=k;
8772: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8773: Fixed[k]= 1;
8774: Dummy[k]= 1;
8775: modell[k].maintype= VTYPE;
8776: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
8777: ncovv++; /* Varying variables without age */
8778: TvarV[ncovv]=Tvar[k];
8779: TvarVind[ncovv]=k;
8780: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8781: Fixed[k]= 1;
8782: Dummy[k]= 1;
8783: modell[k].maintype= VTYPE;
8784: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
8785: ncovv++; /* Varying variables without age */
8786: TvarV[ncovv]=Tvar[k];
8787: TvarVind[ncovv]=k;
8788: ncovv++; /* Varying variables without age */
8789: TvarV[ncovv]=Tvar[k];
8790: TvarVind[ncovv]=k;
8791: }
1.227 brouard 8792: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 8793: if(Tvard[k1][2] <=ncovcol){
8794: Fixed[k]= 1;
8795: Dummy[k]= 1;
8796: modell[k].maintype= VTYPE;
8797: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
8798: ncovv++; /* Varying variables without age */
8799: TvarV[ncovv]=Tvar[k];
8800: TvarVind[ncovv]=k;
8801: }else if(Tvard[k1][2] <=ncovcol+nqv){
8802: Fixed[k]= 1;
8803: Dummy[k]= 1;
8804: modell[k].maintype= VTYPE;
8805: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
8806: ncovv++; /* Varying variables without age */
8807: TvarV[ncovv]=Tvar[k];
8808: TvarVind[ncovv]=k;
8809: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8810: Fixed[k]= 1;
8811: Dummy[k]= 0;
8812: modell[k].maintype= VTYPE;
8813: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
8814: ncovv++; /* Varying variables without age */
8815: TvarV[ncovv]=Tvar[k];
8816: TvarVind[ncovv]=k;
8817: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8818: Fixed[k]= 1;
8819: Dummy[k]= 1;
8820: modell[k].maintype= VTYPE;
8821: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
8822: ncovv++; /* Varying variables without age */
8823: TvarV[ncovv]=Tvar[k];
8824: TvarVind[ncovv]=k;
8825: }
1.227 brouard 8826: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 8827: if(Tvard[k1][2] <=ncovcol){
8828: Fixed[k]= 1;
8829: Dummy[k]= 1;
8830: modell[k].maintype= VTYPE;
8831: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
8832: ncovv++; /* Varying variables without age */
8833: TvarV[ncovv]=Tvar[k];
8834: TvarVind[ncovv]=k;
8835: }else if(Tvard[k1][2] <=ncovcol+nqv){
8836: Fixed[k]= 1;
8837: Dummy[k]= 1;
8838: modell[k].maintype= VTYPE;
8839: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
8840: ncovv++; /* Varying variables without age */
8841: TvarV[ncovv]=Tvar[k];
8842: TvarVind[ncovv]=k;
8843: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
8844: Fixed[k]= 1;
8845: Dummy[k]= 1;
8846: modell[k].maintype= VTYPE;
8847: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
8848: ncovv++; /* Varying variables without age */
8849: TvarV[ncovv]=Tvar[k];
8850: TvarVind[ncovv]=k;
8851: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
8852: Fixed[k]= 1;
8853: Dummy[k]= 1;
8854: modell[k].maintype= VTYPE;
8855: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
8856: ncovv++; /* Varying variables without age */
8857: TvarV[ncovv]=Tvar[k];
8858: TvarVind[ncovv]=k;
8859: }
1.227 brouard 8860: }else{
1.240 brouard 8861: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8862: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
8863: } /*end k1*/
1.225 brouard 8864: }else{
1.226 brouard 8865: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
8866: 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 8867: }
1.227 brouard 8868: 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 8869: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 8870: 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]);
8871: }
8872: /* Searching for doublons in the model */
8873: for(k1=1; k1<= cptcovt;k1++){
8874: for(k2=1; k2 <k1;k2++){
8875: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 8876: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
8877: if(Tvar[k1]==Tvar[k2]){
8878: 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]]);
8879: 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);
8880: return(1);
8881: }
8882: }else if (Typevar[k1] ==2){
8883: k3=Tposprod[k1];
8884: k4=Tposprod[k2];
8885: 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])) ){
8886: 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]]);
8887: 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);
8888: return(1);
8889: }
8890: }
1.227 brouard 8891: }
8892: }
1.225 brouard 8893: }
8894: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
8895: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 8896: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
8897: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 8898: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 8899: /*endread:*/
1.225 brouard 8900: printf("Exiting decodemodel: ");
8901: return (1);
1.136 brouard 8902: }
8903:
1.169 brouard 8904: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 8905: {/* Check ages at death */
1.136 brouard 8906: int i, m;
1.218 brouard 8907: int firstone=0;
8908:
1.136 brouard 8909: for (i=1; i<=imx; i++) {
8910: for(m=2; (m<= maxwav); m++) {
8911: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
8912: anint[m][i]=9999;
1.216 brouard 8913: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
8914: s[m][i]=-1;
1.136 brouard 8915: }
8916: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.169 brouard 8917: *nberr = *nberr + 1;
1.218 brouard 8918: if(firstone == 0){
8919: firstone=1;
8920: 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);
8921: }
8922: 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 8923: s[m][i]=-1;
8924: }
8925: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 8926: (*nberr)++;
1.136 brouard 8927: 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]);
8928: 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]);
8929: s[m][i]=-1; /* We prefer to skip it (and to skip it in version 0.8a1 too */
8930: }
8931: }
8932: }
8933:
8934: for (i=1; i<=imx; i++) {
8935: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
8936: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 8937: 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 8938: if (s[m][i] >= nlstate+1) {
1.169 brouard 8939: if(agedc[i]>0){
8940: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 8941: agev[m][i]=agedc[i];
1.214 brouard 8942: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 8943: }else {
1.136 brouard 8944: if ((int)andc[i]!=9999){
8945: nbwarn++;
8946: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
8947: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
8948: agev[m][i]=-1;
8949: }
8950: }
1.169 brouard 8951: } /* agedc > 0 */
1.214 brouard 8952: } /* end if */
1.136 brouard 8953: else if(s[m][i] !=9){ /* Standard case, age in fractional
8954: years but with the precision of a month */
8955: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
8956: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
8957: agev[m][i]=1;
8958: else if(agev[m][i] < *agemin){
8959: *agemin=agev[m][i];
8960: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
8961: }
8962: else if(agev[m][i] >*agemax){
8963: *agemax=agev[m][i];
1.156 brouard 8964: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 8965: }
8966: /*agev[m][i]=anint[m][i]-annais[i];*/
8967: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 8968: } /* en if 9*/
1.136 brouard 8969: else { /* =9 */
1.214 brouard 8970: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 8971: agev[m][i]=1;
8972: s[m][i]=-1;
8973: }
8974: }
1.214 brouard 8975: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 8976: agev[m][i]=1;
1.214 brouard 8977: else{
8978: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8979: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
8980: agev[m][i]=0;
8981: }
8982: } /* End for lastpass */
8983: }
1.136 brouard 8984:
8985: for (i=1; i<=imx; i++) {
8986: for(m=firstpass; (m<=lastpass); m++){
8987: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 8988: (*nberr)++;
1.136 brouard 8989: 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);
8990: 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);
8991: return 1;
8992: }
8993: }
8994: }
8995:
8996: /*for (i=1; i<=imx; i++){
8997: for (m=firstpass; (m<lastpass); m++){
8998: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
8999: }
9000:
9001: }*/
9002:
9003:
1.139 brouard 9004: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9005: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9006:
9007: return (0);
1.164 brouard 9008: /* endread:*/
1.136 brouard 9009: printf("Exiting calandcheckages: ");
9010: return (1);
9011: }
9012:
1.172 brouard 9013: #if defined(_MSC_VER)
9014: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9015: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9016: //#include "stdafx.h"
9017: //#include <stdio.h>
9018: //#include <tchar.h>
9019: //#include <windows.h>
9020: //#include <iostream>
9021: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9022:
9023: LPFN_ISWOW64PROCESS fnIsWow64Process;
9024:
9025: BOOL IsWow64()
9026: {
9027: BOOL bIsWow64 = FALSE;
9028:
9029: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9030: // (HANDLE, PBOOL);
9031:
9032: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9033:
9034: HMODULE module = GetModuleHandle(_T("kernel32"));
9035: const char funcName[] = "IsWow64Process";
9036: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9037: GetProcAddress(module, funcName);
9038:
9039: if (NULL != fnIsWow64Process)
9040: {
9041: if (!fnIsWow64Process(GetCurrentProcess(),
9042: &bIsWow64))
9043: //throw std::exception("Unknown error");
9044: printf("Unknown error\n");
9045: }
9046: return bIsWow64 != FALSE;
9047: }
9048: #endif
1.177 brouard 9049:
1.191 brouard 9050: void syscompilerinfo(int logged)
1.167 brouard 9051: {
9052: /* #include "syscompilerinfo.h"*/
1.185 brouard 9053: /* command line Intel compiler 32bit windows, XP compatible:*/
9054: /* /GS /W3 /Gy
9055: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9056: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9057: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9058: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9059: */
9060: /* 64 bits */
1.185 brouard 9061: /*
9062: /GS /W3 /Gy
9063: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9064: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9065: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9066: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9067: /* Optimization are useless and O3 is slower than O2 */
9068: /*
9069: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9070: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9071: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9072: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9073: */
1.186 brouard 9074: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9075: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9076: /PDB:"visual studio
9077: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9078: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9079: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9080: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9081: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9082: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9083: uiAccess='false'"
9084: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9085: /NOLOGO /TLBID:1
9086: */
1.177 brouard 9087: #if defined __INTEL_COMPILER
1.178 brouard 9088: #if defined(__GNUC__)
9089: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9090: #endif
1.177 brouard 9091: #elif defined(__GNUC__)
1.179 brouard 9092: #ifndef __APPLE__
1.174 brouard 9093: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9094: #endif
1.177 brouard 9095: struct utsname sysInfo;
1.178 brouard 9096: int cross = CROSS;
9097: if (cross){
9098: printf("Cross-");
1.191 brouard 9099: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9100: }
1.174 brouard 9101: #endif
9102:
1.171 brouard 9103: #include <stdint.h>
1.178 brouard 9104:
1.191 brouard 9105: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9106: #if defined(__clang__)
1.191 brouard 9107: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9108: #endif
9109: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9110: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9111: #endif
9112: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9113: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9114: #endif
9115: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9116: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9117: #endif
9118: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9119: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9120: #endif
9121: #if defined(_MSC_VER)
1.191 brouard 9122: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9123: #endif
9124: #if defined(__PGI)
1.191 brouard 9125: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9126: #endif
9127: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9128: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9129: #endif
1.191 brouard 9130: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9131:
1.167 brouard 9132: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9133: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9134: // Windows (x64 and x86)
1.191 brouard 9135: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9136: #elif __unix__ // all unices, not all compilers
9137: // Unix
1.191 brouard 9138: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9139: #elif __linux__
9140: // linux
1.191 brouard 9141: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9142: #elif __APPLE__
1.174 brouard 9143: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9144: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9145: #endif
9146:
9147: /* __MINGW32__ */
9148: /* __CYGWIN__ */
9149: /* __MINGW64__ */
9150: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9151: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9152: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9153: /* _WIN64 // Defined for applications for Win64. */
9154: /* _M_X64 // Defined for compilations that target x64 processors. */
9155: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9156:
1.167 brouard 9157: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9158: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9159: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9160: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9161: #else
