Annotation of imach/src/imach.c, revision 1.267
1.267 ! brouard 1: /* $Id: imach.c,v 1.266 2017/05/13 07:26:12 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.267 ! brouard 4: Revision 1.266 2017/05/13 07:26:12 brouard
! 5: Summary: Version 0.99r13 (improvements and bugs fixed)
! 6:
1.266 brouard 7: Revision 1.265 2017/04/26 16:22:11 brouard
8: Summary: imach 0.99r13 Some bugs fixed
9:
1.265 brouard 10: Revision 1.264 2017/04/26 06:01:29 brouard
11: Summary: Labels in graphs
12:
1.264 brouard 13: Revision 1.263 2017/04/24 15:23:15 brouard
14: Summary: to save
15:
1.263 brouard 16: Revision 1.262 2017/04/18 16:48:12 brouard
17: *** empty log message ***
18:
1.262 brouard 19: Revision 1.261 2017/04/05 10:14:09 brouard
20: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
21:
1.261 brouard 22: Revision 1.260 2017/04/04 17:46:59 brouard
23: Summary: Gnuplot indexations fixed (humm)
24:
1.260 brouard 25: Revision 1.259 2017/04/04 13:01:16 brouard
26: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
27:
1.259 brouard 28: Revision 1.258 2017/04/03 10:17:47 brouard
29: Summary: Version 0.99r12
30:
31: Some cleanings, conformed with updated documentation.
32:
1.258 brouard 33: Revision 1.257 2017/03/29 16:53:30 brouard
34: Summary: Temp
35:
1.257 brouard 36: Revision 1.256 2017/03/27 05:50:23 brouard
37: Summary: Temporary
38:
1.256 brouard 39: Revision 1.255 2017/03/08 16:02:28 brouard
40: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
41:
1.255 brouard 42: Revision 1.254 2017/03/08 07:13:00 brouard
43: Summary: Fixing data parameter line
44:
1.254 brouard 45: Revision 1.253 2016/12/15 11:59:41 brouard
46: Summary: 0.99 in progress
47:
1.253 brouard 48: Revision 1.252 2016/09/15 21:15:37 brouard
49: *** empty log message ***
50:
1.252 brouard 51: Revision 1.251 2016/09/15 15:01:13 brouard
52: Summary: not working
53:
1.251 brouard 54: Revision 1.250 2016/09/08 16:07:27 brouard
55: Summary: continue
56:
1.250 brouard 57: Revision 1.249 2016/09/07 17:14:18 brouard
58: Summary: Starting values from frequencies
59:
1.249 brouard 60: Revision 1.248 2016/09/07 14:10:18 brouard
61: *** empty log message ***
62:
1.248 brouard 63: Revision 1.247 2016/09/02 11:11:21 brouard
64: *** empty log message ***
65:
1.247 brouard 66: Revision 1.246 2016/09/02 08:49:22 brouard
67: *** empty log message ***
68:
1.246 brouard 69: Revision 1.245 2016/09/02 07:25:01 brouard
70: *** empty log message ***
71:
1.245 brouard 72: Revision 1.244 2016/09/02 07:17:34 brouard
73: *** empty log message ***
74:
1.244 brouard 75: Revision 1.243 2016/09/02 06:45:35 brouard
76: *** empty log message ***
77:
1.243 brouard 78: Revision 1.242 2016/08/30 15:01:20 brouard
79: Summary: Fixing a lots
80:
1.242 brouard 81: Revision 1.241 2016/08/29 17:17:25 brouard
82: Summary: gnuplot problem in Back projection to fix
83:
1.241 brouard 84: Revision 1.240 2016/08/29 07:53:18 brouard
85: Summary: Better
86:
1.240 brouard 87: Revision 1.239 2016/08/26 15:51:03 brouard
88: Summary: Improvement in Powell output in order to copy and paste
89:
90: Author:
91:
1.239 brouard 92: Revision 1.238 2016/08/26 14:23:35 brouard
93: Summary: Starting tests of 0.99
94:
1.238 brouard 95: Revision 1.237 2016/08/26 09:20:19 brouard
96: Summary: to valgrind
97:
1.237 brouard 98: Revision 1.236 2016/08/25 10:50:18 brouard
99: *** empty log message ***
100:
1.236 brouard 101: Revision 1.235 2016/08/25 06:59:23 brouard
102: *** empty log message ***
103:
1.235 brouard 104: Revision 1.234 2016/08/23 16:51:20 brouard
105: *** empty log message ***
106:
1.234 brouard 107: Revision 1.233 2016/08/23 07:40:50 brouard
108: Summary: not working
109:
1.233 brouard 110: Revision 1.232 2016/08/22 14:20:21 brouard
111: Summary: not working
112:
1.232 brouard 113: Revision 1.231 2016/08/22 07:17:15 brouard
114: Summary: not working
115:
1.231 brouard 116: Revision 1.230 2016/08/22 06:55:53 brouard
117: Summary: Not working
118:
1.230 brouard 119: Revision 1.229 2016/07/23 09:45:53 brouard
120: Summary: Completing for func too
121:
1.229 brouard 122: Revision 1.228 2016/07/22 17:45:30 brouard
123: Summary: Fixing some arrays, still debugging
124:
1.227 brouard 125: Revision 1.226 2016/07/12 18:42:34 brouard
126: Summary: temp
127:
1.226 brouard 128: Revision 1.225 2016/07/12 08:40:03 brouard
129: Summary: saving but not running
130:
1.225 brouard 131: Revision 1.224 2016/07/01 13:16:01 brouard
132: Summary: Fixes
133:
1.224 brouard 134: Revision 1.223 2016/02/19 09:23:35 brouard
135: Summary: temporary
136:
1.223 brouard 137: Revision 1.222 2016/02/17 08:14:50 brouard
138: Summary: Probably last 0.98 stable version 0.98r6
139:
1.222 brouard 140: Revision 1.221 2016/02/15 23:35:36 brouard
141: Summary: minor bug
142:
1.220 brouard 143: Revision 1.219 2016/02/15 00:48:12 brouard
144: *** empty log message ***
145:
1.219 brouard 146: Revision 1.218 2016/02/12 11:29:23 brouard
147: Summary: 0.99 Back projections
148:
1.218 brouard 149: Revision 1.217 2015/12/23 17:18:31 brouard
150: Summary: Experimental backcast
151:
1.217 brouard 152: Revision 1.216 2015/12/18 17:32:11 brouard
153: Summary: 0.98r4 Warning and status=-2
154:
155: Version 0.98r4 is now:
156: - displaying an error when status is -1, date of interview unknown and date of death known;
157: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
158: Older changes concerning s=-2, dating from 2005 have been supersed.
159:
1.216 brouard 160: Revision 1.215 2015/12/16 08:52:24 brouard
161: Summary: 0.98r4 working
162:
1.215 brouard 163: Revision 1.214 2015/12/16 06:57:54 brouard
164: Summary: temporary not working
165:
1.214 brouard 166: Revision 1.213 2015/12/11 18:22:17 brouard
167: Summary: 0.98r4
168:
1.213 brouard 169: Revision 1.212 2015/11/21 12:47:24 brouard
170: Summary: minor typo
171:
1.212 brouard 172: Revision 1.211 2015/11/21 12:41:11 brouard
173: Summary: 0.98r3 with some graph of projected cross-sectional
174:
175: Author: Nicolas Brouard
176:
1.211 brouard 177: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 178: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 179: Summary: Adding ftolpl parameter
180: Author: N Brouard
181:
182: We had difficulties to get smoothed confidence intervals. It was due
183: to the period prevalence which wasn't computed accurately. The inner
184: parameter ftolpl is now an outer parameter of the .imach parameter
185: file after estepm. If ftolpl is small 1.e-4 and estepm too,
186: computation are long.
187:
1.209 brouard 188: Revision 1.208 2015/11/17 14:31:57 brouard
189: Summary: temporary
190:
1.208 brouard 191: Revision 1.207 2015/10/27 17:36:57 brouard
192: *** empty log message ***
193:
1.207 brouard 194: Revision 1.206 2015/10/24 07:14:11 brouard
195: *** empty log message ***
196:
1.206 brouard 197: Revision 1.205 2015/10/23 15:50:53 brouard
198: Summary: 0.98r3 some clarification for graphs on likelihood contributions
199:
1.205 brouard 200: Revision 1.204 2015/10/01 16:20:26 brouard
201: Summary: Some new graphs of contribution to likelihood
202:
1.204 brouard 203: Revision 1.203 2015/09/30 17:45:14 brouard
204: Summary: looking at better estimation of the hessian
205:
206: Also a better criteria for convergence to the period prevalence And
207: therefore adding the number of years needed to converge. (The
208: prevalence in any alive state shold sum to one
209:
1.203 brouard 210: Revision 1.202 2015/09/22 19:45:16 brouard
211: Summary: Adding some overall graph on contribution to likelihood. Might change
212:
1.202 brouard 213: Revision 1.201 2015/09/15 17:34:58 brouard
214: Summary: 0.98r0
215:
216: - Some new graphs like suvival functions
217: - Some bugs fixed like model=1+age+V2.
218:
1.201 brouard 219: Revision 1.200 2015/09/09 16:53:55 brouard
220: Summary: Big bug thanks to Flavia
221:
222: Even model=1+age+V2. did not work anymore
223:
1.200 brouard 224: Revision 1.199 2015/09/07 14:09:23 brouard
225: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
226:
1.199 brouard 227: Revision 1.198 2015/09/03 07:14:39 brouard
228: Summary: 0.98q5 Flavia
229:
1.198 brouard 230: Revision 1.197 2015/09/01 18:24:39 brouard
231: *** empty log message ***
232:
1.197 brouard 233: Revision 1.196 2015/08/18 23:17:52 brouard
234: Summary: 0.98q5
235:
1.196 brouard 236: Revision 1.195 2015/08/18 16:28:39 brouard
237: Summary: Adding a hack for testing purpose
238:
239: After reading the title, ftol and model lines, if the comment line has
240: a q, starting with #q, the answer at the end of the run is quit. It
241: permits to run test files in batch with ctest. The former workaround was
242: $ echo q | imach foo.imach
243:
1.195 brouard 244: Revision 1.194 2015/08/18 13:32:00 brouard
245: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
246:
1.194 brouard 247: Revision 1.193 2015/08/04 07:17:42 brouard
248: Summary: 0.98q4
249:
1.193 brouard 250: Revision 1.192 2015/07/16 16:49:02 brouard
251: Summary: Fixing some outputs
252:
1.192 brouard 253: Revision 1.191 2015/07/14 10:00:33 brouard
254: Summary: Some fixes
255:
1.191 brouard 256: Revision 1.190 2015/05/05 08:51:13 brouard
257: Summary: Adding digits in output parameters (7 digits instead of 6)
258:
259: Fix 1+age+.
260:
1.190 brouard 261: Revision 1.189 2015/04/30 14:45:16 brouard
262: Summary: 0.98q2
263:
1.189 brouard 264: Revision 1.188 2015/04/30 08:27:53 brouard
265: *** empty log message ***
266:
1.188 brouard 267: Revision 1.187 2015/04/29 09:11:15 brouard
268: *** empty log message ***
269:
1.187 brouard 270: Revision 1.186 2015/04/23 12:01:52 brouard
271: Summary: V1*age is working now, version 0.98q1
272:
273: Some codes had been disabled in order to simplify and Vn*age was
274: working in the optimization phase, ie, giving correct MLE parameters,
275: but, as usual, outputs were not correct and program core dumped.
276:
1.186 brouard 277: Revision 1.185 2015/03/11 13:26:42 brouard
278: Summary: Inclusion of compile and links command line for Intel Compiler
279:
1.185 brouard 280: Revision 1.184 2015/03/11 11:52:39 brouard
281: Summary: Back from Windows 8. Intel Compiler
282:
1.184 brouard 283: Revision 1.183 2015/03/10 20:34:32 brouard
284: Summary: 0.98q0, trying with directest, mnbrak fixed
285:
286: We use directest instead of original Powell test; probably no
287: incidence on the results, but better justifications;
288: We fixed Numerical Recipes mnbrak routine which was wrong and gave
289: wrong results.
290:
1.183 brouard 291: Revision 1.182 2015/02/12 08:19:57 brouard
292: Summary: Trying to keep directest which seems simpler and more general
293: Author: Nicolas Brouard
294:
1.182 brouard 295: Revision 1.181 2015/02/11 23:22:24 brouard
296: Summary: Comments on Powell added
297:
298: Author:
299:
1.181 brouard 300: Revision 1.180 2015/02/11 17:33:45 brouard
301: Summary: Finishing move from main to function (hpijx and prevalence_limit)
302:
1.180 brouard 303: Revision 1.179 2015/01/04 09:57:06 brouard
304: Summary: back to OS/X
305:
1.179 brouard 306: Revision 1.178 2015/01/04 09:35:48 brouard
307: *** empty log message ***
308:
1.178 brouard 309: Revision 1.177 2015/01/03 18:40:56 brouard
310: Summary: Still testing ilc32 on OSX
311:
1.177 brouard 312: Revision 1.176 2015/01/03 16:45:04 brouard
313: *** empty log message ***
314:
1.176 brouard 315: Revision 1.175 2015/01/03 16:33:42 brouard
316: *** empty log message ***
317:
1.175 brouard 318: Revision 1.174 2015/01/03 16:15:49 brouard
319: Summary: Still in cross-compilation
320:
1.174 brouard 321: Revision 1.173 2015/01/03 12:06:26 brouard
322: Summary: trying to detect cross-compilation
323:
1.173 brouard 324: Revision 1.172 2014/12/27 12:07:47 brouard
325: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
326:
1.172 brouard 327: Revision 1.171 2014/12/23 13:26:59 brouard
328: Summary: Back from Visual C
329:
330: Still problem with utsname.h on Windows
331:
1.171 brouard 332: Revision 1.170 2014/12/23 11:17:12 brouard
333: Summary: Cleaning some \%% back to %%
334:
335: The escape was mandatory for a specific compiler (which one?), but too many warnings.
336:
1.170 brouard 337: Revision 1.169 2014/12/22 23:08:31 brouard
338: Summary: 0.98p
339:
340: Outputs some informations on compiler used, OS etc. Testing on different platforms.
341:
1.169 brouard 342: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 343: Summary: update
1.169 brouard 344:
1.168 brouard 345: Revision 1.167 2014/12/22 13:50:56 brouard
346: Summary: Testing uname and compiler version and if compiled 32 or 64
347:
348: Testing on Linux 64
349:
1.167 brouard 350: Revision 1.166 2014/12/22 11:40:47 brouard
351: *** empty log message ***
352:
1.166 brouard 353: Revision 1.165 2014/12/16 11:20:36 brouard
354: Summary: After compiling on Visual C
355:
356: * imach.c (Module): Merging 1.61 to 1.162
357:
1.165 brouard 358: Revision 1.164 2014/12/16 10:52:11 brouard
359: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
360:
361: * imach.c (Module): Merging 1.61 to 1.162
362:
1.164 brouard 363: Revision 1.163 2014/12/16 10:30:11 brouard
364: * imach.c (Module): Merging 1.61 to 1.162
365:
1.163 brouard 366: Revision 1.162 2014/09/25 11:43:39 brouard
367: Summary: temporary backup 0.99!
368:
1.162 brouard 369: Revision 1.1 2014/09/16 11:06:58 brouard
370: Summary: With some code (wrong) for nlopt
371:
372: Author:
373:
374: Revision 1.161 2014/09/15 20:41:41 brouard
375: Summary: Problem with macro SQR on Intel compiler
376:
1.161 brouard 377: Revision 1.160 2014/09/02 09:24:05 brouard
378: *** empty log message ***
379:
1.160 brouard 380: Revision 1.159 2014/09/01 10:34:10 brouard
381: Summary: WIN32
382: Author: Brouard
383:
1.159 brouard 384: Revision 1.158 2014/08/27 17:11:51 brouard
385: *** empty log message ***
386:
1.158 brouard 387: Revision 1.157 2014/08/27 16:26:55 brouard
388: Summary: Preparing windows Visual studio version
389: Author: Brouard
390:
391: In order to compile on Visual studio, time.h is now correct and time_t
392: and tm struct should be used. difftime should be used but sometimes I
393: just make the differences in raw time format (time(&now).
394: Trying to suppress #ifdef LINUX
395: Add xdg-open for __linux in order to open default browser.
396:
1.157 brouard 397: Revision 1.156 2014/08/25 20:10:10 brouard
398: *** empty log message ***
399:
1.156 brouard 400: Revision 1.155 2014/08/25 18:32:34 brouard
401: Summary: New compile, minor changes
402: Author: Brouard
403:
1.155 brouard 404: Revision 1.154 2014/06/20 17:32:08 brouard
405: Summary: Outputs now all graphs of convergence to period prevalence
406:
1.154 brouard 407: Revision 1.153 2014/06/20 16:45:46 brouard
408: Summary: If 3 live state, convergence to period prevalence on same graph
409: Author: Brouard
410:
1.153 brouard 411: Revision 1.152 2014/06/18 17:54:09 brouard
412: Summary: open browser, use gnuplot on same dir than imach if not found in the path
413:
1.152 brouard 414: Revision 1.151 2014/06/18 16:43:30 brouard
415: *** empty log message ***
416:
1.151 brouard 417: Revision 1.150 2014/06/18 16:42:35 brouard
418: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
419: Author: brouard
420:
1.150 brouard 421: Revision 1.149 2014/06/18 15:51:14 brouard
422: Summary: Some fixes in parameter files errors
423: Author: Nicolas Brouard
424:
1.149 brouard 425: Revision 1.148 2014/06/17 17:38:48 brouard
426: Summary: Nothing new
427: Author: Brouard
428:
429: Just a new packaging for OS/X version 0.98nS
430:
1.148 brouard 431: Revision 1.147 2014/06/16 10:33:11 brouard
432: *** empty log message ***
433:
1.147 brouard 434: Revision 1.146 2014/06/16 10:20:28 brouard
435: Summary: Merge
436: Author: Brouard
437:
438: Merge, before building revised version.
439:
1.146 brouard 440: Revision 1.145 2014/06/10 21:23:15 brouard
441: Summary: Debugging with valgrind
442: Author: Nicolas Brouard
443:
444: Lot of changes in order to output the results with some covariates
445: After the Edimburgh REVES conference 2014, it seems mandatory to
446: improve the code.
447: No more memory valgrind error but a lot has to be done in order to
448: continue the work of splitting the code into subroutines.
449: Also, decodemodel has been improved. Tricode is still not
450: optimal. nbcode should be improved. Documentation has been added in
451: the source code.
452:
1.144 brouard 453: Revision 1.143 2014/01/26 09:45:38 brouard
454: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
455:
456: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
457: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
458:
1.143 brouard 459: Revision 1.142 2014/01/26 03:57:36 brouard
460: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
461:
462: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
463:
1.142 brouard 464: Revision 1.141 2014/01/26 02:42:01 brouard
465: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
466:
1.141 brouard 467: Revision 1.140 2011/09/02 10:37:54 brouard
468: Summary: times.h is ok with mingw32 now.
469:
1.140 brouard 470: Revision 1.139 2010/06/14 07:50:17 brouard
471: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
472: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
473:
1.139 brouard 474: Revision 1.138 2010/04/30 18:19:40 brouard
475: *** empty log message ***
476:
1.138 brouard 477: Revision 1.137 2010/04/29 18:11:38 brouard
478: (Module): Checking covariates for more complex models
479: than V1+V2. A lot of change to be done. Unstable.
480:
1.137 brouard 481: Revision 1.136 2010/04/26 20:30:53 brouard
482: (Module): merging some libgsl code. Fixing computation
483: of likelione (using inter/intrapolation if mle = 0) in order to
484: get same likelihood as if mle=1.
485: Some cleaning of code and comments added.
486:
1.136 brouard 487: Revision 1.135 2009/10/29 15:33:14 brouard
488: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
489:
1.135 brouard 490: Revision 1.134 2009/10/29 13:18:53 brouard
491: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
492:
1.134 brouard 493: Revision 1.133 2009/07/06 10:21:25 brouard
494: just nforces
495:
1.133 brouard 496: Revision 1.132 2009/07/06 08:22:05 brouard
497: Many tings
498:
1.132 brouard 499: Revision 1.131 2009/06/20 16:22:47 brouard
500: Some dimensions resccaled
501:
1.131 brouard 502: Revision 1.130 2009/05/26 06:44:34 brouard
503: (Module): Max Covariate is now set to 20 instead of 8. A
504: lot of cleaning with variables initialized to 0. Trying to make
505: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
506:
1.130 brouard 507: Revision 1.129 2007/08/31 13:49:27 lievre
508: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
509:
1.129 lievre 510: Revision 1.128 2006/06/30 13:02:05 brouard
511: (Module): Clarifications on computing e.j
512:
1.128 brouard 513: Revision 1.127 2006/04/28 18:11:50 brouard
514: (Module): Yes the sum of survivors was wrong since
515: imach-114 because nhstepm was no more computed in the age
516: loop. Now we define nhstepma in the age loop.
517: (Module): In order to speed up (in case of numerous covariates) we
518: compute health expectancies (without variances) in a first step
519: and then all the health expectancies with variances or standard
520: deviation (needs data from the Hessian matrices) which slows the
521: computation.
522: In the future we should be able to stop the program is only health
523: expectancies and graph are needed without standard deviations.
524:
1.127 brouard 525: Revision 1.126 2006/04/28 17:23:28 brouard
526: (Module): Yes the sum of survivors was wrong since
527: imach-114 because nhstepm was no more computed in the age
528: loop. Now we define nhstepma in the age loop.
529: Version 0.98h
530:
1.126 brouard 531: Revision 1.125 2006/04/04 15:20:31 lievre
532: Errors in calculation of health expectancies. Age was not initialized.
533: Forecasting file added.
534:
535: Revision 1.124 2006/03/22 17:13:53 lievre
536: Parameters are printed with %lf instead of %f (more numbers after the comma).
537: The log-likelihood is printed in the log file
538:
539: Revision 1.123 2006/03/20 10:52:43 brouard
540: * imach.c (Module): <title> changed, corresponds to .htm file
541: name. <head> headers where missing.
542:
543: * imach.c (Module): Weights can have a decimal point as for
544: English (a comma might work with a correct LC_NUMERIC environment,
545: otherwise the weight is truncated).
546: Modification of warning when the covariates values are not 0 or
547: 1.
548: Version 0.98g
549:
550: Revision 1.122 2006/03/20 09:45:41 brouard
551: (Module): Weights can have a decimal point as for
552: English (a comma might work with a correct LC_NUMERIC environment,
553: otherwise the weight is truncated).
554: Modification of warning when the covariates values are not 0 or
555: 1.
556: Version 0.98g
557:
558: Revision 1.121 2006/03/16 17:45:01 lievre
559: * imach.c (Module): Comments concerning covariates added
560:
561: * imach.c (Module): refinements in the computation of lli if
562: status=-2 in order to have more reliable computation if stepm is
563: not 1 month. Version 0.98f
564:
565: Revision 1.120 2006/03/16 15:10:38 lievre
566: (Module): refinements in the computation of lli if
567: status=-2 in order to have more reliable computation if stepm is
568: not 1 month. Version 0.98f
569:
570: Revision 1.119 2006/03/15 17:42:26 brouard
571: (Module): Bug if status = -2, the loglikelihood was
572: computed as likelihood omitting the logarithm. Version O.98e
573:
574: Revision 1.118 2006/03/14 18:20:07 brouard
575: (Module): varevsij Comments added explaining the second
576: table of variances if popbased=1 .
577: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
578: (Module): Function pstamp added
579: (Module): Version 0.98d
580:
581: Revision 1.117 2006/03/14 17:16:22 brouard
582: (Module): varevsij Comments added explaining the second
583: table of variances if popbased=1 .
584: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
585: (Module): Function pstamp added
586: (Module): Version 0.98d
587:
588: Revision 1.116 2006/03/06 10:29:27 brouard
589: (Module): Variance-covariance wrong links and
590: varian-covariance of ej. is needed (Saito).
591:
592: Revision 1.115 2006/02/27 12:17:45 brouard
593: (Module): One freematrix added in mlikeli! 0.98c
594:
595: Revision 1.114 2006/02/26 12:57:58 brouard
596: (Module): Some improvements in processing parameter
597: filename with strsep.
598:
599: Revision 1.113 2006/02/24 14:20:24 brouard
600: (Module): Memory leaks checks with valgrind and:
601: datafile was not closed, some imatrix were not freed and on matrix
602: allocation too.
603:
604: Revision 1.112 2006/01/30 09:55:26 brouard
605: (Module): Back to gnuplot.exe instead of wgnuplot.exe
606:
607: Revision 1.111 2006/01/25 20:38:18 brouard
608: (Module): Lots of cleaning and bugs added (Gompertz)
609: (Module): Comments can be added in data file. Missing date values
610: can be a simple dot '.'.
611:
612: Revision 1.110 2006/01/25 00:51:50 brouard
613: (Module): Lots of cleaning and bugs added (Gompertz)
614:
615: Revision 1.109 2006/01/24 19:37:15 brouard
616: (Module): Comments (lines starting with a #) are allowed in data.
617:
618: Revision 1.108 2006/01/19 18:05:42 lievre
619: Gnuplot problem appeared...
620: To be fixed
621:
622: Revision 1.107 2006/01/19 16:20:37 brouard
623: Test existence of gnuplot in imach path
624:
625: Revision 1.106 2006/01/19 13:24:36 brouard
626: Some cleaning and links added in html output
627:
628: Revision 1.105 2006/01/05 20:23:19 lievre
629: *** empty log message ***
630:
631: Revision 1.104 2005/09/30 16:11:43 lievre
632: (Module): sump fixed, loop imx fixed, and simplifications.
633: (Module): If the status is missing at the last wave but we know
634: that the person is alive, then we can code his/her status as -2
635: (instead of missing=-1 in earlier versions) and his/her
636: contributions to the likelihood is 1 - Prob of dying from last
637: health status (= 1-p13= p11+p12 in the easiest case of somebody in
638: the healthy state at last known wave). Version is 0.98
639:
640: Revision 1.103 2005/09/30 15:54:49 lievre
641: (Module): sump fixed, loop imx fixed, and simplifications.
642:
643: Revision 1.102 2004/09/15 17:31:30 brouard
644: Add the possibility to read data file including tab characters.
645:
646: Revision 1.101 2004/09/15 10:38:38 brouard
647: Fix on curr_time
648:
649: Revision 1.100 2004/07/12 18:29:06 brouard
650: Add version for Mac OS X. Just define UNIX in Makefile
651:
652: Revision 1.99 2004/06/05 08:57:40 brouard
653: *** empty log message ***
654:
655: Revision 1.98 2004/05/16 15:05:56 brouard
656: New version 0.97 . First attempt to estimate force of mortality
657: directly from the data i.e. without the need of knowing the health
658: state at each age, but using a Gompertz model: log u =a + b*age .
659: This is the basic analysis of mortality and should be done before any
660: other analysis, in order to test if the mortality estimated from the
661: cross-longitudinal survey is different from the mortality estimated
662: from other sources like vital statistic data.
663:
664: The same imach parameter file can be used but the option for mle should be -3.
665:
1.133 brouard 666: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 667: former routines in order to include the new code within the former code.
668:
669: The output is very simple: only an estimate of the intercept and of
670: the slope with 95% confident intervals.
671:
672: Current limitations:
673: A) Even if you enter covariates, i.e. with the
674: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
675: B) There is no computation of Life Expectancy nor Life Table.
676:
677: Revision 1.97 2004/02/20 13:25:42 lievre
678: Version 0.96d. Population forecasting command line is (temporarily)
679: suppressed.
680:
681: Revision 1.96 2003/07/15 15:38:55 brouard
682: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
683: rewritten within the same printf. Workaround: many printfs.
684:
685: Revision 1.95 2003/07/08 07:54:34 brouard
686: * imach.c (Repository):
687: (Repository): Using imachwizard code to output a more meaningful covariance
688: matrix (cov(a12,c31) instead of numbers.
689:
690: Revision 1.94 2003/06/27 13:00:02 brouard
691: Just cleaning
692:
693: Revision 1.93 2003/06/25 16:33:55 brouard
694: (Module): On windows (cygwin) function asctime_r doesn't
695: exist so I changed back to asctime which exists.
696: (Module): Version 0.96b
697:
698: Revision 1.92 2003/06/25 16:30:45 brouard
699: (Module): On windows (cygwin) function asctime_r doesn't
700: exist so I changed back to asctime which exists.
701:
702: Revision 1.91 2003/06/25 15:30:29 brouard
703: * imach.c (Repository): Duplicated warning errors corrected.
704: (Repository): Elapsed time after each iteration is now output. It
705: helps to forecast when convergence will be reached. Elapsed time
706: is stamped in powell. We created a new html file for the graphs
707: concerning matrix of covariance. It has extension -cov.htm.
708:
709: Revision 1.90 2003/06/24 12:34:15 brouard
710: (Module): Some bugs corrected for windows. Also, when
711: mle=-1 a template is output in file "or"mypar.txt with the design
712: of the covariance matrix to be input.
713:
714: Revision 1.89 2003/06/24 12:30:52 brouard
715: (Module): Some bugs corrected for windows. Also, when
716: mle=-1 a template is output in file "or"mypar.txt with the design
717: of the covariance matrix to be input.
718:
719: Revision 1.88 2003/06/23 17:54:56 brouard
720: * 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.
721:
722: Revision 1.87 2003/06/18 12:26:01 brouard
723: Version 0.96
724:
725: Revision 1.86 2003/06/17 20:04:08 brouard
726: (Module): Change position of html and gnuplot routines and added
727: routine fileappend.
728:
729: Revision 1.85 2003/06/17 13:12:43 brouard
730: * imach.c (Repository): Check when date of death was earlier that
731: current date of interview. It may happen when the death was just
732: prior to the death. In this case, dh was negative and likelihood
733: was wrong (infinity). We still send an "Error" but patch by
734: assuming that the date of death was just one stepm after the
735: interview.
736: (Repository): Because some people have very long ID (first column)
737: we changed int to long in num[] and we added a new lvector for
738: memory allocation. But we also truncated to 8 characters (left
739: truncation)
740: (Repository): No more line truncation errors.
741:
742: Revision 1.84 2003/06/13 21:44:43 brouard
743: * imach.c (Repository): Replace "freqsummary" at a correct
744: place. It differs from routine "prevalence" which may be called
745: many times. Probs is memory consuming and must be used with
746: parcimony.
747: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
748:
749: Revision 1.83 2003/06/10 13:39:11 lievre
750: *** empty log message ***
751:
752: Revision 1.82 2003/06/05 15:57:20 brouard
753: Add log in imach.c and fullversion number is now printed.
754:
755: */
756: /*
757: Interpolated Markov Chain
758:
759: Short summary of the programme:
760:
1.227 brouard 761: This program computes Healthy Life Expectancies or State-specific
762: (if states aren't health statuses) Expectancies from
763: cross-longitudinal data. Cross-longitudinal data consist in:
764:
765: -1- a first survey ("cross") where individuals from different ages
766: are interviewed on their health status or degree of disability (in
767: the case of a health survey which is our main interest)
768:
769: -2- at least a second wave of interviews ("longitudinal") which
770: measure each change (if any) in individual health status. Health
771: expectancies are computed from the time spent in each health state
772: according to a model. More health states you consider, more time is
773: necessary to reach the Maximum Likelihood of the parameters involved
774: in the model. The simplest model is the multinomial logistic model
775: where pij is the probability to be observed in state j at the second
776: wave conditional to be observed in state i at the first
777: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
778: etc , where 'age' is age and 'sex' is a covariate. If you want to
779: have a more complex model than "constant and age", you should modify
780: the program where the markup *Covariates have to be included here
781: again* invites you to do it. More covariates you add, slower the
1.126 brouard 782: convergence.
783:
784: The advantage of this computer programme, compared to a simple
785: multinomial logistic model, is clear when the delay between waves is not
786: identical for each individual. Also, if a individual missed an
787: intermediate interview, the information is lost, but taken into
788: account using an interpolation or extrapolation.
789:
790: hPijx is the probability to be observed in state i at age x+h
791: conditional to the observed state i at age x. The delay 'h' can be
792: split into an exact number (nh*stepm) of unobserved intermediate
793: states. This elementary transition (by month, quarter,
794: semester or year) is modelled as a multinomial logistic. The hPx
795: matrix is simply the matrix product of nh*stepm elementary matrices
796: and the contribution of each individual to the likelihood is simply
797: hPijx.
798:
799: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 800: of the life expectancies. It also computes the period (stable) prevalence.
801:
802: Back prevalence and projections:
1.227 brouard 803:
804: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
805: double agemaxpar, double ftolpl, int *ncvyearp, double
806: dateprev1,double dateprev2, int firstpass, int lastpass, int
807: mobilavproj)
808:
809: Computes the back prevalence limit for any combination of
810: covariate values k at any age between ageminpar and agemaxpar and
811: returns it in **bprlim. In the loops,
812:
813: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
814: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
815:
816: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 817: Computes for any combination of covariates k and any age between bage and fage
818: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
819: oldm=oldms;savm=savms;
1.227 brouard 820:
1.267 ! brouard 821: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 822: Computes the transition matrix starting at age 'age' over
823: 'nhstepm*hstepm*stepm' months (i.e. until
824: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 825: nhstepm*hstepm matrices.
826:
827: Returns p3mat[i][j][h] after calling
828: p3mat[i][j][h]=matprod2(newm,
829: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
830: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
831: oldm);
1.226 brouard 832:
833: Important routines
834:
835: - func (or funcone), computes logit (pij) distinguishing
836: o fixed variables (single or product dummies or quantitative);
837: o varying variables by:
838: (1) wave (single, product dummies, quantitative),
839: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
840: % fixed dummy (treated) or quantitative (not done because time-consuming);
841: % varying dummy (not done) or quantitative (not done);
842: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
843: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
844: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
845: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
846: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 847:
1.226 brouard 848:
849:
1.133 brouard 850: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
851: Institut national d'études démographiques, Paris.
1.126 brouard 852: This software have been partly granted by Euro-REVES, a concerted action
853: from the European Union.
854: It is copyrighted identically to a GNU software product, ie programme and
855: software can be distributed freely for non commercial use. Latest version
856: can be accessed at http://euroreves.ined.fr/imach .
857:
858: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
859: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
860:
861: **********************************************************************/
862: /*
863: main
864: read parameterfile
865: read datafile
866: concatwav
867: freqsummary
868: if (mle >= 1)
869: mlikeli
870: print results files
871: if mle==1
872: computes hessian
873: read end of parameter file: agemin, agemax, bage, fage, estepm
874: begin-prev-date,...
875: open gnuplot file
876: open html file
1.145 brouard 877: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
878: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
879: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
880: freexexit2 possible for memory heap.
881:
882: h Pij x | pij_nom ficrestpij
883: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
884: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
885: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
886:
887: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
888: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
889: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
890: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
891: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
892:
1.126 brouard 893: forecasting if prevfcast==1 prevforecast call prevalence()
894: health expectancies
895: Variance-covariance of DFLE
896: prevalence()
897: movingaverage()
898: varevsij()
899: if popbased==1 varevsij(,popbased)
900: total life expectancies
901: Variance of period (stable) prevalence
902: end
903: */
904:
1.187 brouard 905: /* #define DEBUG */
906: /* #define DEBUGBRENT */
1.203 brouard 907: /* #define DEBUGLINMIN */
908: /* #define DEBUGHESS */
909: #define DEBUGHESSIJ
1.224 brouard 910: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 911: #define POWELL /* Instead of NLOPT */
1.224 brouard 912: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 913: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
914: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 915:
916: #include <math.h>
917: #include <stdio.h>
918: #include <stdlib.h>
919: #include <string.h>
1.226 brouard 920: #include <ctype.h>
1.159 brouard 921:
922: #ifdef _WIN32
923: #include <io.h>
1.172 brouard 924: #include <windows.h>
925: #include <tchar.h>
1.159 brouard 926: #else
1.126 brouard 927: #include <unistd.h>
1.159 brouard 928: #endif
1.126 brouard 929:
930: #include <limits.h>
931: #include <sys/types.h>
1.171 brouard 932:
933: #if defined(__GNUC__)
934: #include <sys/utsname.h> /* Doesn't work on Windows */
935: #endif
936:
1.126 brouard 937: #include <sys/stat.h>
938: #include <errno.h>
1.159 brouard 939: /* extern int errno; */
1.126 brouard 940:
1.157 brouard 941: /* #ifdef LINUX */
942: /* #include <time.h> */
943: /* #include "timeval.h" */
944: /* #else */
945: /* #include <sys/time.h> */
946: /* #endif */
947:
1.126 brouard 948: #include <time.h>
949:
1.136 brouard 950: #ifdef GSL
951: #include <gsl/gsl_errno.h>
952: #include <gsl/gsl_multimin.h>
953: #endif
954:
1.167 brouard 955:
1.162 brouard 956: #ifdef NLOPT
957: #include <nlopt.h>
958: typedef struct {
959: double (* function)(double [] );
960: } myfunc_data ;
961: #endif
962:
1.126 brouard 963: /* #include <libintl.h> */
964: /* #define _(String) gettext (String) */
965:
1.251 brouard 966: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 967:
968: #define GNUPLOTPROGRAM "gnuplot"
969: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
970: #define FILENAMELENGTH 132
971:
972: #define GLOCK_ERROR_NOPATH -1 /* empty path */
973: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
974:
1.144 brouard 975: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
976: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 977:
978: #define NINTERVMAX 8
1.144 brouard 979: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
980: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
981: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 982: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 983: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
984: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.126 brouard 985: #define MAXN 20000
1.144 brouard 986: #define YEARM 12. /**< Number of months per year */
1.218 brouard 987: /* #define AGESUP 130 */
988: #define AGESUP 150
989: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 990: #define AGEBASE 40
1.194 brouard 991: #define AGEOVERFLOW 1.e20
1.164 brouard 992: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 993: #ifdef _WIN32
994: #define DIRSEPARATOR '\\'
995: #define CHARSEPARATOR "\\"
996: #define ODIRSEPARATOR '/'
997: #else
1.126 brouard 998: #define DIRSEPARATOR '/'
999: #define CHARSEPARATOR "/"
1000: #define ODIRSEPARATOR '\\'
1001: #endif
1002:
1.267 ! brouard 1003: /* $Id: imach.c,v 1.266 2017/05/13 07:26:12 brouard Exp $ */
1.126 brouard 1004: /* $State: Exp $ */
1.196 brouard 1005: #include "version.h"
1006: char version[]=__IMACH_VERSION__;
1.224 brouard 1007: 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";
1.267 ! brouard 1008: char fullversion[]="$Revision: 1.266 $ $Date: 2017/05/13 07:26:12 $";
1.126 brouard 1009: char strstart[80];
1010: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1011: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1012: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1013: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1014: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1015: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1016: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1017: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1018: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1019: int cptcovprodnoage=0; /**< Number of covariate products without age */
1020: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1021: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1022: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1023: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1024: int nsd=0; /**< Total number of single dummy variables (output) */
1025: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1026: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1027: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1028: int ntveff=0; /**< ntveff number of effective time varying variables */
1029: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1030: int cptcov=0; /* Working variable */
1.218 brouard 1031: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.126 brouard 1032: int npar=NPARMAX;
1033: int nlstate=2; /* Number of live states */
1034: int ndeath=1; /* Number of dead states */
1.130 brouard 1035: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1036: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1037: int popbased=0;
1038:
1039: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1040: int maxwav=0; /* Maxim number of waves */
1041: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1042: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1043: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1044: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1045: int mle=1, weightopt=0;
1.126 brouard 1046: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1047: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1048: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1049: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1050: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1051: int selected(int kvar); /* Is covariate kvar selected for printing results */
1052:
1.130 brouard 1053: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1054: double **matprod2(); /* test */
1.126 brouard 1055: double **oldm, **newm, **savm; /* Working pointers to matrices */
1056: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1057: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1058:
1.136 brouard 1059: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1060: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1061: FILE *ficlog, *ficrespow;
1.130 brouard 1062: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1063: double fretone; /* Only one call to likelihood */
1.130 brouard 1064: long ipmx=0; /* Number of contributions */
1.126 brouard 1065: double sw; /* Sum of weights */
1066: char filerespow[FILENAMELENGTH];
1067: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1068: FILE *ficresilk;
1069: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1070: FILE *ficresprobmorprev;
1071: FILE *fichtm, *fichtmcov; /* Html File */
1072: FILE *ficreseij;
1073: char filerese[FILENAMELENGTH];
1074: FILE *ficresstdeij;
1075: char fileresstde[FILENAMELENGTH];
1076: FILE *ficrescveij;
1077: char filerescve[FILENAMELENGTH];
1078: FILE *ficresvij;
1079: char fileresv[FILENAMELENGTH];
1080: FILE *ficresvpl;
1081: char fileresvpl[FILENAMELENGTH];
1082: char title[MAXLINE];
1.234 brouard 1083: char model[MAXLINE]; /**< The model line */
1.217 brouard 1084: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1085: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1086: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1087: char command[FILENAMELENGTH];
1088: int outcmd=0;
1089:
1.217 brouard 1090: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1091: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1092: char filelog[FILENAMELENGTH]; /* Log file */
1093: char filerest[FILENAMELENGTH];
1094: char fileregp[FILENAMELENGTH];
1095: char popfile[FILENAMELENGTH];
1096:
1097: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1098:
1.157 brouard 1099: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1100: /* struct timezone tzp; */
1101: /* extern int gettimeofday(); */
1102: struct tm tml, *gmtime(), *localtime();
1103:
1104: extern time_t time();
1105:
1106: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1107: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1108: struct tm tm;
1109:
1.126 brouard 1110: char strcurr[80], strfor[80];
1111:
1112: char *endptr;
1113: long lval;
1114: double dval;
1115:
1116: #define NR_END 1
1117: #define FREE_ARG char*
1118: #define FTOL 1.0e-10
1119:
1120: #define NRANSI
1.240 brouard 1121: #define ITMAX 200
1122: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1123:
1124: #define TOL 2.0e-4
1125:
1126: #define CGOLD 0.3819660
1127: #define ZEPS 1.0e-10
1128: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1129:
1130: #define GOLD 1.618034
1131: #define GLIMIT 100.0
1132: #define TINY 1.0e-20
1133:
1134: static double maxarg1,maxarg2;
1135: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1136: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1137:
1138: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1139: #define rint(a) floor(a+0.5)
1.166 brouard 1140: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1141: #define mytinydouble 1.0e-16
1.166 brouard 1142: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1143: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1144: /* static double dsqrarg; */
1145: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1146: static double sqrarg;
1147: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1148: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1149: int agegomp= AGEGOMP;
1150:
1151: int imx;
1152: int stepm=1;
1153: /* Stepm, step in month: minimum step interpolation*/
1154:
1155: int estepm;
1156: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1157:
1158: int m,nb;
1159: long *num;
1.197 brouard 1160: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1161: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1162: covariate for which somebody answered excluding
1163: undefined. Usually 2: 0 and 1. */
1164: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1165: covariate for which somebody answered including
1166: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1167: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1168: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1169: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1170: double *ageexmed,*agecens;
1171: double dateintmean=0;
1172:
1173: double *weight;
1174: int **s; /* Status */
1.141 brouard 1175: double *agedc;
1.145 brouard 1176: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1177: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1178: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.225 brouard 1179: double **coqvar; /* Fixed quantitative covariate iqv */
1180: double ***cotvar; /* Time varying covariate itv */
1181: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1182: double idx;
1183: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1184: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1185: /*k 1 2 3 4 5 6 7 8 9 */
1186: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1187: /* Tndvar[k] 1 2 3 4 5 */
1188: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1189: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1190: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1191: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1192: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1193: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1194: /* Tprod[i]=k 4 7 */
1195: /* Tage[i]=k 5 8 */
1196: /* */
1197: /* Type */
1198: /* V 1 2 3 4 5 */
1199: /* F F V V V */
1200: /* D Q D D Q */
1201: /* */
1202: int *TvarsD;
1203: int *TvarsDind;
1204: int *TvarsQ;
1205: int *TvarsQind;
1206:
1.235 brouard 1207: #define MAXRESULTLINES 10
1208: int nresult=0;
1.258 brouard 1209: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1210: int TKresult[MAXRESULTLINES];
1.237 brouard 1211: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1212: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1213: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1214: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1215: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1216: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1217:
1.234 brouard 1218: /* 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 1219: 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 */
1220: 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 */
1221: 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 */
1222: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1223: 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 */
1224: 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 1225: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1226: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1227: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1228: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1229: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1230: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1231: 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 */
1232: 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 */
1233:
1.230 brouard 1234: int *Tvarsel; /**< Selected covariates for output */
1235: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1236: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1237: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1238: 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 1239: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1240: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1241: int *Tage;
1.227 brouard 1242: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1243: 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 1244: 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*/
1245: 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 1246: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1247: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1248: int **Tvard;
1249: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1250: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1251: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1252: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1253: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1254: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1255: double *lsurv, *lpop, *tpop;
1256:
1.231 brouard 1257: #define FD 1; /* Fixed dummy covariate */
1258: #define FQ 2; /* Fixed quantitative covariate */
1259: #define FP 3; /* Fixed product covariate */
1260: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1261: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1262: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1263: #define VD 10; /* Varying dummy covariate */
1264: #define VQ 11; /* Varying quantitative covariate */
1265: #define VP 12; /* Varying product covariate */
1266: #define VPDD 13; /* Varying product dummy*dummy covariate */
1267: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1268: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1269: #define APFD 16; /* Age product * fixed dummy covariate */
1270: #define APFQ 17; /* Age product * fixed quantitative covariate */
1271: #define APVD 18; /* Age product * varying dummy covariate */
1272: #define APVQ 19; /* Age product * varying quantitative covariate */
1273:
1274: #define FTYPE 1; /* Fixed covariate */
1275: #define VTYPE 2; /* Varying covariate (loop in wave) */
1276: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1277:
1278: struct kmodel{
1279: int maintype; /* main type */
1280: int subtype; /* subtype */
1281: };
1282: struct kmodel modell[NCOVMAX];
1283:
1.143 brouard 1284: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1285: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1286:
1287: /**************** split *************************/
1288: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1289: {
1290: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1291: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1292: */
1293: char *ss; /* pointer */
1.186 brouard 1294: int l1=0, l2=0; /* length counters */
1.126 brouard 1295:
1296: l1 = strlen(path ); /* length of path */
1297: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1298: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1299: if ( ss == NULL ) { /* no directory, so determine current directory */
1300: strcpy( name, path ); /* we got the fullname name because no directory */
1301: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1302: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1303: /* get current working directory */
1304: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1305: #ifdef WIN32
1306: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1307: #else
1308: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1309: #endif
1.126 brouard 1310: return( GLOCK_ERROR_GETCWD );
1311: }
1312: /* got dirc from getcwd*/
1313: printf(" DIRC = %s \n",dirc);
1.205 brouard 1314: } else { /* strip directory from path */
1.126 brouard 1315: ss++; /* after this, the filename */
1316: l2 = strlen( ss ); /* length of filename */
1317: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1318: strcpy( name, ss ); /* save file name */
1319: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1320: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1321: printf(" DIRC2 = %s \n",dirc);
1322: }
1323: /* We add a separator at the end of dirc if not exists */
1324: l1 = strlen( dirc ); /* length of directory */
1325: if( dirc[l1-1] != DIRSEPARATOR ){
1326: dirc[l1] = DIRSEPARATOR;
1327: dirc[l1+1] = 0;
1328: printf(" DIRC3 = %s \n",dirc);
1329: }
1330: ss = strrchr( name, '.' ); /* find last / */
1331: if (ss >0){
1332: ss++;
1333: strcpy(ext,ss); /* save extension */
1334: l1= strlen( name);
1335: l2= strlen(ss)+1;
1336: strncpy( finame, name, l1-l2);
1337: finame[l1-l2]= 0;
1338: }
1339:
1340: return( 0 ); /* we're done */
1341: }
1342:
1343:
1344: /******************************************/
1345:
1346: void replace_back_to_slash(char *s, char*t)
1347: {
1348: int i;
1349: int lg=0;
1350: i=0;
1351: lg=strlen(t);
1352: for(i=0; i<= lg; i++) {
1353: (s[i] = t[i]);
1354: if (t[i]== '\\') s[i]='/';
1355: }
1356: }
1357:
1.132 brouard 1358: char *trimbb(char *out, char *in)
1.137 brouard 1359: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1360: char *s;
1361: s=out;
1362: while (*in != '\0'){
1.137 brouard 1363: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1364: in++;
1365: }
1366: *out++ = *in++;
1367: }
1368: *out='\0';
1369: return s;
1370: }
1371:
1.187 brouard 1372: /* char *substrchaine(char *out, char *in, char *chain) */
1373: /* { */
1374: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1375: /* char *s, *t; */
1376: /* t=in;s=out; */
1377: /* while ((*in != *chain) && (*in != '\0')){ */
1378: /* *out++ = *in++; */
1379: /* } */
1380:
1381: /* /\* *in matches *chain *\/ */
1382: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1383: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1384: /* } */
1385: /* in--; chain--; */
1386: /* while ( (*in != '\0')){ */
1387: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1388: /* *out++ = *in++; */
1389: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1390: /* } */
1391: /* *out='\0'; */
1392: /* out=s; */
1393: /* return out; */
1394: /* } */
1395: char *substrchaine(char *out, char *in, char *chain)
1396: {
1397: /* Substract chain 'chain' from 'in', return and output 'out' */
1398: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1399:
1400: char *strloc;
1401:
1402: strcpy (out, in);
1403: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1404: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1405: if(strloc != NULL){
1406: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1407: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1408: /* strcpy (strloc, strloc +strlen(chain));*/
1409: }
1410: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1411: return out;
1412: }
1413:
1414:
1.145 brouard 1415: char *cutl(char *blocc, char *alocc, char *in, char occ)
1416: {
1.187 brouard 1417: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1418: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.187 brouard 1419: gives blocc="abcdef" and alocc="ghi2j".
1.145 brouard 1420: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1421: */
1.160 brouard 1422: char *s, *t;
1.145 brouard 1423: t=in;s=in;
1424: while ((*in != occ) && (*in != '\0')){
1425: *alocc++ = *in++;
1426: }
1427: if( *in == occ){
1428: *(alocc)='\0';
1429: s=++in;
1430: }
1431:
1432: if (s == t) {/* occ not found */
1433: *(alocc-(in-s))='\0';
1434: in=s;
1435: }
1436: while ( *in != '\0'){
1437: *blocc++ = *in++;
1438: }
1439:
1440: *blocc='\0';
1441: return t;
1442: }
1.137 brouard 1443: char *cutv(char *blocc, char *alocc, char *in, char occ)
1444: {
1.187 brouard 1445: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1446: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1447: gives blocc="abcdef2ghi" and alocc="j".
1448: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1449: */
1450: char *s, *t;
1451: t=in;s=in;
1452: while (*in != '\0'){
1453: while( *in == occ){
1454: *blocc++ = *in++;
1455: s=in;
1456: }
1457: *blocc++ = *in++;
1458: }
1459: if (s == t) /* occ not found */
1460: *(blocc-(in-s))='\0';
1461: else
1462: *(blocc-(in-s)-1)='\0';
1463: in=s;
1464: while ( *in != '\0'){
1465: *alocc++ = *in++;
1466: }
1467:
1468: *alocc='\0';
1469: return s;
1470: }
1471:
1.126 brouard 1472: int nbocc(char *s, char occ)
1473: {
1474: int i,j=0;
1475: int lg=20;
1476: i=0;
1477: lg=strlen(s);
1478: for(i=0; i<= lg; i++) {
1.234 brouard 1479: if (s[i] == occ ) j++;
1.126 brouard 1480: }
1481: return j;
1482: }
1483:
1.137 brouard 1484: /* void cutv(char *u,char *v, char*t, char occ) */
1485: /* { */
1486: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1487: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1488: /* gives u="abcdef2ghi" and v="j" *\/ */
1489: /* int i,lg,j,p=0; */
1490: /* i=0; */
1491: /* lg=strlen(t); */
1492: /* for(j=0; j<=lg-1; j++) { */
1493: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1494: /* } */
1.126 brouard 1495:
1.137 brouard 1496: /* for(j=0; j<p; j++) { */
1497: /* (u[j] = t[j]); */
1498: /* } */
1499: /* u[p]='\0'; */
1.126 brouard 1500:
1.137 brouard 1501: /* for(j=0; j<= lg; j++) { */
1502: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1503: /* } */
1504: /* } */
1.126 brouard 1505:
1.160 brouard 1506: #ifdef _WIN32
1507: char * strsep(char **pp, const char *delim)
1508: {
1509: char *p, *q;
1510:
1511: if ((p = *pp) == NULL)
1512: return 0;
1513: if ((q = strpbrk (p, delim)) != NULL)
1514: {
1515: *pp = q + 1;
1516: *q = '\0';
1517: }
1518: else
1519: *pp = 0;
1520: return p;
1521: }
1522: #endif
1523:
1.126 brouard 1524: /********************** nrerror ********************/
1525:
1526: void nrerror(char error_text[])
1527: {
1528: fprintf(stderr,"ERREUR ...\n");
1529: fprintf(stderr,"%s\n",error_text);
1530: exit(EXIT_FAILURE);
1531: }
1532: /*********************** vector *******************/
1533: double *vector(int nl, int nh)
1534: {
1535: double *v;
1536: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1537: if (!v) nrerror("allocation failure in vector");
1538: return v-nl+NR_END;
1539: }
1540:
1541: /************************ free vector ******************/
1542: void free_vector(double*v, int nl, int nh)
1543: {
1544: free((FREE_ARG)(v+nl-NR_END));
1545: }
1546:
1547: /************************ivector *******************************/
1548: int *ivector(long nl,long nh)
1549: {
1550: int *v;
1551: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1552: if (!v) nrerror("allocation failure in ivector");
1553: return v-nl+NR_END;
1554: }
1555:
1556: /******************free ivector **************************/
1557: void free_ivector(int *v, long nl, long nh)
1558: {
1559: free((FREE_ARG)(v+nl-NR_END));
1560: }
1561:
1562: /************************lvector *******************************/
1563: long *lvector(long nl,long nh)
1564: {
1565: long *v;
1566: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1567: if (!v) nrerror("allocation failure in ivector");
1568: return v-nl+NR_END;
1569: }
1570:
1571: /******************free lvector **************************/
1572: void free_lvector(long *v, long nl, long nh)
1573: {
1574: free((FREE_ARG)(v+nl-NR_END));
1575: }
1576:
1577: /******************* imatrix *******************************/
1578: int **imatrix(long nrl, long nrh, long ncl, long nch)
1579: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1580: {
1581: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1582: int **m;
1583:
1584: /* allocate pointers to rows */
1585: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1586: if (!m) nrerror("allocation failure 1 in matrix()");
1587: m += NR_END;
1588: m -= nrl;
1589:
1590:
1591: /* allocate rows and set pointers to them */
1592: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1593: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1594: m[nrl] += NR_END;
1595: m[nrl] -= ncl;
1596:
1597: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1598:
1599: /* return pointer to array of pointers to rows */
1600: return m;
1601: }
1602:
1603: /****************** free_imatrix *************************/
1604: void free_imatrix(m,nrl,nrh,ncl,nch)
1605: int **m;
1606: long nch,ncl,nrh,nrl;
1607: /* free an int matrix allocated by imatrix() */
1608: {
1609: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1610: free((FREE_ARG) (m+nrl-NR_END));
1611: }
1612:
1613: /******************* matrix *******************************/
1614: double **matrix(long nrl, long nrh, long ncl, long nch)
1615: {
1616: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1617: double **m;
1618:
1619: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1620: if (!m) nrerror("allocation failure 1 in matrix()");
1621: m += NR_END;
1622: m -= nrl;
1623:
1624: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1625: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1626: m[nrl] += NR_END;
1627: m[nrl] -= ncl;
1628:
1629: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1630: return m;
1.145 brouard 1631: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1632: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1633: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1634: */
1635: }
1636:
1637: /*************************free matrix ************************/
1638: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1639: {
1640: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1641: free((FREE_ARG)(m+nrl-NR_END));
1642: }
1643:
1644: /******************* ma3x *******************************/
1645: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1646: {
1647: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1648: double ***m;
1649:
1650: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1651: if (!m) nrerror("allocation failure 1 in matrix()");
1652: m += NR_END;
1653: m -= nrl;
1654:
1655: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1656: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1657: m[nrl] += NR_END;
1658: m[nrl] -= ncl;
1659:
1660: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1661:
1662: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1663: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1664: m[nrl][ncl] += NR_END;
1665: m[nrl][ncl] -= nll;
1666: for (j=ncl+1; j<=nch; j++)
1667: m[nrl][j]=m[nrl][j-1]+nlay;
1668:
1669: for (i=nrl+1; i<=nrh; i++) {
1670: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1671: for (j=ncl+1; j<=nch; j++)
1672: m[i][j]=m[i][j-1]+nlay;
1673: }
1674: return m;
1675: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1676: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1677: */
1678: }
1679:
1680: /*************************free ma3x ************************/
1681: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1682: {
1683: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1684: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1685: free((FREE_ARG)(m+nrl-NR_END));
1686: }
1687:
1688: /*************** function subdirf ***********/
1689: char *subdirf(char fileres[])
1690: {
1691: /* Caution optionfilefiname is hidden */
1692: strcpy(tmpout,optionfilefiname);
1693: strcat(tmpout,"/"); /* Add to the right */
1694: strcat(tmpout,fileres);
1695: return tmpout;
1696: }
1697:
1698: /*************** function subdirf2 ***********/
1699: char *subdirf2(char fileres[], char *preop)
1700: {
1701:
1702: /* Caution optionfilefiname is hidden */
1703: strcpy(tmpout,optionfilefiname);
1704: strcat(tmpout,"/");
1705: strcat(tmpout,preop);
1706: strcat(tmpout,fileres);
1707: return tmpout;
1708: }
1709:
1710: /*************** function subdirf3 ***********/
1711: char *subdirf3(char fileres[], char *preop, char *preop2)
1712: {
1713:
1714: /* Caution optionfilefiname is hidden */
1715: strcpy(tmpout,optionfilefiname);
1716: strcat(tmpout,"/");
1717: strcat(tmpout,preop);
1718: strcat(tmpout,preop2);
1719: strcat(tmpout,fileres);
1720: return tmpout;
1721: }
1.213 brouard 1722:
1723: /*************** function subdirfext ***********/
1724: char *subdirfext(char fileres[], char *preop, char *postop)
1725: {
1726:
1727: strcpy(tmpout,preop);
1728: strcat(tmpout,fileres);
1729: strcat(tmpout,postop);
1730: return tmpout;
1731: }
1.126 brouard 1732:
1.213 brouard 1733: /*************** function subdirfext3 ***********/
1734: char *subdirfext3(char fileres[], char *preop, char *postop)
1735: {
1736:
1737: /* Caution optionfilefiname is hidden */
1738: strcpy(tmpout,optionfilefiname);
1739: strcat(tmpout,"/");
1740: strcat(tmpout,preop);
1741: strcat(tmpout,fileres);
1742: strcat(tmpout,postop);
1743: return tmpout;
1744: }
1745:
1.162 brouard 1746: char *asc_diff_time(long time_sec, char ascdiff[])
1747: {
1748: long sec_left, days, hours, minutes;
1749: days = (time_sec) / (60*60*24);
1750: sec_left = (time_sec) % (60*60*24);
1751: hours = (sec_left) / (60*60) ;
1752: sec_left = (sec_left) %(60*60);
1753: minutes = (sec_left) /60;
1754: sec_left = (sec_left) % (60);
1755: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1756: return ascdiff;
1757: }
1758:
1.126 brouard 1759: /***************** f1dim *************************/
1760: extern int ncom;
1761: extern double *pcom,*xicom;
1762: extern double (*nrfunc)(double []);
1763:
1764: double f1dim(double x)
1765: {
1766: int j;
1767: double f;
1768: double *xt;
1769:
1770: xt=vector(1,ncom);
1771: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1772: f=(*nrfunc)(xt);
1773: free_vector(xt,1,ncom);
1774: return f;
1775: }
1776:
1777: /*****************brent *************************/
1778: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1779: {
1780: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1781: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1782: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1783: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1784: * returned function value.
1785: */
1.126 brouard 1786: int iter;
1787: double a,b,d,etemp;
1.159 brouard 1788: double fu=0,fv,fw,fx;
1.164 brouard 1789: double ftemp=0.;
1.126 brouard 1790: double p,q,r,tol1,tol2,u,v,w,x,xm;
1791: double e=0.0;
1792:
1793: a=(ax < cx ? ax : cx);
1794: b=(ax > cx ? ax : cx);
1795: x=w=v=bx;
1796: fw=fv=fx=(*f)(x);
1797: for (iter=1;iter<=ITMAX;iter++) {
1798: xm=0.5*(a+b);
1799: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1800: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1801: printf(".");fflush(stdout);
1802: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1803: #ifdef DEBUGBRENT
1.126 brouard 1804: 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);
1805: 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);
1806: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1807: #endif
1808: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1809: *xmin=x;
1810: return fx;
1811: }
1812: ftemp=fu;
1813: if (fabs(e) > tol1) {
1814: r=(x-w)*(fx-fv);
1815: q=(x-v)*(fx-fw);
1816: p=(x-v)*q-(x-w)*r;
1817: q=2.0*(q-r);
1818: if (q > 0.0) p = -p;
1819: q=fabs(q);
1820: etemp=e;
1821: e=d;
1822: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 1823: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 1824: else {
1.224 brouard 1825: d=p/q;
1826: u=x+d;
1827: if (u-a < tol2 || b-u < tol2)
1828: d=SIGN(tol1,xm-x);
1.126 brouard 1829: }
1830: } else {
1831: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1832: }
1833: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
1834: fu=(*f)(u);
1835: if (fu <= fx) {
1836: if (u >= x) a=x; else b=x;
1837: SHFT(v,w,x,u)
1.183 brouard 1838: SHFT(fv,fw,fx,fu)
1839: } else {
1840: if (u < x) a=u; else b=u;
1841: if (fu <= fw || w == x) {
1.224 brouard 1842: v=w;
1843: w=u;
1844: fv=fw;
1845: fw=fu;
1.183 brouard 1846: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 1847: v=u;
1848: fv=fu;
1.183 brouard 1849: }
1850: }
1.126 brouard 1851: }
1852: nrerror("Too many iterations in brent");
1853: *xmin=x;
1854: return fx;
1855: }
1856:
1857: /****************** mnbrak ***********************/
1858:
1859: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
1860: double (*func)(double))
1.183 brouard 1861: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
1862: the downhill direction (defined by the function as evaluated at the initial points) and returns
1863: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
1864: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
1865: */
1.126 brouard 1866: double ulim,u,r,q, dum;
1867: double fu;
1.187 brouard 1868:
1869: double scale=10.;
1870: int iterscale=0;
1871:
1872: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
1873: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
1874:
1875:
1876: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
1877: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
1878: /* *bx = *ax - (*ax - *bx)/scale; */
1879: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
1880: /* } */
1881:
1.126 brouard 1882: if (*fb > *fa) {
1883: SHFT(dum,*ax,*bx,dum)
1.183 brouard 1884: SHFT(dum,*fb,*fa,dum)
1885: }
1.126 brouard 1886: *cx=(*bx)+GOLD*(*bx-*ax);
1887: *fc=(*func)(*cx);
1.183 brouard 1888: #ifdef DEBUG
1.224 brouard 1889: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1890: 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 1891: #endif
1.224 brouard 1892: 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 1893: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 1894: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 1895: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 1896: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
1897: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
1898: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 1899: fu=(*func)(u);
1.163 brouard 1900: #ifdef DEBUG
1901: /* f(x)=A(x-u)**2+f(u) */
1902: double A, fparabu;
1903: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1904: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 1905: 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);
1906: 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 1907: /* And thus,it can be that fu > *fc even if fparabu < *fc */
1908: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
1909: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
1910: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 1911: #endif
1.184 brouard 1912: #ifdef MNBRAKORIGINAL
1.183 brouard 1913: #else
1.191 brouard 1914: /* if (fu > *fc) { */
1915: /* #ifdef DEBUG */
1916: /* printf("mnbrak4 fu > fc \n"); */
1917: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
1918: /* #endif */
1919: /* /\* 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 *\\/ *\/ */
1920: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
1921: /* dum=u; /\* Shifting c and u *\/ */
1922: /* u = *cx; */
1923: /* *cx = dum; */
1924: /* dum = fu; */
1925: /* fu = *fc; */
1926: /* *fc =dum; */
1927: /* } else { /\* end *\/ */
1928: /* #ifdef DEBUG */
1929: /* printf("mnbrak3 fu < fc \n"); */
1930: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
1931: /* #endif */
1932: /* dum=u; /\* Shifting c and u *\/ */
1933: /* u = *cx; */
1934: /* *cx = dum; */
1935: /* dum = fu; */
1936: /* fu = *fc; */
1937: /* *fc =dum; */
1938: /* } */
1.224 brouard 1939: #ifdef DEBUGMNBRAK
1940: double A, fparabu;
1941: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
1942: fparabu= *fa - A*(*ax-u)*(*ax-u);
1943: 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);
1944: 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 1945: #endif
1.191 brouard 1946: dum=u; /* Shifting c and u */
1947: u = *cx;
1948: *cx = dum;
1949: dum = fu;
1950: fu = *fc;
1951: *fc =dum;
1.183 brouard 1952: #endif
1.162 brouard 1953: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 1954: #ifdef DEBUG
1.224 brouard 1955: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1956: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 1957: #endif
1.126 brouard 1958: fu=(*func)(u);
1959: if (fu < *fc) {
1.183 brouard 1960: #ifdef DEBUG
1.224 brouard 1961: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1962: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
1963: #endif
1964: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
1965: SHFT(*fb,*fc,fu,(*func)(u))
1966: #ifdef DEBUG
1967: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 1968: #endif
1969: }
1.162 brouard 1970: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 1971: #ifdef DEBUG
1.224 brouard 1972: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1973: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 1974: #endif
1.126 brouard 1975: u=ulim;
1976: fu=(*func)(u);
1.183 brouard 1977: } else { /* u could be left to b (if r > q parabola has a maximum) */
1978: #ifdef DEBUG
1.224 brouard 1979: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1980: 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 1981: #endif
1.126 brouard 1982: u=(*cx)+GOLD*(*cx-*bx);
1983: fu=(*func)(u);
1.224 brouard 1984: #ifdef DEBUG
1985: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1986: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
1987: #endif
1.183 brouard 1988: } /* end tests */
1.126 brouard 1989: SHFT(*ax,*bx,*cx,u)
1.183 brouard 1990: SHFT(*fa,*fb,*fc,fu)
1991: #ifdef DEBUG
1.224 brouard 1992: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1993: 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 1994: #endif
1995: } /* 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 1996: }
1997:
1998: /*************** linmin ************************/
1.162 brouard 1999: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2000: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2001: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2002: the value of func at the returned location p . This is actually all accomplished by calling the
2003: routines mnbrak and brent .*/
1.126 brouard 2004: int ncom;
2005: double *pcom,*xicom;
2006: double (*nrfunc)(double []);
2007:
1.224 brouard 2008: #ifdef LINMINORIGINAL
1.126 brouard 2009: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2010: #else
2011: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2012: #endif
1.126 brouard 2013: {
2014: double brent(double ax, double bx, double cx,
2015: double (*f)(double), double tol, double *xmin);
2016: double f1dim(double x);
2017: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2018: double *fc, double (*func)(double));
2019: int j;
2020: double xx,xmin,bx,ax;
2021: double fx,fb,fa;
1.187 brouard 2022:
1.203 brouard 2023: #ifdef LINMINORIGINAL
2024: #else
2025: double scale=10., axs, xxs; /* Scale added for infinity */
2026: #endif
2027:
1.126 brouard 2028: ncom=n;
2029: pcom=vector(1,n);
2030: xicom=vector(1,n);
2031: nrfunc=func;
2032: for (j=1;j<=n;j++) {
2033: pcom[j]=p[j];
1.202 brouard 2034: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2035: }
1.187 brouard 2036:
1.203 brouard 2037: #ifdef LINMINORIGINAL
2038: xx=1.;
2039: #else
2040: axs=0.0;
2041: xxs=1.;
2042: do{
2043: xx= xxs;
2044: #endif
1.187 brouard 2045: ax=0.;
2046: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2047: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2048: /* 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)) */
2049: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2050: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2051: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2052: /* 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 2053: #ifdef LINMINORIGINAL
2054: #else
2055: if (fx != fx){
1.224 brouard 2056: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2057: printf("|");
2058: fprintf(ficlog,"|");
1.203 brouard 2059: #ifdef DEBUGLINMIN
1.224 brouard 2060: 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 2061: #endif
2062: }
1.224 brouard 2063: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2064: #endif
2065:
1.191 brouard 2066: #ifdef DEBUGLINMIN
2067: 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 2068: 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 2069: #endif
1.224 brouard 2070: #ifdef LINMINORIGINAL
2071: #else
2072: if(fb == fx){ /* Flat function in the direction */
2073: xmin=xx;
2074: *flat=1;
2075: }else{
2076: *flat=0;
2077: #endif
2078: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2079: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2080: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2081: /* fmin = f(p[j] + xmin * xi[j]) */
2082: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2083: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2084: #ifdef DEBUG
1.224 brouard 2085: 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);
2086: 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);
2087: #endif
2088: #ifdef LINMINORIGINAL
2089: #else
2090: }
1.126 brouard 2091: #endif
1.191 brouard 2092: #ifdef DEBUGLINMIN
2093: printf("linmin end ");
1.202 brouard 2094: fprintf(ficlog,"linmin end ");
1.191 brouard 2095: #endif
1.126 brouard 2096: for (j=1;j<=n;j++) {
1.203 brouard 2097: #ifdef LINMINORIGINAL
2098: xi[j] *= xmin;
2099: #else
2100: #ifdef DEBUGLINMIN
2101: if(xxs <1.0)
2102: printf(" before xi[%d]=%12.8f", j,xi[j]);
2103: #endif
2104: 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) */
2105: #ifdef DEBUGLINMIN
2106: if(xxs <1.0)
2107: 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 );
2108: #endif
2109: #endif
1.187 brouard 2110: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2111: }
1.191 brouard 2112: #ifdef DEBUGLINMIN
1.203 brouard 2113: printf("\n");
1.191 brouard 2114: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2115: 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 2116: for (j=1;j<=n;j++) {
1.202 brouard 2117: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2118: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2119: if(j % ncovmodel == 0){
1.191 brouard 2120: printf("\n");
1.202 brouard 2121: fprintf(ficlog,"\n");
2122: }
1.191 brouard 2123: }
1.203 brouard 2124: #else
1.191 brouard 2125: #endif
1.126 brouard 2126: free_vector(xicom,1,n);
2127: free_vector(pcom,1,n);
2128: }
2129:
2130:
2131: /*************** powell ************************/
1.162 brouard 2132: /*
2133: Minimization of a function func of n variables. Input consists of an initial starting point
2134: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2135: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2136: such that failure to decrease by more than this amount on one iteration signals doneness. On
2137: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2138: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2139: */
1.224 brouard 2140: #ifdef LINMINORIGINAL
2141: #else
2142: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2143: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2144: #endif
1.126 brouard 2145: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2146: double (*func)(double []))
2147: {
1.224 brouard 2148: #ifdef LINMINORIGINAL
2149: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2150: double (*func)(double []));
1.224 brouard 2151: #else
1.241 brouard 2152: void linmin(double p[], double xi[], int n, double *fret,
2153: double (*func)(double []),int *flat);
1.224 brouard 2154: #endif
1.239 brouard 2155: int i,ibig,j,jk,k;
1.126 brouard 2156: double del,t,*pt,*ptt,*xit;
1.181 brouard 2157: double directest;
1.126 brouard 2158: double fp,fptt;
2159: double *xits;
2160: int niterf, itmp;
1.224 brouard 2161: #ifdef LINMINORIGINAL
2162: #else
2163:
2164: flatdir=ivector(1,n);
2165: for (j=1;j<=n;j++) flatdir[j]=0;
2166: #endif
1.126 brouard 2167:
2168: pt=vector(1,n);
2169: ptt=vector(1,n);
2170: xit=vector(1,n);
2171: xits=vector(1,n);
2172: *fret=(*func)(p);
2173: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2174: rcurr_time = time(NULL);
1.126 brouard 2175: for (*iter=1;;++(*iter)) {
1.187 brouard 2176: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2177: ibig=0;
2178: del=0.0;
1.157 brouard 2179: rlast_time=rcurr_time;
2180: /* (void) gettimeofday(&curr_time,&tzp); */
2181: rcurr_time = time(NULL);
2182: curr_time = *localtime(&rcurr_time);
2183: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2184: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2185: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2186: for (i=1;i<=n;i++) {
1.126 brouard 2187: fprintf(ficrespow," %.12lf", p[i]);
2188: }
1.239 brouard 2189: fprintf(ficrespow,"\n");fflush(ficrespow);
2190: printf("\n#model= 1 + age ");
2191: fprintf(ficlog,"\n#model= 1 + age ");
2192: if(nagesqr==1){
1.241 brouard 2193: printf(" + age*age ");
2194: fprintf(ficlog," + age*age ");
1.239 brouard 2195: }
2196: for(j=1;j <=ncovmodel-2;j++){
2197: if(Typevar[j]==0) {
2198: printf(" + V%d ",Tvar[j]);
2199: fprintf(ficlog," + V%d ",Tvar[j]);
2200: }else if(Typevar[j]==1) {
2201: printf(" + V%d*age ",Tvar[j]);
2202: fprintf(ficlog," + V%d*age ",Tvar[j]);
2203: }else if(Typevar[j]==2) {
2204: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2205: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2206: }
2207: }
1.126 brouard 2208: printf("\n");
1.239 brouard 2209: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2210: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2211: fprintf(ficlog,"\n");
1.239 brouard 2212: for(i=1,jk=1; i <=nlstate; i++){
2213: for(k=1; k <=(nlstate+ndeath); k++){
2214: if (k != i) {
2215: printf("%d%d ",i,k);
2216: fprintf(ficlog,"%d%d ",i,k);
2217: for(j=1; j <=ncovmodel; j++){
2218: printf("%12.7f ",p[jk]);
2219: fprintf(ficlog,"%12.7f ",p[jk]);
2220: jk++;
2221: }
2222: printf("\n");
2223: fprintf(ficlog,"\n");
2224: }
2225: }
2226: }
1.241 brouard 2227: if(*iter <=3 && *iter >1){
1.157 brouard 2228: tml = *localtime(&rcurr_time);
2229: strcpy(strcurr,asctime(&tml));
2230: rforecast_time=rcurr_time;
1.126 brouard 2231: itmp = strlen(strcurr);
2232: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2233: strcurr[itmp-1]='\0';
1.162 brouard 2234: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2235: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2236: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2237: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2238: forecast_time = *localtime(&rforecast_time);
2239: strcpy(strfor,asctime(&forecast_time));
2240: itmp = strlen(strfor);
2241: if(strfor[itmp-1]=='\n')
2242: strfor[itmp-1]='\0';
2243: 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);
2244: 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 2245: }
2246: }
1.187 brouard 2247: for (i=1;i<=n;i++) { /* For each direction i */
2248: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2249: fptt=(*fret);
2250: #ifdef DEBUG
1.203 brouard 2251: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2252: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2253: #endif
1.203 brouard 2254: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2255: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2256: #ifdef LINMINORIGINAL
1.188 brouard 2257: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2258: #else
2259: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2260: flatdir[i]=flat; /* Function is vanishing in that direction i */
2261: #endif
2262: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2263: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2264: /* because that direction will be replaced unless the gain del is small */
2265: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2266: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2267: /* with the new direction. */
2268: del=fabs(fptt-(*fret));
2269: ibig=i;
1.126 brouard 2270: }
2271: #ifdef DEBUG
2272: printf("%d %.12e",i,(*fret));
2273: fprintf(ficlog,"%d %.12e",i,(*fret));
2274: for (j=1;j<=n;j++) {
1.224 brouard 2275: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2276: printf(" x(%d)=%.12e",j,xit[j]);
2277: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2278: }
2279: for(j=1;j<=n;j++) {
1.225 brouard 2280: printf(" p(%d)=%.12e",j,p[j]);
2281: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2282: }
2283: printf("\n");
2284: fprintf(ficlog,"\n");
2285: #endif
1.187 brouard 2286: } /* end loop on each direction i */
2287: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2288: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2289: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2290: for(j=1;j<=n;j++) {
1.225 brouard 2291: if(flatdir[j] >0){
2292: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2293: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2294: }
2295: /* printf("\n"); */
2296: /* fprintf(ficlog,"\n"); */
2297: }
1.243 brouard 2298: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2299: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2300: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2301: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2302: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2303: /* decreased of more than 3.84 */
2304: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2305: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2306: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2307:
1.188 brouard 2308: /* Starting the program with initial values given by a former maximization will simply change */
2309: /* the scales of the directions and the directions, because the are reset to canonical directions */
2310: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2311: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2312: #ifdef DEBUG
2313: int k[2],l;
2314: k[0]=1;
2315: k[1]=-1;
2316: printf("Max: %.12e",(*func)(p));
2317: fprintf(ficlog,"Max: %.12e",(*func)(p));
2318: for (j=1;j<=n;j++) {
2319: printf(" %.12e",p[j]);
2320: fprintf(ficlog," %.12e",p[j]);
2321: }
2322: printf("\n");
2323: fprintf(ficlog,"\n");
2324: for(l=0;l<=1;l++) {
2325: for (j=1;j<=n;j++) {
2326: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2327: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2328: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2329: }
2330: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2331: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2332: }
2333: #endif
2334:
1.224 brouard 2335: #ifdef LINMINORIGINAL
2336: #else
2337: free_ivector(flatdir,1,n);
2338: #endif
1.126 brouard 2339: free_vector(xit,1,n);
2340: free_vector(xits,1,n);
2341: free_vector(ptt,1,n);
2342: free_vector(pt,1,n);
2343: return;
1.192 brouard 2344: } /* enough precision */
1.240 brouard 2345: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2346: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2347: ptt[j]=2.0*p[j]-pt[j];
2348: xit[j]=p[j]-pt[j];
2349: pt[j]=p[j];
2350: }
1.181 brouard 2351: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2352: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2353: if (*iter <=4) {
1.225 brouard 2354: #else
2355: #endif
1.224 brouard 2356: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2357: #else
1.161 brouard 2358: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2359: #endif
1.162 brouard 2360: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2361: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2362: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2363: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2364: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2365: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2366: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2367: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2368: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2369: /* Even if f3 <f1, directest can be negative and t >0 */
2370: /* mu² and del² are equal when f3=f1 */
2371: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2372: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2373: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2374: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2375: #ifdef NRCORIGINAL
2376: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2377: #else
2378: 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 2379: t= t- del*SQR(fp-fptt);
1.183 brouard 2380: #endif
1.202 brouard 2381: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2382: #ifdef DEBUG
1.181 brouard 2383: 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);
2384: 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 2385: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2386: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2387: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2388: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2389: 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);
2390: 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);
2391: #endif
1.183 brouard 2392: #ifdef POWELLORIGINAL
2393: if (t < 0.0) { /* Then we use it for new direction */
2394: #else
1.182 brouard 2395: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2396: 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 2397: 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 2398: 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 2399: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2400: }
1.181 brouard 2401: if (directest < 0.0) { /* Then we use it for new direction */