1.191 brouard 9162: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9163: #endif
9164:
1.169 brouard 9165: #if defined(__GNUC__)
9166: # if defined(__GNUC_PATCHLEVEL__)
9167: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9168: + __GNUC_MINOR__ * 100 \
9169: + __GNUC_PATCHLEVEL__)
9170: # else
9171: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9172: + __GNUC_MINOR__ * 100)
9173: # endif
1.174 brouard 9174: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9175: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9176:
9177: if (uname(&sysInfo) != -1) {
9178: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9179: 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 9180: }
9181: else
9182: perror("uname() error");
1.179 brouard 9183: //#ifndef __INTEL_COMPILER
9184: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9185: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9186: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9187: #endif
1.169 brouard 9188: #endif
1.172 brouard 9189:
9190: // void main()
9191: // {
1.169 brouard 9192: #if defined(_MSC_VER)
1.174 brouard 9193: if (IsWow64()){
1.191 brouard 9194: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9195: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9196: }
9197: else{
1.191 brouard 9198: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9199: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9200: }
1.172 brouard 9201: // printf("\nPress Enter to continue...");
9202: // getchar();
9203: // }
9204:
1.169 brouard 9205: #endif
9206:
1.167 brouard 9207:
1.219 brouard 9208: }
1.136 brouard 9209:
1.219 brouard 9210: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9211: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9212: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9213: /* double ftolpl = 1.e-10; */
1.180 brouard 9214: double age, agebase, agelim;
1.203 brouard 9215: double tot;
1.180 brouard 9216:
1.202 brouard 9217: strcpy(filerespl,"PL_");
9218: strcat(filerespl,fileresu);
9219: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9220: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9221: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9222: }
1.227 brouard 9223: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9224: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9225: pstamp(ficrespl);
1.203 brouard 9226: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9227: fprintf(ficrespl,"#Age ");
9228: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9229: fprintf(ficrespl,"\n");
1.180 brouard 9230:
1.219 brouard 9231: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9232:
1.219 brouard 9233: agebase=ageminpar;
9234: agelim=agemaxpar;
1.180 brouard 9235:
1.227 brouard 9236: /* i1=pow(2,ncoveff); */
1.234 brouard 9237: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9238: if (cptcovn < 1){i1=1;}
1.180 brouard 9239:
1.238 brouard 9240: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9241: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9242: if(TKresult[nres]!= k)
9243: continue;
1.235 brouard 9244:
1.238 brouard 9245: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9246: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9247: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9248: /* k=k+1; */
9249: /* to clean */
9250: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9251: fprintf(ficrespl,"#******");
9252: printf("#******");
9253: fprintf(ficlog,"#******");
9254: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9255: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9256: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9257: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9258: }
9259: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9260: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9261: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9262: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9263: }
9264: fprintf(ficrespl,"******\n");
9265: printf("******\n");
9266: fprintf(ficlog,"******\n");
9267: if(invalidvarcomb[k]){
9268: printf("\nCombination (%d) ignored because no case \n",k);
9269: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9270: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9271: continue;
9272: }
1.219 brouard 9273:
1.238 brouard 9274: fprintf(ficrespl,"#Age ");
9275: for(j=1;j<=cptcoveff;j++) {
9276: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9277: }
9278: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9279: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9280:
1.238 brouard 9281: for (age=agebase; age<=agelim; age++){
9282: /* for (age=agebase; age<=agebase; age++){ */
9283: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9284: fprintf(ficrespl,"%.0f ",age );
9285: for(j=1;j<=cptcoveff;j++)
9286: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9287: tot=0.;
9288: for(i=1; i<=nlstate;i++){
9289: tot += prlim[i][i];
9290: fprintf(ficrespl," %.5f", prlim[i][i]);
9291: }
9292: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9293: } /* Age */
9294: /* was end of cptcod */
9295: } /* cptcov */
9296: } /* nres */
1.219 brouard 9297: return 0;
1.180 brouard 9298: }
9299:
1.218 brouard 9300: 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){
9301: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9302:
9303: /* Computes the back prevalence limit for any combination of covariate values
9304: * at any age between ageminpar and agemaxpar
9305: */
1.235 brouard 9306: int i, j, k, i1, nres=0 ;
1.217 brouard 9307: /* double ftolpl = 1.e-10; */
9308: double age, agebase, agelim;
9309: double tot;
1.218 brouard 9310: /* double ***mobaverage; */
9311: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9312:
9313: strcpy(fileresplb,"PLB_");
9314: strcat(fileresplb,fileresu);
9315: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9316: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9317: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9318: }
9319: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9320: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9321: pstamp(ficresplb);
9322: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9323: fprintf(ficresplb,"#Age ");
9324: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9325: fprintf(ficresplb,"\n");
9326:
1.218 brouard 9327:
9328: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9329:
9330: agebase=ageminpar;
9331: agelim=agemaxpar;
9332:
9333:
1.227 brouard 9334: i1=pow(2,cptcoveff);
1.218 brouard 9335: if (cptcovn < 1){i1=1;}
1.227 brouard 9336:
1.238 brouard 9337: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9338: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9339: if(TKresult[nres]!= k)
9340: continue;
9341: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9342: fprintf(ficresplb,"#******");
9343: printf("#******");
9344: fprintf(ficlog,"#******");
9345: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9346: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9347: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9348: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9349: }
9350: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9351: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9352: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9353: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9354: }
9355: fprintf(ficresplb,"******\n");
9356: printf("******\n");
9357: fprintf(ficlog,"******\n");
9358: if(invalidvarcomb[k]){
9359: printf("\nCombination (%d) ignored because no cases \n",k);
9360: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9361: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9362: continue;
9363: }
1.218 brouard 9364:
1.238 brouard 9365: fprintf(ficresplb,"#Age ");
9366: for(j=1;j<=cptcoveff;j++) {
9367: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9368: }
9369: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9370: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9371:
9372:
1.238 brouard 9373: for (age=agebase; age<=agelim; age++){
9374: /* for (age=agebase; age<=agebase; age++){ */
9375: if(mobilavproj > 0){
9376: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9377: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9378: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9379: }else if (mobilavproj == 0){
9380: 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);
9381: 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);
9382: exit(1);
9383: }else{
9384: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9385: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.238 brouard 9386: }
9387: fprintf(ficresplb,"%.0f ",age );
9388: for(j=1;j<=cptcoveff;j++)
9389: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9390: tot=0.;
9391: for(i=1; i<=nlstate;i++){
9392: tot += bprlim[i][i];
9393: fprintf(ficresplb," %.5f", bprlim[i][i]);
9394: }
9395: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9396: } /* Age */
9397: /* was end of cptcod */
9398: } /* end of any combination */
9399: } /* end of nres */
1.218 brouard 9400: /* hBijx(p, bage, fage); */
9401: /* fclose(ficrespijb); */
9402:
9403: return 0;
1.217 brouard 9404: }
1.218 brouard 9405:
1.180 brouard 9406: int hPijx(double *p, int bage, int fage){
9407: /*------------- h Pij x at various ages ------------*/
9408:
9409: int stepsize;
9410: int agelim;
9411: int hstepm;
9412: int nhstepm;
1.235 brouard 9413: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9414:
9415: double agedeb;
9416: double ***p3mat;
9417:
1.201 brouard 9418: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9419: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9420: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9421: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9422: }
9423: printf("Computing pij: result on file '%s' \n", filerespij);
9424: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9425:
9426: stepsize=(int) (stepm+YEARM-1)/YEARM;
9427: /*if (stepm<=24) stepsize=2;*/
9428:
9429: agelim=AGESUP;
9430: hstepm=stepsize*YEARM; /* Every year of age */
9431: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9432:
1.180 brouard 9433: /* hstepm=1; aff par mois*/
9434: pstamp(ficrespij);
9435: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9436: i1= pow(2,cptcoveff);
1.218 brouard 9437: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9438: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9439: /* k=k+1; */
1.235 brouard 9440: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9441: for(k=1; k<=i1;k++){
9442: if(TKresult[nres]!= k)
9443: continue;
1.183 brouard 9444: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9445: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9446: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9447: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9448: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9449: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9450: }
1.183 brouard 9451: fprintf(ficrespij,"******\n");
9452:
9453: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9454: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9455: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9456:
9457: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 9458:
1.183 brouard 9459: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9460: oldm=oldms;savm=savms;
1.235 brouard 9461: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 9462: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
9463: for(i=1; i<=nlstate;i++)
9464: for(j=1; j<=nlstate+ndeath;j++)
9465: fprintf(ficrespij," %1d-%1d",i,j);
9466: fprintf(ficrespij,"\n");
9467: for (h=0; h<=nhstepm; h++){
9468: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9469: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 9470: for(i=1; i<=nlstate;i++)
9471: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 9472: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 9473: fprintf(ficrespij,"\n");
9474: }
1.183 brouard 9475: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9476: fprintf(ficrespij,"\n");
9477: }
1.180 brouard 9478: /*}*/
9479: }
1.218 brouard 9480: return 0;
1.180 brouard 9481: }
1.218 brouard 9482:
9483: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 9484: /*------------- h Bij x at various ages ------------*/
9485:
9486: int stepsize;
1.218 brouard 9487: /* int agelim; */
9488: int ageminl;
1.217 brouard 9489: int hstepm;
9490: int nhstepm;
1.238 brouard 9491: int h, i, i1, j, k, nres;
1.218 brouard 9492:
1.217 brouard 9493: double agedeb;
9494: double ***p3mat;
1.218 brouard 9495:
9496: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
9497: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
9498: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9499: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
9500: }
9501: printf("Computing pij back: result on file '%s' \n", filerespijb);
9502: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
9503:
9504: stepsize=(int) (stepm+YEARM-1)/YEARM;
9505: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 9506:
1.218 brouard 9507: /* agelim=AGESUP; */
9508: ageminl=30;
9509: hstepm=stepsize*YEARM; /* Every year of age */
9510: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
9511:
9512: /* hstepm=1; aff par mois*/
9513: pstamp(ficrespijb);
9514: fprintf(ficrespijb,"#****** h Pij x Back Probability to be in state i at age x-h being in j at x ");
1.227 brouard 9515: i1= pow(2,cptcoveff);
1.218 brouard 9516: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9517: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9518: /* k=k+1; */
1.238 brouard 9519: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9520: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
9521: if(TKresult[nres]!= k)
9522: continue;
9523: fprintf(ficrespijb,"\n#****** ");
9524: for(j=1;j<=cptcoveff;j++)
9525: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9526: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9527: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9528: }
9529: fprintf(ficrespijb,"******\n");
9530: if(invalidvarcomb[k]){
9531: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
9532: continue;
9533: }
9534:
9535: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