2402: #endif
1.191 brouard 2403: #ifdef DEBUGLINMIN
1.234 brouard 2404: printf("Before linmin in direction P%d-P0\n",n);
2405: for (j=1;j<=n;j++) {
2406: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2407: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2408: if(j % ncovmodel == 0){
2409: printf("\n");
2410: fprintf(ficlog,"\n");
2411: }
2412: }
1.224 brouard 2413: #endif
2414: #ifdef LINMINORIGINAL
1.234 brouard 2415: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2416: #else
1.234 brouard 2417: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2418: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2419: #endif
1.234 brouard 2420:
1.191 brouard 2421: #ifdef DEBUGLINMIN
1.234 brouard 2422: for (j=1;j<=n;j++) {
2423: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2424: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2425: if(j % ncovmodel == 0){
2426: printf("\n");
2427: fprintf(ficlog,"\n");
2428: }
2429: }
1.224 brouard 2430: #endif
1.234 brouard 2431: for (j=1;j<=n;j++) {
2432: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2433: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2434: }
1.224 brouard 2435: #ifdef LINMINORIGINAL
2436: #else
1.234 brouard 2437: for (j=1, flatd=0;j<=n;j++) {
2438: if(flatdir[j]>0)
2439: flatd++;
2440: }
2441: if(flatd >0){
1.255 brouard 2442: printf("%d flat directions: ",flatd);
2443: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2444: for (j=1;j<=n;j++) {
2445: if(flatdir[j]>0){
2446: printf("%d ",j);
2447: fprintf(ficlog,"%d ",j);
2448: }
2449: }
2450: printf("\n");
2451: fprintf(ficlog,"\n");
2452: }
1.191 brouard 2453: #endif
1.234 brouard 2454: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2455: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2456:
1.126 brouard 2457: #ifdef DEBUG
1.234 brouard 2458: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2459: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2460: for(j=1;j<=n;j++){
2461: printf(" %lf",xit[j]);
2462: fprintf(ficlog," %lf",xit[j]);
2463: }
2464: printf("\n");
2465: fprintf(ficlog,"\n");
1.126 brouard 2466: #endif
1.192 brouard 2467: } /* end of t or directest negative */
1.224 brouard 2468: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2469: #else
1.234 brouard 2470: } /* end if (fptt < fp) */
1.192 brouard 2471: #endif
1.225 brouard 2472: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2473: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2474: #else
1.224 brouard 2475: #endif
1.234 brouard 2476: } /* loop iteration */
1.126 brouard 2477: }
1.234 brouard 2478:
1.126 brouard 2479: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2480:
1.235 brouard 2481: 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 2482: {
1.235 brouard 2483: /* Computes the prevalence limit in each live state at age x and for covariate combination ij
2484: (and selected quantitative values in nres)
2485: by left multiplying the unit
1.234 brouard 2486: matrix by transitions matrix until convergence is reached with precision ftolpl */
1.206 brouard 2487: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2488: /* Wx is row vector: population in state 1, population in state 2, population dead */
2489: /* or prevalence in state 1, prevalence in state 2, 0 */
2490: /* newm is the matrix after multiplications, its rows are identical at a factor */
2491: /* Initial matrix pimij */
2492: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2493: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2494: /* 0, 0 , 1} */
2495: /*
2496: * and after some iteration: */
2497: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2498: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2499: /* 0, 0 , 1} */
2500: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2501: /* {0.51571254859325999, 0.4842874514067399, */
2502: /* 0.51326036147820708, 0.48673963852179264} */
2503: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2504:
1.126 brouard 2505: int i, ii,j,k;
1.209 brouard 2506: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2507: /* double **matprod2(); */ /* test */
1.218 brouard 2508: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2509: double **newm;
1.209 brouard 2510: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2511: int ncvloop=0;
1.169 brouard 2512:
1.209 brouard 2513: min=vector(1,nlstate);
2514: max=vector(1,nlstate);
2515: meandiff=vector(1,nlstate);
2516:
1.218 brouard 2517: /* Starting with matrix unity */
1.126 brouard 2518: for (ii=1;ii<=nlstate+ndeath;ii++)
2519: for (j=1;j<=nlstate+ndeath;j++){
2520: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2521: }
1.169 brouard 2522:
2523: cov[1]=1.;
2524:
2525: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2526: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2527: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2528: ncvloop++;
1.126 brouard 2529: newm=savm;
2530: /* Covariates have to be included here again */
1.138 brouard 2531: cov[2]=agefin;
1.187 brouard 2532: if(nagesqr==1)
2533: cov[3]= agefin*agefin;;
1.234 brouard 2534: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2535: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2536: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2537: /* 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 2538: }
2539: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2540: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2541: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2542: /* 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 2543: }
1.237 brouard 2544: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2545: if(Dummy[Tvar[Tage[k]]]){
2546: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2547: } else{
1.235 brouard 2548: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2549: }
1.235 brouard 2550: /* 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 2551: }
1.237 brouard 2552: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2553: /* 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 2554: if(Dummy[Tvard[k][1]==0]){
2555: if(Dummy[Tvard[k][2]==0]){
2556: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2557: }else{
2558: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2559: }
2560: }else{
2561: if(Dummy[Tvard[k][2]==0]){
2562: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2563: }else{
2564: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2565: }
2566: }
1.234 brouard 2567: }
1.138 brouard 2568: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2569: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2570: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2571: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2572: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2573: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2574: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2575:
1.126 brouard 2576: savm=oldm;
2577: oldm=newm;
1.209 brouard 2578:
2579: for(j=1; j<=nlstate; j++){
2580: max[j]=0.;
2581: min[j]=1.;
2582: }
2583: for(i=1;i<=nlstate;i++){
2584: sumnew=0;
2585: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2586: for(j=1; j<=nlstate; j++){
2587: prlim[i][j]= newm[i][j]/(1-sumnew);
2588: max[j]=FMAX(max[j],prlim[i][j]);
2589: min[j]=FMIN(min[j],prlim[i][j]);
2590: }
2591: }
2592:
1.126 brouard 2593: maxmax=0.;
1.209 brouard 2594: for(j=1; j<=nlstate; j++){
2595: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2596: maxmax=FMAX(maxmax,meandiff[j]);
2597: /* 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 2598: } /* j loop */
1.203 brouard 2599: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2600: /* 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 2601: if(maxmax < ftolpl){
1.209 brouard 2602: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2603: free_vector(min,1,nlstate);
2604: free_vector(max,1,nlstate);
2605: free_vector(meandiff,1,nlstate);
1.126 brouard 2606: return prlim;
2607: }
1.169 brouard 2608: } /* age loop */
1.208 brouard 2609: /* After some age loop it doesn't converge */
1.209 brouard 2610: 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 2611: 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 2612: /* 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); */
2613: free_vector(min,1,nlstate);
2614: free_vector(max,1,nlstate);
2615: free_vector(meandiff,1,nlstate);
1.208 brouard 2616:
1.169 brouard 2617: return prlim; /* should not reach here */
1.126 brouard 2618: }
2619:
1.217 brouard 2620:
2621: /**** Back Prevalence limit (stable or period prevalence) ****************/
2622:
1.218 brouard 2623: /* 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) */
2624: /* 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 2625: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2626: {
1.264 brouard 2627: /* Computes the prevalence limit in each live state at age x and for covariate combination ij (<=2**cptcoveff) by left multiplying the unit
1.217 brouard 2628: matrix by transitions matrix until convergence is reached with precision ftolpl */
2629: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2630: /* Wx is row vector: population in state 1, population in state 2, population dead */
2631: /* or prevalence in state 1, prevalence in state 2, 0 */
2632: /* newm is the matrix after multiplications, its rows are identical at a factor */
2633: /* Initial matrix pimij */
2634: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2635: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2636: /* 0, 0 , 1} */
2637: /*
2638: * and after some iteration: */
2639: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2640: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2641: /* 0, 0 , 1} */
2642: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2643: /* {0.51571254859325999, 0.4842874514067399, */
2644: /* 0.51326036147820708, 0.48673963852179264} */
2645: /* If we start from prlim again, prlim tends to a constant matrix */
2646:
2647: int i, ii,j,k;
1.247 brouard 2648: int first=0;
1.217 brouard 2649: double *min, *max, *meandiff, maxmax,sumnew=0.;
2650: /* double **matprod2(); */ /* test */
2651: double **out, cov[NCOVMAX+1], **bmij();
2652: double **newm;
1.218 brouard 2653: double **dnewm, **doldm, **dsavm; /* for use */
2654: double **oldm, **savm; /* for use */
2655:
1.217 brouard 2656: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2657: int ncvloop=0;
2658:
2659: min=vector(1,nlstate);
2660: max=vector(1,nlstate);
2661: meandiff=vector(1,nlstate);
2662:
1.266 brouard 2663: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2664: oldm=oldms; savm=savms;
2665:
2666: /* Starting with matrix unity */
2667: for (ii=1;ii<=nlstate+ndeath;ii++)
2668: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2669: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2670: }
2671:
2672: cov[1]=1.;
2673:
2674: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2675: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2676: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2677: for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2678: ncvloop++;
1.218 brouard 2679: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2680: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2681: /* Covariates have to be included here again */
2682: cov[2]=agefin;
2683: if(nagesqr==1)
2684: cov[3]= agefin*agefin;;
1.242 brouard 2685: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2686: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2687: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2688: /* printf("bprevalim Dummy agefin=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agefin,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
1.242 brouard 2689: }
2690: /* for (k=1; k<=cptcovn;k++) { */
2691: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2692: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2693: /* /\* 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])]); *\/ */
2694: /* } */
2695: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2696: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2697: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2698: /* 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]); */
2699: }
2700: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2701: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2702: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2703: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2704: for (k=1; k<=cptcovage;k++){ /* For product with age */
2705: if(Dummy[Tvar[Tage[k]]]){
2706: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2707: } else{
2708: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2709: }
2710: /* 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]); */
2711: }
2712: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2713: /* 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]); */
2714: if(Dummy[Tvard[k][1]==0]){
2715: if(Dummy[Tvard[k][2]==0]){
2716: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2717: }else{
2718: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2719: }
2720: }else{
2721: if(Dummy[Tvard[k][2]==0]){
2722: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2723: }else{
2724: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2725: }
2726: }
1.217 brouard 2727: }
2728:
2729: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2730: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2731: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2732: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2733: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2734: /* ij should be linked to the correct index of cov */
2735: /* age and covariate values ij are in 'cov', but we need to pass
2736: * ij for the observed prevalence at age and status and covariate
2737: * number: prevacurrent[(int)agefin][ii][ij]
2738: */
2739: /* 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 *\/ */
2740: /* 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 *\/ */
2741: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.266 brouard 2742: /* if((int)age == 70){ */
2743: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2744: /* for(i=1; i<=nlstate+ndeath; i++) { */
2745: /* printf("%d newm= ",i); */
2746: /* for(j=1;j<=nlstate+ndeath;j++) { */
2747: /* printf("%f ",newm[i][j]); */
2748: /* } */
2749: /* printf("oldm * "); */
2750: /* for(j=1;j<=nlstate+ndeath;j++) { */
2751: /* printf("%f ",oldm[i][j]); */
2752: /* } */
2753: /* printf(" pmmij "); */
2754: /* for(j=1;j<=nlstate+ndeath;j++) { */
2755: /* printf("%f ",pmmij[i][j]); */
2756: /* } */
2757: /* printf("\n"); */
2758: /* } */
2759: /* } */
1.217 brouard 2760: savm=oldm;
2761: oldm=newm;
1.266 brouard 2762:
1.217 brouard 2763: for(j=1; j<=nlstate; j++){
2764: max[j]=0.;
2765: min[j]=1.;
2766: }
2767: for(j=1; j<=nlstate; j++){
2768: for(i=1;i<=nlstate;i++){
1.234 brouard 2769: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2770: bprlim[i][j]= newm[i][j];
2771: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2772: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2773: }
2774: }
1.218 brouard 2775:
1.217 brouard 2776: maxmax=0.;
2777: for(i=1; i<=nlstate; i++){
2778: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2779: maxmax=FMAX(maxmax,meandiff[i]);
2780: /* 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); */
2781: } /* j loop */
2782: *ncvyear= -( (int)age- (int)agefin);
1.218 brouard 2783: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear);*/
1.217 brouard 2784: if(maxmax < ftolpl){
1.220 brouard 2785: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2786: free_vector(min,1,nlstate);
2787: free_vector(max,1,nlstate);
2788: free_vector(meandiff,1,nlstate);
2789: return bprlim;
2790: }
2791: } /* age loop */
2792: /* After some age loop it doesn't converge */
1.247 brouard 2793: if(first){
2794: first=1;
2795: 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\
2796: 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);
2797: }
2798: 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 2799: 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);
2800: /* 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); */
2801: free_vector(min,1,nlstate);
2802: free_vector(max,1,nlstate);
2803: free_vector(meandiff,1,nlstate);
2804:
2805: return bprlim; /* should not reach here */
2806: }
2807:
1.126 brouard 2808: /*************** transition probabilities ***************/
2809:
2810: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
2811: {
1.138 brouard 2812: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 2813: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 2814: model to the ncovmodel covariates (including constant and age).
2815: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2816: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2817: ncth covariate in the global vector x is given by the formula:
2818: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2819: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2820: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2821: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 2822: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 2823: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 2824: Sum on j ps[i][j] should equal to 1.
1.138 brouard 2825: */
2826: double s1, lnpijopii;
1.126 brouard 2827: /*double t34;*/
1.164 brouard 2828: int i,j, nc, ii, jj;
1.126 brouard 2829:
1.223 brouard 2830: for(i=1; i<= nlstate; i++){
2831: for(j=1; j<i;j++){
2832: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2833: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2834: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2835: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2836: }
2837: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2838: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2839: }
2840: for(j=i+1; j<=nlstate+ndeath;j++){
2841: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2842: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
2843: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
2844: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
2845: }
2846: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2847: }
2848: }
1.218 brouard 2849:
1.223 brouard 2850: for(i=1; i<= nlstate; i++){
2851: s1=0;
2852: for(j=1; j<i; j++){
2853: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2854: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2855: }
2856: for(j=i+1; j<=nlstate+ndeath; j++){
2857: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
2858: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
2859: }
2860: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
2861: ps[i][i]=1./(s1+1.);
2862: /* Computing other pijs */
2863: for(j=1; j<i; j++)
2864: ps[i][j]= exp(ps[i][j])*ps[i][i];
2865: for(j=i+1; j<=nlstate+ndeath; j++)
2866: ps[i][j]= exp(ps[i][j])*ps[i][i];
2867: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
2868: } /* end i */
1.218 brouard 2869:
1.223 brouard 2870: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
2871: for(jj=1; jj<= nlstate+ndeath; jj++){
2872: ps[ii][jj]=0;
2873: ps[ii][ii]=1;
2874: }
2875: }
1.218 brouard 2876:
2877:
1.223 brouard 2878: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
2879: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
2880: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
2881: /* } */
2882: /* printf("\n "); */
2883: /* } */
2884: /* printf("\n ");printf("%lf ",cov[2]);*/
2885: /*
2886: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 2887: goto end;*/
1.266 brouard 2888: return ps; /* Pointer is unchanged since its call */
1.126 brouard 2889: }
2890:
1.218 brouard 2891: /*************** backward transition probabilities ***************/
2892:
2893: /* 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 ) */
2894: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
2895: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
2896: {
1.266 brouard 2897: /* Computes the backward probability at age agefin and covariate combination ij. In fact cov is already filled and x too.
2898: * Call to pmij(cov and x), call to cross prevalence, sums and inverses, left multiply, and returns in **ps as well as **bmij.
1.222 brouard 2899: */
1.218 brouard 2900: int i, ii, j,k;
1.222 brouard 2901:
2902: double **out, **pmij();
2903: double sumnew=0.;
1.218 brouard 2904: double agefin;
1.222 brouard 2905:
2906: double **dnewm, **dsavm, **doldm;
2907: double **bbmij;
2908:
1.218 brouard 2909: doldm=ddoldms; /* global pointers */
1.222 brouard 2910: dnewm=ddnewms;
2911: dsavm=ddsavms;
2912:
2913: agefin=cov[2];
2914: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 2915: the observed prevalence (with this covariate ij) at beginning of transition */
2916: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2917: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
2918: /* outputs pmmij which is a stochastic matrix */
1.222 brouard 2919: /* We do have the matrix Px in savm and we need pij */
2920: for (j=1;j<=nlstate+ndeath;j++){
1.266 brouard 2921: sumnew=0.; /* w1 p11 + w2 p21 only on live states N1./N..*N11/N1. + N2./N..*N21/N2.=(N11+N21)/N..=N.1/N.. */
1.222 brouard 2922: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 2923: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
2924: sumnew+=pmmij[ii][j]*prevacurrent[(int)agefin][ii][ij]; /* Yes prevalence at beginning of transition */
1.222 brouard 2925: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.266 brouard 2926: if(sumnew >= 1.e-10){
2927: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 2928: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
2929: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2930: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
2931: /* doldm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
2932: /* }else */
2933: doldm[ii][j]=(ii==j ? 1./sumnew : 0.0);
1.266 brouard 2934: } /*End ii */
2935: }else{ /* We put the identity matrix */
2936: for (ii=1;ii<=nlstate+ndeath;ii++){
2937: doldm[ii][j]=(ii==j ? 1. : 0.0);
2938: } /*End ii */
2939: /* printf("Problem internal bmij A: sum_i w_i*p_ij=N.j/N.. <1.e-10 i=%d, j=%d, sumnew=%lf,agefin=%d\n",ii,j,sumnew, (int)agefin); */
2940: }
2941: } /* End j, At the end doldm is diag[1/(w_1p1i+w_2 p2i)] or identity*/
2942: /* left Product of this diag matrix by dsavm=Px (dnewm=dsavm*doldm) */
2943: /* bbmij=matprod2(dnewm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /\* Bug Valgrind *\/ */
2944: bbmij=matprod2(dnewm, pmmij,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, doldm); /* Bug Valgrind */
1.222 brouard 2945: /* dsavm=doldm; /\* dsavm is now diag [1/(w_1p1i+w_2 p2i)] but can be overwritten*\/ */
2946: /* doldm=dnewm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2947: /* dnewm=dsavm; /\* doldm is now Px * diag [1/(w_1p1i+w_2 p2i)] *\/ */
2948: /* left Product of this matrix by diag matrix of prevalences (savm) */
2949: for (j=1;j<=nlstate+ndeath;j++){
1.266 brouard 2950: sumnew=0.;
1.222 brouard 2951: for (ii=1;ii<=nlstate+ndeath;ii++){
1.266 brouard 2952: sumnew+=prevacurrent[(int)agefin][ii][ij];
1.222 brouard 2953: dsavm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij] : 0.0);
2954: }
1.266 brouard 2955: /* if(sumnew <0.9){ */
2956: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
2957: /* } */
2958: } /* End j, At the end dsavm is diag[(w_i)] */
2959: /* What if dsavm doesn't sum ii to 1? */
2960: /* ps=matprod2(doldm, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /\* Bug Valgrind *\/ */
2961: ps=matprod2(ps, dsavm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dnewm); /* Bug Valgrind */
1.222 brouard 2962: /* newm or out is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
2963: /* end bmij */
1.266 brouard 2964: return ps; /*pointer is unchanged */
1.218 brouard 2965: }
1.217 brouard 2966: /*************** transition probabilities ***************/
2967:
1.218 brouard 2968: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 2969: {
2970: /* According to parameters values stored in x and the covariate's values stored in cov,
2971: computes the probability to be observed in state j being in state i by appying the
2972: model to the ncovmodel covariates (including constant and age).
2973: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
2974: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
2975: ncth covariate in the global vector x is given by the formula:
2976: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
2977: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
2978: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
2979: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
2980: Outputs ps[i][j] the probability to be observed in j being in j according to
2981: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
2982: */
2983: double s1, lnpijopii;
2984: /*double t34;*/
2985: int i,j, nc, ii, jj;
2986:
1.234 brouard 2987: for(i=1; i<= nlstate; i++){
2988: for(j=1; j<i;j++){
2989: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2990: /*lnpijopii += param[i][j][nc]*cov[nc];*/
2991: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
2992: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2993: }
2994: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
2995: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
2996: }
2997: for(j=i+1; j<=nlstate+ndeath;j++){
2998: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
2999: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3000: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3001: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3002: }
3003: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3004: }
3005: }
3006:
3007: for(i=1; i<= nlstate; i++){
3008: s1=0;
3009: for(j=1; j<i; j++){
3010: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3011: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3012: }
3013: for(j=i+1; j<=nlstate+ndeath; j++){
3014: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3015: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3016: }
3017: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3018: ps[i][i]=1./(s1+1.);
3019: /* Computing other pijs */
3020: for(j=1; j<i; j++)
3021: ps[i][j]= exp(ps[i][j])*ps[i][i];
3022: for(j=i+1; j<=nlstate+ndeath; j++)
3023: ps[i][j]= exp(ps[i][j])*ps[i][i];
3024: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3025: } /* end i */
3026:
3027: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3028: for(jj=1; jj<= nlstate+ndeath; jj++){
3029: ps[ii][jj]=0;
3030: ps[ii][ii]=1;
3031: }
3032: }
3033: /* Added for backcast */ /* Transposed matrix too */
3034: for(jj=1; jj<= nlstate+ndeath; jj++){
3035: s1=0.;
3036: for(ii=1; ii<= nlstate+ndeath; ii++){
3037: s1+=ps[ii][jj];
3038: }
3039: for(ii=1; ii<= nlstate; ii++){
3040: ps[ii][jj]=ps[ii][jj]/s1;
3041: }
3042: }
3043: /* Transposition */
3044: for(jj=1; jj<= nlstate+ndeath; jj++){
3045: for(ii=jj; ii<= nlstate+ndeath; ii++){
3046: s1=ps[ii][jj];
3047: ps[ii][jj]=ps[jj][ii];
3048: ps[jj][ii]=s1;
3049: }
3050: }
3051: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3052: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3053: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3054: /* } */
3055: /* printf("\n "); */
3056: /* } */
3057: /* printf("\n ");printf("%lf ",cov[2]);*/
3058: /*
3059: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3060: goto end;*/
3061: return ps;
1.217 brouard 3062: }
3063:
3064:
1.126 brouard 3065: /**************** Product of 2 matrices ******************/
3066:
1.145 brouard 3067: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3068: {
3069: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3070: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3071: /* in, b, out are matrice of pointers which should have been initialized
3072: before: only the contents of out is modified. The function returns
3073: a pointer to pointers identical to out */
1.145 brouard 3074: int i, j, k;
1.126 brouard 3075: for(i=nrl; i<= nrh; i++)
1.145 brouard 3076: for(k=ncolol; k<=ncoloh; k++){
3077: out[i][k]=0.;
3078: for(j=ncl; j<=nch; j++)
3079: out[i][k] +=in[i][j]*b[j][k];
3080: }
1.126 brouard 3081: return out;
3082: }
3083:
3084:
3085: /************* Higher Matrix Product ***************/
3086:
1.235 brouard 3087: 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 3088: {
1.218 brouard 3089: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3090: 'nhstepm*hstepm*stepm' months (i.e. until
3091: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3092: nhstepm*hstepm matrices.
3093: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3094: (typically every 2 years instead of every month which is too big
3095: for the memory).
3096: Model is determined by parameters x and covariates have to be
3097: included manually here.
3098:
3099: */
3100:
3101: int i, j, d, h, k;
1.131 brouard 3102: double **out, cov[NCOVMAX+1];
1.126 brouard 3103: double **newm;
1.187 brouard 3104: double agexact;
1.214 brouard 3105: double agebegin, ageend;
1.126 brouard 3106:
3107: /* Hstepm could be zero and should return the unit matrix */
3108: for (i=1;i<=nlstate+ndeath;i++)
3109: for (j=1;j<=nlstate+ndeath;j++){
3110: oldm[i][j]=(i==j ? 1.0 : 0.0);
3111: po[i][j][0]=(i==j ? 1.0 : 0.0);
3112: }
3113: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3114: for(h=1; h <=nhstepm; h++){
3115: for(d=1; d <=hstepm; d++){
3116: newm=savm;
3117: /* Covariates have to be included here again */
3118: cov[1]=1.;
1.214 brouard 3119: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3120: cov[2]=agexact;
3121: if(nagesqr==1)
1.227 brouard 3122: cov[3]= agexact*agexact;
1.235 brouard 3123: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3124: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3125: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3126: /* 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)); */
3127: }
3128: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3129: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3130: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3131: /* 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]); */
3132: }
3133: for (k=1; k<=cptcovage;k++){
3134: if(Dummy[Tvar[Tage[k]]]){
3135: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3136: } else{
3137: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3138: }
3139: /* 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]); */
3140: }
3141: for (k=1; k<=cptcovprod;k++){ /* */
3142: /* 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]); */
3143: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3144: }
3145: /* for (k=1; k<=cptcovn;k++) */
3146: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3147: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3148: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3149: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3150: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3151:
3152:
1.126 brouard 3153: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3154: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3155: /* right multiplication of oldm by the current matrix */
1.126 brouard 3156: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3157: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3158: /* if((int)age == 70){ */
3159: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3160: /* for(i=1; i<=nlstate+ndeath; i++) { */
3161: /* printf("%d pmmij ",i); */
3162: /* for(j=1;j<=nlstate+ndeath;j++) { */
3163: /* printf("%f ",pmmij[i][j]); */
3164: /* } */
3165: /* printf(" oldm "); */
3166: /* for(j=1;j<=nlstate+ndeath;j++) { */
3167: /* printf("%f ",oldm[i][j]); */
3168: /* } */
3169: /* printf("\n"); */
3170: /* } */
3171: /* } */
1.126 brouard 3172: savm=oldm;
3173: oldm=newm;
3174: }
3175: for(i=1; i<=nlstate+ndeath; i++)
3176: for(j=1;j<=nlstate+ndeath;j++) {
1.267 ! brouard 3177: po[i][j][h]=newm[i][j];
! 3178: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3179: }
1.128 brouard 3180: /*printf("h=%d ",h);*/
1.126 brouard 3181: } /* end h */
1.267 ! brouard 3182: /* printf("\n H=%d \n",h); */
1.126 brouard 3183: return po;
3184: }
3185:
1.217 brouard 3186: /************* Higher Back Matrix Product ***************/
1.218 brouard 3187: /* 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 ) */
1.267 ! brouard 3188: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij, int nres )
1.217 brouard 3189: {
1.266 brouard 3190: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3191: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3192: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3193: nhstepm*hstepm matrices.
3194: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3195: (typically every 2 years instead of every month which is too big
1.217 brouard 3196: for the memory).
1.218 brouard 3197: Model is determined by parameters x and covariates have to be
1.266 brouard 3198: included manually here. Then we use a call to bmij(x and cov)
3199: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3200: */
1.217 brouard 3201:
3202: int i, j, d, h, k;
1.266 brouard 3203: double **out, cov[NCOVMAX+1], **bmij();
3204: double **newm, ***newmm;
1.217 brouard 3205: double agexact;
3206: double agebegin, ageend;
1.222 brouard 3207: double **oldm, **savm;
1.217 brouard 3208:
1.266 brouard 3209: newmm=po; /* To be saved */
3210: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3211: /* Hstepm could be zero and should return the unit matrix */
3212: for (i=1;i<=nlstate+ndeath;i++)
3213: for (j=1;j<=nlstate+ndeath;j++){
3214: oldm[i][j]=(i==j ? 1.0 : 0.0);
3215: po[i][j][0]=(i==j ? 1.0 : 0.0);
3216: }
3217: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3218: for(h=1; h <=nhstepm; h++){
3219: for(d=1; d <=hstepm; d++){
3220: newm=savm;
3221: /* Covariates have to be included here again */
3222: cov[1]=1.;
1.266 brouard 3223: agexact=age-((h-1)*hstepm + (d))*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3224: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3225: cov[2]=agexact;
3226: if(nagesqr==1)
1.222 brouard 3227: cov[3]= agexact*agexact;
1.266 brouard 3228: for (k=1; k<=cptcovn;k++){
3229: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3230: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3231: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3232: /* printf("hbxij Dummy agexact=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agexact,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
3233: }
1.267 ! brouard 3234: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
! 3235: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
! 3236: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
! 3237: /* 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]); */
! 3238: }
! 3239: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
! 3240: if(Dummy[Tvar[Tage[k]]]){
! 3241: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
! 3242: } else{
! 3243: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
! 3244: }
! 3245: /* printf("hBxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); */
! 3246: }
! 3247: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3248: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 ! brouard 3249: }
1.217 brouard 3250: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3251: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 ! brouard 3252:
1.218 brouard 3253: /* Careful transposed matrix */
1.266 brouard 3254: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3255: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3256: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3257: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3258: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3259: /* if((int)age == 70){ */
3260: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3261: /* for(i=1; i<=nlstate+ndeath; i++) { */
3262: /* printf("%d pmmij ",i); */
3263: /* for(j=1;j<=nlstate+ndeath;j++) { */
3264: /* printf("%f ",pmmij[i][j]); */
3265: /* } */
3266: /* printf(" oldm "); */
3267: /* for(j=1;j<=nlstate+ndeath;j++) { */
3268: /* printf("%f ",oldm[i][j]); */
3269: /* } */
3270: /* printf("\n"); */
3271: /* } */
3272: /* } */
3273: savm=oldm;
3274: oldm=newm;
3275: }
3276: for(i=1; i<=nlstate+ndeath; i++)
3277: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3278: po[i][j][h]=newm[i][j];
3279: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.217 brouard 3280: }
3281: /*printf("h=%d ",h);*/
3282: } /* end h */
1.222 brouard 3283: /* printf("\n H=%d \n",h); */
1.217 brouard 3284: return po;
3285: }
3286:
3287:
1.162 brouard 3288: #ifdef NLOPT
3289: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3290: double fret;
3291: double *xt;
3292: int j;
3293: myfunc_data *d2 = (myfunc_data *) pd;
3294: /* xt = (p1-1); */
3295: xt=vector(1,n);
3296: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3297:
3298: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3299: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3300: printf("Function = %.12lf ",fret);
3301: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3302: printf("\n");
3303: free_vector(xt,1,n);
3304: return fret;
3305: }
3306: #endif
1.126 brouard 3307:
3308: /*************** log-likelihood *************/
3309: double func( double *x)
3310: {
1.226 brouard 3311: int i, ii, j, k, mi, d, kk;
3312: int ioffset=0;
3313: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3314: double **out;
3315: double lli; /* Individual log likelihood */
3316: int s1, s2;
1.228 brouard 3317: 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 3318: double bbh, survp;
3319: long ipmx;
3320: double agexact;
3321: /*extern weight */
3322: /* We are differentiating ll according to initial status */
3323: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3324: /*for(i=1;i<imx;i++)
3325: printf(" %d\n",s[4][i]);
3326: */
1.162 brouard 3327:
1.226 brouard 3328: ++countcallfunc;
1.162 brouard 3329:
1.226 brouard 3330: cov[1]=1.;
1.126 brouard 3331:
1.226 brouard 3332: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3333: ioffset=0;
1.226 brouard 3334: if(mle==1){
3335: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3336: /* Computes the values of the ncovmodel covariates of the model
3337: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3338: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3339: to be observed in j being in i according to the model.
3340: */
1.243 brouard 3341: ioffset=2+nagesqr ;
1.233 brouard 3342: /* Fixed */
1.234 brouard 3343: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3344: 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)*/
3345: }
1.226 brouard 3346: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3347: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3348: has been calculated etc */
3349: /* For an individual i, wav[i] gives the number of effective waves */
3350: /* We compute the contribution to Likelihood of each effective transition
3351: mw[mi][i] is real wave of the mi th effectve wave */
3352: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3353: s2=s[mw[mi+1][i]][i];
3354: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3355: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3356: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3357: */
3358: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3359: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3360: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3361: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3362: }
3363: for (ii=1;ii<=nlstate+ndeath;ii++)
3364: for (j=1;j<=nlstate+ndeath;j++){
3365: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3366: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3367: }
3368: for(d=0; d<dh[mi][i]; d++){
3369: newm=savm;
3370: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3371: cov[2]=agexact;
3372: if(nagesqr==1)
3373: cov[3]= agexact*agexact; /* Should be changed here */
3374: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3375: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3376: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3377: else
3378: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3379: }
3380: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3381: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3382: savm=oldm;
3383: oldm=newm;
3384: } /* end mult */
3385:
3386: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3387: /* But now since version 0.9 we anticipate for bias at large stepm.
3388: * If stepm is larger than one month (smallest stepm) and if the exact delay
3389: * (in months) between two waves is not a multiple of stepm, we rounded to
3390: * the nearest (and in case of equal distance, to the lowest) interval but now
3391: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3392: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3393: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3394: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3395: * -stepm/2 to stepm/2 .
3396: * For stepm=1 the results are the same as for previous versions of Imach.
3397: * For stepm > 1 the results are less biased than in previous versions.
3398: */
1.234 brouard 3399: s1=s[mw[mi][i]][i];
3400: s2=s[mw[mi+1][i]][i];
3401: bbh=(double)bh[mi][i]/(double)stepm;
3402: /* bias bh is positive if real duration
3403: * is higher than the multiple of stepm and negative otherwise.
3404: */
3405: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3406: if( s2 > nlstate){
3407: /* i.e. if s2 is a death state and if the date of death is known
3408: then the contribution to the likelihood is the probability to
3409: die between last step unit time and current step unit time,
3410: which is also equal to probability to die before dh
3411: minus probability to die before dh-stepm .
3412: In version up to 0.92 likelihood was computed
3413: as if date of death was unknown. Death was treated as any other
3414: health state: the date of the interview describes the actual state
3415: and not the date of a change in health state. The former idea was
3416: to consider that at each interview the state was recorded
3417: (healthy, disable or death) and IMaCh was corrected; but when we
3418: introduced the exact date of death then we should have modified
3419: the contribution of an exact death to the likelihood. This new
3420: contribution is smaller and very dependent of the step unit
3421: stepm. It is no more the probability to die between last interview
3422: and month of death but the probability to survive from last
3423: interview up to one month before death multiplied by the
3424: probability to die within a month. Thanks to Chris
3425: Jackson for correcting this bug. Former versions increased
3426: mortality artificially. The bad side is that we add another loop
3427: which slows down the processing. The difference can be up to 10%
3428: lower mortality.
3429: */
3430: /* If, at the beginning of the maximization mostly, the
3431: cumulative probability or probability to be dead is
3432: constant (ie = 1) over time d, the difference is equal to
3433: 0. out[s1][3] = savm[s1][3]: probability, being at state
3434: s1 at precedent wave, to be dead a month before current
3435: wave is equal to probability, being at state s1 at
3436: precedent wave, to be dead at mont of the current
3437: wave. Then the observed probability (that this person died)
3438: is null according to current estimated parameter. In fact,
3439: it should be very low but not zero otherwise the log go to
3440: infinity.
3441: */
1.183 brouard 3442: /* #ifdef INFINITYORIGINAL */
3443: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3444: /* #else */
3445: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3446: /* lli=log(mytinydouble); */
3447: /* else */
3448: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3449: /* #endif */
1.226 brouard 3450: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3451:
1.226 brouard 3452: } else if ( s2==-1 ) { /* alive */
3453: for (j=1,survp=0. ; j<=nlstate; j++)
3454: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3455: /*survp += out[s1][j]; */
3456: lli= log(survp);
3457: }
3458: else if (s2==-4) {
3459: for (j=3,survp=0. ; j<=nlstate; j++)
3460: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3461: lli= log(survp);
3462: }
3463: else if (s2==-5) {
3464: for (j=1,survp=0. ; j<=2; j++)
3465: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3466: lli= log(survp);
3467: }
3468: else{
3469: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3470: /* 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 */
3471: }
3472: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3473: /*if(lli ==000.0)*/
3474: /*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); */
3475: ipmx +=1;
3476: sw += weight[i];
3477: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3478: /* if (lli < log(mytinydouble)){ */
3479: /* 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); */
3480: /* 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]); */
3481: /* } */
3482: } /* end of wave */
3483: } /* end of individual */
3484: } else if(mle==2){
3485: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3486: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3487: for(mi=1; mi<= wav[i]-1; mi++){
3488: for (ii=1;ii<=nlstate+ndeath;ii++)
3489: for (j=1;j<=nlstate+ndeath;j++){
3490: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3491: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3492: }
3493: for(d=0; d<=dh[mi][i]; d++){
3494: newm=savm;
3495: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3496: cov[2]=agexact;
3497: if(nagesqr==1)
3498: cov[3]= agexact*agexact;
3499: for (kk=1; kk<=cptcovage;kk++) {
3500: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3501: }
3502: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3503: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3504: savm=oldm;
3505: oldm=newm;
3506: } /* end mult */
3507:
3508: s1=s[mw[mi][i]][i];
3509: s2=s[mw[mi+1][i]][i];
3510: bbh=(double)bh[mi][i]/(double)stepm;
3511: 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 */
3512: ipmx +=1;
3513: sw += weight[i];
3514: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3515: } /* end of wave */
3516: } /* end of individual */
3517: } else if(mle==3){ /* exponential inter-extrapolation */
3518: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3519: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3520: for(mi=1; mi<= wav[i]-1; mi++){
3521: for (ii=1;ii<=nlstate+ndeath;ii++)
3522: for (j=1;j<=nlstate+ndeath;j++){
3523: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3524: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3525: }
3526: for(d=0; d<dh[mi][i]; d++){
3527: newm=savm;
3528: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3529: cov[2]=agexact;
3530: if(nagesqr==1)
3531: cov[3]= agexact*agexact;
3532: for (kk=1; kk<=cptcovage;kk++) {
3533: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3534: }
3535: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3536: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3537: savm=oldm;
3538: oldm=newm;
3539: } /* end mult */
3540:
3541: s1=s[mw[mi][i]][i];
3542: s2=s[mw[mi+1][i]][i];
3543: bbh=(double)bh[mi][i]/(double)stepm;
3544: 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 */
3545: ipmx +=1;
3546: sw += weight[i];
3547: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3548: } /* end of wave */
3549: } /* end of individual */
3550: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3551: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3552: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3553: for(mi=1; mi<= wav[i]-1; mi++){
3554: for (ii=1;ii<=nlstate+ndeath;ii++)
3555: for (j=1;j<=nlstate+ndeath;j++){
3556: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3557: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3558: }
3559: for(d=0; d<dh[mi][i]; d++){
3560: newm=savm;
3561: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3562: cov[2]=agexact;
3563: if(nagesqr==1)
3564: cov[3]= agexact*agexact;
3565: for (kk=1; kk<=cptcovage;kk++) {
3566: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3567: }
1.126 brouard 3568:
1.226 brouard 3569: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3570: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3571: savm=oldm;
3572: oldm=newm;
3573: } /* end mult */
3574:
3575: s1=s[mw[mi][i]][i];
3576: s2=s[mw[mi+1][i]][i];
3577: if( s2 > nlstate){
3578: lli=log(out[s1][s2] - savm[s1][s2]);
3579: } else if ( s2==-1 ) { /* alive */
3580: for (j=1,survp=0. ; j<=nlstate; j++)
3581: survp += out[s1][j];
3582: lli= log(survp);
3583: }else{
3584: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3585: }
3586: ipmx +=1;
3587: sw += weight[i];
3588: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3589: /* 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 3590: } /* end of wave */
3591: } /* end of individual */
3592: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3593: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3594: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3595: for(mi=1; mi<= wav[i]-1; mi++){
3596: for (ii=1;ii<=nlstate+ndeath;ii++)
3597: for (j=1;j<=nlstate+ndeath;j++){
3598: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3599: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3600: }
3601: for(d=0; d<dh[mi][i]; d++){
3602: newm=savm;
3603: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3604: cov[2]=agexact;
3605: if(nagesqr==1)
3606: cov[3]= agexact*agexact;
3607: for (kk=1; kk<=cptcovage;kk++) {
3608: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3609: }
1.126 brouard 3610:
1.226 brouard 3611: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3612: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3613: savm=oldm;
3614: oldm=newm;
3615: } /* end mult */
3616:
3617: s1=s[mw[mi][i]][i];
3618: s2=s[mw[mi+1][i]][i];
3619: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3620: ipmx +=1;
3621: sw += weight[i];
3622: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3623: /*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]);*/
3624: } /* end of wave */
3625: } /* end of individual */
3626: } /* End of if */
3627: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3628: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3629: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3630: return -l;
1.126 brouard 3631: }
3632:
3633: /*************** log-likelihood *************/
3634: double funcone( double *x)
3635: {
1.228 brouard 3636: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3637: int i, ii, j, k, mi, d, kk;
1.228 brouard 3638: int ioffset=0;
1.131 brouard 3639: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3640: double **out;
3641: double lli; /* Individual log likelihood */
3642: double llt;
3643: int s1, s2;
1.228 brouard 3644: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3645:
1.126 brouard 3646: double bbh, survp;
1.187 brouard 3647: double agexact;
1.214 brouard 3648: double agebegin, ageend;
1.126 brouard 3649: /*extern weight */
3650: /* We are differentiating ll according to initial status */
3651: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3652: /*for(i=1;i<imx;i++)
3653: printf(" %d\n",s[4][i]);
3654: */
3655: cov[1]=1.;
3656:
3657: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3658: ioffset=0;
3659: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3660: /* ioffset=2+nagesqr+cptcovage; */
3661: ioffset=2+nagesqr;
1.232 brouard 3662: /* Fixed */
1.224 brouard 3663: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3664: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
3665: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3666: 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)*/
3667: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3668: /* cov[2+6]=covar[Tvar[6]][i]; */
3669: /* cov[2+6]=covar[2][i]; V2 */
3670: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3671: /* cov[2+7]=covar[Tvar[7]][i]; */
3672: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3673: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3674: /* cov[2+9]=covar[Tvar[9]][i]; */
3675: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3676: }
1.232 brouard 3677: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3678: /* 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?)*\/ */
3679: /* } */
1.231 brouard 3680: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3681: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3682: /* } */
1.225 brouard 3683:
1.233 brouard 3684:
3685: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3686: /* Wave varying (but not age varying) */
3687: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3688: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3689: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3690: }
1.232 brouard 3691: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3692: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3693: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3694: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3695: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3696: /* 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 3697: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3698: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3699: /* /\* 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]); *\/ */
3700: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3701: /* } */
1.126 brouard 3702: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3703: for (j=1;j<=nlstate+ndeath;j++){
3704: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3705: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3706: }
1.214 brouard 3707:
3708: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3709: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3710: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3711: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3712: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3713: and mw[mi+1][i]. dh depends on stepm.*/
3714: newm=savm;
1.247 brouard 3715: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3716: cov[2]=agexact;
3717: if(nagesqr==1)
3718: cov[3]= agexact*agexact;
3719: for (kk=1; kk<=cptcovage;kk++) {
3720: if(!FixedV[Tvar[Tage[kk]]])
3721: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3722: else
3723: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3724: }
3725: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3726: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3727: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3728: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3729: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3730: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3731: savm=oldm;
3732: oldm=newm;
1.126 brouard 3733: } /* end mult */
3734:
3735: s1=s[mw[mi][i]][i];
3736: s2=s[mw[mi+1][i]][i];
1.217 brouard 3737: /* if(s2==-1){ */
3738: /* printf(" s1=%d, s2=%d i=%d \n", s1, s2, i); */
3739: /* /\* exit(1); *\/ */
3740: /* } */
1.126 brouard 3741: bbh=(double)bh[mi][i]/(double)stepm;
3742: /* bias is positive if real duration
3743: * is higher than the multiple of stepm and negative otherwise.
3744: */
3745: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3746: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3747: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3748: for (j=1,survp=0. ; j<=nlstate; j++)
3749: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3750: lli= log(survp);
1.126 brouard 3751: }else if (mle==1){
1.242 brouard 3752: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3753: } else if(mle==2){
1.242 brouard 3754: 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 3755: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3756: 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 3757: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3758: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3759: } else{ /* mle=0 back to 1 */
1.242 brouard 3760: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3761: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3762: } /* End of if */
3763: ipmx +=1;
3764: sw += weight[i];
3765: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3766: /*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 3767: if(globpr){
1.246 brouard 3768: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3769: %11.6f %11.6f %11.6f ", \
1.242 brouard 3770: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
3771: 2*weight[i]*lli,out[s1][s2],savm[s1][s2]);
3772: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3773: llt +=ll[k]*gipmx/gsw;
3774: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3775: }
3776: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3777: }
1.232 brouard 3778: } /* end of wave */
3779: } /* end of individual */
3780: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3781: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3782: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3783: if(globpr==0){ /* First time we count the contributions and weights */
3784: gipmx=ipmx;
3785: gsw=sw;
3786: }
3787: return -l;
1.126 brouard 3788: }
3789:
3790:
3791: /*************** function likelione ***********/
3792: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*funcone)(double []))
3793: {
3794: /* This routine should help understanding what is done with
3795: the selection of individuals/waves and
3796: to check the exact contribution to the likelihood.
3797: Plotting could be done.
3798: */
3799: int k;
3800:
3801: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 3802: strcpy(fileresilk,"ILK_");
1.202 brouard 3803: strcat(fileresilk,fileresu);
1.126 brouard 3804: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
3805: printf("Problem with resultfile: %s\n", fileresilk);
3806: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
3807: }
1.214 brouard 3808: 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");
3809: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 3810: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
3811: for(k=1; k<=nlstate; k++)
3812: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
3813: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
3814: }
3815:
3816: *fretone=(*funcone)(p);
3817: if(*globpri !=0){
3818: fclose(ficresilk);
1.205 brouard 3819: if (mle ==0)
3820: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
3821: else if(mle >=1)
3822: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
3823: 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 3824:
1.208 brouard 3825:
3826: for (k=1; k<= nlstate ; k++) {
1.211 brouard 3827: 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 3828: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
3829: }
1.207 brouard 3830: 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 3831: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3832: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 3833: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 3834: fflush(fichtm);
1.205 brouard 3835: }
1.126 brouard 3836: return;
3837: }
3838:
3839:
3840: /*********** Maximum Likelihood Estimation ***************/
3841:
3842: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
3843: {
1.165 brouard 3844: int i,j, iter=0;
1.126 brouard 3845: double **xi;
3846: double fret;
3847: double fretone; /* Only one call to likelihood */
3848: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 3849:
3850: #ifdef NLOPT
3851: int creturn;
3852: nlopt_opt opt;
3853: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
3854: double *lb;
3855: double minf; /* the minimum objective value, upon return */
3856: double * p1; /* Shifted parameters from 0 instead of 1 */
3857: myfunc_data dinst, *d = &dinst;
3858: #endif
3859:
3860:
1.126 brouard 3861: xi=matrix(1,npar,1,npar);
3862: for (i=1;i<=npar;i++)
3863: for (j=1;j<=npar;j++)
3864: xi[i][j]=(i==j ? 1.0 : 0.0);
3865: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 3866: strcpy(filerespow,"POW_");
1.126 brouard 3867: strcat(filerespow,fileres);
3868: if((ficrespow=fopen(filerespow,"w"))==NULL) {
3869: printf("Problem with resultfile: %s\n", filerespow);
3870: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
3871: }
3872: fprintf(ficrespow,"# Powell\n# iter -2*LL");
3873: for (i=1;i<=nlstate;i++)
3874: for(j=1;j<=nlstate+ndeath;j++)
3875: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
3876: fprintf(ficrespow,"\n");
1.162 brouard 3877: #ifdef POWELL
1.126 brouard 3878: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 3879: #endif
1.126 brouard 3880:
1.162 brouard 3881: #ifdef NLOPT
3882: #ifdef NEWUOA
3883: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
3884: #else
3885: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
3886: #endif
3887: lb=vector(0,npar-1);
3888: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
3889: nlopt_set_lower_bounds(opt, lb);
3890: nlopt_set_initial_step1(opt, 0.1);
3891:
3892: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
3893: d->function = func;
3894: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
3895: nlopt_set_min_objective(opt, myfunc, d);
3896: nlopt_set_xtol_rel(opt, ftol);
3897: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
3898: printf("nlopt failed! %d\n",creturn);
3899: }
3900: else {
3901: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
3902: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
3903: iter=1; /* not equal */
3904: }
3905: nlopt_destroy(opt);
3906: #endif
1.126 brouard 3907: free_matrix(xi,1,npar,1,npar);
3908: fclose(ficrespow);
1.203 brouard 3909: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
3910: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 3911: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 3912:
3913: }
3914:
3915: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 3916: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 3917: {
3918: double **a,**y,*x,pd;
1.203 brouard 3919: /* double **hess; */
1.164 brouard 3920: int i, j;
1.126 brouard 3921: int *indx;
3922:
3923: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 3924: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 3925: void lubksb(double **a, int npar, int *indx, double b[]) ;
3926: void ludcmp(double **a, int npar, int *indx, double *d) ;
3927: double gompertz(double p[]);
1.203 brouard 3928: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 3929:
3930: printf("\nCalculation of the hessian matrix. Wait...\n");
3931: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
3932: for (i=1;i<=npar;i++){
1.203 brouard 3933: printf("%d-",i);fflush(stdout);
3934: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 3935:
3936: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
3937:
3938: /* printf(" %f ",p[i]);
3939: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
3940: }
3941:
3942: for (i=1;i<=npar;i++) {
3943: for (j=1;j<=npar;j++) {
3944: if (j>i) {
1.203 brouard 3945: printf(".%d-%d",i,j);fflush(stdout);
3946: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
3947: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 3948:
3949: hess[j][i]=hess[i][j];
3950: /*printf(" %lf ",hess[i][j]);*/
3951: }
3952: }
3953: }
3954: printf("\n");
3955: fprintf(ficlog,"\n");
3956:
3957: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
3958: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
3959:
3960: a=matrix(1,npar,1,npar);
3961: y=matrix(1,npar,1,npar);
3962: x=vector(1,npar);
3963: indx=ivector(1,npar);
3964: for (i=1;i<=npar;i++)
3965: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
3966: ludcmp(a,npar,indx,&pd);
3967:
3968: for (j=1;j<=npar;j++) {
3969: for (i=1;i<=npar;i++) x[i]=0;
3970: x[j]=1;
3971: lubksb(a,npar,indx,x);
3972: for (i=1;i<=npar;i++){
3973: matcov[i][j]=x[i];
3974: }
3975: }
3976:
3977: printf("\n#Hessian matrix#\n");
3978: fprintf(ficlog,"\n#Hessian matrix#\n");
3979: for (i=1;i<=npar;i++) {
3980: for (j=1;j<=npar;j++) {
1.203 brouard 3981: printf("%.6e ",hess[i][j]);
3982: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 3983: }
3984: printf("\n");
3985: fprintf(ficlog,"\n");
3986: }
3987:
1.203 brouard 3988: /* printf("\n#Covariance matrix#\n"); */
3989: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
3990: /* for (i=1;i<=npar;i++) { */
3991: /* for (j=1;j<=npar;j++) { */
3992: /* printf("%.6e ",matcov[i][j]); */
3993: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
3994: /* } */
3995: /* printf("\n"); */
3996: /* fprintf(ficlog,"\n"); */
3997: /* } */
3998:
1.126 brouard 3999: /* Recompute Inverse */
1.203 brouard 4000: /* for (i=1;i<=npar;i++) */
4001: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4002: /* ludcmp(a,npar,indx,&pd); */
4003:
4004: /* printf("\n#Hessian matrix recomputed#\n"); */
4005:
4006: /* for (j=1;j<=npar;j++) { */
4007: /* for (i=1;i<=npar;i++) x[i]=0; */
4008: /* x[j]=1; */
4009: /* lubksb(a,npar,indx,x); */
4010: /* for (i=1;i<=npar;i++){ */
4011: /* y[i][j]=x[i]; */
4012: /* printf("%.3e ",y[i][j]); */
4013: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4014: /* } */
4015: /* printf("\n"); */
4016: /* fprintf(ficlog,"\n"); */
4017: /* } */
4018:
4019: /* Verifying the inverse matrix */
4020: #ifdef DEBUGHESS
4021: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4022:
1.203 brouard 4023: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4024: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4025:
4026: for (j=1;j<=npar;j++) {
4027: for (i=1;i<=npar;i++){
1.203 brouard 4028: printf("%.2f ",y[i][j]);
4029: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4030: }
4031: printf("\n");
4032: fprintf(ficlog,"\n");
4033: }
1.203 brouard 4034: #endif
1.126 brouard 4035:
4036: free_matrix(a,1,npar,1,npar);
4037: free_matrix(y,1,npar,1,npar);
4038: free_vector(x,1,npar);
4039: free_ivector(indx,1,npar);
1.203 brouard 4040: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4041:
4042:
4043: }
4044:
4045: /*************** hessian matrix ****************/
4046: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4047: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4048: int i;
4049: int l=1, lmax=20;
1.203 brouard 4050: double k1,k2, res, fx;
1.132 brouard 4051: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4052: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4053: int k=0,kmax=10;
4054: double l1;
4055:
4056: fx=func(x);
4057: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4058: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4059: l1=pow(10,l);
4060: delts=delt;
4061: for(k=1 ; k <kmax; k=k+1){
4062: delt = delta*(l1*k);
4063: p2[theta]=x[theta] +delt;
1.145 brouard 4064: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4065: p2[theta]=x[theta]-delt;
4066: k2=func(p2)-fx;
4067: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4068: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4069:
1.203 brouard 4070: #ifdef DEBUGHESSII
1.126 brouard 4071: 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);
4072: 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);
4073: #endif
4074: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4075: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4076: k=kmax;
4077: }
4078: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4079: k=kmax; l=lmax*10;
1.126 brouard 4080: }
4081: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4082: delts=delt;
4083: }
1.203 brouard 4084: } /* End loop k */
1.126 brouard 4085: }
4086: delti[theta]=delts;
4087: return res;
4088:
4089: }
4090:
1.203 brouard 4091: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4092: {
4093: int i;
1.164 brouard 4094: int l=1, lmax=20;
1.126 brouard 4095: double k1,k2,k3,k4,res,fx;
1.132 brouard 4096: double p2[MAXPARM+1];
1.203 brouard 4097: int k, kmax=1;
4098: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4099:
4100: int firstime=0;
1.203 brouard 4101:
1.126 brouard 4102: fx=func(x);
1.203 brouard 4103: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4104: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4105: p2[thetai]=x[thetai]+delti[thetai]*k;
4106: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4107: k1=func(p2)-fx;
4108:
1.203 brouard 4109: p2[thetai]=x[thetai]+delti[thetai]*k;
4110: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4111: k2=func(p2)-fx;
4112:
1.203 brouard 4113: p2[thetai]=x[thetai]-delti[thetai]*k;
4114: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4115: k3=func(p2)-fx;
4116:
1.203 brouard 4117: p2[thetai]=x[thetai]-delti[thetai]*k;
4118: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4119: k4=func(p2)-fx;
1.203 brouard 4120: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4121: if(k1*k2*k3*k4 <0.){
1.208 brouard 4122: firstime=1;
1.203 brouard 4123: kmax=kmax+10;
1.208 brouard 4124: }
4125: if(kmax >=10 || firstime ==1){
1.246 brouard 4126: 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);
4127: 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 4128: 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);
4129: 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);
4130: }
4131: #ifdef DEBUGHESSIJ
4132: v1=hess[thetai][thetai];
4133: v2=hess[thetaj][thetaj];
4134: cv12=res;
4135: /* Computing eigen value of Hessian matrix */
4136: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4137: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4138: if ((lc2 <0) || (lc1 <0) ){
4139: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4140: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4141: 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);
4142: 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);
4143: }
1.126 brouard 4144: #endif
4145: }
4146: return res;
4147: }
4148:
1.203 brouard 4149: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4150: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4151: /* { */
4152: /* int i; */
4153: /* int l=1, lmax=20; */
4154: /* double k1,k2,k3,k4,res,fx; */
4155: /* double p2[MAXPARM+1]; */
4156: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4157: /* int k=0,kmax=10; */
4158: /* double l1; */
4159:
4160: /* fx=func(x); */
4161: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4162: /* l1=pow(10,l); */
4163: /* delts=delt; */
4164: /* for(k=1 ; k <kmax; k=k+1){ */
4165: /* delt = delti*(l1*k); */
4166: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4167: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4168: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4169: /* k1=func(p2)-fx; */
4170:
4171: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4172: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4173: /* k2=func(p2)-fx; */
4174:
4175: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4176: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4177: /* k3=func(p2)-fx; */
4178:
4179: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4180: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4181: /* k4=func(p2)-fx; */
4182: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4183: /* #ifdef DEBUGHESSIJ */
4184: /* 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); */
4185: /* 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); */
4186: /* #endif */
4187: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4188: /* k=kmax; */
4189: /* } */
4190: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4191: /* k=kmax; l=lmax*10; */
4192: /* } */
4193: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4194: /* delts=delt; */
4195: /* } */
4196: /* } /\* End loop k *\/ */
4197: /* } */
4198: /* delti[theta]=delts; */
4199: /* return res; */
4200: /* } */
4201:
4202:
1.126 brouard 4203: /************** Inverse of matrix **************/
4204: void ludcmp(double **a, int n, int *indx, double *d)
4205: {
4206: int i,imax,j,k;
4207: double big,dum,sum,temp;
4208: double *vv;
4209:
4210: vv=vector(1,n);
4211: *d=1.0;
4212: for (i=1;i<=n;i++) {
4213: big=0.0;
4214: for (j=1;j<=n;j++)
4215: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4216: if (big == 0.0){
4217: printf(" Singular Hessian matrix at row %d:\n",i);
4218: for (j=1;j<=n;j++) {
4219: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4220: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4221: }
4222: fflush(ficlog);
4223: fclose(ficlog);
4224: nrerror("Singular matrix in routine ludcmp");
4225: }
1.126 brouard 4226: vv[i]=1.0/big;
4227: }
4228: for (j=1;j<=n;j++) {
4229: for (i=1;i<j;i++) {
4230: sum=a[i][j];
4231: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4232: a[i][j]=sum;
4233: }
4234: big=0.0;
4235: for (i=j;i<=n;i++) {
4236: sum=a[i][j];
4237: for (k=1;k<j;k++)
4238: sum -= a[i][k]*a[k][j];
4239: a[i][j]=sum;
4240: if ( (dum=vv[i]*fabs(sum)) >= big) {
4241: big=dum;
4242: imax=i;
4243: }
4244: }
4245: if (j != imax) {
4246: for (k=1;k<=n;k++) {
4247: dum=a[imax][k];
4248: a[imax][k]=a[j][k];
4249: a[j][k]=dum;
4250: }
4251: *d = -(*d);
4252: vv[imax]=vv[j];
4253: }
4254: indx[j]=imax;
4255: if (a[j][j] == 0.0) a[j][j]=TINY;
4256: if (j != n) {
4257: dum=1.0/(a[j][j]);
4258: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4259: }
4260: }
4261: free_vector(vv,1,n); /* Doesn't work */
4262: ;
4263: }
4264:
4265: void lubksb(double **a, int n, int *indx, double b[])
4266: {
4267: int i,ii=0,ip,j;
4268: double sum;
4269:
4270: for (i=1;i<=n;i++) {
4271: ip=indx[i];
4272: sum=b[ip];
4273: b[ip]=b[i];
4274: if (ii)
4275: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4276: else if (sum) ii=i;
4277: b[i]=sum;
4278: }
4279: for (i=n;i>=1;i--) {
4280: sum=b[i];
4281: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4282: b[i]=sum/a[i][i];
4283: }
4284: }
4285:
4286: void pstamp(FILE *fichier)
4287: {
1.196 brouard 4288: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4289: }
4290:
1.253 brouard 4291: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
4292:
4293: /* y=a+bx regression */
4294: double sumx = 0.0; /* sum of x */
4295: double sumx2 = 0.0; /* sum of x**2 */
4296: double sumxy = 0.0; /* sum of x * y */
4297: double sumy = 0.0; /* sum of y */
4298: double sumy2 = 0.0; /* sum of y**2 */
4299: double sume2; /* sum of square or residuals */
4300: double yhat;
4301:
4302: double denom=0;
4303: int i;
4304: int ne=*no;
4305:
4306: for ( i=ifi, ne=0;i<=ila;i++) {
4307: if(!isfinite(x[i]) || !isfinite(y[i])){
4308: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4309: continue;
4310: }
4311: ne=ne+1;
4312: sumx += x[i];
4313: sumx2 += x[i]*x[i];
4314: sumxy += x[i] * y[i];
4315: sumy += y[i];
4316: sumy2 += y[i]*y[i];
4317: denom = (ne * sumx2 - sumx*sumx);
4318: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4319: }
4320:
4321: denom = (ne * sumx2 - sumx*sumx);
4322: if (denom == 0) {
4323: // vertical, slope m is infinity
4324: *b = INFINITY;
4325: *a = 0;
4326: if (r) *r = 0;
4327: return 1;
4328: }
4329:
4330: *b = (ne * sumxy - sumx * sumy) / denom;
4331: *a = (sumy * sumx2 - sumx * sumxy) / denom;
4332: if (r!=NULL) {
4333: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
4334: sqrt((sumx2 - sumx*sumx/ne) *
4335: (sumy2 - sumy*sumy/ne));
4336: }
4337: *no=ne;
4338: for ( i=ifi, ne=0;i<=ila;i++) {
4339: if(!isfinite(x[i]) || !isfinite(y[i])){
4340: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
4341: continue;
4342: }
4343: ne=ne+1;
4344: yhat = y[i] - *a -*b* x[i];
4345: sume2 += yhat * yhat ;
4346:
4347: denom = (ne * sumx2 - sumx*sumx);
4348: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
4349: }
4350: *sb = sqrt(sume2/(ne-2)/(sumx2 - sumx * sumx /ne));
4351: *sa= *sb * sqrt(sumx2/ne);
4352:
4353: return 0;
4354: }
4355:
1.126 brouard 4356: /************ Frequencies ********************/
1.251 brouard 4357: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4358: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4359: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4360: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4361:
1.265 brouard 4362: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4363: int iind=0, iage=0;
4364: int mi; /* Effective wave */
4365: int first;
4366: double ***freq; /* Frequencies */
1.253 brouard 4367: double *x, *y, a,b,r, sa, sb; /* for regression, y=b+m*x and r is the correlation coefficient */
4368: int no;
1.226 brouard 4369: double *meanq;
4370: double **meanqt;
4371: double *pp, **prop, *posprop, *pospropt;
4372: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4373: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4374: double agebegin, ageend;
4375:
4376: pp=vector(1,nlstate);
1.251 brouard 4377: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4378: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4379: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4380: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4381: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
4382: meanqt=matrix(1,lastpass,1,nqtveff);
4383: strcpy(fileresp,"P_");
4384: strcat(fileresp,fileresu);
4385: /*strcat(fileresphtm,fileresu);*/
4386: if((ficresp=fopen(fileresp,"w"))==NULL) {
4387: printf("Problem with prevalence resultfile: %s\n", fileresp);
4388: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4389: exit(0);
4390: }
1.240 brouard 4391:
1.226 brouard 4392: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4393: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4394: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4395: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4396: fflush(ficlog);
4397: exit(70);
4398: }
4399: else{
4400: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4401: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4402: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4403: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4404: }
1.237 brouard 4405: 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 4406:
1.226 brouard 4407: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4408: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4409: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4410: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4411: fflush(ficlog);
4412: exit(70);
1.240 brouard 4413: } else{
1.226 brouard 4414: 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 4415: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4416: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4417: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4418: }
1.240 brouard 4419: 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);
4420:
1.253 brouard 4421: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4422: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4423: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4424: j1=0;
1.126 brouard 4425:
1.227 brouard 4426: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4427: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4428: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4429:
4430:
1.226 brouard 4431: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4432: reference=low_education V1=0,V2=0
4433: med_educ V1=1 V2=0,
4434: high_educ V1=0 V2=1
4435: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4436: */
1.249 brouard 4437: dateintsum=0;
4438: k2cpt=0;
4439:
1.253 brouard 4440: if(cptcoveff == 0 )
1.265 brouard 4441: nl=1; /* Constant and age model only */
1.253 brouard 4442: else
4443: nl=2;
1.265 brouard 4444:
4445: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4446: /* Loop on nj=1 or 2 if dummy covariates j!=0
4447: * Loop on j1(1 to 2**cptcoveff) covariate combination
4448: * freq[s1][s2][iage] =0.
4449: * Loop on iind
4450: * ++freq[s1][s2][iage] weighted
4451: * end iind
4452: * if covariate and j!0
4453: * headers Variable on one line
4454: * endif cov j!=0
4455: * header of frequency table by age
4456: * Loop on age
4457: * pp[s1]+=freq[s1][s2][iage] weighted
4458: * pos+=freq[s1][s2][iage] weighted
4459: * Loop on s1 initial state
4460: * fprintf(ficresp
4461: * end s1
4462: * end age
4463: * if j!=0 computes starting values
4464: * end compute starting values
4465: * end j1
4466: * end nl
4467: */
1.253 brouard 4468: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4469: if(nj==1)
4470: j=0; /* First pass for the constant */
1.265 brouard 4471: else{
1.253 brouard 4472: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4473: }
1.251 brouard 4474: first=1;
1.265 brouard 4475: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all covariates combination of the model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 4476: posproptt=0.;
4477: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4478: scanf("%d", i);*/
4479: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4480: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4481: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4482: freq[i][s2][m]=0;
1.251 brouard 4483:
4484: for (i=1; i<=nlstate; i++) {
1.240 brouard 4485: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4486: prop[i][m]=0;
4487: posprop[i]=0;
4488: pospropt[i]=0;
4489: }
4490: /* for (z1=1; z1<= nqfveff; z1++) { */
4491: /* meanq[z1]+=0.; */
4492: /* for(m=1;m<=lastpass;m++){ */
4493: /* meanqt[m][z1]=0.; */
4494: /* } */
4495: /* } */
4496:
4497: /* dateintsum=0; */
4498: /* k2cpt=0; */
4499:
1.265 brouard 4500: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4501: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4502: bool=1;
4503: if(j !=0){
4504: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4505: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4506: /* for (z1=1; z1<= nqfveff; z1++) { */
4507: /* meanq[z1]+=coqvar[Tvar[z1]][iind]; /\* Computes mean of quantitative with selected filter *\/ */
4508: /* } */
4509: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4510: /* if(Tvaraff[z1] ==-20){ */
4511: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4512: /* }else if(Tvaraff[z1] ==-10){ */
4513: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4514: /* }else */
4515: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4516: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4517: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4518: /* 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",
4519: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4520: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4521: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4522: } /* Onlyf fixed */
4523: } /* end z1 */
4524: } /* cptcovn > 0 */
4525: } /* end any */
4526: }/* end j==0 */
1.265 brouard 4527: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4528: /* for(m=firstpass; m<=lastpass; m++){ */
4529: for(mi=1; mi<wav[iind];mi++){ /* For that wave */
4530: m=mw[mi][iind];
4531: if(j!=0){
4532: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4533: for (z1=1; z1<=cptcoveff; z1++) {
4534: if( Fixed[Tmodelind[z1]]==1){
4535: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4536: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4537: value is -1, we don't select. It differs from the
4538: constant and age model which counts them. */
4539: bool=0; /* not selected */
4540: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4541: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4542: bool=0;
4543: }
4544: }
4545: }
4546: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4547: } /* end j==0 */
4548: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
4549: if(bool==1){
4550: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4551: and mw[mi+1][iind]. dh depends on stepm. */
4552: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4553: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4554: if(m >=firstpass && m <=lastpass){
4555: k2=anint[m][iind]+(mint[m][iind]/12.);
4556: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4557: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4558: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4559: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4560: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4561: if (m<lastpass) {
4562: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4563: /* 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]); */
4564: if(s[m][iind]==-1)
4565: 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.));
4566: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4567: /* if((int)agev[m][iind] == 55) */
4568: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4569: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4570: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 4571: }
1.251 brouard 4572: } /* end if between passes */
4573: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4574: dateintsum=dateintsum+k2; /* on all covariates ?*/
4575: k2cpt++;
4576: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4577: }
1.251 brouard 4578: }else{
4579: bool=1;
4580: }/* end bool 2 */
4581: } /* end m */
4582: } /* end bool */
4583: } /* end iind = 1 to imx */
4584: /* prop[s][age] is feeded for any initial and valid live state as well as
4585: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4586:
4587:
4588: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4589: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4590: pstamp(ficresp);
1.251 brouard 4591: if (cptcoveff>0 && j!=0){
1.265 brouard 4592: pstamp(ficresp);
1.251 brouard 4593: printf( "\n#********** Variable ");
4594: fprintf(ficresp, "\n#********** Variable ");
4595: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4596: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4597: fprintf(ficlog, "\n#********** Variable ");
4598: for (z1=1; z1<=cptcoveff; z1++){
4599: if(!FixedV[Tvaraff[z1]]){
4600: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4601: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4602: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4603: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4604: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4605: }else{
1.251 brouard 4606: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4607: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4608: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4609: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4610: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4611: }
4612: }
4613: printf( "**********\n#");
4614: fprintf(ficresp, "**********\n#");
4615: fprintf(ficresphtm, "**********</h3>\n");
4616: fprintf(ficresphtmfr, "**********</h3>\n");
4617: fprintf(ficlog, "**********\n");
4618: }
4619: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4620: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4621: fprintf(ficresp, " Age");
4622: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 4623: for(i=1; i<=nlstate;i++) {
1.265 brouard 4624: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4625: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4626: }
1.265 brouard 4627: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4628: fprintf(ficresphtm, "\n");
4629:
4630: /* Header of frequency table by age */
4631: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4632: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4633: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4634: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4635: if(s2!=0 && m!=0)
4636: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4637: }
1.226 brouard 4638: }
1.251 brouard 4639: fprintf(ficresphtmfr, "\n");
4640:
4641: /* For each age */
4642: for(iage=iagemin; iage <= iagemax+3; iage++){
4643: fprintf(ficresphtm,"<tr>");
4644: if(iage==iagemax+1){
4645: fprintf(ficlog,"1");
4646: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4647: }else if(iage==iagemax+2){
4648: fprintf(ficlog,"0");
4649: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4650: }else if(iage==iagemax+3){
4651: fprintf(ficlog,"Total");
4652: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4653: }else{
1.240 brouard 4654: if(first==1){
1.251 brouard 4655: first=0;
4656: printf("See log file for details...\n");
4657: }
4658: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4659: fprintf(ficlog,"Age %d", iage);
4660: }
1.265 brouard 4661: for(s1=1; s1 <=nlstate ; s1++){
4662: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4663: pp[s1] += freq[s1][m][iage];
1.251 brouard 4664: }
1.265 brouard 4665: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4666: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4667: pos += freq[s1][m][iage];
4668: if(pp[s1]>=1.e-10){
1.251 brouard 4669: if(first==1){
1.265 brouard 4670: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4671: }
1.265 brouard 4672: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4673: }else{
4674: if(first==1)
1.265 brouard 4675: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4676: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4677: }
4678: }
4679:
1.265 brouard 4680: for(s1=1; s1 <=nlstate ; s1++){
4681: /* posprop[s1]=0; */
4682: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4683: pp[s1] += freq[s1][m][iage];
4684: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4685:
4686: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4687: pos += pp[s1]; /* pos is the total number of transitions until this age */
4688: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4689: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4690: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4691: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4692: }
4693:
4694: /* Writing ficresp */
4695: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4696: if( iage <= iagemax){
4697: fprintf(ficresp," %d",iage);
4698: }
4699: }else if( nj==2){
4700: if( iage <= iagemax){
4701: fprintf(ficresp," %d",iage);
4702: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4703: }
1.240 brouard 4704: }
1.265 brouard 4705: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4706: if(pos>=1.e-5){
1.251 brouard 4707: if(first==1)
1.265 brouard 4708: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4709: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4710: }else{
4711: if(first==1)
1.265 brouard 4712: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4713: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4714: }
4715: if( iage <= iagemax){
4716: if(pos>=1.e-5){
1.265 brouard 4717: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4718: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4719: }else if( nj==2){
4720: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4721: }
4722: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4723: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4724: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4725: } else{
4726: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4727: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4728: }
1.240 brouard 4729: }
1.265 brouard 4730: pospropt[s1] +=posprop[s1];
4731: } /* end loop s1 */
1.251 brouard 4732: /* pospropt=0.; */
1.265 brouard 4733: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4734: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4735: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4736: if(first==1){
1.265 brouard 4737: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4738: }
1.265 brouard 4739: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4740: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4741: }
1.265 brouard 4742: if(s1!=0 && m!=0)
4743: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4744: }
1.265 brouard 4745: } /* end loop s1 */
1.251 brouard 4746: posproptt=0.;
1.265 brouard 4747: for(s1=1; s1 <=nlstate; s1++){
4748: posproptt += pospropt[s1];
1.251 brouard 4749: }
4750: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4751: fprintf(ficresphtm,"</tr>\n");
4752: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4753: if(iage <= iagemax)
4754: fprintf(ficresp,"\n");
1.240 brouard 4755: }
1.251 brouard 4756: if(first==1)
4757: printf("Others in log...\n");
4758: fprintf(ficlog,"\n");
4759: } /* end loop age iage */
1.265 brouard 4760:
1.251 brouard 4761: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4762: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4763: if(posproptt < 1.e-5){
1.265 brouard 4764: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4765: }else{
1.265 brouard 4766: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4767: }
1.226 brouard 4768: }
1.251 brouard 4769: fprintf(ficresphtm,"</tr>\n");
4770: fprintf(ficresphtm,"</table>\n");
4771: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4772: if(posproptt < 1.e-5){
1.251 brouard 4773: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4774: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4775: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4776: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4777: invalidvarcomb[j1]=1;
1.226 brouard 4778: }else{
1.251 brouard 4779: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4780: invalidvarcomb[j1]=0;
1.226 brouard 4781: }
1.251 brouard 4782: fprintf(ficresphtmfr,"</table>\n");
4783: fprintf(ficlog,"\n");
4784: if(j!=0){
4785: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4786: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4787: for(k=1; k <=(nlstate+ndeath); k++){
4788: if (k != i) {
1.265 brouard 4789: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4790: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4791: if(j1==1){ /* All dummy covariates to zero */
4792: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4793: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4794: printf("%d%d ",i,k);
4795: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4796: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
4797: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4798: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4799: }
1.253 brouard 4800: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4801: for(iage=iagemin; iage <= iagemax+3; iage++){
4802: x[iage]= (double)iage;
4803: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4804: /* printf("i=%d, k=%d, s1=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,s1,j1,jj, iage, y[iage]); */
1.253 brouard 4805: }
4806: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 4807: pstart[s1]=b;
4808: pstart[s1-1]=a;
1.252 brouard 4809: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
4810: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
4811: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.265 brouard 4812: pstart[s1]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 4813: printf("%d%d ",i,k);
4814: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4815: printf("s1=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",s1,i,k,s1,p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
1.251 brouard 4816: }else{ /* Other cases, like quantitative fixed or varying covariates */
4817: ;
4818: }
4819: /* printf("%12.7f )", param[i][jj][k]); */
4820: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4821: s1++;
1.251 brouard 4822: } /* end jj */
4823: } /* end k!= i */
4824: } /* end k */
1.265 brouard 4825: } /* end i, s1 */
1.251 brouard 4826: } /* end j !=0 */
4827: } /* end selected combination of covariate j1 */
4828: if(j==0){ /* We can estimate starting values from the occurences in each case */
4829: printf("#Freqsummary: Starting values for the constants:\n");
4830: fprintf(ficlog,"\n");
1.265 brouard 4831: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4832: for(k=1; k <=(nlstate+ndeath); k++){
4833: if (k != i) {
4834: printf("%d%d ",i,k);
4835: fprintf(ficlog,"%d%d ",i,k);
4836: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 4837: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 4838: if(jj==1){ /* Age has to be done */
1.265 brouard 4839: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
4840: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
4841: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
1.251 brouard 4842: }
4843: /* printf("%12.7f )", param[i][jj][k]); */
4844: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 4845: s1++;
1.250 brouard 4846: }
1.251 brouard 4847: printf("\n");
4848: fprintf(ficlog,"\n");
1.250 brouard 4849: }
4850: }
4851: }
1.251 brouard 4852: printf("#Freqsummary\n");
4853: fprintf(ficlog,"\n");
1.265 brouard 4854: for(s1=-1; s1 <=nlstate+ndeath; s1++){
4855: for(s2=-1; s2 <=nlstate+ndeath; s2++){
4856: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
4857: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4858: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
4859: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
4860: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
4861: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 4862: /* } */
4863: }
1.265 brouard 4864: } /* end loop s1 */
1.251 brouard 4865:
4866: printf("\n");
4867: fprintf(ficlog,"\n");
4868: } /* end j=0 */
1.249 brouard 4869: } /* end j */
1.252 brouard 4870:
1.253 brouard 4871: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 4872: for(i=1, jk=1; i <=nlstate; i++){
4873: for(j=1; j <=nlstate+ndeath; j++){
4874: if(j!=i){
4875: /*ca[0]= k+'a'-1;ca[1]='\0';*/
4876: printf("%1d%1d",i,j);
4877: fprintf(ficparo,"%1d%1d",i,j);
4878: for(k=1; k<=ncovmodel;k++){
4879: /* printf(" %lf",param[i][j][k]); */
4880: /* fprintf(ficparo," %lf",param[i][j][k]); */
4881: p[jk]=pstart[jk];
4882: printf(" %f ",pstart[jk]);
4883: fprintf(ficparo," %f ",pstart[jk]);
4884: jk++;
4885: }
4886: printf("\n");
4887: fprintf(ficparo,"\n");
4888: }
4889: }
4890: }
4891: } /* end mle=-2 */
1.226 brouard 4892: dateintmean=dateintsum/k2cpt;
1.240 brouard 4893:
1.226 brouard 4894: fclose(ficresp);
4895: fclose(ficresphtm);
4896: fclose(ficresphtmfr);
4897: free_vector(meanq,1,nqfveff);
4898: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 4899: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
4900: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 4901: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4902: free_vector(pospropt,1,nlstate);
4903: free_vector(posprop,1,nlstate);
1.251 brouard 4904: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 4905: free_vector(pp,1,nlstate);
4906: /* End of freqsummary */
4907: }
1.126 brouard 4908:
4909: /************ Prevalence ********************/
1.227 brouard 4910: 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)
4911: {
4912: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
4913: in each health status at the date of interview (if between dateprev1 and dateprev2).
4914: We still use firstpass and lastpass as another selection.
4915: */
1.126 brouard 4916:
1.227 brouard 4917: int i, m, jk, j1, bool, z1,j, iv;
4918: int mi; /* Effective wave */
4919: int iage;
4920: double agebegin, ageend;
4921:
4922: double **prop;
4923: double posprop;
4924: double y2; /* in fractional years */
4925: int iagemin, iagemax;
4926: int first; /** to stop verbosity which is redirected to log file */
4927:
4928: iagemin= (int) agemin;
4929: iagemax= (int) agemax;
4930: /*pp=vector(1,nlstate);*/
1.251 brouard 4931: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 4932: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
4933: j1=0;
1.222 brouard 4934:
1.227 brouard 4935: /*j=cptcoveff;*/
4936: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 4937:
1.227 brouard 4938: first=1;
4939: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
4940: for (i=1; i<=nlstate; i++)
1.251 brouard 4941: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 4942: prop[i][iage]=0.0;
4943: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
4944: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
4945: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
4946:
4947: for (i=1; i<=imx; i++) { /* Each individual */
4948: bool=1;
4949: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
4950: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
4951: m=mw[mi][i];
4952: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
4953: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
4954: for (z1=1; z1<=cptcoveff; z1++){
4955: if( Fixed[Tmodelind[z1]]==1){
4956: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4957: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
4958: bool=0;
4959: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
4960: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4961: bool=0;
4962: }
4963: }
4964: if(bool==1){ /* Otherwise we skip that wave/person */
4965: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
4966: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
4967: if(m >=firstpass && m <=lastpass){
4968: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
4969: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
4970: if(agev[m][i]==0) agev[m][i]=iagemax+1;
4971: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 4972: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 4973: 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);
4974: exit(1);
4975: }
4976: if (s[m][i]>0 && s[m][i]<=nlstate) {
4977: /*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]]);*/
4978: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
4979: prop[s[m][i]][iagemax+3] += weight[i];
4980: } /* end valid statuses */
4981: } /* end selection of dates */
4982: } /* end selection of waves */
4983: } /* end bool */
4984: } /* end wave */
4985: } /* end individual */
4986: for(i=iagemin; i <= iagemax+3; i++){
4987: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
4988: posprop += prop[jk][i];
4989: }
4990:
4991: for(jk=1; jk <=nlstate ; jk++){
4992: if( i <= iagemax){
4993: if(posprop>=1.e-5){
4994: probs[i][jk][j1]= prop[jk][i]/posprop;
4995: } else{
4996: if(first==1){
4997: first=0;
1.266 brouard 4998: printf("Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
4999: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
5000: }else{
5001: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 5002: }
5003: }
5004: }
5005: }/* end jk */
5006: }/* end i */
1.222 brouard 5007: /*} *//* end i1 */
1.227 brouard 5008: } /* end j1 */
1.222 brouard 5009:
1.227 brouard 5010: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5011: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5012: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5013: } /* End of prevalence */
1.126 brouard 5014:
5015: /************* Waves Concatenation ***************/
5016:
5017: 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)
5018: {
5019: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
5020: Death is a valid wave (if date is known).