9536: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
9537: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
9538: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9539: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
9540:
9541: /* nhstepm=nhstepm*YEARM; aff par mois*/
9542:
9543: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9544: /* oldm=oldms;savm=savms; */
9545: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9546: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k);
9547: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
9548: fprintf(ficrespijb,"# Cov Agex agex-h hpijx with i,j=");
1.217 brouard 9549: for(i=1; i<=nlstate;i++)
9550: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 9551: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 9552: fprintf(ficrespijb,"\n");
1.238 brouard 9553: for (h=0; h<=nhstepm; h++){
9554: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
9555: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
9556: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
9557: for(i=1; i<=nlstate;i++)
9558: for(j=1; j<=nlstate+ndeath;j++)
9559: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
9560: fprintf(ficrespijb,"\n");
9561: }
9562: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9563: fprintf(ficrespijb,"\n");
9564: } /* end age deb */
9565: } /* end combination */
9566: } /* end nres */
1.218 brouard 9567: return 0;
9568: } /* hBijx */
1.217 brouard 9569:
1.180 brouard 9570:
1.136 brouard 9571: /***********************************************/
9572: /**************** Main Program *****************/
9573: /***********************************************/
9574:
9575: int main(int argc, char *argv[])
9576: {
9577: #ifdef GSL
9578: const gsl_multimin_fminimizer_type *T;
9579: size_t iteri = 0, it;
9580: int rval = GSL_CONTINUE;
9581: int status = GSL_SUCCESS;
9582: double ssval;
9583: #endif
9584: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 9585: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 9586: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 9587: int jj, ll, li, lj, lk;
1.136 brouard 9588: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 9589: int num_filled;
1.136 brouard 9590: int itimes;
9591: int NDIM=2;
9592: int vpopbased=0;
1.235 brouard 9593: int nres=0;
1.136 brouard 9594:
1.164 brouard 9595: char ca[32], cb[32];
1.136 brouard 9596: /* FILE *fichtm; *//* Html File */
9597: /* FILE *ficgp;*/ /*Gnuplot File */
9598: struct stat info;
1.191 brouard 9599: double agedeb=0.;
1.194 brouard 9600:
9601: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 9602: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 9603:
1.165 brouard 9604: double fret;
1.191 brouard 9605: double dum=0.; /* Dummy variable */
1.136 brouard 9606: double ***p3mat;
1.218 brouard 9607: /* double ***mobaverage; */
1.164 brouard 9608:
9609: char line[MAXLINE];
1.197 brouard 9610: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
9611:
1.234 brouard 9612: char modeltemp[MAXLINE];
1.230 brouard 9613: char resultline[MAXLINE];
9614:
1.136 brouard 9615: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 9616: char *tok, *val; /* pathtot */
1.136 brouard 9617: int firstobs=1, lastobs=10;
1.195 brouard 9618: int c, h , cpt, c2;
1.191 brouard 9619: int jl=0;
9620: int i1, j1, jk, stepsize=0;
1.194 brouard 9621: int count=0;
9622:
1.164 brouard 9623: int *tab;
1.136 brouard 9624: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 9625: int backcast=0;
1.136 brouard 9626: int mobilav=0,popforecast=0;
1.191 brouard 9627: int hstepm=0, nhstepm=0;
1.136 brouard 9628: int agemortsup;
9629: float sumlpop=0.;
9630: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
9631: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
9632:
1.191 brouard 9633: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 9634: double ftolpl=FTOL;
9635: double **prlim;
1.217 brouard 9636: double **bprlim;
1.136 brouard 9637: double ***param; /* Matrix of parameters */
9638: double *p;
9639: double **matcov; /* Matrix of covariance */
1.203 brouard 9640: double **hess; /* Hessian matrix */
1.136 brouard 9641: double ***delti3; /* Scale */
9642: double *delti; /* Scale */
9643: double ***eij, ***vareij;
9644: double **varpl; /* Variances of prevalence limits by age */
9645: double *epj, vepp;
1.164 brouard 9646:
1.136 brouard 9647: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 9648: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
9649:
1.136 brouard 9650: double **ximort;
1.145 brouard 9651: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 9652: int *dcwave;
9653:
1.164 brouard 9654: char z[1]="c";
1.136 brouard 9655:
9656: /*char *strt;*/
9657: char strtend[80];
1.126 brouard 9658:
1.164 brouard 9659:
1.126 brouard 9660: /* setlocale (LC_ALL, ""); */
9661: /* bindtextdomain (PACKAGE, LOCALEDIR); */
9662: /* textdomain (PACKAGE); */
9663: /* setlocale (LC_CTYPE, ""); */
9664: /* setlocale (LC_MESSAGES, ""); */
9665:
9666: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 9667: rstart_time = time(NULL);
9668: /* (void) gettimeofday(&start_time,&tzp);*/
9669: start_time = *localtime(&rstart_time);
1.126 brouard 9670: curr_time=start_time;
1.157 brouard 9671: /*tml = *localtime(&start_time.tm_sec);*/
9672: /* strcpy(strstart,asctime(&tml)); */
9673: strcpy(strstart,asctime(&start_time));
1.126 brouard 9674:
9675: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 9676: /* tp.tm_sec = tp.tm_sec +86400; */
9677: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 9678: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
9679: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
9680: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 9681: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 9682: /* strt=asctime(&tmg); */
9683: /* printf("Time(after) =%s",strstart); */
9684: /* (void) time (&time_value);
9685: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
9686: * tm = *localtime(&time_value);
9687: * strstart=asctime(&tm);
9688: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
9689: */
9690:
9691: nberr=0; /* Number of errors and warnings */
9692: nbwarn=0;
1.184 brouard 9693: #ifdef WIN32
9694: _getcwd(pathcd, size);
9695: #else
1.126 brouard 9696: getcwd(pathcd, size);
1.184 brouard 9697: #endif
1.191 brouard 9698: syscompilerinfo(0);
1.196 brouard 9699: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 9700: if(argc <=1){
9701: printf("\nEnter the parameter file name: ");
1.205 brouard 9702: if(!fgets(pathr,FILENAMELENGTH,stdin)){
9703: printf("ERROR Empty parameter file name\n");
9704: goto end;
9705: }
1.126 brouard 9706: i=strlen(pathr);
9707: if(pathr[i-1]=='\n')
9708: pathr[i-1]='\0';
1.156 brouard 9709: i=strlen(pathr);
1.205 brouard 9710: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 9711: pathr[i-1]='\0';
1.205 brouard 9712: }
9713: i=strlen(pathr);
9714: if( i==0 ){
9715: printf("ERROR Empty parameter file name\n");
9716: goto end;
9717: }
9718: for (tok = pathr; tok != NULL; ){
1.126 brouard 9719: printf("Pathr |%s|\n",pathr);
9720: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
9721: printf("val= |%s| pathr=%s\n",val,pathr);
9722: strcpy (pathtot, val);
9723: if(pathr[0] == '\0') break; /* Dirty */
9724: }
9725: }
9726: else{
9727: strcpy(pathtot,argv[1]);
9728: }
9729: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
9730: /*cygwin_split_path(pathtot,path,optionfile);
9731: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
9732: /* cutv(path,optionfile,pathtot,'\\');*/
9733:
9734: /* Split argv[0], imach program to get pathimach */
9735: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
9736: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9737: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
9738: /* strcpy(pathimach,argv[0]); */
9739: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
9740: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
9741: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 9742: #ifdef WIN32
9743: _chdir(path); /* Can be a relative path */
9744: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
9745: #else
1.126 brouard 9746: chdir(path); /* Can be a relative path */
1.184 brouard 9747: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
9748: #endif
9749: printf("Current directory %s!\n",pathcd);
1.126 brouard 9750: strcpy(command,"mkdir ");
9751: strcat(command,optionfilefiname);
9752: if((outcmd=system(command)) != 0){
1.169 brouard 9753: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 9754: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
9755: /* fclose(ficlog); */
9756: /* exit(1); */
9757: }
9758: /* if((imk=mkdir(optionfilefiname))<0){ */
9759: /* perror("mkdir"); */
9760: /* } */
9761:
9762: /*-------- arguments in the command line --------*/
9763:
1.186 brouard 9764: /* Main Log file */
1.126 brouard 9765: strcat(filelog, optionfilefiname);
9766: strcat(filelog,".log"); /* */
9767: if((ficlog=fopen(filelog,"w"))==NULL) {
9768: printf("Problem with logfile %s\n",filelog);
9769: goto end;
9770: }
9771: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 9772: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 9773: fprintf(ficlog,"\nEnter the parameter file name: \n");
9774: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
9775: path=%s \n\
9776: optionfile=%s\n\
9777: optionfilext=%s\n\
1.156 brouard 9778: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 9779:
1.197 brouard 9780: syscompilerinfo(1);
1.167 brouard 9781:
1.126 brouard 9782: printf("Local time (at start):%s",strstart);
9783: fprintf(ficlog,"Local time (at start): %s",strstart);
9784: fflush(ficlog);
9785: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 9786: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 9787:
9788: /* */
9789: strcpy(fileres,"r");
9790: strcat(fileres, optionfilefiname);
1.201 brouard 9791: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 9792: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 9793: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 9794:
1.186 brouard 9795: /* Main ---------arguments file --------*/
1.126 brouard 9796:
9797: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 9798: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
9799: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 9800: fflush(ficlog);
1.149 brouard 9801: /* goto end; */
9802: exit(70);
1.126 brouard 9803: }
9804:
9805:
9806:
9807: strcpy(filereso,"o");
1.201 brouard 9808: strcat(filereso,fileresu);
1.126 brouard 9809: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
9810: printf("Problem with Output resultfile: %s\n", filereso);
9811: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
9812: fflush(ficlog);
9813: goto end;
9814: }
9815:
9816: /* Reads comments: lines beginning with '#' */
9817: numlinepar=0;
1.197 brouard 9818:
9819: /* First parameter line */
9820: while(fgets(line, MAXLINE, ficpar)) {
9821: /* If line starts with a # it is a comment */
9822: if (line[0] == '#') {
9823: numlinepar++;
9824: fputs(line,stdout);
9825: fputs(line,ficparo);
9826: fputs(line,ficlog);
9827: continue;
9828: }else
9829: break;
9830: }
9831: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
9832: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
9833: if (num_filled != 5) {
9834: printf("Should be 5 parameters\n");
9835: }
1.126 brouard 9836: numlinepar++;
1.197 brouard 9837: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
9838: }
9839: /* Second parameter line */
9840: while(fgets(line, MAXLINE, ficpar)) {
9841: /* If line starts with a # it is a comment */
9842: if (line[0] == '#') {
9843: numlinepar++;
9844: fputs(line,stdout);
9845: fputs(line,ficparo);
9846: fputs(line,ficlog);
9847: continue;
9848: }else
9849: break;
9850: }
1.223 brouard 9851: 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", \
9852: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
9853: if (num_filled != 11) {
9854: 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 9855: printf("but line=%s\n",line);
1.197 brouard 9856: }
1.223 brouard 9857: 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 9858: }
1.203 brouard 9859: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 9860: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 9861: /* Third parameter line */
9862: while(fgets(line, MAXLINE, ficpar)) {
9863: /* If line starts with a # it is a comment */
9864: if (line[0] == '#') {
9865: numlinepar++;
9866: fputs(line,stdout);
9867: fputs(line,ficparo);
9868: fputs(line,ficlog);
9869: continue;
9870: }else
9871: break;
9872: }
1.201 brouard 9873: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
9874: if (num_filled == 0)
9875: model[0]='\0';
9876: else if (num_filled != 1){
1.197 brouard 9877: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9878: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
9879: model[0]='\0';
9880: goto end;
9881: }
9882: else{
9883: if (model[0]=='+'){
9884: for(i=1; i<=strlen(model);i++)
9885: modeltemp[i-1]=model[i];
1.201 brouard 9886: strcpy(model,modeltemp);
1.197 brouard 9887: }
9888: }
1.199 brouard 9889: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 9890: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 9891: }
9892: /* 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); */
9893: /* numlinepar=numlinepar+3; /\* In general *\/ */
9894: /* 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 9895: 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);
9896: 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 9897: fflush(ficlog);
1.190 brouard 9898: /* if(model[0]=='#'|| model[0]== '\0'){ */
9899: if(model[0]=='#'){
1.187 brouard 9900: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
9901: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
9902: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
9903: if(mle != -1){
9904: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
9905: exit(1);
9906: }
9907: }
1.126 brouard 9908: while((c=getc(ficpar))=='#' && c!= EOF){
9909: ungetc(c,ficpar);
9910: fgets(line, MAXLINE, ficpar);
9911: numlinepar++;
1.195 brouard 9912: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
9913: z[0]=line[1];
9914: }
9915: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 9916: fputs(line, stdout);
9917: //puts(line);
1.126 brouard 9918: fputs(line,ficparo);
9919: fputs(line,ficlog);
9920: }
9921: ungetc(c,ficpar);
9922:
9923:
1.145 brouard 9924: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 9925: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 9926: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 9927: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 9928: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
9929: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
9930: v1+v2*age+v2*v3 makes cptcovn = 3
9931: */
9932: if (strlen(model)>1)
1.187 brouard 9933: 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 9934: else
1.187 brouard 9935: ncovmodel=2; /* Constant and age */
1.133 brouard 9936: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
9937: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 9938: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
9939: 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);
9940: 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);
9941: fflush(stdout);
9942: fclose (ficlog);
9943: goto end;
9944: }
1.126 brouard 9945: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9946: delti=delti3[1][1];
9947: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
9948: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 9949: /* We could also provide initial parameters values giving by simple logistic regression
9950: * only one way, that is without matrix product. We will have nlstate maximizations */
9951: /* for(i=1;i<nlstate;i++){ */
9952: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
9953: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
9954: /* } */
1.126 brouard 9955: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 9956: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
9957: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9958: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
9959: fclose (ficparo);
9960: fclose (ficlog);
9961: goto end;
9962: exit(0);
1.248 brouard 9963: } else if(mle==-2) { /* Guessing from means */
9964: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
9965: printf(" You chose mle=-2, look at file %s for a template of covariance matrix \n",filereso);
9966: fprintf(ficlog," You chose mle=-2, look at file %s for a template of covariance matrix \n",filereso);
9967:
1.220 brouard 9968: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 9969: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 9970: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
9971: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 9972: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9973: matcov=matrix(1,npar,1,npar);
1.203 brouard 9974: hess=matrix(1,npar,1,npar);
1.220 brouard 9975: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 9976: /* Read guessed parameters */
1.126 brouard 9977: /* Reads comments: lines beginning with '#' */
9978: while((c=getc(ficpar))=='#' && c!= EOF){
9979: ungetc(c,ficpar);
9980: fgets(line, MAXLINE, ficpar);
9981: numlinepar++;
1.141 brouard 9982: fputs(line,stdout);
1.126 brouard 9983: fputs(line,ficparo);
9984: fputs(line,ficlog);
9985: }
9986: ungetc(c,ficpar);
9987:
9988: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
9989: for(i=1; i <=nlstate; i++){
1.234 brouard 9990: j=0;
1.126 brouard 9991: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 9992: if(jj==i) continue;
9993: j++;
9994: fscanf(ficpar,"%1d%1d",&i1,&j1);
9995: if ((i1 != i) || (j1 != jj)){
9996: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 9997: It might be a problem of design; if ncovcol and the model are correct\n \
9998: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 9999: exit(1);
10000: }
10001: fprintf(ficparo,"%1d%1d",i1,j1);
10002: if(mle==1)
10003: printf("%1d%1d",i,jj);
10004: fprintf(ficlog,"%1d%1d",i,jj);
10005: for(k=1; k<=ncovmodel;k++){
10006: fscanf(ficpar," %lf",¶m[i][j][k]);
10007: if(mle==1){
10008: printf(" %lf",param[i][j][k]);
10009: fprintf(ficlog," %lf",param[i][j][k]);
10010: }
10011: else
10012: fprintf(ficlog," %lf",param[i][j][k]);
10013: fprintf(ficparo," %lf",param[i][j][k]);
10014: }
10015: fscanf(ficpar,"\n");
10016: numlinepar++;
10017: if(mle==1)
10018: printf("\n");
10019: fprintf(ficlog,"\n");
10020: fprintf(ficparo,"\n");
1.126 brouard 10021: }
10022: }
10023: fflush(ficlog);
1.234 brouard 10024:
1.145 brouard 10025: /* Reads scales values */
1.126 brouard 10026: p=param[1][1];
10027:
10028: /* Reads comments: lines beginning with '#' */
10029: while((c=getc(ficpar))=='#' && c!= EOF){
10030: ungetc(c,ficpar);
10031: fgets(line, MAXLINE, ficpar);
10032: numlinepar++;
1.141 brouard 10033: fputs(line,stdout);
1.126 brouard 10034: fputs(line,ficparo);
10035: fputs(line,ficlog);
10036: }
10037: ungetc(c,ficpar);
10038:
10039: for(i=1; i <=nlstate; i++){
10040: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10041: fscanf(ficpar,"%1d%1d",&i1,&j1);
10042: if ( (i1-i) * (j1-j) != 0){
10043: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10044: exit(1);
10045: }
10046: printf("%1d%1d",i,j);
10047: fprintf(ficparo,"%1d%1d",i1,j1);
10048: fprintf(ficlog,"%1d%1d",i1,j1);
10049: for(k=1; k<=ncovmodel;k++){
10050: fscanf(ficpar,"%le",&delti3[i][j][k]);
10051: printf(" %le",delti3[i][j][k]);
10052: fprintf(ficparo," %le",delti3[i][j][k]);
10053: fprintf(ficlog," %le",delti3[i][j][k]);
10054: }
10055: fscanf(ficpar,"\n");
10056: numlinepar++;
10057: printf("\n");
10058: fprintf(ficparo,"\n");
10059: fprintf(ficlog,"\n");
1.126 brouard 10060: }
10061: }
10062: fflush(ficlog);
1.234 brouard 10063:
1.145 brouard 10064: /* Reads covariance matrix */
1.126 brouard 10065: delti=delti3[1][1];
1.220 brouard 10066:
10067:
1.126 brouard 10068: /* 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 10069:
1.126 brouard 10070: /* Reads comments: lines beginning with '#' */
10071: while((c=getc(ficpar))=='#' && c!= EOF){
10072: ungetc(c,ficpar);
10073: fgets(line, MAXLINE, ficpar);
10074: numlinepar++;
1.141 brouard 10075: fputs(line,stdout);
1.126 brouard 10076: fputs(line,ficparo);
10077: fputs(line,ficlog);
10078: }
10079: ungetc(c,ficpar);
1.220 brouard 10080:
1.126 brouard 10081: matcov=matrix(1,npar,1,npar);
1.203 brouard 10082: hess=matrix(1,npar,1,npar);
1.131 brouard 10083: for(i=1; i <=npar; i++)
10084: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10085:
1.194 brouard 10086: /* Scans npar lines */
1.126 brouard 10087: for(i=1; i <=npar; i++){
1.226 brouard 10088: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10089: if(count != 3){
1.226 brouard 10090: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10091: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10092: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10093: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10094: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10095: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10096: exit(1);
1.220 brouard 10097: }else{
1.226 brouard 10098: if(mle==1)
10099: printf("%1d%1d%d",i1,j1,jk);
10100: }
10101: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10102: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10103: for(j=1; j <=i; j++){
1.226 brouard 10104: fscanf(ficpar," %le",&matcov[i][j]);
10105: if(mle==1){
10106: printf(" %.5le",matcov[i][j]);
10107: }
10108: fprintf(ficlog," %.5le",matcov[i][j]);
10109: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10110: }
10111: fscanf(ficpar,"\n");
10112: numlinepar++;
10113: if(mle==1)
1.220 brouard 10114: printf("\n");
1.126 brouard 10115: fprintf(ficlog,"\n");
10116: fprintf(ficparo,"\n");
10117: }
1.194 brouard 10118: /* End of read covariance matrix npar lines */
1.126 brouard 10119: for(i=1; i <=npar; i++)
10120: for(j=i+1;j<=npar;j++)
1.226 brouard 10121: matcov[i][j]=matcov[j][i];
1.126 brouard 10122:
10123: if(mle==1)
10124: printf("\n");
10125: fprintf(ficlog,"\n");
10126:
10127: fflush(ficlog);
10128:
10129: /*-------- Rewriting parameter file ----------*/
10130: strcpy(rfileres,"r"); /* "Rparameterfile */
10131: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10132: strcat(rfileres,"."); /* */
10133: strcat(rfileres,optionfilext); /* Other files have txt extension */
10134: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10135: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10136: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10137: }
10138: fprintf(ficres,"#%s\n",version);
10139: } /* End of mle != -3 */
1.218 brouard 10140:
1.186 brouard 10141: /* Main data
10142: */
1.126 brouard 10143: n= lastobs;
10144: num=lvector(1,n);
10145: moisnais=vector(1,n);
10146: annais=vector(1,n);
10147: moisdc=vector(1,n);
10148: andc=vector(1,n);
1.220 brouard 10149: weight=vector(1,n);
1.126 brouard 10150: agedc=vector(1,n);
10151: cod=ivector(1,n);
1.220 brouard 10152: for(i=1;i<=n;i++){
1.234 brouard 10153: num[i]=0;
10154: moisnais[i]=0;
10155: annais[i]=0;
10156: moisdc[i]=0;
10157: andc[i]=0;
10158: agedc[i]=0;
10159: cod[i]=0;
10160: weight[i]=1.0; /* Equal weights, 1 by default */
10161: }
1.126 brouard 10162: mint=matrix(1,maxwav,1,n);
10163: anint=matrix(1,maxwav,1,n);
1.131 brouard 10164: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10165: tab=ivector(1,NCOVMAX);
1.144 brouard 10166: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10167: 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 10168:
1.136 brouard 10169: /* Reads data from file datafile */
10170: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10171: goto end;
10172:
10173: /* Calculation of the number of parameters from char model */
1.234 brouard 10174: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10175: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10176: k=3 V4 Tvar[k=3]= 4 (from V4)
10177: k=2 V1 Tvar[k=2]= 1 (from V1)
10178: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10179: */
10180:
10181: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10182: TvarsDind=ivector(1,NCOVMAX); /* */
10183: TvarsD=ivector(1,NCOVMAX); /* */
10184: TvarsQind=ivector(1,NCOVMAX); /* */
10185: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10186: TvarF=ivector(1,NCOVMAX); /* */
10187: TvarFind=ivector(1,NCOVMAX); /* */
10188: TvarV=ivector(1,NCOVMAX); /* */
10189: TvarVind=ivector(1,NCOVMAX); /* */
10190: TvarA=ivector(1,NCOVMAX); /* */
10191: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10192: TvarFD=ivector(1,NCOVMAX); /* */
10193: TvarFDind=ivector(1,NCOVMAX); /* */
10194: TvarFQ=ivector(1,NCOVMAX); /* */
10195: TvarFQind=ivector(1,NCOVMAX); /* */
10196: TvarVD=ivector(1,NCOVMAX); /* */
10197: TvarVDind=ivector(1,NCOVMAX); /* */
10198: TvarVQ=ivector(1,NCOVMAX); /* */
10199: TvarVQind=ivector(1,NCOVMAX); /* */
10200:
1.230 brouard 10201: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10202: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10203: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10204: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10205: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10206: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10207: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10208: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10209: */
10210: /* For model-covariate k tells which data-covariate to use but
10211: because this model-covariate is a construction we invent a new column
10212: ncovcol + k1
10213: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10214: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10215: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10216: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10217: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10218: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10219: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10220: */
1.145 brouard 10221: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10222: 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 10223: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10224: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10225: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10226: 4 covariates (3 plus signs)
10227: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10228: */
1.230 brouard 10229: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10230: * individual dummy, fixed or varying:
10231: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10232: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10233: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10234: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10235: * Tmodelind[1]@9={9,0,3,2,}*/
10236: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10237: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10238: * individual quantitative, fixed or varying:
10239: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10240: * 3, 1, 0, 0, 0, 0, 0, 0},
10241: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10242: /* Main decodemodel */
10243:
1.187 brouard 10244:
1.223 brouard 10245: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10246: goto end;
10247:
1.137 brouard 10248: if((double)(lastobs-imx)/(double)imx > 1.10){
10249: nbwarn++;
10250: 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);
10251: 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);
10252: }
1.136 brouard 10253: /* if(mle==1){*/
1.137 brouard 10254: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10255: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10256: }
10257:
10258: /*-calculation of age at interview from date of interview and age at death -*/
10259: agev=matrix(1,maxwav,1,imx);
10260:
10261: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10262: goto end;
10263:
1.126 brouard 10264:
1.136 brouard 10265: agegomp=(int)agemin;
10266: free_vector(moisnais,1,n);
10267: free_vector(annais,1,n);
1.126 brouard 10268: /* free_matrix(mint,1,maxwav,1,n);
10269: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10270: /* free_vector(moisdc,1,n); */
10271: /* free_vector(andc,1,n); */
1.145 brouard 10272: /* */
10273:
1.126 brouard 10274: wav=ivector(1,imx);
1.214 brouard 10275: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10276: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10277: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10278: 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.*/
10279: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10280: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10281:
10282: /* Concatenates waves */
1.214 brouard 10283: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10284: Death is a valid wave (if date is known).