5021: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5022: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
5023: and mw[mi+1][i]. dh depends on stepm.
1.227 brouard 5024: */
1.126 brouard 5025:
1.224 brouard 5026: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5027: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5028: double sum=0., jmean=0.;*/
1.224 brouard 5029: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5030: int j, k=0,jk, ju, jl;
5031: double sum=0.;
5032: first=0;
1.214 brouard 5033: firstwo=0;
1.217 brouard 5034: firsthree=0;
1.218 brouard 5035: firstfour=0;
1.164 brouard 5036: jmin=100000;
1.126 brouard 5037: jmax=-1;
5038: jmean=0.;
1.224 brouard 5039:
5040: /* Treating live states */
1.214 brouard 5041: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5042: mi=0; /* First valid wave */
1.227 brouard 5043: mli=0; /* Last valid wave */
1.126 brouard 5044: m=firstpass;
1.214 brouard 5045: while(s[m][i] <= nlstate){ /* a live state */
1.227 brouard 5046: 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 */
5047: mli=m-1;/* mw[++mi][i]=m-1; */
5048: }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 */
5049: mw[++mi][i]=m;
5050: mli=m;
1.224 brouard 5051: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5052: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5053: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5054: }
1.227 brouard 5055: else{ /* m >= lastpass, eventual special issue with warning */
1.224 brouard 5056: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5057: break;
1.224 brouard 5058: #else
1.227 brouard 5059: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){
5060: if(firsthree == 0){
1.262 brouard 5061: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p%d%d .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 5062: firsthree=1;
5063: }
1.262 brouard 5064: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p%d%d .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 5065: mw[++mi][i]=m;
5066: mli=m;
5067: }
5068: if(s[m][i]==-2){ /* Vital status is really unknown */
5069: nbwarn++;
5070: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified? */
5071: 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);
5072: 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);
5073: }
5074: break;
5075: }
5076: break;
1.224 brouard 5077: #endif
1.227 brouard 5078: }/* End m >= lastpass */
1.126 brouard 5079: }/* end while */
1.224 brouard 5080:
1.227 brouard 5081: /* 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 5082: /* After last pass */
1.224 brouard 5083: /* Treating death states */
1.214 brouard 5084: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5085: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5086: /* } */
1.126 brouard 5087: mi++; /* Death is another wave */
5088: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5089: /* Only death is a correct wave */
1.126 brouard 5090: mw[mi][i]=m;
1.257 brouard 5091: } /* else not in a death state */
1.224 brouard 5092: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5093: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5094: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.227 brouard 5095: 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 */
5096: nbwarn++;
5097: if(firstfiv==0){
5098: 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 );
5099: firstfiv=1;
5100: }else{
5101: 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 );
5102: }
5103: }else{ /* Death occured afer last wave potential bias */
5104: nberr++;
5105: if(firstwo==0){
1.257 brouard 5106: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.227 brouard 5107: firstwo=1;
5108: }
1.257 brouard 5109: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictive wave at the date of last vital status scan, with a dead status or alive but unknown state status (-1). See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], i,m );
1.227 brouard 5110: }
1.257 brouard 5111: }else{ /* if date of interview is unknown */
1.227 brouard 5112: /* death is known but not confirmed by death status at any wave */
5113: if(firstfour==0){
5114: 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 );
5115: firstfour=1;
5116: }
5117: 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 5118: }
1.224 brouard 5119: } /* end if date of death is known */
5120: #endif
5121: wav[i]=mi; /* mi should be the last effective wave (or mli) */
5122: /* wav[i]=mw[mi][i]; */
1.126 brouard 5123: if(mi==0){
5124: nbwarn++;
5125: if(first==0){
1.227 brouard 5126: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5127: first=1;
1.126 brouard 5128: }
5129: if(first==1){
1.227 brouard 5130: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5131: }
5132: } /* end mi==0 */
5133: } /* End individuals */
1.214 brouard 5134: /* wav and mw are no more changed */
1.223 brouard 5135:
1.214 brouard 5136:
1.126 brouard 5137: for(i=1; i<=imx; i++){
5138: for(mi=1; mi<wav[i];mi++){
5139: if (stepm <=0)
1.227 brouard 5140: dh[mi][i]=1;
1.126 brouard 5141: else{
1.260 brouard 5142: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5143: if (agedc[i] < 2*AGESUP) {
5144: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5145: if(j==0) j=1; /* Survives at least one month after exam */
5146: else if(j<0){
5147: nberr++;
5148: 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]);
5149: j=1; /* Temporary Dangerous patch */
5150: 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);
5151: 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]);
5152: 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);
5153: }
5154: k=k+1;
5155: if (j >= jmax){
5156: jmax=j;
5157: ijmax=i;
5158: }
5159: if (j <= jmin){
5160: jmin=j;
5161: ijmin=i;
5162: }
5163: sum=sum+j;
5164: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5165: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5166: }
5167: }
5168: else{
5169: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5170: /* 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 5171:
1.227 brouard 5172: k=k+1;
5173: if (j >= jmax) {
5174: jmax=j;
5175: ijmax=i;
5176: }
5177: else if (j <= jmin){
5178: jmin=j;
5179: ijmin=i;
5180: }
5181: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5182: /*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]);*/
5183: if(j<0){
5184: nberr++;
5185: 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]);
5186: 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]);
5187: }
5188: sum=sum+j;
5189: }
5190: jk= j/stepm;
5191: jl= j -jk*stepm;
5192: ju= j -(jk+1)*stepm;
5193: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5194: if(jl==0){
5195: dh[mi][i]=jk;
5196: bh[mi][i]=0;
5197: }else{ /* We want a negative bias in order to only have interpolation ie
5198: * to avoid the price of an extra matrix product in likelihood */
5199: dh[mi][i]=jk+1;
5200: bh[mi][i]=ju;
5201: }
5202: }else{
5203: if(jl <= -ju){
5204: dh[mi][i]=jk;
5205: bh[mi][i]=jl; /* bias is positive if real duration
5206: * is higher than the multiple of stepm and negative otherwise.
5207: */
5208: }
5209: else{
5210: dh[mi][i]=jk+1;
5211: bh[mi][i]=ju;
5212: }
5213: if(dh[mi][i]==0){
5214: dh[mi][i]=1; /* At least one step */
5215: bh[mi][i]=ju; /* At least one step */
5216: /* 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);*/
5217: }
5218: } /* end if mle */
1.126 brouard 5219: }
5220: } /* end wave */
5221: }
5222: jmean=sum/k;
5223: 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 5224: 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 5225: }
1.126 brouard 5226:
5227: /*********** Tricode ****************************/
1.220 brouard 5228: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5229: {
5230: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5231: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5232: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5233: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5234: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5235: */
1.130 brouard 5236:
1.242 brouard 5237: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5238: int modmaxcovj=0; /* Modality max of covariates j */
5239: int cptcode=0; /* Modality max of covariates j */
5240: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5241:
5242:
1.242 brouard 5243: /* cptcoveff=0; */
5244: /* *cptcov=0; */
1.126 brouard 5245:
1.242 brouard 5246: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.126 brouard 5247:
1.242 brouard 5248: /* Loop on covariates without age and products and no quantitative variable */
5249: /* for (j=1; j<=(cptcovs); j++) { /\* From model V1 + V2*age+ V3 + V3*V4 keeps V1 + V3 = 2 only *\/ */
5250: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5251: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5252: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5253: switch(Fixed[k]) {
5254: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
5255: 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*/
5256: ij=(int)(covar[Tvar[k]][i]);
5257: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5258: * If product of Vn*Vm, still boolean *:
5259: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5260: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5261: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5262: modality of the nth covariate of individual i. */
5263: if (ij > modmaxcovj)
5264: modmaxcovj=ij;
5265: else if (ij < modmincovj)
5266: modmincovj=ij;
5267: if ((ij < -1) && (ij > NCOVMAX)){
5268: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5269: exit(1);
5270: }else
5271: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5272: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5273: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5274: /* getting the maximum value of the modality of the covariate
5275: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5276: female ies 1, then modmaxcovj=1.
5277: */
5278: } /* end for loop on individuals i */
5279: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5280: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5281: cptcode=modmaxcovj;
5282: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5283: /*for (i=0; i<=cptcode; i++) {*/
5284: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5285: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5286: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5287: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5288: if( j != -1){
5289: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5290: covariate for which somebody answered excluding
5291: undefined. Usually 2: 0 and 1. */
5292: }
5293: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5294: covariate for which somebody answered including
5295: undefined. Usually 3: -1, 0 and 1. */
5296: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5297: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5298: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5299:
1.242 brouard 5300: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5301: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5302: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5303: /* modmincovj=3; modmaxcovj = 7; */
5304: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5305: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5306: /* defining two dummy variables: variables V1_1 and V1_2.*/
5307: /* nbcode[Tvar[j]][ij]=k; */
5308: /* nbcode[Tvar[j]][1]=0; */
5309: /* nbcode[Tvar[j]][2]=1; */
5310: /* nbcode[Tvar[j]][3]=2; */
5311: /* To be continued (not working yet). */
5312: ij=0; /* ij is similar to i but can jump over null modalities */
5313: 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*/
5314: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5315: break;
5316: }
5317: ij++;
5318: 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*/
5319: cptcode = ij; /* New max modality for covar j */
5320: } /* end of loop on modality i=-1 to 1 or more */
5321: break;
5322: case 1: /* Testing on varying covariate, could be simple and
5323: * should look at waves or product of fixed *
5324: * varying. No time to test -1, assuming 0 and 1 only */
5325: ij=0;
5326: for(i=0; i<=1;i++){
5327: nbcode[Tvar[k]][++ij]=i;
5328: }
5329: break;
5330: default:
5331: break;
5332: } /* end switch */
5333: } /* end dummy test */
5334:
5335: /* for (k=0; k<= cptcode; k++) { /\* k=-1 ? k=0 to 1 *\//\* Could be 1 to 4 *\//\* cptcode=modmaxcovj *\/ */
5336: /* /\*recode from 0 *\/ */
5337: /* k is a modality. If we have model=V1+V1*sex */
5338: /* then: nbcode[1][1]=0 ; nbcode[1][2]=1; nbcode[2][1]=0 ; nbcode[2][2]=1; */
5339: /* But if some modality were not used, it is recoded from 0 to a newer modmaxcovj=cptcode *\/ */
5340: /* } */
5341: /* /\* cptcode = ij; *\/ /\* New max modality for covar j *\/ */
5342: /* if (ij > ncodemax[j]) { */
5343: /* printf( " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5344: /* fprintf(ficlog, " Error ij=%d > ncodemax[%d]=%d\n", ij, j, ncodemax[j]); */
5345: /* break; */
5346: /* } */
5347: /* } /\* end of loop on modality k *\/ */
5348: } /* end of loop on model-covariate j. nbcode[Tvarj][1]=0 and nbcode[Tvarj][2]=1 sets the value of covariate j*/
5349:
5350: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5351: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5352: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5353: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5354: 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 */
5355: 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 */
5356: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5357: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5358:
5359: ij=0;
5360: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5361: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5362: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5363: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5364: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5365: /* If product not in single variable we don't print results */
5366: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5367: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5368: 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*/
5369: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5370: 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 */
5371: if(Fixed[k]!=0)
5372: anyvaryingduminmodel=1;
5373: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5374: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5375: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5376: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5377: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5378: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5379: }
5380: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5381: /* ij--; */
5382: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5383: *cptcov=ij; /*Number of total real effective covariates: effective
5384: * because they can be excluded from the model and real
5385: * if in the model but excluded because missing values, but how to get k from ij?*/
5386: for(j=ij+1; j<= cptcovt; j++){
5387: Tvaraff[j]=0;
5388: Tmodelind[j]=0;
5389: }
5390: for(j=ntveff+1; j<= cptcovt; j++){
5391: TmodelInvind[j]=0;
5392: }
5393: /* To be sorted */
5394: ;
5395: }
1.126 brouard 5396:
1.145 brouard 5397:
1.126 brouard 5398: /*********** Health Expectancies ****************/
5399:
1.235 brouard 5400: 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 5401:
5402: {
5403: /* Health expectancies, no variances */
1.164 brouard 5404: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5405: int nhstepma, nstepma; /* Decreasing with age */
5406: double age, agelim, hf;
5407: double ***p3mat;
5408: double eip;
5409:
1.238 brouard 5410: /* pstamp(ficreseij); */
1.126 brouard 5411: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5412: fprintf(ficreseij,"# Age");
5413: for(i=1; i<=nlstate;i++){
5414: for(j=1; j<=nlstate;j++){
5415: fprintf(ficreseij," e%1d%1d ",i,j);
5416: }
5417: fprintf(ficreseij," e%1d. ",i);
5418: }
5419: fprintf(ficreseij,"\n");
5420:
5421:
5422: if(estepm < stepm){
5423: printf ("Problem %d lower than %d\n",estepm, stepm);
5424: }
5425: else hstepm=estepm;
5426: /* We compute the life expectancy from trapezoids spaced every estepm months
5427: * This is mainly to measure the difference between two models: for example
5428: * if stepm=24 months pijx are given only every 2 years and by summing them
5429: * we are calculating an estimate of the Life Expectancy assuming a linear
5430: * progression in between and thus overestimating or underestimating according
5431: * to the curvature of the survival function. If, for the same date, we
5432: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5433: * to compare the new estimate of Life expectancy with the same linear
5434: * hypothesis. A more precise result, taking into account a more precise
5435: * curvature will be obtained if estepm is as small as stepm. */
5436:
5437: /* For example we decided to compute the life expectancy with the smallest unit */
5438: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5439: nhstepm is the number of hstepm from age to agelim
5440: nstepm is the number of stepm from age to agelin.
5441: Look at hpijx to understand the reason of that which relies in memory size
5442: and note for a fixed period like estepm months */
5443: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5444: survival function given by stepm (the optimization length). Unfortunately it
5445: means that if the survival funtion is printed only each two years of age and if
5446: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5447: results. So we changed our mind and took the option of the best precision.
5448: */
5449: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5450:
5451: agelim=AGESUP;
5452: /* If stepm=6 months */
5453: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5454: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5455:
5456: /* nhstepm age range expressed in number of stepm */
5457: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5458: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5459: /* if (stepm >= YEARM) hstepm=1;*/
5460: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5461: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5462:
5463: for (age=bage; age<=fage; age ++){
5464: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5465: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5466: /* if (stepm >= YEARM) hstepm=1;*/
5467: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5468:
5469: /* If stepm=6 months */
5470: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5471: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5472:
1.235 brouard 5473: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5474:
5475: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5476:
5477: printf("%d|",(int)age);fflush(stdout);
5478: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5479:
5480: /* Computing expectancies */
5481: for(i=1; i<=nlstate;i++)
5482: for(j=1; j<=nlstate;j++)
5483: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5484: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5485:
5486: /* 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]);*/
5487:
5488: }
5489:
5490: fprintf(ficreseij,"%3.0f",age );
5491: for(i=1; i<=nlstate;i++){
5492: eip=0;
5493: for(j=1; j<=nlstate;j++){
5494: eip +=eij[i][j][(int)age];
5495: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5496: }
5497: fprintf(ficreseij,"%9.4f", eip );
5498: }
5499: fprintf(ficreseij,"\n");
5500:
5501: }
5502: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5503: printf("\n");
5504: fprintf(ficlog,"\n");
5505:
5506: }
5507:
1.235 brouard 5508: 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 5509:
5510: {
5511: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5512: to initial status i, ei. .
1.126 brouard 5513: */
5514: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5515: int nhstepma, nstepma; /* Decreasing with age */
5516: double age, agelim, hf;
5517: double ***p3matp, ***p3matm, ***varhe;
5518: double **dnewm,**doldm;
5519: double *xp, *xm;
5520: double **gp, **gm;
5521: double ***gradg, ***trgradg;
5522: int theta;
5523:
5524: double eip, vip;
5525:
5526: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5527: xp=vector(1,npar);
5528: xm=vector(1,npar);
5529: dnewm=matrix(1,nlstate*nlstate,1,npar);
5530: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5531:
5532: pstamp(ficresstdeij);
5533: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5534: fprintf(ficresstdeij,"# Age");
5535: for(i=1; i<=nlstate;i++){
5536: for(j=1; j<=nlstate;j++)
5537: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5538: fprintf(ficresstdeij," e%1d. ",i);
5539: }
5540: fprintf(ficresstdeij,"\n");
5541:
5542: pstamp(ficrescveij);
5543: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5544: fprintf(ficrescveij,"# Age");
5545: for(i=1; i<=nlstate;i++)
5546: for(j=1; j<=nlstate;j++){
5547: cptj= (j-1)*nlstate+i;
5548: for(i2=1; i2<=nlstate;i2++)
5549: for(j2=1; j2<=nlstate;j2++){
5550: cptj2= (j2-1)*nlstate+i2;
5551: if(cptj2 <= cptj)
5552: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5553: }
5554: }
5555: fprintf(ficrescveij,"\n");
5556:
5557: if(estepm < stepm){
5558: printf ("Problem %d lower than %d\n",estepm, stepm);
5559: }
5560: else hstepm=estepm;
5561: /* We compute the life expectancy from trapezoids spaced every estepm months
5562: * This is mainly to measure the difference between two models: for example
5563: * if stepm=24 months pijx are given only every 2 years and by summing them
5564: * we are calculating an estimate of the Life Expectancy assuming a linear
5565: * progression in between and thus overestimating or underestimating according
5566: * to the curvature of the survival function. If, for the same date, we
5567: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5568: * to compare the new estimate of Life expectancy with the same linear
5569: * hypothesis. A more precise result, taking into account a more precise
5570: * curvature will be obtained if estepm is as small as stepm. */
5571:
5572: /* For example we decided to compute the life expectancy with the smallest unit */
5573: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5574: nhstepm is the number of hstepm from age to agelim
5575: nstepm is the number of stepm from age to agelin.
5576: Look at hpijx to understand the reason of that which relies in memory size
5577: and note for a fixed period like estepm months */
5578: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5579: survival function given by stepm (the optimization length). Unfortunately it
5580: means that if the survival funtion is printed only each two years of age and if
5581: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5582: results. So we changed our mind and took the option of the best precision.
5583: */
5584: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5585:
5586: /* If stepm=6 months */
5587: /* nhstepm age range expressed in number of stepm */
5588: agelim=AGESUP;
5589: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5590: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5591: /* if (stepm >= YEARM) hstepm=1;*/
5592: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5593:
5594: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5595: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5596: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5597: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5598: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5599: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5600:
5601: for (age=bage; age<=fage; age ++){
5602: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5603: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5604: /* if (stepm >= YEARM) hstepm=1;*/
5605: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5606:
1.126 brouard 5607: /* If stepm=6 months */
5608: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5609: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5610:
5611: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5612:
1.126 brouard 5613: /* Computing Variances of health expectancies */
5614: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5615: decrease memory allocation */
5616: for(theta=1; theta <=npar; theta++){
5617: for(i=1; i<=npar; i++){
1.222 brouard 5618: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5619: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5620: }
1.235 brouard 5621: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5622: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5623:
1.126 brouard 5624: for(j=1; j<= nlstate; j++){
1.222 brouard 5625: for(i=1; i<=nlstate; i++){
5626: for(h=0; h<=nhstepm-1; h++){
5627: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5628: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5629: }
5630: }
1.126 brouard 5631: }
1.218 brouard 5632:
1.126 brouard 5633: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5634: for(h=0; h<=nhstepm-1; h++){
5635: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5636: }
1.126 brouard 5637: }/* End theta */
5638:
5639:
5640: for(h=0; h<=nhstepm-1; h++)
5641: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5642: for(theta=1; theta <=npar; theta++)
5643: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5644:
1.218 brouard 5645:
1.222 brouard 5646: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5647: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5648: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5649:
1.222 brouard 5650: printf("%d|",(int)age);fflush(stdout);
5651: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5652: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5653: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5654: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5655: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5656: for(ij=1;ij<=nlstate*nlstate;ij++)
5657: for(ji=1;ji<=nlstate*nlstate;ji++)
5658: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5659: }
5660: }
1.218 brouard 5661:
1.126 brouard 5662: /* Computing expectancies */
1.235 brouard 5663: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5664: for(i=1; i<=nlstate;i++)
5665: for(j=1; j<=nlstate;j++)
1.222 brouard 5666: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5667: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5668:
1.222 brouard 5669: /* 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 5670:
1.222 brouard 5671: }
1.218 brouard 5672:
1.126 brouard 5673: fprintf(ficresstdeij,"%3.0f",age );
5674: for(i=1; i<=nlstate;i++){
5675: eip=0.;
5676: vip=0.;
5677: for(j=1; j<=nlstate;j++){
1.222 brouard 5678: eip += eij[i][j][(int)age];
5679: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5680: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5681: 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 5682: }
5683: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5684: }
5685: fprintf(ficresstdeij,"\n");
1.218 brouard 5686:
1.126 brouard 5687: fprintf(ficrescveij,"%3.0f",age );
5688: for(i=1; i<=nlstate;i++)
5689: for(j=1; j<=nlstate;j++){
1.222 brouard 5690: cptj= (j-1)*nlstate+i;
5691: for(i2=1; i2<=nlstate;i2++)
5692: for(j2=1; j2<=nlstate;j2++){
5693: cptj2= (j2-1)*nlstate+i2;
5694: if(cptj2 <= cptj)
5695: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5696: }
1.126 brouard 5697: }
5698: fprintf(ficrescveij,"\n");
1.218 brouard 5699:
1.126 brouard 5700: }
5701: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5702: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5703: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5704: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5705: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5706: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5707: printf("\n");
5708: fprintf(ficlog,"\n");
1.218 brouard 5709:
1.126 brouard 5710: free_vector(xm,1,npar);
5711: free_vector(xp,1,npar);
5712: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5713: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5714: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5715: }
1.218 brouard 5716:
1.126 brouard 5717: /************ Variance ******************/
1.235 brouard 5718: 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 5719: {
5720: /* Variance of health expectancies */
5721: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);*/
5722: /* double **newm;*/
5723: /* int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)*/
5724:
5725: /* int movingaverage(); */
5726: double **dnewm,**doldm;
5727: double **dnewmp,**doldmp;
5728: int i, j, nhstepm, hstepm, h, nstepm ;
5729: int k;
5730: double *xp;
5731: double **gp, **gm; /* for var eij */
5732: double ***gradg, ***trgradg; /*for var eij */
5733: double **gradgp, **trgradgp; /* for var p point j */
5734: double *gpp, *gmp; /* for var p point j */
5735: double **varppt; /* for var p point j nlstate to nlstate+ndeath */
5736: double ***p3mat;
5737: double age,agelim, hf;
5738: /* double ***mobaverage; */
5739: int theta;
5740: char digit[4];
5741: char digitp[25];
5742:
5743: char fileresprobmorprev[FILENAMELENGTH];
5744:
5745: if(popbased==1){
5746: if(mobilav!=0)
5747: strcpy(digitp,"-POPULBASED-MOBILAV_");
5748: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
5749: }
5750: else
5751: strcpy(digitp,"-STABLBASED_");
1.126 brouard 5752:
1.218 brouard 5753: /* if (mobilav!=0) { */
5754: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
5755: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
5756: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
5757: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
5758: /* } */
5759: /* } */
5760:
5761: strcpy(fileresprobmorprev,"PRMORPREV-");
5762: sprintf(digit,"%-d",ij);
5763: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
5764: strcat(fileresprobmorprev,digit); /* Tvar to be done */
5765: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
5766: strcat(fileresprobmorprev,fileresu);
5767: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
5768: printf("Problem with resultfile: %s\n", fileresprobmorprev);
5769: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
5770: }
5771: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5772: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
5773: pstamp(ficresprobmorprev);
5774: 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 5775: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
5776: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
5777: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
5778: }
5779: for(j=1;j<=cptcoveff;j++)
5780: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
5781: fprintf(ficresprobmorprev,"\n");
5782:
1.218 brouard 5783: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
5784: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5785: fprintf(ficresprobmorprev," p.%-d SE",j);
5786: for(i=1; i<=nlstate;i++)
5787: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
5788: }
5789: fprintf(ficresprobmorprev,"\n");
5790:
5791: fprintf(ficgp,"\n# Routine varevsij");
5792: fprintf(ficgp,"\nunset title \n");
5793: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
5794: 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");
5795: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
5796: /* } */
5797: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5798: pstamp(ficresvij);
5799: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
5800: if(popbased==1)
5801: 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);
5802: else
5803: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
5804: fprintf(ficresvij,"# Age");
5805: for(i=1; i<=nlstate;i++)
5806: for(j=1; j<=nlstate;j++)
5807: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
5808: fprintf(ficresvij,"\n");
5809:
5810: xp=vector(1,npar);
5811: dnewm=matrix(1,nlstate,1,npar);
5812: doldm=matrix(1,nlstate,1,nlstate);
5813: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
5814: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
5815:
5816: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
5817: gpp=vector(nlstate+1,nlstate+ndeath);
5818: gmp=vector(nlstate+1,nlstate+ndeath);
5819: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 5820:
1.218 brouard 5821: if(estepm < stepm){
5822: printf ("Problem %d lower than %d\n",estepm, stepm);
5823: }
5824: else hstepm=estepm;
5825: /* For example we decided to compute the life expectancy with the smallest unit */
5826: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5827: nhstepm is the number of hstepm from age to agelim
5828: nstepm is the number of stepm from age to agelim.
5829: Look at function hpijx to understand why because of memory size limitations,
5830: we decided (b) to get a life expectancy respecting the most precise curvature of the
5831: survival function given by stepm (the optimization length). Unfortunately it
5832: means that if the survival funtion is printed every two years of age and if
5833: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5834: results. So we changed our mind and took the option of the best precision.
5835: */
5836: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5837: agelim = AGESUP;
5838: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
5839: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
5840: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5841: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5842: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
5843: gp=matrix(0,nhstepm,1,nlstate);
5844: gm=matrix(0,nhstepm,1,nlstate);
5845:
5846:
5847: for(theta=1; theta <=npar; theta++){
5848: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
5849: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5850: }
5851:
1.242 brouard 5852: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5853:
5854: if (popbased==1) {
5855: if(mobilav ==0){
5856: for(i=1; i<=nlstate;i++)
5857: prlim[i][i]=probs[(int)age][i][ij];
5858: }else{ /* mobilav */
5859: for(i=1; i<=nlstate;i++)
5860: prlim[i][i]=mobaverage[(int)age][i][ij];
5861: }
5862: }
5863:
1.235 brouard 5864: 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 5865: for(j=1; j<= nlstate; j++){
5866: for(h=0; h<=nhstepm; h++){
5867: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
5868: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
5869: }
5870: }
5871: /* Next for computing probability of death (h=1 means
5872: computed over hstepm matrices product = hstepm*stepm months)
5873: as a weighted average of prlim.
5874: */
5875: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5876: for(i=1,gpp[j]=0.; i<= nlstate; i++)
5877: gpp[j] += prlim[i][i]*p3mat[i][j][1];
5878: }
5879: /* end probability of death */
5880:
5881: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
5882: xp[i] = x[i] - (i==theta ?delti[theta]:0);
5883:
1.242 brouard 5884: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 5885:
5886: if (popbased==1) {
5887: if(mobilav ==0){
5888: for(i=1; i<=nlstate;i++)
5889: prlim[i][i]=probs[(int)age][i][ij];
5890: }else{ /* mobilav */
5891: for(i=1; i<=nlstate;i++)
5892: prlim[i][i]=mobaverage[(int)age][i][ij];
5893: }
5894: }
5895:
1.235 brouard 5896: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 5897:
5898: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
5899: for(h=0; h<=nhstepm; h++){
5900: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
5901: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
5902: }
5903: }
5904: /* This for computing probability of death (h=1 means
5905: computed over hstepm matrices product = hstepm*stepm months)
5906: as a weighted average of prlim.
5907: */
5908: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5909: for(i=1,gmp[j]=0.; i<= nlstate; i++)
5910: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5911: }
5912: /* end probability of death */
5913:
5914: for(j=1; j<= nlstate; j++) /* vareij */
5915: for(h=0; h<=nhstepm; h++){
5916: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
5917: }
5918:
5919: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu */
5920: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
5921: }
5922:
5923: } /* End theta */
5924:
5925: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
5926:
5927: for(h=0; h<=nhstepm; h++) /* veij */
5928: for(j=1; j<=nlstate;j++)
5929: for(theta=1; theta <=npar; theta++)
5930: trgradg[h][j][theta]=gradg[h][theta][j];
5931:
5932: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
5933: for(theta=1; theta <=npar; theta++)
5934: trgradgp[j][theta]=gradgp[theta][j];
5935:
5936:
5937: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5938: for(i=1;i<=nlstate;i++)
5939: for(j=1;j<=nlstate;j++)
5940: vareij[i][j][(int)age] =0.;
5941:
5942: for(h=0;h<=nhstepm;h++){
5943: for(k=0;k<=nhstepm;k++){
5944: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
5945: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
5946: for(i=1;i<=nlstate;i++)
5947: for(j=1;j<=nlstate;j++)
5948: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
5949: }
5950: }
5951:
5952: /* pptj */
5953: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
5954: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
5955: for(j=nlstate+1;j<=nlstate+ndeath;j++)
5956: for(i=nlstate+1;i<=nlstate+ndeath;i++)
5957: varppt[j][i]=doldmp[j][i];
5958: /* end ppptj */
5959: /* x centered again */
5960:
1.242 brouard 5961: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 5962:
5963: if (popbased==1) {
5964: if(mobilav ==0){
5965: for(i=1; i<=nlstate;i++)
5966: prlim[i][i]=probs[(int)age][i][ij];
5967: }else{ /* mobilav */
5968: for(i=1; i<=nlstate;i++)
5969: prlim[i][i]=mobaverage[(int)age][i][ij];
5970: }
5971: }
5972:
5973: /* This for computing probability of death (h=1 means
5974: computed over hstepm (estepm) matrices product = hstepm*stepm months)
5975: as a weighted average of prlim.