10285: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10286: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10287: and mw[mi+1][i]. dh depends on stepm.
10288: */
10289:
1.126 brouard 10290: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10291: /* Concatenates waves */
1.145 brouard 10292:
1.215 brouard 10293: free_vector(moisdc,1,n);
10294: free_vector(andc,1,n);
10295:
1.126 brouard 10296: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10297: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10298: ncodemax[1]=1;
1.145 brouard 10299: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10300: cptcoveff=0;
1.220 brouard 10301: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10302: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10303: }
10304:
10305: ncovcombmax=pow(2,cptcoveff);
10306: invalidvarcomb=ivector(1, ncovcombmax);
10307: for(i=1;i<ncovcombmax;i++)
10308: invalidvarcomb[i]=0;
10309:
1.211 brouard 10310: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10311: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10312: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10313:
1.200 brouard 10314: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10315: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10316: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10317: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10318: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10319: * (currently 0 or 1) in the data.
10320: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10321: * corresponding modality (h,j).
10322: */
10323:
1.145 brouard 10324: h=0;
10325: /*if (cptcovn > 0) */
1.126 brouard 10326: m=pow(2,cptcoveff);
10327:
1.144 brouard 10328: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10329: * For k=4 covariates, h goes from 1 to m=2**k
10330: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10331: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10332: * h\k 1 2 3 4
1.143 brouard 10333: *______________________________
10334: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10335: * 2 2 1 1 1
10336: * 3 i=2 1 2 1 1
10337: * 4 2 2 1 1
10338: * 5 i=3 1 i=2 1 2 1
10339: * 6 2 1 2 1
10340: * 7 i=4 1 2 2 1
10341: * 8 2 2 2 1
1.197 brouard 10342: * 9 i=5 1 i=3 1 i=2 1 2
10343: * 10 2 1 1 2
10344: * 11 i=6 1 2 1 2
10345: * 12 2 2 1 2
10346: * 13 i=7 1 i=4 1 2 2
10347: * 14 2 1 2 2
10348: * 15 i=8 1 2 2 2
10349: * 16 2 2 2 2
1.143 brouard 10350: */
1.212 brouard 10351: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10352: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10353: * and the value of each covariate?
10354: * V1=1, V2=1, V3=2, V4=1 ?
10355: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10356: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10357: * In order to get the real value in the data, we use nbcode
10358: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10359: * We are keeping this crazy system in order to be able (in the future?)
10360: * to have more than 2 values (0 or 1) for a covariate.
10361: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10362: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10363: * bbbbbbbb
10364: * 76543210
10365: * h-1 00000101 (6-1=5)
1.219 brouard 10366: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10367: * &
10368: * 1 00000001 (1)
1.219 brouard 10369: * 00000000 = 1 & ((h-1) >> (k-1))
10370: * +1= 00000001 =1
1.211 brouard 10371: *
10372: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10373: * h' 1101 =2^3+2^2+0x2^1+2^0
10374: * >>k' 11
10375: * & 00000001
10376: * = 00000001
10377: * +1 = 00000010=2 = codtabm(14,3)
10378: * Reverse h=6 and m=16?
10379: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10380: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10381: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10382: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10383: * V3=decodtabm(14,3,2**4)=2
10384: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10385: *(h-1) >> (j-1) 0011 =13 >> 2
10386: * &1 000000001
10387: * = 000000001
10388: * +1= 000000010 =2
10389: * 2211
10390: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10391: * V3=2
1.220 brouard 10392: * codtabm and decodtabm are identical
1.211 brouard 10393: */
10394:
1.145 brouard 10395:
10396: free_ivector(Ndum,-1,NCOVMAX);
10397:
10398:
1.126 brouard 10399:
1.186 brouard 10400: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10401: strcpy(optionfilegnuplot,optionfilefiname);
10402: if(mle==-3)
1.201 brouard 10403: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10404: strcat(optionfilegnuplot,".gp");
10405:
10406: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10407: printf("Problem with file %s",optionfilegnuplot);
10408: }
10409: else{
1.204 brouard 10410: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10411: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10412: //fprintf(ficgp,"set missing 'NaNq'\n");
10413: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10414: }
10415: /* fclose(ficgp);*/
1.186 brouard 10416:
10417:
10418: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10419:
10420: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10421: if(mle==-3)
1.201 brouard 10422: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10423: strcat(optionfilehtm,".htm");
10424: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10425: printf("Problem with %s \n",optionfilehtm);
10426: exit(0);
1.126 brouard 10427: }
10428:
10429: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10430: strcat(optionfilehtmcov,"-cov.htm");
10431: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10432: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10433: }
10434: else{
10435: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10436: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10437: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10438: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10439: }
10440:
1.213 brouard 10441: 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 10442: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10443: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10444: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10445: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10446: \n\
10447: <hr size=\"2\" color=\"#EC5E5E\">\
10448: <ul><li><h4>Parameter files</h4>\n\
10449: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10450: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10451: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10452: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
10453: - Date and time at start: %s</ul>\n",\
10454: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
10455: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
10456: fileres,fileres,\
10457: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
10458: fflush(fichtm);
10459:
10460: strcpy(pathr,path);
10461: strcat(pathr,optionfilefiname);
1.184 brouard 10462: #ifdef WIN32
10463: _chdir(optionfilefiname); /* Move to directory named optionfile */
10464: #else
1.126 brouard 10465: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 10466: #endif
10467:
1.126 brouard 10468:
1.220 brouard 10469: /* Calculates basic frequencies. Computes observed prevalence at single age
10470: and for any valid combination of covariates
1.126 brouard 10471: and prints on file fileres'p'. */
1.250 ! brouard 10472: freqsummary(fileres, p, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 10473: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 10474:
10475: fprintf(fichtm,"\n");
10476: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
10477: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
10478: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
10479: imx,agemin,agemax,jmin,jmax,jmean);
10480: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 10481: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10482: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10483: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
10484: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 10485:
1.126 brouard 10486: /* For Powell, parameters are in a vector p[] starting at p[1]
10487: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
10488: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
10489:
10490: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 10491: /* For mortality only */
1.126 brouard 10492: if (mle==-3){
1.136 brouard 10493: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 10494: for(i=1;i<=NDIM;i++)
10495: for(j=1;j<=NDIM;j++)
10496: ximort[i][j]=0.;
1.186 brouard 10497: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 10498: cens=ivector(1,n);
10499: ageexmed=vector(1,n);
10500: agecens=vector(1,n);
10501: dcwave=ivector(1,n);
1.223 brouard 10502:
1.126 brouard 10503: for (i=1; i<=imx; i++){
10504: dcwave[i]=-1;
10505: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 10506: if (s[m][i]>nlstate) {
10507: dcwave[i]=m;
10508: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
10509: break;
10510: }
1.126 brouard 10511: }
1.226 brouard 10512:
1.126 brouard 10513: for (i=1; i<=imx; i++) {
10514: if (wav[i]>0){
1.226 brouard 10515: ageexmed[i]=agev[mw[1][i]][i];
10516: j=wav[i];
10517: agecens[i]=1.;
10518:
10519: if (ageexmed[i]> 1 && wav[i] > 0){
10520: agecens[i]=agev[mw[j][i]][i];
10521: cens[i]= 1;
10522: }else if (ageexmed[i]< 1)
10523: cens[i]= -1;
10524: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
10525: cens[i]=0 ;
1.126 brouard 10526: }
10527: else cens[i]=-1;
10528: }
10529:
10530: for (i=1;i<=NDIM;i++) {
10531: for (j=1;j<=NDIM;j++)
1.226 brouard 10532: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 10533: }
10534:
1.145 brouard 10535: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 10536: /*printf("%lf %lf", p[1], p[2]);*/
10537:
10538:
1.136 brouard 10539: #ifdef GSL
10540: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 10541: #else
1.126 brouard 10542: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 10543: #endif
1.201 brouard 10544: strcpy(filerespow,"POW-MORT_");
10545: strcat(filerespow,fileresu);
1.126 brouard 10546: if((ficrespow=fopen(filerespow,"w"))==NULL) {
10547: printf("Problem with resultfile: %s\n", filerespow);
10548: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
10549: }
1.136 brouard 10550: #ifdef GSL
10551: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 10552: #else
1.126 brouard 10553: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 10554: #endif
1.126 brouard 10555: /* for (i=1;i<=nlstate;i++)
10556: for(j=1;j<=nlstate+ndeath;j++)
10557: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
10558: */
10559: fprintf(ficrespow,"\n");
1.136 brouard 10560: #ifdef GSL
10561: /* gsl starts here */
10562: T = gsl_multimin_fminimizer_nmsimplex;
10563: gsl_multimin_fminimizer *sfm = NULL;
10564: gsl_vector *ss, *x;
10565: gsl_multimin_function minex_func;
10566:
10567: /* Initial vertex size vector */
10568: ss = gsl_vector_alloc (NDIM);
10569:
10570: if (ss == NULL){
10571: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
10572: }
10573: /* Set all step sizes to 1 */
10574: gsl_vector_set_all (ss, 0.001);
10575:
10576: /* Starting point */
1.126 brouard 10577:
1.136 brouard 10578: x = gsl_vector_alloc (NDIM);
10579:
10580: if (x == NULL){
10581: gsl_vector_free(ss);
10582: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
10583: }
10584:
10585: /* Initialize method and iterate */
10586: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 10587: /* gsl_vector_set(x, 0, 0.0268); */
10588: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 10589: gsl_vector_set(x, 0, p[1]);
10590: gsl_vector_set(x, 1, p[2]);
10591:
10592: minex_func.f = &gompertz_f;
10593: minex_func.n = NDIM;
10594: minex_func.params = (void *)&p; /* ??? */
10595:
10596: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
10597: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
10598:
10599: printf("Iterations beginning .....\n\n");
10600: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
10601:
10602: iteri=0;
10603: while (rval == GSL_CONTINUE){
10604: iteri++;
10605: status = gsl_multimin_fminimizer_iterate(sfm);
10606:
10607: if (status) printf("error: %s\n", gsl_strerror (status));
10608: fflush(0);
10609:
10610: if (status)
10611: break;
10612:
10613: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
10614: ssval = gsl_multimin_fminimizer_size (sfm);
10615:
10616: if (rval == GSL_SUCCESS)
10617: printf ("converged to a local maximum at\n");
10618:
10619: printf("%5d ", iteri);
10620: for (it = 0; it < NDIM; it++){
10621: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
10622: }
10623: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
10624: }
10625:
10626: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
10627:
10628: gsl_vector_free(x); /* initial values */
10629: gsl_vector_free(ss); /* inital step size */
10630: for (it=0; it<NDIM; it++){
10631: p[it+1]=gsl_vector_get(sfm->x,it);
10632: fprintf(ficrespow," %.12lf", p[it]);
10633: }
10634: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
10635: #endif
10636: #ifdef POWELL