5976: */
1.235 brouard 5977: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 5978: for(j=nlstate+1;j<=nlstate+ndeath;j++){
5979: for(i=1,gmp[j]=0.;i<= nlstate; i++)
5980: gmp[j] += prlim[i][i]*p3mat[i][j][1];
5981: }
5982: /* end probability of death */
5983:
5984: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
5985: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
5986: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
5987: for(i=1; i<=nlstate;i++){
5988: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
5989: }
5990: }
5991: fprintf(ficresprobmorprev,"\n");
5992:
5993: fprintf(ficresvij,"%.0f ",age );
5994: for(i=1; i<=nlstate;i++)
5995: for(j=1; j<=nlstate;j++){
5996: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
5997: }
5998: fprintf(ficresvij,"\n");
5999: free_matrix(gp,0,nhstepm,1,nlstate);
6000: free_matrix(gm,0,nhstepm,1,nlstate);
6001: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6002: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6003: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6004: } /* End age */
6005: free_vector(gpp,nlstate+1,nlstate+ndeath);
6006: free_vector(gmp,nlstate+1,nlstate+ndeath);
6007: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6008: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6009: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6010: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6011: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6012: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6013: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6014: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6015: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6016: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6017: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6018: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6019: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6020: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6021: 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);
6022: /* 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 6023: */
1.218 brouard 6024: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6025: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6026:
1.218 brouard 6027: free_vector(xp,1,npar);
6028: free_matrix(doldm,1,nlstate,1,nlstate);
6029: free_matrix(dnewm,1,nlstate,1,npar);
6030: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6031: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6032: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6033: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6034: fclose(ficresprobmorprev);
6035: fflush(ficgp);
6036: fflush(fichtm);
6037: } /* end varevsij */
1.126 brouard 6038:
6039: /************ Variance of prevlim ******************/
1.235 brouard 6040: 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 6041: {
1.205 brouard 6042: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6043: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6044:
1.126 brouard 6045: double **dnewm,**doldm;
6046: int i, j, nhstepm, hstepm;
6047: double *xp;
6048: double *gp, *gm;
6049: double **gradg, **trgradg;
1.208 brouard 6050: double **mgm, **mgp;
1.126 brouard 6051: double age,agelim;
6052: int theta;
6053:
6054: pstamp(ficresvpl);
6055: fprintf(ficresvpl,"# Standard deviation of period (stable) prevalences \n");
1.241 brouard 6056: fprintf(ficresvpl,"# Age ");
6057: if(nresult >=1)
6058: fprintf(ficresvpl," Result# ");
1.126 brouard 6059: for(i=1; i<=nlstate;i++)
6060: fprintf(ficresvpl," %1d-%1d",i,i);
6061: fprintf(ficresvpl,"\n");
6062:
6063: xp=vector(1,npar);
6064: dnewm=matrix(1,nlstate,1,npar);
6065: doldm=matrix(1,nlstate,1,nlstate);
6066:
6067: hstepm=1*YEARM; /* Every year of age */
6068: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6069: agelim = AGESUP;
6070: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6071: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6072: if (stepm >= YEARM) hstepm=1;
6073: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6074: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6075: mgp=matrix(1,npar,1,nlstate);
6076: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6077: gp=vector(1,nlstate);
6078: gm=vector(1,nlstate);
6079:
6080: for(theta=1; theta <=npar; theta++){
6081: for(i=1; i<=npar; i++){ /* Computes gradient */
6082: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6083: }
1.209 brouard 6084: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6085: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6086: else
1.235 brouard 6087: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6088: for(i=1;i<=nlstate;i++){
1.126 brouard 6089: gp[i] = prlim[i][i];
1.208 brouard 6090: mgp[theta][i] = prlim[i][i];
6091: }
1.126 brouard 6092: for(i=1; i<=npar; i++) /* Computes gradient */
6093: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.209 brouard 6094: if((int)age==79 ||(int)age== 80 ||(int)age== 81 )
1.235 brouard 6095: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.209 brouard 6096: else
1.235 brouard 6097: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6098: for(i=1;i<=nlstate;i++){
1.126 brouard 6099: gm[i] = prlim[i][i];
1.208 brouard 6100: mgm[theta][i] = prlim[i][i];
6101: }
1.126 brouard 6102: for(i=1;i<=nlstate;i++)
6103: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6104: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6105: } /* End theta */
6106:
6107: trgradg =matrix(1,nlstate,1,npar);
6108:
6109: for(j=1; j<=nlstate;j++)
6110: for(theta=1; theta <=npar; theta++)
6111: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6112: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6113: /* printf("\nmgm mgp %d ",(int)age); */
6114: /* for(j=1; j<=nlstate;j++){ */
6115: /* printf(" %d ",j); */
6116: /* for(theta=1; theta <=npar; theta++) */
6117: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6118: /* printf("\n "); */
6119: /* } */
6120: /* } */
6121: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6122: /* printf("\n gradg %d ",(int)age); */
6123: /* for(j=1; j<=nlstate;j++){ */
6124: /* printf("%d ",j); */
6125: /* for(theta=1; theta <=npar; theta++) */
6126: /* printf("%d %lf ",theta,gradg[theta][j]); */
6127: /* printf("\n "); */
6128: /* } */
6129: /* } */
1.126 brouard 6130:
6131: for(i=1;i<=nlstate;i++)
6132: varpl[i][(int)age] =0.;
1.209 brouard 6133: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.205 brouard 6134: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6135: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
6136: }else{
1.126 brouard 6137: matprod2(dnewm,trgradg,1,nlstate,1,npar,1,npar,matcov);
6138: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6139: }
1.126 brouard 6140: for(i=1;i<=nlstate;i++)
6141: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6142:
6143: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6144: if(nresult >=1)
6145: fprintf(ficresvpl,"%d ",nres );
1.126 brouard 6146: for(i=1; i<=nlstate;i++)
6147: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
6148: fprintf(ficresvpl,"\n");
6149: free_vector(gp,1,nlstate);
6150: free_vector(gm,1,nlstate);
1.208 brouard 6151: free_matrix(mgm,1,npar,1,nlstate);
6152: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6153: free_matrix(gradg,1,npar,1,nlstate);
6154: free_matrix(trgradg,1,nlstate,1,npar);
6155: } /* End age */
6156:
6157: free_vector(xp,1,npar);
6158: free_matrix(doldm,1,nlstate,1,npar);
6159: free_matrix(dnewm,1,nlstate,1,nlstate);
6160:
6161: }
6162:
6163: /************ Variance of one-step probabilities ******************/
6164: 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 6165: {
6166: int i, j=0, k1, l1, tj;
6167: int k2, l2, j1, z1;
6168: int k=0, l;
6169: int first=1, first1, first2;
6170: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6171: double **dnewm,**doldm;
6172: double *xp;
6173: double *gp, *gm;
6174: double **gradg, **trgradg;
6175: double **mu;
6176: double age, cov[NCOVMAX+1];
6177: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6178: int theta;
6179: char fileresprob[FILENAMELENGTH];
6180: char fileresprobcov[FILENAMELENGTH];
6181: char fileresprobcor[FILENAMELENGTH];
6182: double ***varpij;
6183:
6184: strcpy(fileresprob,"PROB_");
6185: strcat(fileresprob,fileres);
6186: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6187: printf("Problem with resultfile: %s\n", fileresprob);
6188: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6189: }
6190: strcpy(fileresprobcov,"PROBCOV_");
6191: strcat(fileresprobcov,fileresu);
6192: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6193: printf("Problem with resultfile: %s\n", fileresprobcov);
6194: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6195: }
6196: strcpy(fileresprobcor,"PROBCOR_");
6197: strcat(fileresprobcor,fileresu);
6198: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6199: printf("Problem with resultfile: %s\n", fileresprobcor);
6200: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6201: }
6202: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6203: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6204: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6205: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6206: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6207: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6208: pstamp(ficresprob);
6209: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6210: fprintf(ficresprob,"# Age");
6211: pstamp(ficresprobcov);
6212: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6213: fprintf(ficresprobcov,"# Age");
6214: pstamp(ficresprobcor);
6215: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6216: fprintf(ficresprobcor,"# Age");
1.126 brouard 6217:
6218:
1.222 brouard 6219: for(i=1; i<=nlstate;i++)
6220: for(j=1; j<=(nlstate+ndeath);j++){
6221: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6222: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6223: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6224: }
6225: /* fprintf(ficresprob,"\n");
6226: fprintf(ficresprobcov,"\n");
6227: fprintf(ficresprobcor,"\n");
6228: */
6229: xp=vector(1,npar);
6230: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6231: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6232: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6233: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6234: first=1;
6235: fprintf(ficgp,"\n# Routine varprob");
6236: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6237: fprintf(fichtm,"\n");
6238:
1.266 brouard 6239: 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. %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 6240: 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);
6241: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6242: and drawn. It helps understanding how is the covariance between two incidences.\
6243: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6244: 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 6245: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6246: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6247: standard deviations wide on each axis. <br>\
6248: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6249: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6250: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6251:
1.222 brouard 6252: cov[1]=1;
6253: /* tj=cptcoveff; */
1.225 brouard 6254: tj = (int) pow(2,cptcoveff);
1.222 brouard 6255: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6256: j1=0;
1.224 brouard 6257: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6258: if (cptcovn>0) {
6259: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6260: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6261: fprintf(ficresprob, "**********\n#\n");
6262: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6263: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6264: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6265:
1.222 brouard 6266: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6267: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6268: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6269:
6270:
1.222 brouard 6271: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6272: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6273: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6274:
1.222 brouard 6275: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6276: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6277: fprintf(ficresprobcor, "**********\n#");
6278: if(invalidvarcomb[j1]){
6279: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6280: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6281: continue;
6282: }
6283: }
6284: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6285: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6286: gp=vector(1,(nlstate)*(nlstate+ndeath));
6287: gm=vector(1,(nlstate)*(nlstate+ndeath));
6288: for (age=bage; age<=fage; age ++){
6289: cov[2]=age;
6290: if(nagesqr==1)
6291: cov[3]= age*age;
6292: for (k=1; k<=cptcovn;k++) {
6293: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6294: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6295: * 1 1 1 1 1
6296: * 2 2 1 1 1
6297: * 3 1 2 1 1
6298: */
6299: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6300: }
6301: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6302: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6303: for (k=1; k<=cptcovprod;k++)
6304: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6305:
6306:
1.222 brouard 6307: for(theta=1; theta <=npar; theta++){
6308: for(i=1; i<=npar; i++)
6309: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6310:
1.222 brouard 6311: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6312:
1.222 brouard 6313: k=0;
6314: for(i=1; i<= (nlstate); i++){
6315: for(j=1; j<=(nlstate+ndeath);j++){
6316: k=k+1;
6317: gp[k]=pmmij[i][j];
6318: }
6319: }
1.220 brouard 6320:
1.222 brouard 6321: for(i=1; i<=npar; i++)
6322: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6323:
1.222 brouard 6324: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6325: k=0;
6326: for(i=1; i<=(nlstate); i++){
6327: for(j=1; j<=(nlstate+ndeath);j++){
6328: k=k+1;
6329: gm[k]=pmmij[i][j];
6330: }
6331: }
1.220 brouard 6332:
1.222 brouard 6333: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6334: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6335: }
1.126 brouard 6336:
1.222 brouard 6337: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6338: for(theta=1; theta <=npar; theta++)
6339: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6340:
1.222 brouard 6341: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6342: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6343:
1.222 brouard 6344: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6345:
1.222 brouard 6346: k=0;
6347: for(i=1; i<=(nlstate); i++){
6348: for(j=1; j<=(nlstate+ndeath);j++){
6349: k=k+1;
6350: mu[k][(int) age]=pmmij[i][j];
6351: }
6352: }
6353: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6354: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6355: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6356:
1.222 brouard 6357: /*printf("\n%d ",(int)age);
6358: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6359: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6360: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6361: }*/
1.220 brouard 6362:
1.222 brouard 6363: fprintf(ficresprob,"\n%d ",(int)age);
6364: fprintf(ficresprobcov,"\n%d ",(int)age);
6365: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6366:
1.222 brouard 6367: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6368: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6369: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6370: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6371: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6372: }
6373: i=0;
6374: for (k=1; k<=(nlstate);k++){
6375: for (l=1; l<=(nlstate+ndeath);l++){
6376: i++;
6377: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6378: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6379: for (j=1; j<=i;j++){
6380: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6381: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6382: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6383: }
6384: }
6385: }/* end of loop for state */
6386: } /* end of loop for age */
6387: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6388: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6389: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6390: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6391:
6392: /* Confidence intervalle of pij */
6393: /*
6394: fprintf(ficgp,"\nunset parametric;unset label");
6395: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6396: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6397: 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);
6398: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6399: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6400: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6401: */
6402:
6403: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6404: first1=1;first2=2;
6405: for (k2=1; k2<=(nlstate);k2++){
6406: for (l2=1; l2<=(nlstate+ndeath);l2++){
6407: if(l2==k2) continue;
6408: j=(k2-1)*(nlstate+ndeath)+l2;
6409: for (k1=1; k1<=(nlstate);k1++){
6410: for (l1=1; l1<=(nlstate+ndeath);l1++){
6411: if(l1==k1) continue;
6412: i=(k1-1)*(nlstate+ndeath)+l1;
6413: if(i<=j) continue;
6414: for (age=bage; age<=fage; age ++){
6415: if ((int)age %5==0){
6416: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6417: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6418: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6419: mu1=mu[i][(int) age]/stepm*YEARM ;
6420: mu2=mu[j][(int) age]/stepm*YEARM;
6421: c12=cv12/sqrt(v1*v2);
6422: /* Computing eigen value of matrix of covariance */
6423: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6424: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6425: if ((lc2 <0) || (lc1 <0) ){
6426: if(first2==1){
6427: first1=0;
6428: 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);
6429: }
6430: 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);
6431: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6432: /* lc2=fabs(lc2); */
6433: }
1.220 brouard 6434:
1.222 brouard 6435: /* Eigen vectors */
6436: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
6437: /*v21=sqrt(1.-v11*v11); *//* error */
6438: v21=(lc1-v1)/cv12*v11;
6439: v12=-v21;
6440: v22=v11;
6441: tnalp=v21/v11;
6442: if(first1==1){
6443: first1=0;
6444: 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);
6445: }
6446: 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);
6447: /*printf(fignu*/
6448: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6449: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6450: if(first==1){
6451: first=0;
6452: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6453: fprintf(ficgp,"\nset parametric;unset label");
6454: 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);
6455: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6456: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6457: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6458: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6459: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6460: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6461: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6462: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6463: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6464: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6465: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6466: 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", \
1.266 brouard 6467: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6468: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6469: }else{
6470: first=0;
6471: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6472: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6473: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6474: 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", \
1.266 brouard 6475: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6476: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6477: }/* if first */
6478: } /* age mod 5 */
6479: } /* end loop age */
6480: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6481: first=1;
6482: } /*l12 */
6483: } /* k12 */
6484: } /*l1 */
6485: }/* k1 */
6486: } /* loop on combination of covariates j1 */
6487: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6488: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6489: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6490: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6491: free_vector(xp,1,npar);
6492: fclose(ficresprob);
6493: fclose(ficresprobcov);
6494: fclose(ficresprobcor);
6495: fflush(ficgp);
6496: fflush(fichtmcov);
6497: }
1.126 brouard 6498:
6499:
6500: /******************* Printing html file ***********/
1.201 brouard 6501: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6502: int lastpass, int stepm, int weightopt, char model[],\
6503: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.258 brouard 6504: int popforecast, int mobilav, int prevfcast, int mobilavproj, int backcast, int estepm , \
1.213 brouard 6505: double jprev1, double mprev1,double anprev1, double dateprev1, \
6506: double jprev2, double mprev2,double anprev2, double dateprev2){
1.237 brouard 6507: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6508:
6509: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6510: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6511: </ul>");
1.237 brouard 6512: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6513: </ul>", model);
1.214 brouard 6514: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6515: 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",
6516: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6517: 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 6518: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6519: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6520: fprintf(fichtm,"\
6521: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6522: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6523: fprintf(fichtm,"\
1.217 brouard 6524: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6525: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6526: fprintf(fichtm,"\
1.126 brouard 6527: - Period (stable) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6528: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6529: fprintf(fichtm,"\
1.217 brouard 6530: - Period (stable) back prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
6531: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6532: fprintf(fichtm,"\
1.211 brouard 6533: - (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 6534: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6535: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6536: if(prevfcast==1){
6537: fprintf(fichtm,"\
6538: - Prevalence projections by age and states: \
1.201 brouard 6539: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6540: }
1.126 brouard 6541:
6542:
1.225 brouard 6543: m=pow(2,cptcoveff);
1.222 brouard 6544: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6545:
1.264 brouard 6546: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6547:
6548: jj1=0;
6549:
6550: fprintf(fichtm," \n<ul>");
6551: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6552: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6553: if(m != 1 && TKresult[nres]!= k1)
6554: continue;
6555: jj1++;
6556: if (cptcovn > 0) {
6557: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6558: for (cpt=1; cpt<=cptcoveff;cpt++){
6559: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6560: }
6561: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6562: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6563: }
6564: fprintf(fichtm,"\">");
6565:
6566: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
6567: fprintf(fichtm,"************ Results for covariates");
6568: for (cpt=1; cpt<=cptcoveff;cpt++){
6569: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6570: }
6571: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6572: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6573: }
6574: if(invalidvarcomb[k1]){
6575: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
6576: continue;
6577: }
6578: fprintf(fichtm,"</a></li>");
6579: } /* cptcovn >0 */
6580: }
6581: fprintf(fichtm," \n</ul>");
6582:
1.222 brouard 6583: jj1=0;
1.237 brouard 6584:
6585: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 6586: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 6587: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6588: continue;
1.220 brouard 6589:
1.222 brouard 6590: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6591: jj1++;
6592: if (cptcovn > 0) {
1.264 brouard 6593: fprintf(fichtm,"\n<p><a name=\"rescov");
6594: for (cpt=1; cpt<=cptcoveff;cpt++){
6595: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6596: }
6597: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6598: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6599: }
6600: fprintf(fichtm,"\"</a>");
6601:
1.222 brouard 6602: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6603: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 6604: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6605: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
6606: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6607: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 6608: }
1.237 brouard 6609: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6610: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6611: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
6612: }
6613:
1.230 brouard 6614: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 6615: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
6616: if(invalidvarcomb[k1]){
6617: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
6618: printf("\nCombination (%d) ignored because no cases \n",k1);
6619: continue;
6620: }
6621: }
6622: /* aij, bij */
1.259 brouard 6623: fprintf(fichtm,"<br>- Logit model (yours is: logit(pij)=log(pij/pii)= aij+ bij age+%s) as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
1.241 brouard 6624: <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 6625: /* Pij */
1.241 brouard 6626: 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> \
6627: <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 6628: /* Quasi-incidences */
6629: 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 6630: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 6631: 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 6632: 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> \
6633: <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 6634: /* Survival functions (period) in state j */
6635: for(cpt=1; cpt<=nlstate;cpt++){
1.241 brouard 6636: 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> \
6637: <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 6638: }
6639: /* State specific survival functions (period) */
6640: for(cpt=1; cpt<=nlstate;cpt++){
6641: fprintf(fichtm,"<br>\n- Survival functions from state %d in each live state and total.\
1.220 brouard 6642: Or probability to survive in various states (1 to %d) being in state %d at different ages. \
1.241 brouard 6643: <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 6644: }
6645: /* Period (stable) prevalence in each health state */
6646: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6647: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6648: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 6649: }
6650: if(backcast==1){
6651: /* Period (stable) back prevalence in each health state */
6652: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 6653: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 6654: <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 6655: }
1.217 brouard 6656: }
1.222 brouard 6657: if(prevfcast==1){
6658: /* Projection of prevalence up to period (stable) prevalence in each health state */
6659: for(cpt=1; cpt<=nlstate;cpt++){
1.258 brouard 6660: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d) up to period (stable) prevalence in state %d. Or probability to be in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
6661: <img src=\"%s_%d-%d-%d.svg\">", dateprev1, dateprev2, mobilavproj, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 6662: }
6663: }
1.220 brouard 6664:
1.222 brouard 6665: for(cpt=1; cpt<=nlstate;cpt++) {
1.241 brouard 6666: 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> \
6667: <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 6668: }
6669: /* } /\* end i1 *\/ */
6670: }/* End k1 */
6671: fprintf(fichtm,"</ul>");
1.126 brouard 6672:
1.222 brouard 6673: fprintf(fichtm,"\
1.126 brouard 6674: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 6675: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 6676: - 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 6677: But because parameters are usually highly correlated (a higher incidence of disability \
6678: and a higher incidence of recovery can give very close observed transition) it might \
6679: be very useful to look not only at linear confidence intervals estimated from the \
6680: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
6681: (parameters) of the logistic regression, it might be more meaningful to visualize the \
6682: covariance matrix of the one-step probabilities. \
6683: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 6684:
1.222 brouard 6685: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
6686: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
6687: fprintf(fichtm,"\
1.126 brouard 6688: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6689: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 6690:
1.222 brouard 6691: fprintf(fichtm,"\
1.126 brouard 6692: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 6693: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
6694: fprintf(fichtm,"\
1.126 brouard 6695: - 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): \
6696: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6697: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 6698: fprintf(fichtm,"\
1.126 brouard 6699: - (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): \
6700: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 6701: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 6702: fprintf(fichtm,"\
1.128 brouard 6703: - 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 6704: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
6705: fprintf(fichtm,"\
1.128 brouard 6706: - 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 6707: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
6708: fprintf(fichtm,"\
1.126 brouard 6709: - Standard deviation of period (stable) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 6710: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 6711:
6712: /* if(popforecast==1) fprintf(fichtm,"\n */
6713: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
6714: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
6715: /* <br>",fileres,fileres,fileres,fileres); */
6716: /* else */
6717: /* 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 6718: fflush(fichtm);
6719: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 6720:
1.225 brouard 6721: m=pow(2,cptcoveff);
1.222 brouard 6722: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6723:
1.222 brouard 6724: jj1=0;
1.237 brouard 6725:
1.241 brouard 6726: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 6727: for(k1=1; k1<=m;k1++){
1.253 brouard 6728: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 6729: continue;
1.222 brouard 6730: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
6731: jj1++;
1.126 brouard 6732: if (cptcovn > 0) {
6733: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 6734: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 6735: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
6736: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
6737: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6738: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6739: }
6740:
1.126 brouard 6741: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6742:
1.222 brouard 6743: if(invalidvarcomb[k1]){
6744: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
6745: continue;
6746: }
1.126 brouard 6747: }
6748: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 6749: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.241 brouard 6750: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>\n <br>\
1.258 brouard 6751: <img src=\"%s_%d-%d-%d.svg\">",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
1.126 brouard 6752: }
6753: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.128 brouard 6754: health expectancies in states (1) and (2). If popbased=1 the smooth (due to the model) \
6755: true period expectancies (those weighted with period prevalences are also\
6756: drawn in addition to the population based expectancies computed using\
1.241 brouard 6757: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>\n<br>\
6758: <img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 6759: /* } /\* end i1 *\/ */
6760: }/* End k1 */
1.241 brouard 6761: }/* End nres */
1.222 brouard 6762: fprintf(fichtm,"</ul>");
6763: fflush(fichtm);
1.126 brouard 6764: }
6765:
6766: /******************* Gnuplot file **************/
1.266 brouard 6767: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , int prevfcast, int backcast, char pathc[], double p[], int offyear){
1.126 brouard 6768:
6769: char dirfileres[132],optfileres[132];
1.264 brouard 6770: char gplotcondition[132], gplotlabel[132];
1.237 brouard 6771: 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 6772: int lv=0, vlv=0, kl=0;
1.130 brouard 6773: int ng=0;
1.201 brouard 6774: int vpopbased;
1.223 brouard 6775: int ioffset; /* variable offset for columns */
1.235 brouard 6776: int nres=0; /* Index of resultline */
1.266 brouard 6777: int istart=1; /* For starting graphs in projections */
1.219 brouard 6778:
1.126 brouard 6779: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
6780: /* printf("Problem with file %s",optionfilegnuplot); */
6781: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
6782: /* } */
6783:
6784: /*#ifdef windows */
6785: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 6786: /*#endif */
1.225 brouard 6787: m=pow(2,cptcoveff);
1.126 brouard 6788:
1.202 brouard 6789: /* Contribution to likelihood */
6790: /* Plot the probability implied in the likelihood */
1.223 brouard 6791: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
6792: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
6793: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
6794: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 6795: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 6796: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
6797: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 6798: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
6799: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
6800: 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));
6801: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
6802: 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));
6803: for (i=1; i<= nlstate ; i ++) {
6804: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
6805: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
6806: 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);
6807: for (j=2; j<= nlstate+ndeath ; j ++) {
6808: 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);
6809: }
6810: fprintf(ficgp,";\nset out; unset ylabel;\n");
6811: }
6812: /* 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 */
6813: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
6814: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
6815: fprintf(ficgp,"\nset out;unset log\n");
6816: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 6817:
1.126 brouard 6818: strcpy(dirfileres,optionfilefiname);
6819: strcpy(optfileres,"vpl");
1.223 brouard 6820: /* 1eme*/
1.238 brouard 6821: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
6822: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 6823: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 6824: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 6825: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6826: continue;
6827: /* We are interested in selected combination by the resultline */
1.246 brouard 6828: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.238 brouard 6829: fprintf(ficgp,"\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 6830: strcpy(gplotlabel,"(");
1.238 brouard 6831: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
6832: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
6833: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6834: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6835: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6836: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
6837: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 6838: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 6839: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 6840: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 6841: }
6842: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 6843: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 6844: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 6845: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
6846: }
6847: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 6848: /* printf("\n#\n"); */
1.238 brouard 6849: fprintf(ficgp,"\n#\n");
6850: if(invalidvarcomb[k1]){
1.260 brouard 6851: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 6852: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6853: continue;
6854: }
1.235 brouard 6855:
1.241 brouard 6856: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
6857: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.264 brouard 6858: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 6859: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
6860: /* 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); */
6861: /* k1-1 error should be nres-1*/
1.238 brouard 6862: for (i=1; i<= nlstate ; i ++) {
6863: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6864: else fprintf(ficgp," %%*lf (%%*lf)");
6865: }
1.260 brouard 6866: fprintf(ficgp,"\" t\"Period (stable) prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6867: for (i=1; i<= nlstate ; i ++) {
6868: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6869: else fprintf(ficgp," %%*lf (%%*lf)");
6870: }
1.260 brouard 6871: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 6872: for (i=1; i<= nlstate ; i ++) {
6873: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
6874: else fprintf(ficgp," %%*lf (%%*lf)");
6875: }
1.265 brouard 6876: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
6877:
6878: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
6879: if(cptcoveff ==0){
6880: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
6881: }else{
6882: kl=0;
6883: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6884: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6885: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6886: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6887: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6888: vlv= nbcode[Tvaraff[k]][lv];
6889: kl++;
6890: /* 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 *\/ */
6891: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6892: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6893: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6894: if(k==cptcoveff){
6895: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
6896: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
6897: }else{
6898: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6899: kl++;
6900: }
6901: } /* end covariate */
6902: } /* end if no covariate */
6903:
1.238 brouard 6904: if(backcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
6905: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 6906: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 6907: if(cptcoveff ==0){
1.245 brouard 6908: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 6909: }else{
6910: kl=0;
6911: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
6912: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
6913: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6914: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6915: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6916: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 6917: kl++;
1.238 brouard 6918: /* 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 *\/ */
6919: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
6920: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
6921: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
6922: if(k==cptcoveff){
1.245 brouard 6923: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 6924: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 6925: }else{
6926: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
6927: kl++;
6928: }
6929: } /* end covariate */
6930: } /* end if no covariate */
6931: } /* end if backcast */
1.264 brouard 6932: fprintf(ficgp,"\nset out ;unset label;\n");
1.238 brouard 6933: } /* nres */
1.201 brouard 6934: } /* k1 */
6935: } /* cpt */
1.235 brouard 6936:
6937:
1.126 brouard 6938: /*2 eme*/
1.238 brouard 6939: for (k1=1; k1<= m ; k1 ++){
6940: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 6941: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 6942: continue;
6943: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 6944: strcpy(gplotlabel,"(");
1.238 brouard 6945: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 6946: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 6947: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
6948: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
6949: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
6950: vlv= nbcode[Tvaraff[k]][lv];
6951: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 6952: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 6953: }
1.237 brouard 6954: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 6955: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 6956: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 6957: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 6958: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 6959: }
1.264 brouard 6960: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 6961: fprintf(ficgp,"\n#\n");
1.223 brouard 6962: if(invalidvarcomb[k1]){
6963: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
6964: continue;
6965: }
1.219 brouard 6966:
1.241 brouard 6967: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 6968: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 6969: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
6970: if(vpopbased==0){
1.238 brouard 6971: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 6972: }else
1.238 brouard 6973: fprintf(ficgp,"\nreplot ");
6974: for (i=1; i<= nlstate+1 ; i ++) {
6975: k=2*i;
1.261 brouard 6976: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased);
1.238 brouard 6977: for (j=1; j<= nlstate+1 ; j ++) {
6978: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6979: else fprintf(ficgp," %%*lf (%%*lf)");
6980: }
6981: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
6982: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 6983: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 6984: for (j=1; j<= nlstate+1 ; j ++) {
6985: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6986: else fprintf(ficgp," %%*lf (%%*lf)");
6987: }
6988: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 6989: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 6990: for (j=1; j<= nlstate+1 ; j ++) {
6991: if (j==i) fprintf(ficgp," %%lf (%%lf)");
6992: else fprintf(ficgp," %%*lf (%%*lf)");
6993: }
6994: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
6995: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
6996: } /* state */
6997: } /* vpopbased */
1.264 brouard 6998: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; unset label;\n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 6999: } /* end nres */
7000: } /* k1 end 2 eme*/
7001:
7002:
7003: /*3eme*/
7004: for (k1=1; k1<= m ; k1 ++){
7005: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7006: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7007: continue;
7008:
7009: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7010: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7011: strcpy(gplotlabel,"(");
1.238 brouard 7012: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7013: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7014: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7015: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7016: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7017: vlv= nbcode[Tvaraff[k]][lv];
7018: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7019: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7020: }
7021: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7022: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7023: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7024: }
1.264 brouard 7025: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7026: fprintf(ficgp,"\n#\n");
7027: if(invalidvarcomb[k1]){
7028: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7029: continue;
7030: }
7031:
7032: /* k=2+nlstate*(2*cpt-2); */
7033: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7034: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7035: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7036: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7037: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),nres-1,nres-1,k,cpt);
1.238 brouard 7038: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7039: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7040: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7041: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7042: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7043: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7044:
1.238 brouard 7045: */
7046: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7047: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+i,cpt,i+1);
1.238 brouard 7048: /* 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 7049:
1.238 brouard 7050: }
1.261 brouard 7051: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+nlstate,cpt);
1.238 brouard 7052: }
1.264 brouard 7053: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7054: } /* end nres */
7055: } /* end kl 3eme */
1.126 brouard 7056:
1.223 brouard 7057: /* 4eme */
1.201 brouard 7058: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7059: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7060: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7061: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7062: continue;
1.238 brouard 7063: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7064: strcpy(gplotlabel,"(");
1.238 brouard 7065: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7066: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7067: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7068: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7069: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7070: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7071: vlv= nbcode[Tvaraff[k]][lv];
7072: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7073: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7074: }
7075: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7076: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7077: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7078: }
1.264 brouard 7079: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7080: fprintf(ficgp,"\n#\n");
7081: if(invalidvarcomb[k1]){
7082: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7083: continue;
1.223 brouard 7084: }
1.238 brouard 7085:
1.241 brouard 7086: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7087: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 7088: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7089: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7090: k=3;
7091: for (i=1; i<= nlstate ; i ++){
7092: if(i==1){
7093: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7094: }else{
7095: fprintf(ficgp,", '' ");
7096: }
7097: l=(nlstate+ndeath)*(i-1)+1;
7098: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7099: for (j=2; j<= nlstate+ndeath ; j ++)
7100: fprintf(ficgp,"+$%d",k+l+j-1);
7101: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7102: } /* nlstate */
1.264 brouard 7103: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7104: } /* end cpt state*/
7105: } /* end nres */
7106: } /* end covariate k1 */
7107:
1.220 brouard 7108: /* 5eme */
1.201 brouard 7109: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7110: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7111: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7112: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7113: continue;
1.238 brouard 7114: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7115: strcpy(gplotlabel,"(");
1.238 brouard 7116: 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);
7117: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7118: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7119: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7120: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7121: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7122: vlv= nbcode[Tvaraff[k]][lv];
7123: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7124: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7125: }
7126: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7127: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7128: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7129: }
1.264 brouard 7130: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7131: fprintf(ficgp,"\n#\n");
7132: if(invalidvarcomb[k1]){
7133: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7134: continue;
7135: }
1.227 brouard 7136:
1.241 brouard 7137: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7138: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 7139: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7140: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7141: k=3;
7142: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7143: if(j==1)
7144: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7145: else
7146: fprintf(ficgp,", '' ");
7147: l=(nlstate+ndeath)*(cpt-1) +j;
7148: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7149: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7150: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7151: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7152: } /* nlstate */
7153: fprintf(ficgp,", '' ");
7154: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7155: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7156: l=(nlstate+ndeath)*(cpt-1) +j;
7157: if(j < nlstate)
7158: fprintf(ficgp,"$%d +",k+l);
7159: else
7160: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7161: }
1.264 brouard 7162: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7163: } /* end cpt state*/
7164: } /* end covariate */
7165: } /* end nres */
1.227 brouard 7166:
1.220 brouard 7167: /* 6eme */
1.202 brouard 7168: /* CV preval stable (period) for each covariate */
1.237 brouard 7169: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7170: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7171: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7172: continue;
1.255 brouard 7173: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7174: strcpy(gplotlabel,"(");
1.211 brouard 7175: fprintf(ficgp,"\n#\n#\n#CV preval stable (period): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7176: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7177: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7178: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7179: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7180: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7181: vlv= nbcode[Tvaraff[k]][lv];
7182: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7183: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7184: }
1.237 brouard 7185: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7186: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7187: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7188: }
1.264 brouard 7189: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7190: fprintf(ficgp,"\n#\n");
1.223 brouard 7191: if(invalidvarcomb[k1]){
1.227 brouard 7192: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7193: continue;
1.223 brouard 7194: }
1.227 brouard 7195:
1.241 brouard 7196: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7197: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.126 brouard 7198: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7199: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7200: k=3; /* Offset */
1.255 brouard 7201: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7202: if(i==1)
7203: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7204: else
7205: fprintf(ficgp,", '' ");
1.255 brouard 7206: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7207: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7208: for (j=2; j<= nlstate ; j ++)
7209: fprintf(ficgp,"+$%d",k+l+j-1);
7210: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7211: } /* nlstate */
1.264 brouard 7212: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7213: } /* end cpt state*/
7214: } /* end covariate */
1.227 brouard 7215:
7216:
1.220 brouard 7217: /* 7eme */
1.218 brouard 7218: if(backcast == 1){
1.217 brouard 7219: /* CV back preval stable (period) for each covariate */
1.237 brouard 7220: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7221: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7222: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7223: continue;
1.255 brouard 7224: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life ending state */
1.264 brouard 7225: strcpy(gplotlabel,"(");
7226: fprintf(ficgp,"\n#\n#\n#CV Back preval stable (period): 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7227: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7228: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7229: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7230: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7231: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7232: vlv= nbcode[Tvaraff[k]][lv];
7233: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7234: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7235: }
1.237 brouard 7236: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7237: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7238: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7239: }
1.264 brouard 7240: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7241: fprintf(ficgp,"\n#\n");
7242: if(invalidvarcomb[k1]){
7243: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7244: continue;
7245: }
7246:
1.241 brouard 7247: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.264 brouard 7248: fprintf(ficgp,"set label \"Ending alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7249: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7250: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7251: k=3; /* Offset */
1.255 brouard 7252: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7253: if(i==1)
7254: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7255: else
7256: fprintf(ficgp,", '' ");
7257: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7258: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7259: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7260: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
1.255 brouard 7261: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7262: /* for (j=2; j<= nlstate ; j ++) */
7263: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7264: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
7265: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",i,cpt);
7266: } /* nlstate */
1.264 brouard 7267: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7268: } /* end cpt state*/
7269: } /* end covariate */
7270: } /* End if backcast */
7271:
1.223 brouard 7272: /* 8eme */
1.218 brouard 7273: if(prevfcast==1){
7274: /* Projection from cross-sectional to stable (period) for each covariate */
7275:
1.237 brouard 7276: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7277: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7278: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7279: continue;
1.211 brouard 7280: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7281: strcpy(gplotlabel,"(");
1.227 brouard 7282: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to stable (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
7283: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7284: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7285: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7286: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7287: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7288: vlv= nbcode[Tvaraff[k]][lv];
7289: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7290: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7291: }
1.237 brouard 7292: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7293: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7294: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7295: }
1.264 brouard 7296: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7297: fprintf(ficgp,"\n#\n");
7298: if(invalidvarcomb[k1]){
7299: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7300: continue;
7301: }
7302:
7303: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7304: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7305: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 7306: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7307: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7308:
7309: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7310: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7311: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7312: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7313: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7314: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7315: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7316: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7317: if(i==istart){
1.227 brouard 7318: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7319: }else{
7320: fprintf(ficgp,",\\\n '' ");
7321: }
7322: if(cptcoveff ==0){ /* No covariate */
7323: ioffset=2; /* Age is in 2 */
7324: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7325: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7326: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7327: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7328: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7329: if(i==nlstate+1){
7330: fprintf(ficgp," $%d/(1.-$%d)):5 t 'pw.%d' with line lc variable ", \
7331: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7332: fprintf(ficgp,",\\\n '' ");
7333: fprintf(ficgp," u %d:(",ioffset);
7334: fprintf(ficgp," (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", \
7335: offyear, \
1.227 brouard 7336: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
1.266 brouard 7337: }else
1.227 brouard 7338: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7339: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7340: }else{ /* more than 2 covariates */
7341: if(cptcoveff ==1){
7342: ioffset=4; /* Age is in 4 */
7343: }else{
7344: ioffset=6; /* Age is in 6 */
7345: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7346: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7347: }
7348: fprintf(ficgp," u %d:(",ioffset);
7349: kl=0;
7350: strcpy(gplotcondition,"(");
7351: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7352: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7353: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7354: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7355: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7356: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7357: kl++;
7358: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7359: kl++;
7360: if(k <cptcoveff && cptcoveff>1)
7361: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7362: }
7363: strcpy(gplotcondition+strlen(gplotcondition),")");
7364: /* 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 *\/ */
7365: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7366: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7367: /* '' 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*/
7368: if(i==nlstate+1){
1.266 brouard 7369: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):5 t 'p.%d' with line lc variable", gplotcondition, \
7370: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7371: fprintf(ficgp,",\\\n '' ");
7372: fprintf(ficgp," u %d:(",ioffset);
7373: fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \
7374: offyear, \
1.227 brouard 7375: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
1.266 brouard 7376: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
1.227 brouard 7377: }else{
7378: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7379: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7380: }
7381: } /* end if covariate */
7382: } /* nlstate */
1.264 brouard 7383: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7384: } /* end cpt state*/
7385: } /* end covariate */
7386: } /* End if prevfcast */
1.227 brouard 7387:
7388:
1.238 brouard 7389: /* 9eme writing MLE parameters */
7390: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7391: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7392: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7393: for(k=1; k <=(nlstate+ndeath); k++){
7394: if (k != i) {
1.227 brouard 7395: fprintf(ficgp,"# current state %d\n",k);
7396: for(j=1; j <=ncovmodel; j++){
7397: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7398: jk++;
7399: }
7400: fprintf(ficgp,"\n");
1.126 brouard 7401: }
7402: }
1.223 brouard 7403: }
1.187 brouard 7404: fprintf(ficgp,"##############\n#\n");
1.227 brouard 7405:
1.145 brouard 7406: /*goto avoid;*/
1.238 brouard 7407: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
7408: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 7409: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
7410: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
7411: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
7412: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
7413: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7414: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7415: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7416: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
7417: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
7418: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
7419: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
7420: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
7421: fprintf(ficgp,"#\n");
1.223 brouard 7422: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 7423: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 7424: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 7425: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 7426: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
7427: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 7428: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 7429: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7430: continue;
1.264 brouard 7431: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
7432: strcpy(gplotlabel,"(");
7433: sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);
7434: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7435: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7436: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7437: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7438: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7439: vlv= nbcode[Tvaraff[k]][lv];
7440: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7441: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7442: }
1.237 brouard 7443: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7444: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7445: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7446: }
1.264 brouard 7447: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 7448: fprintf(ficgp,"\n#\n");
1.264 brouard 7449: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
7450: fprintf(ficgp,"\nset label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 7451: fprintf(ficgp,"\nset ter svg size 640, 480 ");
7452: if (ng==1){
7453: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
7454: fprintf(ficgp,"\nunset log y");
7455: }else if (ng==2){
7456: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
7457: fprintf(ficgp,"\nset log y");
7458: }else if (ng==3){
7459: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
7460: fprintf(ficgp,"\nset log y");
7461: }else
7462: fprintf(ficgp,"\nunset title ");
7463: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
7464: i=1;
7465: for(k2=1; k2<=nlstate; k2++) {
7466: k3=i;
7467: for(k=1; k<=(nlstate+ndeath); k++) {
7468: if (k != k2){
7469: switch( ng) {
7470: case 1:
7471: if(nagesqr==0)
7472: fprintf(ficgp," p%d+p%d*x",i,i+1);
7473: else /* nagesqr =1 */
7474: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7475: break;
7476: case 2: /* ng=2 */
7477: if(nagesqr==0)
7478: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
7479: else /* nagesqr =1 */
7480: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
7481: break;
7482: case 3:
7483: if(nagesqr==0)
7484: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
7485: else /* nagesqr =1 */
7486: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
7487: break;
7488: }
7489: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 7490: ijp=1; /* product no age */
7491: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
7492: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 7493: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.237 brouard 7494: if(j==Tage[ij]) { /* Product by age */
7495: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.238 brouard 7496: if(DummyV[j]==0){
1.237 brouard 7497: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
7498: }else{ /* quantitative */
7499: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
1.264 brouard 7500: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.237 brouard 7501: }
7502: ij++;
7503: }
7504: }else if(j==Tprod[ijp]) { /* */
7505: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
7506: if(ijp <=cptcovprod) { /* Product */
1.238 brouard 7507: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
7508: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
1.264 brouard 7509: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
1.237 brouard 7510: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
7511: }else{ /* Vn is dummy and Vm is quanti */
1.264 brouard 7512: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
1.237 brouard 7513: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7514: }
7515: }else{ /* Vn*Vm Vn is quanti */
1.238 brouard 7516: if(DummyV[Tvard[ijp][2]]==0){
1.237 brouard 7517: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
7518: }else{ /* Both quanti */
7519: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
7520: }
7521: }
1.238 brouard 7522: ijp++;
1.237 brouard 7523: }
7524: } else{ /* simple covariate */
1.264 brouard 7525: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 7526: if(Dummy[j]==0){
7527: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
7528: }else{ /* quantitative */
7529: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 7530: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7531: }
1.237 brouard 7532: } /* end simple */
7533: } /* end j */
1.223 brouard 7534: }else{
7535: i=i-ncovmodel;
7536: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
7537: fprintf(ficgp," (1.");
7538: }
1.227 brouard 7539:
1.223 brouard 7540: if(ng != 1){
7541: fprintf(ficgp,")/(1");
1.227 brouard 7542:
1.264 brouard 7543: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 7544: if(nagesqr==0)
1.264 brouard 7545: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 7546: else /* nagesqr =1 */
1.264 brouard 7547: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1,k3+(cpt-1)*ncovmodel+1+nagesqr);
1.217 brouard 7548:
1.223 brouard 7549: ij=1;
7550: for(j=3; j <=ncovmodel-nagesqr; j++){
1.237 brouard 7551: if((j-2)==Tage[ij]) { /* Bug valgrind */
7552: if(ij <=cptcovage) { /* Bug valgrind */
1.264 brouard 7553: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
7554: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 7555: ij++;
7556: }
7557: }
7558: else
1.264 brouard 7559: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 7560: }
7561: fprintf(ficgp,")");
7562: }
7563: fprintf(ficgp,")");
7564: if(ng ==2)
7565: fprintf(ficgp," t \"p%d%d\" ", k2,k);
7566: else /* ng= 3 */
7567: fprintf(ficgp," t \"i%d%d\" ", k2,k);
7568: }else{ /* end ng <> 1 */
7569: if( k !=k2) /* logit p11 is hard to draw */
7570: fprintf(ficgp," t \"logit(p%d%d)\" ", k2,k);
7571: }
7572: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
7573: fprintf(ficgp,",");
7574: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
7575: fprintf(ficgp,",");
7576: i=i+ncovmodel;
7577: } /* end k */
7578: } /* end k2 */
1.264 brouard 7579: fprintf(ficgp,"\n set out; unset label;\n");
7580: } /* end k1 */
1.223 brouard 7581: } /* end ng */
7582: /* avoid: */
7583: fflush(ficgp);
1.126 brouard 7584: } /* end gnuplot */
7585:
7586:
7587: /*************** Moving average **************/
1.219 brouard 7588: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 7589: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 7590:
1.222 brouard 7591: int i, cpt, cptcod;
7592: int modcovmax =1;
7593: int mobilavrange, mob;
7594: int iage=0;
7595:
1.266 brouard 7596: double sum=0., sumr=0.;
1.222 brouard 7597: double age;
1.266 brouard 7598: double *sumnewp, *sumnewm, *sumnewmr;
7599: double *agemingood, *agemaxgood;
7600: double *agemingoodr, *agemaxgoodr;
1.222 brouard 7601:
7602:
1.225 brouard 7603: /* modcovmax=2*cptcoveff;/\* Max number of modalities. We suppose */
1.222 brouard 7604: /* a covariate has 2 modalities, should be equal to ncovcombmax *\/ */
7605:
7606: sumnewp = vector(1,ncovcombmax);
7607: sumnewm = vector(1,ncovcombmax);
1.266 brouard 7608: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 7609: agemingood = vector(1,ncovcombmax);
1.266 brouard 7610: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 7611: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 7612: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 7613:
7614: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 7615: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 7616: sumnewp[cptcod]=0.;
1.266 brouard 7617: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
7618: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 7619: }
7620: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
7621:
1.266 brouard 7622: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
7623: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 7624: else mobilavrange=mobilav;
7625: for (age=bage; age<=fage; age++)
7626: for (i=1; i<=nlstate;i++)
7627: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
7628: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7629: /* We keep the original values on the extreme ages bage, fage and for
7630: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
7631: we use a 5 terms etc. until the borders are no more concerned.
7632: */
7633: for (mob=3;mob <=mobilavrange;mob=mob+2){
7634: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 7635: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
7636: sumnewm[cptcod]=0.;
7637: for (i=1; i<=nlstate;i++){
1.222 brouard 7638: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
7639: for (cpt=1;cpt<=(mob-1)/2;cpt++){
7640: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
7641: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
7642: }
7643: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 7644: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7645: } /* end i */
7646: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
7647: } /* end cptcod */
1.222 brouard 7648: }/* end age */
7649: }/* end mob */
1.266 brouard 7650: }else{
7651: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 7652: return -1;
1.266 brouard 7653: }
7654:
7655: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 7656: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
7657: if(invalidvarcomb[cptcod]){
7658: printf("\nCombination (%d) ignored because no cases \n",cptcod);
7659: continue;
7660: }
1.219 brouard 7661:
1.266 brouard 7662: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
7663: sumnewm[cptcod]=0.;
7664: sumnewmr[cptcod]=0.;
7665: for (i=1; i<=nlstate;i++){
7666: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7667: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7668: }
7669: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7670: agemingoodr[cptcod]=age;
7671: }
7672: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7673: agemingood[cptcod]=age;
7674: }
7675: } /* age */
7676: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 7677: sumnewm[cptcod]=0.;
1.266 brouard 7678: sumnewmr[cptcod]=0.;
1.222 brouard 7679: for (i=1; i<=nlstate;i++){
7680: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 7681: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7682: }
7683: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7684: agemaxgoodr[cptcod]=age;
1.222 brouard 7685: }
7686: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 7687: agemaxgood[cptcod]=age;
7688: }
7689: } /* age */
7690: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
7691: /* but they will change */
7692: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
7693: sumnewm[cptcod]=0.;
7694: sumnewmr[cptcod]=0.;
7695: for (i=1; i<=nlstate;i++){
7696: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7697: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7698: }
7699: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
7700: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
7701: agemaxgoodr[cptcod]=age; /* age min */
7702: for (i=1; i<=nlstate;i++)
7703: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7704: }else{ /* bad we change the value with the values of good ages */
7705: for (i=1; i<=nlstate;i++){
7706: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
7707: } /* i */
7708: } /* end bad */
7709: }else{
7710: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7711: agemaxgood[cptcod]=age;
7712: }else{ /* bad we change the value with the values of good ages */
7713: for (i=1; i<=nlstate;i++){
7714: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
7715: } /* i */
7716: } /* end bad */
7717: }/* end else */
7718: sum=0.;sumr=0.;
7719: for (i=1; i<=nlstate;i++){
7720: sum+=mobaverage[(int)age][i][cptcod];
7721: sumr+=probs[(int)age][i][cptcod];
7722: }
7723: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7724: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, bage);
7725: } /* end bad */
7726: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7727: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
7728: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, bage);
1.222 brouard 7729: } /* end bad */
7730: }/* age */
1.266 brouard 7731:
7732: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 7733: sumnewm[cptcod]=0.;
1.266 brouard 7734: sumnewmr[cptcod]=0.;
1.222 brouard 7735: for (i=1; i<=nlstate;i++){
7736: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 7737: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
7738: }
7739: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
7740: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
7741: agemingoodr[cptcod]=age;
7742: for (i=1; i<=nlstate;i++)
7743: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
7744: }else{ /* bad we change the value with the values of good ages */
7745: for (i=1; i<=nlstate;i++){
7746: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
7747: } /* i */
7748: } /* end bad */
7749: }else{
7750: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
7751: agemingood[cptcod]=age;
7752: }else{ /* bad */
7753: for (i=1; i<=nlstate;i++){
7754: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
7755: } /* i */
7756: } /* end bad */
7757: }/* end else */
7758: sum=0.;sumr=0.;
7759: for (i=1; i<=nlstate;i++){
7760: sum+=mobaverage[(int)age][i][cptcod];
7761: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 7762: }
1.266 brouard 7763: if(fabs(sum - 1.) > 1.e-3) { /* bad */
7764: printf("Moving average B1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you decrease fage=%d?\n",cptcod, sum, (int) age, fage);
7765: } /* end bad */
7766: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
7767: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
7768: printf("Moving average B2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase fage=%d\n",cptcod,sumr, (int)age, fage);
1.222 brouard 7769: } /* end bad */
7770: }/* age */
1.266 brouard 7771:
1.222 brouard 7772:
7773: for (age=bage; age<=fage; age++){
1.235 brouard 7774: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 7775: sumnewp[cptcod]=0.;
7776: sumnewm[cptcod]=0.;
7777: for (i=1; i<=nlstate;i++){
7778: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
7779: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
7780: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
7781: }
7782: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
7783: }
7784: /* printf("\n"); */
7785: /* } */
1.266 brouard 7786:
1.222 brouard 7787: /* brutal averaging */
1.266 brouard 7788: /* for (i=1; i<=nlstate;i++){ */
7789: /* for (age=1; age<=bage; age++){ */
7790: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
7791: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
7792: /* } */
7793: /* for (age=fage; age<=AGESUP; age++){ */
7794: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
7795: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
7796: /* } */
7797: /* } /\* end i status *\/ */
7798: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
7799: /* for (age=1; age<=AGESUP; age++){ */
7800: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
7801: /* mobaverage[(int)age][i][cptcod]=0.; */
7802: /* } */
7803: /* } */
1.222 brouard 7804: }/* end cptcod */
1.266 brouard 7805: free_vector(agemaxgoodr,1, ncovcombmax);
7806: free_vector(agemaxgood,1, ncovcombmax);
7807: free_vector(agemingood,1, ncovcombmax);
7808: free_vector(agemingoodr,1, ncovcombmax);
7809: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 7810: free_vector(sumnewm,1, ncovcombmax);
7811: free_vector(sumnewp,1, ncovcombmax);
7812: return 0;
7813: }/* End movingaverage */
1.218 brouard 7814:
1.126 brouard 7815:
7816: /************** Forecasting ******************/
1.267 ! brouard 7817: 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 7818: /* proj1, year, month, day of starting projection
7819: agemin, agemax range of age
7820: dateprev1 dateprev2 range of dates during which prevalence is computed
7821: anproj2 year of en of projection (same day and month as proj1).
7822: */
1.267 ! brouard 7823: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 7824: double agec; /* generic age */
7825: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
7826: double *popeffectif,*popcount;
7827: double ***p3mat;
1.218 brouard 7828: /* double ***mobaverage; */
1.126 brouard 7829: char fileresf[FILENAMELENGTH];
7830:
7831: agelim=AGESUP;
1.211 brouard 7832: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
7833: in each health status at the date of interview (if between dateprev1 and dateprev2).
7834: We still use firstpass and lastpass as another selection.
7835: */
1.214 brouard 7836: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
7837: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 7838:
1.201 brouard 7839: strcpy(fileresf,"F_");
7840: strcat(fileresf,fileresu);
1.126 brouard 7841: if((ficresf=fopen(fileresf,"w"))==NULL) {
7842: printf("Problem with forecast resultfile: %s\n", fileresf);
7843: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
7844: }
1.235 brouard 7845: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
7846: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 7847:
1.225 brouard 7848: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 7849:
7850:
7851: stepsize=(int) (stepm+YEARM-1)/YEARM;
7852: if (stepm<=12) stepsize=1;
7853: if(estepm < stepm){
7854: printf ("Problem %d lower than %d\n",estepm, stepm);
7855: }
7856: else hstepm=estepm;
7857:
7858: hstepm=hstepm/stepm;
7859: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
7860: fractional in yp1 */
7861: anprojmean=yp;
7862: yp2=modf((yp1*12),&yp);
7863: mprojmean=yp;
7864: yp1=modf((yp2*30.5),&yp);
7865: jprojmean=yp;
7866: if(jprojmean==0) jprojmean=1;
7867: if(mprojmean==0) jprojmean=1;
7868:
1.227 brouard 7869: i1=pow(2,cptcoveff);
1.126 brouard 7870: if (cptcovn < 1){i1=1;}
7871:
7872: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
7873:
7874: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 7875:
1.126 brouard 7876: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 7877: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7878: for(k=1; k<=i1;k++){
1.253 brouard 7879: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 7880: continue;
1.227 brouard 7881: if(invalidvarcomb[k]){
7882: printf("\nCombination (%d) projection ignored because no cases \n",k);
7883: continue;
7884: }
7885: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
7886: for(j=1;j<=cptcoveff;j++) {
7887: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7888: }
1.235 brouard 7889: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7890: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 7891: }
1.227 brouard 7892: fprintf(ficresf," yearproj age");
7893: for(j=1; j<=nlstate+ndeath;j++){
7894: for(i=1; i<=nlstate;i++)
7895: fprintf(ficresf," p%d%d",i,j);
7896: fprintf(ficresf," wp.%d",j);
7897: }
7898: for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) {
7899: fprintf(ficresf,"\n");
7900: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jproj1,mproj1,anproj1+yearp);
7901: for (agec=fage; agec>=(ageminpar-1); agec--){
7902: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
7903: nhstepm = nhstepm/hstepm;
7904: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7905: oldm=oldms;savm=savms;
1.235 brouard 7906: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.227 brouard 7907:
7908: for (h=0; h<=nhstepm; h++){
7909: if (h*hstepm/YEARM*stepm ==yearp) {
7910: fprintf(ficresf,"\n");
7911: for(j=1;j<=cptcoveff;j++)
7912: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
7913: fprintf(ficresf,"%.f %.f ",anproj1+yearp,agec+h*hstepm/YEARM*stepm);
7914: }
7915: for(j=1; j<=nlstate+ndeath;j++) {
7916: ppij=0.;
7917: for(i=1; i<=nlstate;i++) {
1.266 brouard 7918: /* if (mobilav>=1) */
1.227 brouard 7919: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
1.266 brouard 7920: /* else { */ /* even if mobilav==-1 we use mobaverage */
7921: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
7922: /* } */
1.227 brouard 7923: if (h*hstepm/YEARM*stepm== yearp) {
7924: fprintf(ficresf," %.3f", p3mat[i][j][h]);
7925: }
7926: } /* end i */
7927: if (h*hstepm/YEARM*stepm==yearp) {
7928: fprintf(ficresf," %.3f", ppij);
7929: }
7930: }/* end j */
7931: } /* end h */
7932: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7933: } /* end agec */
1.266 brouard 7934: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
7935: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 7936: } /* end yearp */
7937: } /* end k */
1.219 brouard 7938:
1.126 brouard 7939: fclose(ficresf);
1.215 brouard 7940: printf("End of Computing forecasting \n");
7941: fprintf(ficlog,"End of Computing forecasting\n");
7942:
1.126 brouard 7943: }
7944:
1.218 brouard 7945: /* /\************** Back Forecasting ******************\/ */
1.267 ! brouard 7946: void prevbackforecast(char fileres[], double ***prevacurrent, 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){
! 7947: /* back1, year, month, day of starting backection
! 7948: agemin, agemax range of age
! 7949: dateprev1 dateprev2 range of dates during which prevalence is computed
! 7950: anback2 year of en of backection (same day and month as back1).
! 7951: */
! 7952: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
! 7953: double agec; /* generic age */
! 7954: double agelim, ppij, yp,yp1,yp2,jprojmean,mprojmean,anprojmean;
! 7955: double *popeffectif,*popcount;
! 7956: double ***p3mat;
! 7957: /* double ***mobaverage; */
! 7958: char fileresfb[FILENAMELENGTH];
! 7959:
! 7960: agelim=AGESUP;
! 7961: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
! 7962: in each health status at the date of interview (if between dateprev1 and dateprev2).
! 7963: We still use firstpass and lastpass as another selection.
! 7964: */
! 7965: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
! 7966: /* firstpass, lastpass, stepm, weightopt, model); */
! 7967:
! 7968: /*Do we need to compute prevalence again?*/
! 7969:
! 7970: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
! 7971:
! 7972: strcpy(fileresfb,"FB_");
! 7973: strcat(fileresfb,fileresu);
! 7974: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
! 7975: printf("Problem with back forecast resultfile: %s\n", fileresfb);
! 7976: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
! 7977: }
! 7978: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
! 7979: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
! 7980:
! 7981: if (cptcoveff==0) ncodemax[cptcoveff]=1;
! 7982:
! 7983:
! 7984: stepsize=(int) (stepm+YEARM-1)/YEARM;
! 7985: if (stepm<=12) stepsize=1;
! 7986: if(estepm < stepm){
! 7987: printf ("Problem %d lower than %d\n",estepm, stepm);
! 7988: }
! 7989: else hstepm=estepm;
! 7990:
! 7991: hstepm=hstepm/stepm;
! 7992: yp1=modf(dateintmean,&yp);/* extracts integral of datemean in yp and
! 7993: fractional in yp1 */
! 7994: anprojmean=yp;
! 7995: yp2=modf((yp1*12),&yp);
! 7996: mprojmean=yp;
! 7997: yp1=modf((yp2*30.5),&yp);
! 7998: jprojmean=yp;
! 7999: if(jprojmean==0) jprojmean=1;
! 8000: if(mprojmean==0) jprojmean=1;
! 8001:
! 8002: i1=pow(2,cptcoveff);
! 8003: if (cptcovn < 1){i1=1;}
! 8004:
! 8005: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jprojmean,mprojmean,anprojmean,dateintmean,dateprev1,dateprev2);
! 8006:
! 8007: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
! 8008:
! 8009: /* if (h==(int)(YEARM*yearp)){ */
! 8010: /* for(cptcov=1, k=0;cptcov<=i1;cptcov++){ */
! 8011: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
! 8012: /* k=k+1; */
! 8013: for(nres=1; nres <= nresult; nres++) /* For each resultline */
! 8014: for(k=1; k<=i1;k++){
! 8015: if(i1 != 1 && TKresult[nres]!= k)
! 8016: continue;
! 8017: if(invalidvarcomb[k]){
! 8018: printf("\nCombination (%d) projection ignored because no cases \n",k);
! 8019: continue;
! 8020: }
! 8021: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hp.jx is weighted by observed prev \n#");
! 8022: for(j=1;j<=cptcoveff;j++) {
! 8023: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
! 8024: }
! 8025: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
! 8026: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
! 8027: }
! 8028: fprintf(ficresfb," yearbproj age");
! 8029: for(j=1; j<=nlstate+ndeath;j++){
! 8030: for(i=1; i<=nlstate;i++)
! 8031: fprintf(ficresfb," p%d%d",i,j);
! 8032: fprintf(ficresfb," p.%d",j);
! 8033: }
! 8034: for (yearp=0; yearp>=(anback2-anback1);yearp -=stepsize) {
! 8035: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
! 8036: fprintf(ficresfb,"\n");
! 8037: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp);
! 8038: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
! 8039: /* nhstepm=(int) rint((agelim-agec)*YEARM/stepm); */
! 8040: for (agec=fage; agec>=fage-20; agec--){ /* testing up to 10 */
! 8041: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
! 8042: nhstepm = nhstepm/hstepm;
! 8043: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
! 8044: oldm=oldms;savm=savms;
! 8045: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
! 8046:
! 8047: for (h=0; h<=nhstepm; h++){
! 8048: if (h*hstepm/YEARM*stepm ==yearp) {
! 8049: fprintf(ficresfb,"\n");
! 8050: for(j=1;j<=cptcoveff;j++)
! 8051: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
! 8052: fprintf(ficresfb,"%.f %.f ",anback1+yearp,agec+h*hstepm/YEARM*stepm);
! 8053: }
! 8054: for(j=1; j<=nlstate+ndeath;j++) {
! 8055: ppij=0.;
! 8056: for(i=1; i<=nlstate;i++) {
! 8057: /* if (mobilav==1) */
! 8058: ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][i][k];
! 8059: /* else { */
! 8060: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
! 8061: /* } */
! 8062: if (h*hstepm/YEARM*stepm== yearp) {
! 8063: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
! 8064: }
! 8065: } /* end i */
! 8066: if (h*hstepm/YEARM*stepm==yearp) {
! 8067: fprintf(ficresfb," %.3f", ppij);
! 8068: }
! 8069: }/* end j */
! 8070: } /* end h */
! 8071: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
! 8072: } /* end agec */
! 8073: } /* end yearp */
! 8074: } /* end k */
1.217 brouard 8075:
1.267 ! brouard 8076: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8077:
1.267 ! brouard 8078: fclose(ficresfb);
! 8079: printf("End of Computing Back forecasting \n");
! 8080: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8081:
1.267 ! brouard 8082: }
1.217 brouard 8083:
1.126 brouard 8084: /************** Forecasting *****not tested NB*************/
1.227 brouard 8085: /* 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 8086:
1.227 brouard 8087: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8088: /* int *popage; */
8089: /* double calagedatem, agelim, kk1, kk2; */
8090: /* double *popeffectif,*popcount; */
8091: /* double ***p3mat,***tabpop,***tabpopprev; */
8092: /* /\* double ***mobaverage; *\/ */
8093: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8094:
1.227 brouard 8095: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8096: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8097: /* agelim=AGESUP; */
8098: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8099:
1.227 brouard 8100: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8101:
8102:
1.227 brouard 8103: /* strcpy(filerespop,"POP_"); */
8104: /* strcat(filerespop,fileresu); */
8105: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8106: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8107: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8108: /* } */
8109: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8110: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8111:
1.227 brouard 8112: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8113:
1.227 brouard 8114: /* /\* if (mobilav!=0) { *\/ */
8115: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8116: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8117: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8118: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8119: /* /\* } *\/ */
8120: /* /\* } *\/ */
1.126 brouard 8121:
1.227 brouard 8122: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8123: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8124:
1.227 brouard 8125: /* agelim=AGESUP; */
1.126 brouard 8126:
1.227 brouard 8127: /* hstepm=1; */
8128: /* hstepm=hstepm/stepm; */
1.218 brouard 8129:
1.227 brouard 8130: /* if (popforecast==1) { */
8131: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8132: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8133: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8134: /* } */
8135: /* popage=ivector(0,AGESUP); */
8136: /* popeffectif=vector(0,AGESUP); */
8137: /* popcount=vector(0,AGESUP); */
1.126 brouard 8138:
1.227 brouard 8139: /* i=1; */
8140: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8141:
1.227 brouard 8142: /* imx=i; */
8143: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8144: /* } */
1.218 brouard 8145:
1.227 brouard 8146: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8147: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8148: /* k=k+1; */
8149: /* fprintf(ficrespop,"\n#******"); */
8150: /* for(j=1;j<=cptcoveff;j++) { */
8151: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8152: /* } */
8153: /* fprintf(ficrespop,"******\n"); */
8154: /* fprintf(ficrespop,"# Age"); */
8155: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8156: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8157:
1.227 brouard 8158: /* for (cpt=0; cpt<=0;cpt++) { */
8159: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8160:
1.227 brouard 8161: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8162: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8163: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8164:
1.227 brouard 8165: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8166: /* oldm=oldms;savm=savms; */
8167: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8168:
1.227 brouard 8169: /* for (h=0; h<=nhstepm; h++){ */
8170: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8171: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8172: /* } */
8173: /* for(j=1; j<=nlstate+ndeath;j++) { */
8174: /* kk1=0.;kk2=0; */
8175: /* for(i=1; i<=nlstate;i++) { */
8176: /* if (mobilav==1) */
8177: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8178: /* else { */
8179: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8180: /* } */
8181: /* } */
8182: /* if (h==(int)(calagedatem+12*cpt)){ */
8183: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8184: /* /\*fprintf(ficrespop," %.3f", kk1); */
8185: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8186: /* } */
8187: /* } */
8188: /* for(i=1; i<=nlstate;i++){ */
8189: /* kk1=0.; */
8190: /* for(j=1; j<=nlstate;j++){ */
8191: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8192: /* } */
8193: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8194: /* } */
1.218 brouard 8195:
1.227 brouard 8196: /* if (h==(int)(calagedatem+12*cpt)) */
8197: /* for(j=1; j<=nlstate;j++) */
8198: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8199: /* } */
8200: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8201: /* } */
8202: /* } */
1.218 brouard 8203:
1.227 brouard 8204: /* /\******\/ */
1.218 brouard 8205:
1.227 brouard 8206: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8207: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8208: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8209: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8210: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8211:
1.227 brouard 8212: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8213: /* oldm=oldms;savm=savms; */
8214: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8215: /* for (h=0; h<=nhstepm; h++){ */
8216: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8217: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8218: /* } */
8219: /* for(j=1; j<=nlstate+ndeath;j++) { */
8220: /* kk1=0.;kk2=0; */
8221: /* for(i=1; i<=nlstate;i++) { */
8222: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8223: /* } */
8224: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8225: /* } */
8226: /* } */
8227: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8228: /* } */
8229: /* } */
8230: /* } */
8231: /* } */
1.218 brouard 8232:
1.227 brouard 8233: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8234:
1.227 brouard 8235: /* if (popforecast==1) { */
8236: /* free_ivector(popage,0,AGESUP); */
8237: /* free_vector(popeffectif,0,AGESUP); */
8238: /* free_vector(popcount,0,AGESUP); */
8239: /* } */
8240: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8241: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8242: /* fclose(ficrespop); */
8243: /* } /\* End of popforecast *\/ */
1.218 brouard 8244:
1.126 brouard 8245: int fileappend(FILE *fichier, char *optionfich)
8246: {
8247: if((fichier=fopen(optionfich,"a"))==NULL) {
8248: printf("Problem with file: %s\n", optionfich);
8249: fprintf(ficlog,"Problem with file: %s\n", optionfich);
8250: return (0);
8251: }
8252: fflush(fichier);
8253: return (1);
8254: }
8255:
8256:
8257: /**************** function prwizard **********************/
8258: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
8259: {
8260:
8261: /* Wizard to print covariance matrix template */
8262:
1.164 brouard 8263: char ca[32], cb[32];
8264: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 8265: int numlinepar;
8266:
8267: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8268: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
8269: for(i=1; i <=nlstate; i++){
8270: jj=0;
8271: for(j=1; j <=nlstate+ndeath; j++){
8272: if(j==i) continue;
8273: jj++;
8274: /*ca[0]= k+'a'-1;ca[1]='\0';*/
8275: printf("%1d%1d",i,j);
8276: fprintf(ficparo,"%1d%1d",i,j);
8277: for(k=1; k<=ncovmodel;k++){
8278: /* printf(" %lf",param[i][j][k]); */
8279: /* fprintf(ficparo," %lf",param[i][j][k]); */
8280: printf(" 0.");
8281: fprintf(ficparo," 0.");
8282: }
8283: printf("\n");
8284: fprintf(ficparo,"\n");
8285: }
8286: }
8287: printf("# Scales (for hessian or gradient estimation)\n");
8288: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
8289: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
8290: for(i=1; i <=nlstate; i++){
8291: jj=0;
8292: for(j=1; j <=nlstate+ndeath; j++){
8293: if(j==i) continue;
8294: jj++;
8295: fprintf(ficparo,"%1d%1d",i,j);
8296: printf("%1d%1d",i,j);
8297: fflush(stdout);
8298: for(k=1; k<=ncovmodel;k++){
8299: /* printf(" %le",delti3[i][j][k]); */
8300: /* fprintf(ficparo," %le",delti3[i][j][k]); */
8301: printf(" 0.");
8302: fprintf(ficparo," 0.");
8303: }
8304: numlinepar++;
8305: printf("\n");
8306: fprintf(ficparo,"\n");
8307: }
8308: }
8309: printf("# Covariance matrix\n");
8310: /* # 121 Var(a12)\n\ */
8311: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8312: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
8313: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
8314: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
8315: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
8316: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
8317: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8318: fflush(stdout);
8319: fprintf(ficparo,"# Covariance matrix\n");
8320: /* # 121 Var(a12)\n\ */
8321: /* # 122 Cov(b12,a12) Var(b12)\n\ */
8322: /* # ...\n\ */
8323: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
8324:
8325: for(itimes=1;itimes<=2;itimes++){
8326: jj=0;
8327: for(i=1; i <=nlstate; i++){
8328: for(j=1; j <=nlstate+ndeath; j++){
8329: if(j==i) continue;
8330: for(k=1; k<=ncovmodel;k++){
8331: jj++;
8332: ca[0]= k+'a'-1;ca[1]='\0';
8333: if(itimes==1){
8334: printf("#%1d%1d%d",i,j,k);
8335: fprintf(ficparo,"#%1d%1d%d",i,j,k);
8336: }else{
8337: printf("%1d%1d%d",i,j,k);
8338: fprintf(ficparo,"%1d%1d%d",i,j,k);
8339: /* printf(" %.5le",matcov[i][j]); */
8340: }
8341: ll=0;
8342: for(li=1;li <=nlstate; li++){
8343: for(lj=1;lj <=nlstate+ndeath; lj++){
8344: if(lj==li) continue;
8345: for(lk=1;lk<=ncovmodel;lk++){
8346: ll++;
8347: if(ll<=jj){
8348: cb[0]= lk +'a'-1;cb[1]='\0';
8349: if(ll<jj){
8350: if(itimes==1){
8351: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8352: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
8353: }else{
8354: printf(" 0.");
8355: fprintf(ficparo," 0.");
8356: }
8357: }else{
8358: if(itimes==1){
8359: printf(" Var(%s%1d%1d)",ca,i,j);
8360: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
8361: }else{
8362: printf(" 0.");
8363: fprintf(ficparo," 0.");
8364: }
8365: }
8366: }
8367: } /* end lk */
8368: } /* end lj */
8369: } /* end li */
8370: printf("\n");
8371: fprintf(ficparo,"\n");
8372: numlinepar++;
8373: } /* end k*/
8374: } /*end j */
8375: } /* end i */
8376: } /* end itimes */
8377:
8378: } /* end of prwizard */
8379: /******************* Gompertz Likelihood ******************************/
8380: double gompertz(double x[])
8381: {
8382: double A,B,L=0.0,sump=0.,num=0.;
8383: int i,n=0; /* n is the size of the sample */
8384:
1.220 brouard 8385: for (i=1;i<=imx ; i++) {
1.126 brouard 8386: sump=sump+weight[i];
8387: /* sump=sump+1;*/
8388: num=num+1;
8389: }
8390:
8391:
8392: /* for (i=0; i<=imx; i++)
8393: 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]);*/
8394:
8395: for (i=1;i<=imx ; i++)
8396: {
8397: if (cens[i] == 1 && wav[i]>1)
8398: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
8399:
8400: if (cens[i] == 0 && wav[i]>1)
8401: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
8402: +log(x[1]/YEARM)+x[2]*(agedc[i]-agegomp)+log(YEARM);
8403:
8404: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8405: if (wav[i] > 1 ) { /* ??? */
8406: L=L+A*weight[i];
8407: /* 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]);*/
8408: }
8409: }
8410:
8411: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8412:
8413: return -2*L*num/sump;
8414: }
8415:
1.136 brouard 8416: #ifdef GSL
8417: /******************* Gompertz_f Likelihood ******************************/
8418: double gompertz_f(const gsl_vector *v, void *params)
8419: {
8420: double A,B,LL=0.0,sump=0.,num=0.;
8421: double *x= (double *) v->data;
8422: int i,n=0; /* n is the size of the sample */
8423:
8424: for (i=0;i<=imx-1 ; i++) {
8425: sump=sump+weight[i];
8426: /* sump=sump+1;*/
8427: num=num+1;
8428: }
8429:
8430:
8431: /* for (i=0; i<=imx; i++)
8432: 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]);*/
8433: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
8434: for (i=1;i<=imx ; i++)
8435: {
8436: if (cens[i] == 1 && wav[i]>1)
8437: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
8438:
8439: if (cens[i] == 0 && wav[i]>1)
8440: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
8441: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
8442:
8443: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
8444: if (wav[i] > 1 ) { /* ??? */
8445: LL=LL+A*weight[i];
8446: /* 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]);*/
8447: }
8448: }
8449:
8450: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
8451: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
8452:
8453: return -2*LL*num/sump;
8454: }
8455: #endif
8456:
1.126 brouard 8457: /******************* Printing html file ***********/
1.201 brouard 8458: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 8459: int lastpass, int stepm, int weightopt, char model[],\
8460: int imx, double p[],double **matcov,double agemortsup){
8461: int i,k;
8462:
8463: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
8464: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
8465: for (i=1;i<=2;i++)
8466: 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 8467: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 8468: fprintf(fichtm,"</ul>");
8469:
8470: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
8471:
8472: 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>");
8473:
8474: for (k=agegomp;k<(agemortsup-2);k++)
8475: 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]);
8476:
8477:
8478: fflush(fichtm);
8479: }
8480:
8481: /******************* Gnuplot file **************/
1.201 brouard 8482: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 8483:
8484: char dirfileres[132],optfileres[132];
1.164 brouard 8485:
1.126 brouard 8486: int ng;
8487:
8488:
8489: /*#ifdef windows */
8490: fprintf(ficgp,"cd \"%s\" \n",pathc);
8491: /*#endif */
8492:
8493:
8494: strcpy(dirfileres,optionfilefiname);
8495: strcpy(optfileres,"vpl");
1.199 brouard 8496: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 8497: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 8498: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 8499: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 8500: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
8501:
8502: }
8503:
1.136 brouard 8504: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
8505: {
1.126 brouard 8506:
1.136 brouard 8507: /*-------- data file ----------*/
8508: FILE *fic;
8509: char dummy[]=" ";
1.240 brouard 8510: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 8511: int lstra;
1.136 brouard 8512: int linei, month, year,iout;
8513: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 8514: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 8515: char *stratrunc;
1.223 brouard 8516:
1.240 brouard 8517: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
8518: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 8519:
1.240 brouard 8520: for(v=1; v <=ncovcol;v++){
8521: DummyV[v]=0;
8522: FixedV[v]=0;
8523: }
8524: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
8525: DummyV[v]=1;
8526: FixedV[v]=0;
8527: }
8528: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
8529: DummyV[v]=0;
8530: FixedV[v]=1;
8531: }
8532: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
8533: DummyV[v]=1;
8534: FixedV[v]=1;
8535: }
8536: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
8537: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
8538: 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]);
8539: }
1.126 brouard 8540:
1.136 brouard 8541: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 8542: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
8543: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 8544: }
1.126 brouard 8545:
1.136 brouard 8546: i=1;
8547: linei=0;
8548: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
8549: linei=linei+1;
8550: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
8551: if(line[j] == '\t')
8552: line[j] = ' ';
8553: }
8554: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
8555: ;
8556: };
8557: line[j+1]=0; /* Trims blanks at end of line */
8558: if(line[0]=='#'){
8559: fprintf(ficlog,"Comment line\n%s\n",line);
8560: printf("Comment line\n%s\n",line);
8561: continue;
8562: }
8563: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 8564: strcpy(line, linetmp);
1.223 brouard 8565:
8566: /* Loops on waves */
8567: for (j=maxwav;j>=1;j--){
8568: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 8569: cutv(stra, strb, line, ' ');
8570: if(strb[0]=='.') { /* Missing value */
8571: lval=-1;
8572: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
8573: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
8574: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
8575: 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);
8576: 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);
8577: return 1;
8578: }
8579: }else{
8580: errno=0;
8581: /* what_kind_of_number(strb); */
8582: dval=strtod(strb,&endptr);
8583: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
8584: /* if(strb != endptr && *endptr == '\0') */
8585: /* dval=dlval; */
8586: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8587: if( strb[0]=='\0' || (*endptr != '\0')){
8588: 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);
8589: 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);
8590: return 1;
8591: }
8592: cotqvar[j][iv][i]=dval;
8593: cotvar[j][ntv+iv][i]=dval;
8594: }
8595: strcpy(line,stra);
1.223 brouard 8596: }/* end loop ntqv */
1.225 brouard 8597:
1.223 brouard 8598: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 8599: cutv(stra, strb, line, ' ');
8600: if(strb[0]=='.') { /* Missing value */
8601: lval=-1;
8602: }else{
8603: errno=0;
8604: lval=strtol(strb,&endptr,10);
8605: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8606: if( strb[0]=='\0' || (*endptr != '\0')){
8607: 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);
8608: 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);
8609: return 1;
8610: }
8611: }
8612: if(lval <-1 || lval >1){
8613: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8614: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8615: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8616: For example, for multinomial values like 1, 2 and 3,\n \
8617: build V1=0 V2=0 for the reference value (1),\n \
8618: V1=1 V2=0 for (2) \n \
1.223 brouard 8619: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8620: output of IMaCh is often meaningless.\n \
1.223 brouard 8621: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 8622: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 8623: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8624: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 8625: For example, for multinomial values like 1, 2 and 3,\n \
8626: build V1=0 V2=0 for the reference value (1),\n \
8627: V1=1 V2=0 for (2) \n \
1.223 brouard 8628: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 8629: output of IMaCh is often meaningless.\n \
1.223 brouard 8630: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 8631: return 1;
8632: }
8633: cotvar[j][iv][i]=(double)(lval);
8634: strcpy(line,stra);
1.223 brouard 8635: }/* end loop ntv */
1.225 brouard 8636:
1.223 brouard 8637: /* Statuses at wave */
1.137 brouard 8638: cutv(stra, strb, line, ' ');
1.223 brouard 8639: if(strb[0]=='.') { /* Missing value */
1.238 brouard 8640: lval=-1;
1.136 brouard 8641: }else{
1.238 brouard 8642: errno=0;
8643: lval=strtol(strb,&endptr,10);
8644: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
8645: if( strb[0]=='\0' || (*endptr != '\0')){
8646: 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);
8647: 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);
8648: return 1;
8649: }
1.136 brouard 8650: }
1.225 brouard 8651:
1.136 brouard 8652: s[j][i]=lval;
1.225 brouard 8653:
1.223 brouard 8654: /* Date of Interview */
1.136 brouard 8655: strcpy(line,stra);
8656: cutv(stra, strb,line,' ');
1.169 brouard 8657: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8658: }
1.169 brouard 8659: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 8660: month=99;
8661: year=9999;
1.136 brouard 8662: }else{
1.225 brouard 8663: 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);
8664: 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);
8665: return 1;
1.136 brouard 8666: }
8667: anint[j][i]= (double) year;
8668: mint[j][i]= (double)month;
8669: strcpy(line,stra);
1.223 brouard 8670: } /* End loop on waves */
1.225 brouard 8671:
1.223 brouard 8672: /* Date of death */
1.136 brouard 8673: cutv(stra, strb,line,' ');
1.169 brouard 8674: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8675: }
1.169 brouard 8676: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 8677: month=99;
8678: year=9999;
8679: }else{
1.141 brouard 8680: 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 8681: 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);
8682: return 1;
1.136 brouard 8683: }
8684: andc[i]=(double) year;
8685: moisdc[i]=(double) month;
8686: strcpy(line,stra);
8687:
1.223 brouard 8688: /* Date of birth */
1.136 brouard 8689: cutv(stra, strb,line,' ');
1.169 brouard 8690: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 8691: }
1.169 brouard 8692: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 8693: month=99;
8694: year=9999;
8695: }else{
1.141 brouard 8696: 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);
8697: 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 8698: return 1;
1.136 brouard 8699: }
8700: if (year==9999) {
1.141 brouard 8701: 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);
8702: 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 8703: return 1;
8704:
1.136 brouard 8705: }
8706: annais[i]=(double)(year);
8707: moisnais[i]=(double)(month);
8708: strcpy(line,stra);
1.225 brouard 8709:
1.223 brouard 8710: /* Sample weight */
1.136 brouard 8711: cutv(stra, strb,line,' ');
8712: errno=0;
8713: dval=strtod(strb,&endptr);
8714: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 8715: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
8716: 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 8717: fflush(ficlog);
8718: return 1;
8719: }
8720: weight[i]=dval;
8721: strcpy(line,stra);
1.225 brouard 8722:
1.223 brouard 8723: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
8724: cutv(stra, strb, line, ' ');
8725: if(strb[0]=='.') { /* Missing value */
1.225 brouard 8726: lval=-1;
1.223 brouard 8727: }else{
1.225 brouard 8728: errno=0;
8729: /* what_kind_of_number(strb); */
8730: dval=strtod(strb,&endptr);
8731: /* if(strb != endptr && *endptr == '\0') */
8732: /* dval=dlval; */
8733: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
8734: if( strb[0]=='\0' || (*endptr != '\0')){
8735: 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);
8736: 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);
8737: return 1;
8738: }
8739: coqvar[iv][i]=dval;
1.226 brouard 8740: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 8741: }
8742: strcpy(line,stra);
8743: }/* end loop nqv */
1.136 brouard 8744:
1.223 brouard 8745: /* Covariate values */
1.136 brouard 8746: for (j=ncovcol;j>=1;j--){
8747: cutv(stra, strb,line,' ');
1.223 brouard 8748: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 8749: lval=-1;
1.136 brouard 8750: }else{
1.225 brouard 8751: errno=0;
8752: lval=strtol(strb,&endptr,10);
8753: if( strb[0]=='\0' || (*endptr != '\0')){
8754: 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);
8755: 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);
8756: return 1;
8757: }
1.136 brouard 8758: }
8759: if(lval <-1 || lval >1){
1.225 brouard 8760: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8761: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8762: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8763: For example, for multinomial values like 1, 2 and 3,\n \
8764: build V1=0 V2=0 for the reference value (1),\n \
8765: V1=1 V2=0 for (2) \n \
1.136 brouard 8766: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8767: output of IMaCh is often meaningless.\n \
1.136 brouard 8768: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 8769: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 8770: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
8771: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 8772: For example, for multinomial values like 1, 2 and 3,\n \
8773: build V1=0 V2=0 for the reference value (1),\n \
8774: V1=1 V2=0 for (2) \n \
1.136 brouard 8775: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 8776: output of IMaCh is often meaningless.\n \
1.136 brouard 8777: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 8778: return 1;
1.136 brouard 8779: }
8780: covar[j][i]=(double)(lval);
8781: strcpy(line,stra);
8782: }
8783: lstra=strlen(stra);
1.225 brouard 8784:
1.136 brouard 8785: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
8786: stratrunc = &(stra[lstra-9]);
8787: num[i]=atol(stratrunc);
8788: }
8789: else
8790: num[i]=atol(stra);
8791: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
8792: 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;}*/
8793:
8794: i=i+1;
8795: } /* End loop reading data */
1.225 brouard 8796:
1.136 brouard 8797: *imax=i-1; /* Number of individuals */
8798: fclose(fic);
1.225 brouard 8799:
1.136 brouard 8800: return (0);
1.164 brouard 8801: /* endread: */
1.225 brouard 8802: printf("Exiting readdata: ");
8803: fclose(fic);
8804: return (1);
1.223 brouard 8805: }
1.126 brouard 8806:
1.234 brouard 8807: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 8808: char *p1 = *stri, *p2 = *stri;
1.235 brouard 8809: while (*p2 == ' ')
1.234 brouard 8810: p2++;
8811: /* while ((*p1++ = *p2++) !=0) */
8812: /* ; */
8813: /* do */
8814: /* while (*p2 == ' ') */
8815: /* p2++; */
8816: /* while (*p1++ == *p2++); */
8817: *stri=p2;
1.145 brouard 8818: }
8819:
1.235 brouard 8820: int decoderesult ( char resultline[], int nres)
1.230 brouard 8821: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
8822: {
1.235 brouard 8823: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 8824: char resultsav[MAXLINE];
1.234 brouard 8825: int resultmodel[MAXLINE];
8826: int modelresult[MAXLINE];
1.230 brouard 8827: char stra[80], strb[80], strc[80], strd[80],stre[80];
8828:
1.234 brouard 8829: removefirstspace(&resultline);
1.233 brouard 8830: printf("decoderesult:%s\n",resultline);
1.230 brouard 8831:
8832: if (strstr(resultline,"v") !=0){
8833: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
8834: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
8835: return 1;
8836: }
8837: trimbb(resultsav, resultline);
8838: if (strlen(resultsav) >1){
8839: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
8840: }
1.253 brouard 8841: if(j == 0){ /* Resultline but no = */
8842: TKresult[nres]=0; /* Combination for the nresult and the model */
8843: return (0);
8844: }
8845:
1.234 brouard 8846: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
8847: 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);
8848: 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);
8849: }
8850: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
8851: if(nbocc(resultsav,'=') >1){
8852: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
8853: resultsav= V4=1 V5=25.1 V3=0 strb=V3=0 stra= V4=1 V5=25.1 */
8854: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
8855: }else
8856: cutl(strc,strd,resultsav,'=');
1.230 brouard 8857: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 8858:
1.230 brouard 8859: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
8860: Tvarsel[k]=atoi(strc);
8861: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
8862: /* cptcovsel++; */
8863: if (nbocc(stra,'=') >0)
8864: strcpy(resultsav,stra); /* and analyzes it */
8865: }
1.235 brouard 8866: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 8867: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8868: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 8869: match=0;
1.236 brouard 8870: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 8871: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 8872: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 8873: match=1;
8874: break;
8875: }
8876: }
8877: if(match == 0){
8878: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8879: }
8880: }
8881: }
1.235 brouard 8882: /* Checking for missing or useless values in comparison of current model needs */
8883: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 8884: match=0;
1.235 brouard 8885: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8886: if(Typevar[k1]==0){ /* Single */
1.237 brouard 8887: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 8888: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 8889: ++match;
8890: }
8891: }
8892: }
8893: if(match == 0){
8894: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
8895: }else if(match > 1){
8896: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
8897: }
8898: }
1.235 brouard 8899:
1.234 brouard 8900: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 8901: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
8902: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
8903: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
8904: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
8905: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
8906: /* 1 0 0 0 */
8907: /* 2 1 0 0 */
8908: /* 3 0 1 0 */
8909: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
8910: /* 5 0 0 1 */
8911: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
8912: /* 7 0 1 1 */
8913: /* 8 1 1 1 */
1.237 brouard 8914: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
8915: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
8916: /* V5*age V5 known which value for nres? */
8917: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 8918: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
8919: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 8920: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 8921: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
8922: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 8923: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
8924: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
8925: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 8926: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
8927: k4++;;
8928: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
8929: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
8930: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 8931: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
8932: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
8933: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 8934: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
8935: k4q++;;
8936: }
8937: }
1.234 brouard 8938:
1.235 brouard 8939: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 8940: return (0);
8941: }
1.235 brouard 8942:
1.230 brouard 8943: int decodemodel( char model[], int lastobs)
8944: /**< This routine decodes the model and returns:
1.224 brouard 8945: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
8946: * - nagesqr = 1 if age*age in the model, otherwise 0.
8947: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
8948: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
8949: * - cptcovage number of covariates with age*products =2
8950: * - cptcovs number of simple covariates
8951: * - 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
8952: * which is a new column after the 9 (ncovcol) variables.
8953: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
8954: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
8955: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
8956: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
8957: */
1.136 brouard 8958: {
1.238 brouard 8959: int i, j, k, ks, v;
1.227 brouard 8960: int j1, k1, k2, k3, k4;
1.136 brouard 8961: char modelsav[80];
1.145 brouard 8962: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 8963: char *strpt;
1.136 brouard 8964:
1.145 brouard 8965: /*removespace(model);*/
1.136 brouard 8966: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 8967: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 8968: if (strstr(model,"AGE") !=0){
1.192 brouard 8969: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
8970: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 8971: return 1;
8972: }
1.141 brouard 8973: if (strstr(model,"v") !=0){
8974: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
8975: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
8976: return 1;
8977: }
1.187 brouard 8978: strcpy(modelsav,model);
8979: if ((strpt=strstr(model,"age*age")) !=0){
8980: printf(" strpt=%s, model=%s\n",strpt, model);
8981: if(strpt != model){
1.234 brouard 8982: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8983: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8984: corresponding column of parameters.\n",model);
1.234 brouard 8985: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 8986: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 8987: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 8988: return 1;
1.225 brouard 8989: }
1.187 brouard 8990: nagesqr=1;
8991: if (strstr(model,"+age*age") !=0)
1.234 brouard 8992: substrchaine(modelsav, model, "+age*age");
1.187 brouard 8993: else if (strstr(model,"age*age+") !=0)
1.234 brouard 8994: substrchaine(modelsav, model, "age*age+");
1.187 brouard 8995: else
1.234 brouard 8996: substrchaine(modelsav, model, "age*age");
1.187 brouard 8997: }else
8998: nagesqr=0;
8999: if (strlen(modelsav) >1){
9000: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9001: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9002: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9003: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9004: * cst, age and age*age
9005: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9006: /* including age products which are counted in cptcovage.