10637: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
10638: #endif
1.126 brouard 10639: fclose(ficrespow);
10640:
1.203 brouard 10641: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 10642:
10643: for(i=1; i <=NDIM; i++)
10644: for(j=i+1;j<=NDIM;j++)
1.220 brouard 10645: matcov[i][j]=matcov[j][i];
1.126 brouard 10646:
10647: printf("\nCovariance matrix\n ");
1.203 brouard 10648: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 10649: for(i=1; i <=NDIM; i++) {
10650: for(j=1;j<=NDIM;j++){
1.220 brouard 10651: printf("%f ",matcov[i][j]);
10652: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 10653: }
1.203 brouard 10654: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 10655: }
10656:
10657: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 10658: for (i=1;i<=NDIM;i++) {
1.126 brouard 10659: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 10660: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
10661: }
1.126 brouard 10662: lsurv=vector(1,AGESUP);
10663: lpop=vector(1,AGESUP);
10664: tpop=vector(1,AGESUP);
10665: lsurv[agegomp]=100000;
10666:
10667: for (k=agegomp;k<=AGESUP;k++) {
10668: agemortsup=k;
10669: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
10670: }
10671:
10672: for (k=agegomp;k<agemortsup;k++)
10673: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
10674:
10675: for (k=agegomp;k<agemortsup;k++){
10676: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
10677: sumlpop=sumlpop+lpop[k];
10678: }
10679:
10680: tpop[agegomp]=sumlpop;
10681: for (k=agegomp;k<(agemortsup-3);k++){
10682: /* tpop[k+1]=2;*/
10683: tpop[k+1]=tpop[k]-lpop[k];
10684: }
10685:
10686:
10687: printf("\nAge lx qx dx Lx Tx e(x)\n");
10688: for (k=agegomp;k<(agemortsup-2);k++)
10689: 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]);
10690:
10691:
10692: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 10693: ageminpar=50;
10694: agemaxpar=100;
1.194 brouard 10695: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
10696: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10697: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10698: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
10699: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
10700: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
10701: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 10702: }else{
10703: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
10704: 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 10705: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 10706: }
1.201 brouard 10707: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 10708: stepm, weightopt,\
10709: model,imx,p,matcov,agemortsup);
10710:
10711: free_vector(lsurv,1,AGESUP);
10712: free_vector(lpop,1,AGESUP);
10713: free_vector(tpop,1,AGESUP);
1.220 brouard 10714: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 10715: free_ivector(cens,1,n);
10716: free_vector(agecens,1,n);
10717: free_ivector(dcwave,1,n);
1.220 brouard 10718: #ifdef GSL
1.136 brouard 10719: #endif
1.186 brouard 10720: } /* Endof if mle==-3 mortality only */
1.205 brouard 10721: /* Standard */
10722: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
10723: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10724: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 10725: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 10726: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10727: for (k=1; k<=npar;k++)
10728: printf(" %d %8.5f",k,p[k]);
10729: printf("\n");
1.205 brouard 10730: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
10731: /* mlikeli uses func not funcone */
1.247 brouard 10732: /* for(i=1;i<nlstate;i++){ */
10733: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10734: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10735: /* } */
1.205 brouard 10736: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
10737: }
10738: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
10739: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
10740: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
10741: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10742: }
10743: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 10744: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
10745: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
10746: for (k=1; k<=npar;k++)
10747: printf(" %d %8.5f",k,p[k]);
10748: printf("\n");
10749:
10750: /*--------- results files --------------*/
1.224 brouard 10751: 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 10752:
10753:
10754: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10755: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10756: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10757: for(i=1,jk=1; i <=nlstate; i++){
10758: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 10759: if (k != i) {
10760: printf("%d%d ",i,k);
10761: fprintf(ficlog,"%d%d ",i,k);
10762: fprintf(ficres,"%1d%1d ",i,k);
10763: for(j=1; j <=ncovmodel; j++){
10764: printf("%12.7f ",p[jk]);
10765: fprintf(ficlog,"%12.7f ",p[jk]);
10766: fprintf(ficres,"%12.7f ",p[jk]);
10767: jk++;
10768: }
10769: printf("\n");
10770: fprintf(ficlog,"\n");
10771: fprintf(ficres,"\n");
10772: }
1.126 brouard 10773: }
10774: }
1.203 brouard 10775: if(mle != 0){
10776: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 10777: ftolhess=ftol; /* Usually correct */
1.203 brouard 10778: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
10779: 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");
10780: 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");
10781: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 10782: for(k=1; k <=(nlstate+ndeath); k++){
10783: if (k != i) {
10784: printf("%d%d ",i,k);
10785: fprintf(ficlog,"%d%d ",i,k);
10786: for(j=1; j <=ncovmodel; j++){
10787: 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]));
10788: 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]));
10789: jk++;
10790: }
10791: printf("\n");
10792: fprintf(ficlog,"\n");
10793: }
10794: }
1.193 brouard 10795: }
1.203 brouard 10796: } /* end of hesscov and Wald tests */
1.225 brouard 10797:
1.203 brouard 10798: /* */
1.126 brouard 10799: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
10800: printf("# Scales (for hessian or gradient estimation)\n");
10801: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
10802: for(i=1,jk=1; i <=nlstate; i++){
10803: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 10804: if (j!=i) {
10805: fprintf(ficres,"%1d%1d",i,j);
10806: printf("%1d%1d",i,j);
10807: fprintf(ficlog,"%1d%1d",i,j);
10808: for(k=1; k<=ncovmodel;k++){
10809: printf(" %.5e",delti[jk]);
10810: fprintf(ficlog," %.5e",delti[jk]);
10811: fprintf(ficres," %.5e",delti[jk]);
10812: jk++;
10813: }
10814: printf("\n");
10815: fprintf(ficlog,"\n");
10816: fprintf(ficres,"\n");
10817: }
1.126 brouard 10818: }
10819: }
10820:
10821: 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 10822: if(mle >= 1) /* To big for the screen */
1.126 brouard 10823: 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");
10824: 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");
10825: /* # 121 Var(a12)\n\ */
10826: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10827: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10828: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10829: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10830: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10831: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10832: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10833:
10834:
10835: /* Just to have a covariance matrix which will be more understandable
10836: even is we still don't want to manage dictionary of variables
10837: */
10838: for(itimes=1;itimes<=2;itimes++){
10839: jj=0;
10840: for(i=1; i <=nlstate; i++){
1.225 brouard 10841: for(j=1; j <=nlstate+ndeath; j++){
10842: if(j==i) continue;
10843: for(k=1; k<=ncovmodel;k++){
10844: jj++;
10845: ca[0]= k+'a'-1;ca[1]='\0';
10846: if(itimes==1){
10847: if(mle>=1)
10848: printf("#%1d%1d%d",i,j,k);
10849: fprintf(ficlog,"#%1d%1d%d",i,j,k);
10850: fprintf(ficres,"#%1d%1d%d",i,j,k);
10851: }else{
10852: if(mle>=1)
10853: printf("%1d%1d%d",i,j,k);
10854: fprintf(ficlog,"%1d%1d%d",i,j,k);
10855: fprintf(ficres,"%1d%1d%d",i,j,k);
10856: }
10857: ll=0;
10858: for(li=1;li <=nlstate; li++){
10859: for(lj=1;lj <=nlstate+ndeath; lj++){
10860: if(lj==li) continue;
10861: for(lk=1;lk<=ncovmodel;lk++){
10862: ll++;
10863: if(ll<=jj){
10864: cb[0]= lk +'a'-1;cb[1]='\0';
10865: if(ll<jj){
10866: if(itimes==1){
10867: if(mle>=1)
10868: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10869: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10870: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10871: }else{
10872: if(mle>=1)
10873: printf(" %.5e",matcov[jj][ll]);
10874: fprintf(ficlog," %.5e",matcov[jj][ll]);
10875: fprintf(ficres," %.5e",matcov[jj][ll]);
10876: }
10877: }else{
10878: if(itimes==1){
10879: if(mle>=1)
10880: printf(" Var(%s%1d%1d)",ca,i,j);
10881: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
10882: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
10883: }else{
10884: if(mle>=1)
10885: printf(" %.7e",matcov[jj][ll]);
10886: fprintf(ficlog," %.7e",matcov[jj][ll]);
10887: fprintf(ficres," %.7e",matcov[jj][ll]);
10888: }
10889: }
10890: }
10891: } /* end lk */
10892: } /* end lj */
10893: } /* end li */
10894: if(mle>=1)
10895: printf("\n");
10896: fprintf(ficlog,"\n");
10897: fprintf(ficres,"\n");
10898: numlinepar++;
10899: } /* end k*/
10900: } /*end j */
1.126 brouard 10901: } /* end i */
10902: } /* end itimes */
10903:
10904: fflush(ficlog);
10905: fflush(ficres);
1.225 brouard 10906: while(fgets(line, MAXLINE, ficpar)) {
10907: /* If line starts with a # it is a comment */
10908: if (line[0] == '#') {
10909: numlinepar++;
10910: fputs(line,stdout);
10911: fputs(line,ficparo);
10912: fputs(line,ficlog);
10913: continue;
10914: }else
10915: break;
10916: }
10917:
1.209 brouard 10918: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
10919: /* ungetc(c,ficpar); */
10920: /* fgets(line, MAXLINE, ficpar); */
10921: /* fputs(line,stdout); */
10922: /* fputs(line,ficparo); */
10923: /* } */
10924: /* ungetc(c,ficpar); */
1.126 brouard 10925:
10926: estepm=0;
1.209 brouard 10927: 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 10928:
10929: if (num_filled != 6) {
10930: 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);
10931: 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);
10932: goto end;
10933: }
10934: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
10935: }
10936: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
10937: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
10938:
1.209 brouard 10939: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 10940: if (estepm==0 || estepm < stepm) estepm=stepm;
10941: if (fage <= 2) {
10942: bage = ageminpar;
10943: fage = agemaxpar;
10944: }
10945:
10946: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 10947: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
10948: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 10949:
1.186 brouard 10950: /* Other stuffs, more or less useful */
1.126 brouard 10951: while((c=getc(ficpar))=='#' && c!= EOF){
10952: ungetc(c,ficpar);
10953: fgets(line, MAXLINE, ficpar);
1.141 brouard 10954: fputs(line,stdout);
1.126 brouard 10955: fputs(line,ficparo);
10956: }
10957: ungetc(c,ficpar);
10958:
10959: 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);
10960: 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);
10961: 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);
10962: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
10963: 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);
10964:
10965: while((c=getc(ficpar))=='#' && c!= EOF){
10966: ungetc(c,ficpar);
10967: fgets(line, MAXLINE, ficpar);
1.141 brouard 10968: fputs(line,stdout);
1.126 brouard 10969: fputs(line,ficparo);
10970: }
10971: ungetc(c,ficpar);
10972:
10973:
10974: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
10975: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
10976:
10977: fscanf(ficpar,"pop_based=%d\n",&popbased);
1.193 brouard 10978: fprintf(ficlog,"pop_based=%d\n",popbased);
1.126 brouard 10979: fprintf(ficparo,"pop_based=%d\n",popbased);
10980: fprintf(ficres,"pop_based=%d\n",popbased);
10981:
10982: while((c=getc(ficpar))=='#' && c!= EOF){
10983: ungetc(c,ficpar);
10984: fgets(line, MAXLINE, ficpar);
1.141 brouard 10985: fputs(line,stdout);
1.238 brouard 10986: fputs(line,ficres);
1.126 brouard 10987: fputs(line,ficparo);
10988: }
10989: ungetc(c,ficpar);
10990:
10991: 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);
10992: 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);
10993: 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);
10994: 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);
10995: 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);
10996: /* day and month of proj2 are not used but only year anproj2.*/
10997:
1.217 brouard 10998: while((c=getc(ficpar))=='#' && c!= EOF){
10999: ungetc(c,ficpar);
11000: fgets(line, MAXLINE, ficpar);
11001: fputs(line,stdout);
11002: fputs(line,ficparo);
1.238 brouard 11003: fputs(line,ficres);
1.217 brouard 11004: }
11005: ungetc(c,ficpar);
11006:
11007: 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 11008: 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);
11009: 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);
11010: 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 11011: /* day and month of proj2 are not used but only year anproj2.*/
1.126 brouard 11012:
1.230 brouard 11013: /* Results */
1.235 brouard 11014: nresult=0;
1.230 brouard 11015: while(fgets(line, MAXLINE, ficpar)) {
11016: /* If line starts with a # it is a comment */
11017: if (line[0] == '#') {
11018: numlinepar++;
11019: fputs(line,stdout);
11020: fputs(line,ficparo);
11021: fputs(line,ficlog);
1.238 brouard 11022: fputs(line,ficres);
1.230 brouard 11023: continue;
11024: }else
11025: break;
11026: }
1.240 brouard 11027: if (!feof(ficpar))
1.230 brouard 11028: while((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
1.240 brouard 11029: if (num_filled == 0){
1.230 brouard 11030: resultline[0]='\0';
1.240 brouard 11031: break;
11032: } else if (num_filled != 1){
1.230 brouard 11033: printf("ERROR %d: result line should be at minimum 'result=' %s\n",num_filled, line);
11034: }
1.235 brouard 11035: nresult++; /* Sum of resultlines */
11036: printf("Result %d: result=%s\n",nresult, resultline);
11037: if(nresult > MAXRESULTLINES){
11038: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11039: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11040: goto end;
11041: }
11042: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.238 brouard 11043: fprintf(ficparo,"result: %s\n",resultline);
11044: fprintf(ficres,"result: %s\n",resultline);
11045: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11046: while(fgets(line, MAXLINE, ficpar)) {
11047: /* If line starts with a # it is a comment */
11048: if (line[0] == '#') {
11049: numlinepar++;
11050: fputs(line,stdout);
11051: fputs(line,ficparo);
1.238 brouard 11052: fputs(line,ficres);
1.230 brouard 11053: fputs(line,ficlog);
11054: continue;
11055: }else
11056: break;
11057: }
11058: if (feof(ficpar))
11059: break;
11060: else{ /* Processess output results for this combination of covariate values */
11061: }
1.240 brouard 11062: } /* end while */
1.230 brouard 11063:
11064:
1.126 brouard 11065:
1.230 brouard 11066: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11067: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11068:
11069: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11070: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11071: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11072: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11073: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11074: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11075: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11076: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11077: }else{
1.218 brouard 11078: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p);
1.220 brouard 11079: }
11080: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.225 brouard 11081: model,imx,jmin,jmax,jmean,rfileres,popforecast,prevfcast,backcast, estepm, \
11082: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11083:
1.225 brouard 11084: /*------------ free_vector -------------*/
11085: /* chdir(path); */
1.220 brouard 11086:
1.215 brouard 11087: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11088: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11089: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11090: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11091: free_lvector(num,1,n);
11092: free_vector(agedc,1,n);
11093: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11094: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11095: fclose(ficparo);
11096: fclose(ficres);
1.220 brouard 11097:
11098:
1.186 brouard 11099: /* Other results (useful)*/
1.220 brouard 11100:
11101:
1.126 brouard 11102: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11103: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11104: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11105: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11106: fclose(ficrespl);
11107:
11108: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11109: /*#include "hpijx.h"*/
11110: hPijx(p, bage, fage);
1.145 brouard 11111: fclose(ficrespij);
1.227 brouard 11112:
1.220 brouard 11113: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11114: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11115: k=1;
1.126 brouard 11116: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11117:
1.219 brouard 11118: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11119: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11120: for(i=1;i<=AGESUP;i++)
1.219 brouard 11121: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11122: for(k=1;k<=ncovcombmax;k++)
11123: probs[i][j][k]=0.;
1.219 brouard 11124: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11125: if (mobilav!=0 ||mobilavproj !=0 ) {
11126: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11127: for(i=1;i<=AGESUP;i++)
11128: for(j=1;j<=nlstate;j++)
11129: for(k=1;k<=ncovcombmax;k++)
11130: mobaverages[i][j][k]=0.;
1.219 brouard 11131: mobaverage=mobaverages;
11132: if (mobilav!=0) {
1.235 brouard 11133: printf("Movingaveraging observed prevalence\n");
1.227 brouard 11134: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11135: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11136: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11137: }
1.219 brouard 11138: }
11139: /* /\* Prevalence for each covariates in probs[age][status][cov] *\/ */
11140: /* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
11141: else if (mobilavproj !=0) {
1.235 brouard 11142: printf("Movingaveraging projected observed prevalence\n");
1.227 brouard 11143: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11144: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11145: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11146: }
1.219 brouard 11147: }
11148: }/* end if moving average */
1.227 brouard 11149:
1.126 brouard 11150: /*---------- Forecasting ------------------*/
11151: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11152: if(prevfcast==1){
11153: /* if(stepm ==1){*/
1.225 brouard 11154: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11155: }
1.217 brouard 11156: if(backcast==1){
1.219 brouard 11157: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11158: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11159: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11160:
11161: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11162:
11163: bprlim=matrix(1,nlstate,1,nlstate);
11164: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11165: fclose(ficresplb);
11166:
1.222 brouard 11167: hBijx(p, bage, fage, mobaverage);
11168: fclose(ficrespijb);
1.219 brouard 11169: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11170:
11171: /* prevbackforecast(fileresu, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
1.225 brouard 11172: bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
1.219 brouard 11173: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11174: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11175: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11176: }
1.217 brouard 11177:
1.186 brouard 11178:
11179: /* ------ Other prevalence ratios------------ */
1.126 brouard 11180:
1.215 brouard 11181: free_ivector(wav,1,imx);
11182: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11183: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11184: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11185:
11186:
1.127 brouard 11187: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11188:
1.201 brouard 11189: strcpy(filerese,"E_");
11190: strcat(filerese,fileresu);
1.126 brouard 11191: if((ficreseij=fopen(filerese,"w"))==NULL) {
11192: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11193: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11194: }
1.208 brouard 11195: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11196: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11197:
11198: pstamp(ficreseij);
1.219 brouard 11199:
1.235 brouard 11200: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11201: if (cptcovn < 1){i1=1;}
11202:
11203: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11204: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11205: if(TKresult[nres]!= k)
11206: continue;
1.219 brouard 11207: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11208: printf("\n#****** ");
1.225 brouard 11209: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11210: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11211: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11212: }
11213: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11214: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11215: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11216: }
11217: fprintf(ficreseij,"******\n");
1.235 brouard 11218: printf("******\n");
1.219 brouard 11219:
11220: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11221: oldm=oldms;savm=savms;
1.235 brouard 11222: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11223:
1.219 brouard 11224: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11225: }
11226: fclose(ficreseij);
1.208 brouard 11227: printf("done evsij\n");fflush(stdout);
11228: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11229:
1.227 brouard 11230: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11231:
11232:
1.201 brouard 11233: strcpy(filerest,"T_");
11234: strcat(filerest,fileresu);
1.127 brouard 11235: if((ficrest=fopen(filerest,"w"))==NULL) {
11236: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11237: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11238: }
1.208 brouard 11239: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11240: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11241:
1.126 brouard 11242:
1.201 brouard 11243: strcpy(fileresstde,"STDE_");
11244: strcat(fileresstde,fileresu);
1.126 brouard 11245: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11246: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11247: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11248: }
1.227 brouard 11249: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11250: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11251:
1.201 brouard 11252: strcpy(filerescve,"CVE_");
11253: strcat(filerescve,fileresu);
1.126 brouard 11254: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11255: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11256: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11257: }
1.227 brouard 11258: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11259: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11260:
1.201 brouard 11261: strcpy(fileresv,"V_");
11262: strcat(fileresv,fileresu);
1.126 brouard 11263: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11264: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11265: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11266: }
1.227 brouard 11267: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11268: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11269:
1.145 brouard 11270: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11271: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11272:
1.235 brouard 11273: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11274: if (cptcovn < 1){i1=1;}
11275:
11276: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11277: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
11278: if(TKresult[nres]!= k)
11279: continue;
1.242 brouard 11280: printf("\n#****** Result for:");
11281: fprintf(ficrest,"\n#****** Result for:");
11282: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11283: for(j=1;j<=cptcoveff;j++){
11284: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11285: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11286: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11287: }
1.235 brouard 11288: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11289: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11290: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11291: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11292: }
1.208 brouard 11293: fprintf(ficrest,"******\n");
1.227 brouard 11294: fprintf(ficlog,"******\n");
11295: printf("******\n");
1.208 brouard 11296:
11297: fprintf(ficresstdeij,"\n#****** ");
11298: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11299: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11300: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11301: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11302: }
1.235 brouard 11303: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11304: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11305: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11306: }
1.208 brouard 11307: fprintf(ficresstdeij,"******\n");
11308: fprintf(ficrescveij,"******\n");
11309:
11310: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11311: /* pstamp(ficresvij); */
1.225 brouard 11312: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11313: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11314: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11315: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11316: }
1.208 brouard 11317: fprintf(ficresvij,"******\n");
11318:
11319: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11320: oldm=oldms;savm=savms;
1.235 brouard 11321: printf(" cvevsij ");
11322: fprintf(ficlog, " cvevsij ");
11323: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11324: printf(" end cvevsij \n ");
11325: fprintf(ficlog, " end cvevsij \n ");
11326:
11327: /*
11328: */
11329: /* goto endfree; */
11330:
11331: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11332: pstamp(ficrest);
11333:
11334:
11335: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11336: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11337: cptcod= 0; /* To be deleted */
11338: printf("varevsij vpopbased=%d \n",vpopbased);
11339: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11340: 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 11341: 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 ");
11342: if(vpopbased==1)
11343: 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);
11344: else
11345: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11346: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11347: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11348: fprintf(ficrest,"\n");
11349: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11350: epj=vector(1,nlstate+1);
11351: printf("Computing age specific period (stable) prevalences in each health state \n");
11352: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11353: for(age=bage; age <=fage ;age++){
1.235 brouard 11354: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11355: if (vpopbased==1) {
11356: if(mobilav ==0){
11357: for(i=1; i<=nlstate;i++)
11358: prlim[i][i]=probs[(int)age][i][k];
11359: }else{ /* mobilav */
11360: for(i=1; i<=nlstate;i++)
11361: prlim[i][i]=mobaverage[(int)age][i][k];
11362: }
11363: }
1.219 brouard 11364:
1.227 brouard 11365: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11366: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11367: /* printf(" age %4.0f ",age); */
11368: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11369: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11370: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11371: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11372: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11373: }
11374: epj[nlstate+1] +=epj[j];
11375: }
11376: /* printf(" age %4.0f \n",age); */
1.219 brouard 11377:
1.227 brouard 11378: for(i=1, vepp=0.;i <=nlstate;i++)
11379: for(j=1;j <=nlstate;j++)
11380: vepp += vareij[i][j][(int)age];
11381: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11382: for(j=1;j <=nlstate;j++){
11383: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11384: }
11385: fprintf(ficrest,"\n");
11386: }
1.208 brouard 11387: } /* End vpopbased */
11388: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11389: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11390: free_vector(epj,1,nlstate+1);
1.235 brouard 11391: printf("done selection\n");fflush(stdout);
11392: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11393:
1.145 brouard 11394: /*}*/
1.235 brouard 11395: } /* End k selection */
1.227 brouard 11396:
11397: printf("done State-specific expectancies\n");fflush(stdout);
11398: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11399:
1.126 brouard 11400: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11401:
1.201 brouard 11402: strcpy(fileresvpl,"VPL_");
11403: strcat(fileresvpl,fileresu);
1.126 brouard 11404: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11405: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11406: exit(0);
11407: }
1.208 brouard 11408: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11409: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11410:
1.145 brouard 11411: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11412: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11413:
1.235 brouard 11414: i1=pow(2,cptcoveff);
11415: if (cptcovn < 1){i1=1;}
11416:
11417: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11418: for(k=1; k<=i1;k++){
11419: if(TKresult[nres]!= k)
11420: continue;
1.227 brouard 11421: fprintf(ficresvpl,"\n#****** ");
11422: printf("\n#****** ");
11423: fprintf(ficlog,"\n#****** ");
11424: for(j=1;j<=cptcoveff;j++) {
11425: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11426: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11427: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11428: }
1.235 brouard 11429: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11430: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11431: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11432: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11433: }
1.227 brouard 11434: fprintf(ficresvpl,"******\n");
11435: printf("******\n");
11436: fprintf(ficlog,"******\n");
11437:
11438: varpl=matrix(1,nlstate,(int) bage, (int) fage);
11439: oldm=oldms;savm=savms;
1.235 brouard 11440: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 11441: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 11442: /*}*/
1.126 brouard 11443: }
1.227 brouard 11444:
1.126 brouard 11445: fclose(ficresvpl);
1.208 brouard 11446: printf("done variance-covariance of period prevalence\n");fflush(stdout);
11447: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 11448:
11449: free_vector(weight,1,n);
11450: free_imatrix(Tvard,1,NCOVMAX,1,2);
11451: free_imatrix(s,1,maxwav+1,1,n);
11452: free_matrix(anint,1,maxwav,1,n);
11453: free_matrix(mint,1,maxwav,1,n);
11454: free_ivector(cod,1,n);
11455: free_ivector(tab,1,NCOVMAX);
11456: fclose(ficresstdeij);
11457: fclose(ficrescveij);
11458: fclose(ficresvij);
11459: fclose(ficrest);
11460: fclose(ficpar);
11461:
11462:
1.126 brouard 11463: /*---------- End : free ----------------*/
1.219 brouard 11464: if (mobilav!=0 ||mobilavproj !=0)
11465: 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 11466: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 11467: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
11468: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 11469: } /* mle==-3 arrives here for freeing */
1.227 brouard 11470: /* endfree:*/
11471: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
11472: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
11473: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
11474: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 11475: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 11476: free_matrix(coqvar,1,maxwav,1,n);
11477: free_matrix(covar,0,NCOVMAX,1,n);
11478: free_matrix(matcov,1,npar,1,npar);
11479: free_matrix(hess,1,npar,1,npar);
11480: /*free_vector(delti,1,npar);*/
11481: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11482: free_matrix(agev,1,maxwav,1,imx);
11483: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11484:
11485: free_ivector(ncodemax,1,NCOVMAX);
11486: free_ivector(ncodemaxwundef,1,NCOVMAX);
11487: free_ivector(Dummy,-1,NCOVMAX);
11488: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 11489: free_ivector(DummyV,1,NCOVMAX);
11490: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 11491: free_ivector(Typevar,-1,NCOVMAX);
11492: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 11493: free_ivector(TvarsQ,1,NCOVMAX);
11494: free_ivector(TvarsQind,1,NCOVMAX);
11495: free_ivector(TvarsD,1,NCOVMAX);
11496: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 11497: free_ivector(TvarFD,1,NCOVMAX);
11498: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 11499: free_ivector(TvarF,1,NCOVMAX);
11500: free_ivector(TvarFind,1,NCOVMAX);
11501: free_ivector(TvarV,1,NCOVMAX);
11502: free_ivector(TvarVind,1,NCOVMAX);
11503: free_ivector(TvarA,1,NCOVMAX);
11504: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 11505: free_ivector(TvarFQ,1,NCOVMAX);
11506: free_ivector(TvarFQind,1,NCOVMAX);
11507: free_ivector(TvarVD,1,NCOVMAX);
11508: free_ivector(TvarVDind,1,NCOVMAX);
11509: free_ivector(TvarVQ,1,NCOVMAX);
11510: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 11511: free_ivector(Tvarsel,1,NCOVMAX);
11512: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 11513: free_ivector(Tposprod,1,NCOVMAX);
11514: free_ivector(Tprod,1,NCOVMAX);
11515: free_ivector(Tvaraff,1,NCOVMAX);
11516: free_ivector(invalidvarcomb,1,ncovcombmax);
11517: free_ivector(Tage,1,NCOVMAX);
11518: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 11519: free_ivector(TmodelInvind,1,NCOVMAX);
11520: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 11521:
11522: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
11523: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 11524: fflush(fichtm);
11525: fflush(ficgp);
11526:
1.227 brouard 11527:
1.126 brouard 11528: if((nberr >0) || (nbwarn>0)){
1.216 brouard 11529: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
11530: 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 11531: }else{
11532: printf("End of Imach\n");
11533: fprintf(ficlog,"End of Imach\n");
11534: }
11535: printf("See log file on %s\n",filelog);
11536: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 11537: /*(void) gettimeofday(&end_time,&tzp);*/
11538: rend_time = time(NULL);
11539: end_time = *localtime(&rend_time);
11540: /* tml = *localtime(&end_time.tm_sec); */
11541: strcpy(strtend,asctime(&end_time));
1.126 brouard 11542: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
11543: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 11544: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 11545:
1.157 brouard 11546: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
11547: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
11548: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 11549: /* printf("Total time was %d uSec.\n", total_usecs);*/
11550: /* if(fileappend(fichtm,optionfilehtm)){ */
11551: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11552: fclose(fichtm);
11553: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
11554: fclose(fichtmcov);
11555: fclose(ficgp);
11556: fclose(ficlog);
11557: /*------ End -----------*/
1.227 brouard 11558:
11559:
11560: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 11561: #ifdef WIN32
1.227 brouard 11562: if (_chdir(pathcd) != 0)
11563: printf("Can't move to directory %s!\n",path);
11564: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 11565: #else
1.227 brouard 11566: if(chdir(pathcd) != 0)
11567: printf("Can't move to directory %s!\n", path);
11568: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 11569: #endif
1.126 brouard 11570: printf("Current directory %s!\n",pathcd);
11571: /*strcat(plotcmd,CHARSEPARATOR);*/
11572: sprintf(plotcmd,"gnuplot");
1.157 brouard 11573: #ifdef _WIN32
1.126 brouard 11574: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
11575: #endif
11576: if(!stat(plotcmd,&info)){
1.158 brouard 11577: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11578: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 11579: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 11580: }else
11581: strcpy(pplotcmd,plotcmd);
1.157 brouard 11582: #ifdef __unix
1.126 brouard 11583: strcpy(plotcmd,GNUPLOTPROGRAM);
11584: if(!stat(plotcmd,&info)){
1.158 brouard 11585: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 11586: }else
11587: strcpy(pplotcmd,plotcmd);
11588: #endif
11589: }else
11590: strcpy(pplotcmd,plotcmd);
11591:
11592: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 11593: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 11594:
1.126 brouard 11595: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 11596: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 11597: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 11598: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 11599: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 11600: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 11601: }
1.158 brouard 11602: printf(" Successful, please wait...");
1.126 brouard 11603: while (z[0] != 'q') {
11604: /* chdir(path); */
1.154 brouard 11605: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 11606: scanf("%s",z);
11607: /* if (z[0] == 'c') system("./imach"); */
11608: if (z[0] == 'e') {
1.158 brouard 11609: #ifdef __APPLE__
1.152 brouard 11610: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 11611: #elif __linux
11612: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 11613: #else
1.152 brouard 11614: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 11615: #endif
11616: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
11617: system(pplotcmd);
1.126 brouard 11618: }
11619: else if (z[0] == 'g') system(plotcmd);
11620: else if (z[0] == 'q') exit(0);
11621: }
1.227 brouard 11622: end:
1.126 brouard 11623: while (z[0] != 'q') {
1.195 brouard 11624: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 11625: scanf("%s",z);
11626: }
11627: }
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