9007: * but the covariates which are products must be treated
9008: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9009: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9010: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9011:
9012:
1.187 brouard 9013: /* Design
9014: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9015: * < ncovcol=8 >
9016: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9017: * k= 1 2 3 4 5 6 7 8
9018: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9019: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9020: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9021: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9022: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9023: * Tage[++cptcovage]=k
9024: * if products, new covar are created after ncovcol with k1
9025: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9026: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9027: * 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
9028: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9029: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9030: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9031: * < ncovcol=8 >
9032: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9033: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9034: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9035: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9036: * p Tprod[1]@2={ 6, 5}
9037: *p Tvard[1][1]@4= {7, 8, 5, 6}
9038: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9039: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9040: *How to reorganize?
9041: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9042: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9043: * {2, 1, 4, 8, 5, 6, 3, 7}
9044: * Struct []
9045: */
1.225 brouard 9046:
1.187 brouard 9047: /* This loop fills the array Tvar from the string 'model'.*/
9048: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9049: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9050: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9051: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9052: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9053: /* k=1 Tvar[1]=2 (from V2) */
9054: /* k=5 Tvar[5] */
9055: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9056: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9057: /* } */
1.198 brouard 9058: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9059: /*
9060: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9061: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9062: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9063: }
1.187 brouard 9064: cptcovage=0;
9065: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9066: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9067: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9068: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9069: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9070: /*scanf("%d",i);*/
9071: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9072: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9073: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9074: /* covar is not filled and then is empty */
9075: cptcovprod--;
9076: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9077: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9078: Typevar[k]=1; /* 1 for age product */
9079: cptcovage++; /* Sums the number of covariates which include age as a product */
9080: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9081: /*printf("stre=%s ", stre);*/
9082: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9083: cptcovprod--;
9084: cutl(stre,strb,strc,'V');
9085: Tvar[k]=atoi(stre);
9086: Typevar[k]=1; /* 1 for age product */
9087: cptcovage++;
9088: Tage[cptcovage]=k;
9089: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9090: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9091: cptcovn++;
9092: cptcovprodnoage++;k1++;
9093: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9094: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9095: because this model-covariate is a construction we invent a new column
9096: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9097: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9098: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9099: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9100: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9101: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9102: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9103: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9104: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9105: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9106: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9107: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9108: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9109: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9110: for (i=1; i<=lastobs;i++){
9111: /* Computes the new covariate which is a product of
9112: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9113: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9114: }
9115: } /* End age is not in the model */
9116: } /* End if model includes a product */
9117: else { /* no more sum */
9118: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9119: /* scanf("%d",i);*/
9120: cutl(strd,strc,strb,'V');
9121: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9122: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9123: Tvar[k]=atoi(strd);
9124: Typevar[k]=0; /* 0 for simple covariates */
9125: }
9126: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9127: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9128: scanf("%d",i);*/
1.187 brouard 9129: } /* end of loop + on total covariates */
9130: } /* end if strlen(modelsave == 0) age*age might exist */
9131: } /* end if strlen(model == 0) */
1.136 brouard 9132:
9133: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9134: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9135:
1.136 brouard 9136: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9137: printf("cptcovprod=%d ", cptcovprod);
9138: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9139: scanf("%d ",i);*/
9140:
9141:
1.230 brouard 9142: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9143: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9144: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9145: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9146: k = 1 2 3 4 5 6 7 8 9
9147: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9148: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9149: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9150: Dummy[k] 1 0 0 0 3 1 1 2 3
9151: Tmodelind[combination of covar]=k;
1.225 brouard 9152: */
9153: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9154: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9155: /* 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 9156: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9157: printf("Model=%s\n\
9158: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9159: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9160: 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);
9161: fprintf(ficlog,"Model=%s\n\
9162: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9163: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9164: 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 9165: for(k=1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9166: 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 */
9167: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 9168: Fixed[k]= 0;
9169: Dummy[k]= 0;
1.225 brouard 9170: ncoveff++;
1.232 brouard 9171: ncovf++;
1.234 brouard 9172: nsd++;
9173: modell[k].maintype= FTYPE;
9174: TvarsD[nsd]=Tvar[k];
9175: TvarsDind[nsd]=k;
9176: TvarF[ncovf]=Tvar[k];
9177: TvarFind[ncovf]=k;
9178: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9179: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9180: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
9181: Fixed[k]= 0;
9182: Dummy[k]= 0;
9183: ncoveff++;
9184: ncovf++;
9185: modell[k].maintype= FTYPE;
9186: TvarF[ncovf]=Tvar[k];
9187: TvarFind[ncovf]=k;
1.230 brouard 9188: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 9189: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 9190: }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 9191: Fixed[k]= 0;
9192: Dummy[k]= 1;
1.230 brouard 9193: nqfveff++;
1.234 brouard 9194: modell[k].maintype= FTYPE;
9195: modell[k].subtype= FQ;
9196: nsq++;
9197: TvarsQ[nsq]=Tvar[k];
9198: TvarsQind[nsq]=k;
1.232 brouard 9199: ncovf++;
1.234 brouard 9200: TvarF[ncovf]=Tvar[k];
9201: TvarFind[ncovf]=k;
1.231 brouard 9202: 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 9203: 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 9204: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 9205: Fixed[k]= 1;
9206: Dummy[k]= 0;
1.225 brouard 9207: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 9208: modell[k].maintype= VTYPE;
9209: modell[k].subtype= VD;
9210: nsd++;
9211: TvarsD[nsd]=Tvar[k];
9212: TvarsDind[nsd]=k;
9213: ncovv++; /* Only simple time varying variables */
9214: TvarV[ncovv]=Tvar[k];
1.242 brouard 9215: 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 9216: 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 */
9217: 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 9218: 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);
9219: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 9220: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 9221: Fixed[k]= 1;
9222: Dummy[k]= 1;
9223: nqtveff++;
9224: modell[k].maintype= VTYPE;
9225: modell[k].subtype= VQ;
9226: ncovv++; /* Only simple time varying variables */
9227: nsq++;
9228: TvarsQ[nsq]=Tvar[k];
9229: TvarsQind[nsq]=k;
9230: TvarV[ncovv]=Tvar[k];
1.242 brouard 9231: 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 9232: 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 */
9233: 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 9234: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
9235: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
9236: 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 9237: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 9238: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 9239: ncova++;
9240: TvarA[ncova]=Tvar[k];
9241: TvarAind[ncova]=k;
1.231 brouard 9242: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 9243: Fixed[k]= 2;
9244: Dummy[k]= 2;
9245: modell[k].maintype= ATYPE;
9246: modell[k].subtype= APFD;
9247: /* ncoveff++; */
1.227 brouard 9248: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 9249: Fixed[k]= 2;
9250: Dummy[k]= 3;
9251: modell[k].maintype= ATYPE;
9252: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
9253: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 9254: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 9255: Fixed[k]= 3;
9256: Dummy[k]= 2;
9257: modell[k].maintype= ATYPE;
9258: modell[k].subtype= APVD; /* Product age * varying dummy */
9259: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 9260: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9261: Fixed[k]= 3;
9262: Dummy[k]= 3;
9263: modell[k].maintype= ATYPE;
9264: modell[k].subtype= APVQ; /* Product age * varying quantitative */
9265: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 9266: }
9267: }else if (Typevar[k] == 2) { /* product without age */
9268: k1=Tposprod[k];
9269: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 9270: if(Tvard[k1][2] <=ncovcol){
9271: Fixed[k]= 1;
9272: Dummy[k]= 0;
9273: modell[k].maintype= FTYPE;
9274: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
9275: ncovf++; /* Fixed variables without age */
9276: TvarF[ncovf]=Tvar[k];
9277: TvarFind[ncovf]=k;
9278: }else if(Tvard[k1][2] <=ncovcol+nqv){
9279: Fixed[k]= 0; /* or 2 ?*/
9280: Dummy[k]= 1;
9281: modell[k].maintype= FTYPE;
9282: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
9283: ncovf++; /* Varying variables without age */
9284: TvarF[ncovf]=Tvar[k];
9285: TvarFind[ncovf]=k;
9286: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9287: Fixed[k]= 1;
9288: Dummy[k]= 0;
9289: modell[k].maintype= VTYPE;
9290: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
9291: ncovv++; /* Varying variables without age */
9292: TvarV[ncovv]=Tvar[k];
9293: TvarVind[ncovv]=k;
9294: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9295: Fixed[k]= 1;
9296: Dummy[k]= 1;
9297: modell[k].maintype= VTYPE;
9298: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
9299: ncovv++; /* Varying variables without age */
9300: TvarV[ncovv]=Tvar[k];
9301: TvarVind[ncovv]=k;
9302: }
1.227 brouard 9303: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 9304: if(Tvard[k1][2] <=ncovcol){
9305: Fixed[k]= 0; /* or 2 ?*/
9306: Dummy[k]= 1;
9307: modell[k].maintype= FTYPE;
9308: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
9309: ncovf++; /* Fixed variables without age */
9310: TvarF[ncovf]=Tvar[k];
9311: TvarFind[ncovf]=k;
9312: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9313: Fixed[k]= 1;
9314: Dummy[k]= 1;
9315: modell[k].maintype= VTYPE;
9316: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
9317: ncovv++; /* Varying variables without age */
9318: TvarV[ncovv]=Tvar[k];
9319: TvarVind[ncovv]=k;
9320: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9321: Fixed[k]= 1;
9322: Dummy[k]= 1;
9323: modell[k].maintype= VTYPE;
9324: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
9325: ncovv++; /* Varying variables without age */
9326: TvarV[ncovv]=Tvar[k];
9327: TvarVind[ncovv]=k;
9328: ncovv++; /* Varying variables without age */
9329: TvarV[ncovv]=Tvar[k];
9330: TvarVind[ncovv]=k;
9331: }
1.227 brouard 9332: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 9333: if(Tvard[k1][2] <=ncovcol){
9334: Fixed[k]= 1;
9335: Dummy[k]= 1;
9336: modell[k].maintype= VTYPE;
9337: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
9338: ncovv++; /* Varying variables without age */
9339: TvarV[ncovv]=Tvar[k];
9340: TvarVind[ncovv]=k;
9341: }else if(Tvard[k1][2] <=ncovcol+nqv){
9342: Fixed[k]= 1;
9343: Dummy[k]= 1;
9344: modell[k].maintype= VTYPE;
9345: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
9346: ncovv++; /* Varying variables without age */
9347: TvarV[ncovv]=Tvar[k];
9348: TvarVind[ncovv]=k;
9349: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9350: Fixed[k]= 1;
9351: Dummy[k]= 0;
9352: modell[k].maintype= VTYPE;
9353: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
9354: ncovv++; /* Varying variables without age */
9355: TvarV[ncovv]=Tvar[k];
9356: TvarVind[ncovv]=k;
9357: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9358: Fixed[k]= 1;
9359: Dummy[k]= 1;
9360: modell[k].maintype= VTYPE;
9361: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
9362: ncovv++; /* Varying variables without age */
9363: TvarV[ncovv]=Tvar[k];
9364: TvarVind[ncovv]=k;
9365: }
1.227 brouard 9366: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 9367: if(Tvard[k1][2] <=ncovcol){
9368: Fixed[k]= 1;
9369: Dummy[k]= 1;
9370: modell[k].maintype= VTYPE;
9371: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
9372: ncovv++; /* Varying variables without age */
9373: TvarV[ncovv]=Tvar[k];
9374: TvarVind[ncovv]=k;
9375: }else if(Tvard[k1][2] <=ncovcol+nqv){
9376: Fixed[k]= 1;
9377: Dummy[k]= 1;
9378: modell[k].maintype= VTYPE;
9379: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
9380: ncovv++; /* Varying variables without age */
9381: TvarV[ncovv]=Tvar[k];
9382: TvarVind[ncovv]=k;
9383: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
9384: Fixed[k]= 1;
9385: Dummy[k]= 1;
9386: modell[k].maintype= VTYPE;
9387: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
9388: ncovv++; /* Varying variables without age */
9389: TvarV[ncovv]=Tvar[k];
9390: TvarVind[ncovv]=k;
9391: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
9392: Fixed[k]= 1;
9393: Dummy[k]= 1;
9394: modell[k].maintype= VTYPE;
9395: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
9396: ncovv++; /* Varying variables without age */
9397: TvarV[ncovv]=Tvar[k];
9398: TvarVind[ncovv]=k;
9399: }
1.227 brouard 9400: }else{
1.240 brouard 9401: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9402: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
9403: } /*end k1*/
1.225 brouard 9404: }else{
1.226 brouard 9405: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
9406: 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 9407: }
1.227 brouard 9408: 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 9409: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 9410: 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]);
9411: }
9412: /* Searching for doublons in the model */
9413: for(k1=1; k1<= cptcovt;k1++){
9414: for(k2=1; k2 <k1;k2++){
9415: if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){
1.234 brouard 9416: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
9417: if(Tvar[k1]==Tvar[k2]){
9418: 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]]);
9419: 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);
9420: return(1);
9421: }
9422: }else if (Typevar[k1] ==2){
9423: k3=Tposprod[k1];
9424: k4=Tposprod[k2];
9425: 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])) ){
9426: 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]]);
9427: 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);
9428: return(1);
9429: }
9430: }
1.227 brouard 9431: }
9432: }
1.225 brouard 9433: }
9434: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
9435: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 9436: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
9437: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 9438: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 9439: /*endread:*/
1.225 brouard 9440: printf("Exiting decodemodel: ");
9441: return (1);
1.136 brouard 9442: }
9443:
1.169 brouard 9444: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 9445: {/* Check ages at death */
1.136 brouard 9446: int i, m;
1.218 brouard 9447: int firstone=0;
9448:
1.136 brouard 9449: for (i=1; i<=imx; i++) {
9450: for(m=2; (m<= maxwav); m++) {
9451: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
9452: anint[m][i]=9999;
1.216 brouard 9453: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
9454: s[m][i]=-1;
1.136 brouard 9455: }
9456: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 9457: *nberr = *nberr + 1;
1.218 brouard 9458: if(firstone == 0){
9459: firstone=1;
1.260 brouard 9460: printf("Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 9461: }
1.262 brouard 9462: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.260 brouard 9463: s[m][i]=-1; /* Droping the death status */
1.136 brouard 9464: }
9465: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 9466: (*nberr)++;
1.259 brouard 9467: printf("Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.262 brouard 9468: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.259 brouard 9469: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 9470: }
9471: }
9472: }
9473:
9474: for (i=1; i<=imx; i++) {
9475: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
9476: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 9477: 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 9478: if (s[m][i] >= nlstate+1) {
1.169 brouard 9479: if(agedc[i]>0){
9480: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 9481: agev[m][i]=agedc[i];
1.214 brouard 9482: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 9483: }else {
1.136 brouard 9484: if ((int)andc[i]!=9999){
9485: nbwarn++;
9486: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
9487: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
9488: agev[m][i]=-1;
9489: }
9490: }
1.169 brouard 9491: } /* agedc > 0 */
1.214 brouard 9492: } /* end if */
1.136 brouard 9493: else if(s[m][i] !=9){ /* Standard case, age in fractional
9494: years but with the precision of a month */
9495: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
9496: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
9497: agev[m][i]=1;
9498: else if(agev[m][i] < *agemin){
9499: *agemin=agev[m][i];
9500: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
9501: }
9502: else if(agev[m][i] >*agemax){
9503: *agemax=agev[m][i];
1.156 brouard 9504: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 9505: }
9506: /*agev[m][i]=anint[m][i]-annais[i];*/
9507: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 9508: } /* en if 9*/
1.136 brouard 9509: else { /* =9 */
1.214 brouard 9510: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 9511: agev[m][i]=1;
9512: s[m][i]=-1;
9513: }
9514: }
1.214 brouard 9515: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 9516: agev[m][i]=1;
1.214 brouard 9517: else{
9518: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9519: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
9520: agev[m][i]=0;
9521: }
9522: } /* End for lastpass */
9523: }
1.136 brouard 9524:
9525: for (i=1; i<=imx; i++) {
9526: for(m=firstpass; (m<=lastpass); m++){
9527: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 9528: (*nberr)++;
1.136 brouard 9529: 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);
9530: 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);
9531: return 1;
9532: }
9533: }
9534: }
9535:
9536: /*for (i=1; i<=imx; i++){
9537: for (m=firstpass; (m<lastpass); m++){
9538: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
9539: }
9540:
9541: }*/
9542:
9543:
1.139 brouard 9544: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
9545: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 9546:
9547: return (0);
1.164 brouard 9548: /* endread:*/
1.136 brouard 9549: printf("Exiting calandcheckages: ");
9550: return (1);
9551: }
9552:
1.172 brouard 9553: #if defined(_MSC_VER)
9554: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9555: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
9556: //#include "stdafx.h"
9557: //#include <stdio.h>
9558: //#include <tchar.h>
9559: //#include <windows.h>
9560: //#include <iostream>
9561: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
9562:
9563: LPFN_ISWOW64PROCESS fnIsWow64Process;
9564:
9565: BOOL IsWow64()
9566: {
9567: BOOL bIsWow64 = FALSE;
9568:
9569: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
9570: // (HANDLE, PBOOL);
9571:
9572: //LPFN_ISWOW64PROCESS fnIsWow64Process;
9573:
9574: HMODULE module = GetModuleHandle(_T("kernel32"));
9575: const char funcName[] = "IsWow64Process";
9576: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
9577: GetProcAddress(module, funcName);
9578:
9579: if (NULL != fnIsWow64Process)
9580: {
9581: if (!fnIsWow64Process(GetCurrentProcess(),
9582: &bIsWow64))
9583: //throw std::exception("Unknown error");
9584: printf("Unknown error\n");
9585: }
9586: return bIsWow64 != FALSE;
9587: }
9588: #endif
1.177 brouard 9589:
1.191 brouard 9590: void syscompilerinfo(int logged)
1.167 brouard 9591: {
9592: /* #include "syscompilerinfo.h"*/
1.185 brouard 9593: /* command line Intel compiler 32bit windows, XP compatible:*/
9594: /* /GS /W3 /Gy
9595: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
9596: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
9597: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 9598: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
9599: */
9600: /* 64 bits */
1.185 brouard 9601: /*
9602: /GS /W3 /Gy
9603: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
9604: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
9605: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
9606: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
9607: /* Optimization are useless and O3 is slower than O2 */
9608: /*
9609: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
9610: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
9611: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
9612: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
9613: */
1.186 brouard 9614: /* Link is */ /* /OUT:"visual studio
1.185 brouard 9615: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
9616: /PDB:"visual studio
9617: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
9618: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
9619: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
9620: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
9621: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
9622: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
9623: uiAccess='false'"
9624: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
9625: /NOLOGO /TLBID:1
9626: */
1.177 brouard 9627: #if defined __INTEL_COMPILER
1.178 brouard 9628: #if defined(__GNUC__)
9629: struct utsname sysInfo; /* For Intel on Linux and OS/X */
9630: #endif
1.177 brouard 9631: #elif defined(__GNUC__)
1.179 brouard 9632: #ifndef __APPLE__
1.174 brouard 9633: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 9634: #endif
1.177 brouard 9635: struct utsname sysInfo;
1.178 brouard 9636: int cross = CROSS;
9637: if (cross){
9638: printf("Cross-");
1.191 brouard 9639: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 9640: }
1.174 brouard 9641: #endif
9642:
1.171 brouard 9643: #include <stdint.h>
1.178 brouard 9644:
1.191 brouard 9645: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 9646: #if defined(__clang__)
1.191 brouard 9647: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 9648: #endif
9649: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 9650: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 9651: #endif
9652: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 9653: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 9654: #endif
9655: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 9656: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 9657: #endif
9658: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 9659: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 9660: #endif
9661: #if defined(_MSC_VER)
1.191 brouard 9662: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 9663: #endif
9664: #if defined(__PGI)
1.191 brouard 9665: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 9666: #endif
9667: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 9668: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 9669: #endif
1.191 brouard 9670: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 9671:
1.167 brouard 9672: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
9673: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
9674: // Windows (x64 and x86)
1.191 brouard 9675: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 9676: #elif __unix__ // all unices, not all compilers
9677: // Unix
1.191 brouard 9678: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 9679: #elif __linux__
9680: // linux
1.191 brouard 9681: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 9682: #elif __APPLE__
1.174 brouard 9683: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 9684: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 9685: #endif
9686:
9687: /* __MINGW32__ */
9688: /* __CYGWIN__ */
9689: /* __MINGW64__ */
9690: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
9691: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
9692: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
9693: /* _WIN64 // Defined for applications for Win64. */
9694: /* _M_X64 // Defined for compilations that target x64 processors. */
9695: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 9696:
1.167 brouard 9697: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 9698: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 9699: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 9700: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 9701: #else
1.191 brouard 9702: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 9703: #endif
9704:
1.169 brouard 9705: #if defined(__GNUC__)
9706: # if defined(__GNUC_PATCHLEVEL__)
9707: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9708: + __GNUC_MINOR__ * 100 \
9709: + __GNUC_PATCHLEVEL__)
9710: # else
9711: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
9712: + __GNUC_MINOR__ * 100)
9713: # endif
1.174 brouard 9714: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 9715: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 9716:
9717: if (uname(&sysInfo) != -1) {
9718: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 9719: 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 9720: }
9721: else
9722: perror("uname() error");
1.179 brouard 9723: //#ifndef __INTEL_COMPILER
9724: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 9725: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 9726: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 9727: #endif
1.169 brouard 9728: #endif
1.172 brouard 9729:
9730: // void main()
9731: // {
1.169 brouard 9732: #if defined(_MSC_VER)
1.174 brouard 9733: if (IsWow64()){
1.191 brouard 9734: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
9735: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 9736: }
9737: else{
1.191 brouard 9738: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
9739: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 9740: }
1.172 brouard 9741: // printf("\nPress Enter to continue...");
9742: // getchar();
9743: // }
9744:
1.169 brouard 9745: #endif
9746:
1.167 brouard 9747:
1.219 brouard 9748: }
1.136 brouard 9749:
1.219 brouard 9750: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.180 brouard 9751: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.235 brouard 9752: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 9753: /* double ftolpl = 1.e-10; */
1.180 brouard 9754: double age, agebase, agelim;
1.203 brouard 9755: double tot;
1.180 brouard 9756:
1.202 brouard 9757: strcpy(filerespl,"PL_");
9758: strcat(filerespl,fileresu);
9759: if((ficrespl=fopen(filerespl,"w"))==NULL) {
9760: printf("Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9761: fprintf(ficlog,"Problem with period (stable) prevalence resultfile: %s\n", filerespl);return 1;
9762: }
1.227 brouard 9763: printf("\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
9764: fprintf(ficlog,"\nComputing period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 9765: pstamp(ficrespl);
1.203 brouard 9766: fprintf(ficrespl,"# Period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 9767: fprintf(ficrespl,"#Age ");
9768: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
9769: fprintf(ficrespl,"\n");
1.180 brouard 9770:
1.219 brouard 9771: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 9772:
1.219 brouard 9773: agebase=ageminpar;
9774: agelim=agemaxpar;
1.180 brouard 9775:
1.227 brouard 9776: /* i1=pow(2,ncoveff); */
1.234 brouard 9777: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 9778: if (cptcovn < 1){i1=1;}
1.180 brouard 9779:
1.238 brouard 9780: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
9781: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 9782: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9783: continue;
1.235 brouard 9784:
1.238 brouard 9785: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9786: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
9787: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
9788: /* k=k+1; */
9789: /* to clean */
9790: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9791: fprintf(ficrespl,"#******");
9792: printf("#******");
9793: fprintf(ficlog,"#******");
9794: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9795: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
9796: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9797: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9798: }
9799: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9800: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9801: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9802: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9803: }
9804: fprintf(ficrespl,"******\n");
9805: printf("******\n");
9806: fprintf(ficlog,"******\n");
9807: if(invalidvarcomb[k]){
9808: printf("\nCombination (%d) ignored because no case \n",k);
9809: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
9810: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
9811: continue;
9812: }
1.219 brouard 9813:
1.238 brouard 9814: fprintf(ficrespl,"#Age ");
9815: for(j=1;j<=cptcoveff;j++) {
9816: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9817: }
9818: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
9819: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 9820:
1.238 brouard 9821: for (age=agebase; age<=agelim; age++){
9822: /* for (age=agebase; age<=agebase; age++){ */
9823: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
9824: fprintf(ficrespl,"%.0f ",age );
9825: for(j=1;j<=cptcoveff;j++)
9826: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9827: tot=0.;
9828: for(i=1; i<=nlstate;i++){
9829: tot += prlim[i][i];
9830: fprintf(ficrespl," %.5f", prlim[i][i]);
9831: }
9832: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
9833: } /* Age */
9834: /* was end of cptcod */
9835: } /* cptcov */
9836: } /* nres */
1.219 brouard 9837: return 0;
1.180 brouard 9838: }
9839:
1.218 brouard 9840: 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){
9841: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
9842:
9843: /* Computes the back prevalence limit for any combination of covariate values
9844: * at any age between ageminpar and agemaxpar
9845: */
1.235 brouard 9846: int i, j, k, i1, nres=0 ;
1.217 brouard 9847: /* double ftolpl = 1.e-10; */
9848: double age, agebase, agelim;
9849: double tot;
1.218 brouard 9850: /* double ***mobaverage; */
9851: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 9852:
9853: strcpy(fileresplb,"PLB_");
9854: strcat(fileresplb,fileresu);
9855: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
9856: printf("Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9857: fprintf(ficlog,"Problem with period (stable) back prevalence resultfile: %s\n", fileresplb);return 1;
9858: }
9859: printf("Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9860: fprintf(ficlog,"Computing period (stable) back prevalence: result on file '%s' \n", fileresplb);
9861: pstamp(ficresplb);
9862: fprintf(ficresplb,"# Period (stable) back prevalence. Precision given by ftolpl=%g \n", ftolpl);
9863: fprintf(ficresplb,"#Age ");
9864: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
9865: fprintf(ficresplb,"\n");
9866:
1.218 brouard 9867:
9868: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
9869:
9870: agebase=ageminpar;
9871: agelim=agemaxpar;
9872:
9873:
1.227 brouard 9874: i1=pow(2,cptcoveff);
1.218 brouard 9875: if (cptcovn < 1){i1=1;}
1.227 brouard 9876:
1.238 brouard 9877: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9878: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 9879: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 9880: continue;
9881: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
9882: fprintf(ficresplb,"#******");
9883: printf("#******");
9884: fprintf(ficlog,"#******");
9885: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
9886: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9887: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9888: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9889: }
9890: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9891: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9892: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9893: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9894: }
9895: fprintf(ficresplb,"******\n");
9896: printf("******\n");
9897: fprintf(ficlog,"******\n");
9898: if(invalidvarcomb[k]){
9899: printf("\nCombination (%d) ignored because no cases \n",k);
9900: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
9901: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
9902: continue;
9903: }
1.218 brouard 9904:
1.238 brouard 9905: fprintf(ficresplb,"#Age ");
9906: for(j=1;j<=cptcoveff;j++) {
9907: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9908: }
9909: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
9910: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 9911:
9912:
1.238 brouard 9913: for (age=agebase; age<=agelim; age++){
9914: /* for (age=agebase; age<=agebase; age++){ */
9915: if(mobilavproj > 0){
9916: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
9917: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9918: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 9919: }else if (mobilavproj == 0){
9920: 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);
9921: 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);
9922: exit(1);
9923: }else{
9924: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 9925: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 9926: /* printf("TOTOT\n"); */
9927: /* exit(1); */
1.238 brouard 9928: }
9929: fprintf(ficresplb,"%.0f ",age );
9930: for(j=1;j<=cptcoveff;j++)
9931: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9932: tot=0.;
9933: for(i=1; i<=nlstate;i++){
9934: tot += bprlim[i][i];
9935: fprintf(ficresplb," %.5f", bprlim[i][i]);
9936: }
9937: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
9938: } /* Age */
9939: /* was end of cptcod */
1.255 brouard 9940: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 9941: } /* end of any combination */
9942: } /* end of nres */
1.218 brouard 9943: /* hBijx(p, bage, fage); */
9944: /* fclose(ficrespijb); */
9945:
9946: return 0;
1.217 brouard 9947: }
1.218 brouard 9948:
1.180 brouard 9949: int hPijx(double *p, int bage, int fage){
9950: /*------------- h Pij x at various ages ------------*/
9951:
9952: int stepsize;
9953: int agelim;
9954: int hstepm;
9955: int nhstepm;
1.235 brouard 9956: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 9957:
9958: double agedeb;
9959: double ***p3mat;
9960:
1.201 brouard 9961: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 9962: if((ficrespij=fopen(filerespij,"w"))==NULL) {
9963: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
9964: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
9965: }
9966: printf("Computing pij: result on file '%s' \n", filerespij);
9967: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
9968:
9969: stepsize=(int) (stepm+YEARM-1)/YEARM;
9970: /*if (stepm<=24) stepsize=2;*/
9971:
9972: agelim=AGESUP;
9973: hstepm=stepsize*YEARM; /* Every year of age */
9974: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 9975:
1.180 brouard 9976: /* hstepm=1; aff par mois*/
9977: pstamp(ficrespij);
9978: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 9979: i1= pow(2,cptcoveff);
1.218 brouard 9980: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
9981: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
9982: /* k=k+1; */
1.235 brouard 9983: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9984: for(k=1; k<=i1;k++){
1.253 brouard 9985: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9986: continue;
1.183 brouard 9987: fprintf(ficrespij,"\n#****** ");
1.227 brouard 9988: for(j=1;j<=cptcoveff;j++)
1.198 brouard 9989: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 9990: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9991: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9992: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9993: }
1.183 brouard 9994: fprintf(ficrespij,"******\n");
9995:
9996: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
9997: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
9998: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
9999:
10000: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10001:
1.183 brouard 10002: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10003: oldm=oldms;savm=savms;
1.235 brouard 10004: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10005: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10006: for(i=1; i<=nlstate;i++)
10007: for(j=1; j<=nlstate+ndeath;j++)
10008: fprintf(ficrespij," %1d-%1d",i,j);
10009: fprintf(ficrespij,"\n");
10010: for (h=0; h<=nhstepm; h++){
10011: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10012: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10013: for(i=1; i<=nlstate;i++)
10014: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10015: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10016: fprintf(ficrespij,"\n");
10017: }
1.183 brouard 10018: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10019: fprintf(ficrespij,"\n");
10020: }
1.180 brouard 10021: /*}*/
10022: }
1.218 brouard 10023: return 0;
1.180 brouard 10024: }
1.218 brouard 10025:
10026: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10027: /*------------- h Bij x at various ages ------------*/
10028:
10029: int stepsize;
1.218 brouard 10030: /* int agelim; */
10031: int ageminl;
1.217 brouard 10032: int hstepm;
10033: int nhstepm;
1.238 brouard 10034: int h, i, i1, j, k, nres;
1.218 brouard 10035:
1.217 brouard 10036: double agedeb;
10037: double ***p3mat;
1.218 brouard 10038:
10039: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10040: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10041: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10042: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10043: }
10044: printf("Computing pij back: result on file '%s' \n", filerespijb);
10045: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10046:
10047: stepsize=(int) (stepm+YEARM-1)/YEARM;
10048: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10049:
1.218 brouard 10050: /* agelim=AGESUP; */
10051: ageminl=30;
10052: hstepm=stepsize*YEARM; /* Every year of age */
10053: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10054:
10055: /* hstepm=1; aff par mois*/
10056: pstamp(ficrespijb);
1.255 brouard 10057: fprintf(ficrespijb,"#****** h Bij x Back probability to be in state i at age x-h being in j at x: B1j+B2j+...=1 ");
1.227 brouard 10058: i1= pow(2,cptcoveff);
1.218 brouard 10059: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10060: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10061: /* k=k+1; */
1.238 brouard 10062: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10063: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10064: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10065: continue;
10066: fprintf(ficrespijb,"\n#****** ");
10067: for(j=1;j<=cptcoveff;j++)
10068: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10069: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10070: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10071: }
10072: fprintf(ficrespijb,"******\n");
1.264 brouard 10073: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10074: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10075: continue;
10076: }
10077:
10078: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10079: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10080: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
10081: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10082: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 */
10083:
10084: /* nhstepm=nhstepm*YEARM; aff par mois*/
10085:
1.266 brouard 10086: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10087: /* and memory limitations if stepm is small */
10088:
1.238 brouard 10089: /* oldm=oldms;savm=savms; */
10090: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 ! brouard 10091: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10092: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10093: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10094: for(i=1; i<=nlstate;i++)
10095: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10096: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10097: fprintf(ficrespijb,"\n");
1.238 brouard 10098: for (h=0; h<=nhstepm; h++){
10099: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10100: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10101: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10102: for(i=1; i<=nlstate;i++)
10103: for(j=1; j<=nlstate+ndeath;j++)
10104: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10105: fprintf(ficrespijb,"\n");
10106: }
10107: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10108: fprintf(ficrespijb,"\n");
10109: } /* end age deb */
10110: } /* end combination */
10111: } /* end nres */
1.218 brouard 10112: return 0;
10113: } /* hBijx */
1.217 brouard 10114:
1.180 brouard 10115:
1.136 brouard 10116: /***********************************************/
10117: /**************** Main Program *****************/
10118: /***********************************************/
10119:
10120: int main(int argc, char *argv[])
10121: {
10122: #ifdef GSL
10123: const gsl_multimin_fminimizer_type *T;
10124: size_t iteri = 0, it;
10125: int rval = GSL_CONTINUE;
10126: int status = GSL_SUCCESS;
10127: double ssval;
10128: #endif
10129: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.164 brouard 10130: int i,j, k, n=MAXN,iter=0,m,size=100, cptcod;
1.209 brouard 10131: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10132: int jj, ll, li, lj, lk;
1.136 brouard 10133: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10134: int num_filled;
1.136 brouard 10135: int itimes;
10136: int NDIM=2;
10137: int vpopbased=0;
1.235 brouard 10138: int nres=0;
1.258 brouard 10139: int endishere=0;
1.136 brouard 10140:
1.164 brouard 10141: char ca[32], cb[32];
1.136 brouard 10142: /* FILE *fichtm; *//* Html File */
10143: /* FILE *ficgp;*/ /*Gnuplot File */
10144: struct stat info;
1.191 brouard 10145: double agedeb=0.;
1.194 brouard 10146:
10147: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10148: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10149:
1.165 brouard 10150: double fret;
1.191 brouard 10151: double dum=0.; /* Dummy variable */
1.136 brouard 10152: double ***p3mat;
1.218 brouard 10153: /* double ***mobaverage; */
1.164 brouard 10154:
10155: char line[MAXLINE];
1.197 brouard 10156: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10157:
1.234 brouard 10158: char modeltemp[MAXLINE];
1.230 brouard 10159: char resultline[MAXLINE];
10160:
1.136 brouard 10161: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 10162: char *tok, *val; /* pathtot */
1.136 brouard 10163: int firstobs=1, lastobs=10;
1.195 brouard 10164: int c, h , cpt, c2;
1.191 brouard 10165: int jl=0;
10166: int i1, j1, jk, stepsize=0;
1.194 brouard 10167: int count=0;
10168:
1.164 brouard 10169: int *tab;
1.136 brouard 10170: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.217 brouard 10171: int backcast=0;
1.136 brouard 10172: int mobilav=0,popforecast=0;
1.191 brouard 10173: int hstepm=0, nhstepm=0;
1.136 brouard 10174: int agemortsup;
10175: float sumlpop=0.;
10176: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
10177: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
10178:
1.191 brouard 10179: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 10180: double ftolpl=FTOL;
10181: double **prlim;
1.217 brouard 10182: double **bprlim;
1.136 brouard 10183: double ***param; /* Matrix of parameters */
1.251 brouard 10184: double ***paramstart; /* Matrix of starting parameter values */
10185: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 10186: double **matcov; /* Matrix of covariance */
1.203 brouard 10187: double **hess; /* Hessian matrix */
1.136 brouard 10188: double ***delti3; /* Scale */
10189: double *delti; /* Scale */
10190: double ***eij, ***vareij;
10191: double **varpl; /* Variances of prevalence limits by age */
10192: double *epj, vepp;
1.164 brouard 10193:
1.136 brouard 10194: double dateprev1, dateprev2,jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000;
1.217 brouard 10195: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000;
10196:
1.136 brouard 10197: double **ximort;
1.145 brouard 10198: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 10199: int *dcwave;
10200:
1.164 brouard 10201: char z[1]="c";
1.136 brouard 10202:
10203: /*char *strt;*/
10204: char strtend[80];
1.126 brouard 10205:
1.164 brouard 10206:
1.126 brouard 10207: /* setlocale (LC_ALL, ""); */
10208: /* bindtextdomain (PACKAGE, LOCALEDIR); */
10209: /* textdomain (PACKAGE); */
10210: /* setlocale (LC_CTYPE, ""); */
10211: /* setlocale (LC_MESSAGES, ""); */
10212:
10213: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 10214: rstart_time = time(NULL);
10215: /* (void) gettimeofday(&start_time,&tzp);*/
10216: start_time = *localtime(&rstart_time);
1.126 brouard 10217: curr_time=start_time;
1.157 brouard 10218: /*tml = *localtime(&start_time.tm_sec);*/
10219: /* strcpy(strstart,asctime(&tml)); */
10220: strcpy(strstart,asctime(&start_time));
1.126 brouard 10221:
10222: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 10223: /* tp.tm_sec = tp.tm_sec +86400; */
10224: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 10225: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
10226: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
10227: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 10228: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 10229: /* strt=asctime(&tmg); */
10230: /* printf("Time(after) =%s",strstart); */
10231: /* (void) time (&time_value);
10232: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
10233: * tm = *localtime(&time_value);
10234: * strstart=asctime(&tm);
10235: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
10236: */
10237:
10238: nberr=0; /* Number of errors and warnings */
10239: nbwarn=0;
1.184 brouard 10240: #ifdef WIN32
10241: _getcwd(pathcd, size);
10242: #else
1.126 brouard 10243: getcwd(pathcd, size);
1.184 brouard 10244: #endif
1.191 brouard 10245: syscompilerinfo(0);
1.196 brouard 10246: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 10247: if(argc <=1){
10248: printf("\nEnter the parameter file name: ");
1.205 brouard 10249: if(!fgets(pathr,FILENAMELENGTH,stdin)){
10250: printf("ERROR Empty parameter file name\n");
10251: goto end;
10252: }
1.126 brouard 10253: i=strlen(pathr);
10254: if(pathr[i-1]=='\n')
10255: pathr[i-1]='\0';
1.156 brouard 10256: i=strlen(pathr);
1.205 brouard 10257: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 10258: pathr[i-1]='\0';
1.205 brouard 10259: }
10260: i=strlen(pathr);
10261: if( i==0 ){
10262: printf("ERROR Empty parameter file name\n");
10263: goto end;
10264: }
10265: for (tok = pathr; tok != NULL; ){
1.126 brouard 10266: printf("Pathr |%s|\n",pathr);
10267: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
10268: printf("val= |%s| pathr=%s\n",val,pathr);
10269: strcpy (pathtot, val);
10270: if(pathr[0] == '\0') break; /* Dirty */
10271: }
10272: }
10273: else{
10274: strcpy(pathtot,argv[1]);
10275: }
10276: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
10277: /*cygwin_split_path(pathtot,path,optionfile);
10278: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
10279: /* cutv(path,optionfile,pathtot,'\\');*/
10280:
10281: /* Split argv[0], imach program to get pathimach */
10282: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
10283: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10284: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
10285: /* strcpy(pathimach,argv[0]); */
10286: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
10287: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
10288: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 10289: #ifdef WIN32
10290: _chdir(path); /* Can be a relative path */
10291: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
10292: #else
1.126 brouard 10293: chdir(path); /* Can be a relative path */
1.184 brouard 10294: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
10295: #endif
10296: printf("Current directory %s!\n",pathcd);
1.126 brouard 10297: strcpy(command,"mkdir ");
10298: strcat(command,optionfilefiname);
10299: if((outcmd=system(command)) != 0){
1.169 brouard 10300: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 10301: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
10302: /* fclose(ficlog); */
10303: /* exit(1); */
10304: }
10305: /* if((imk=mkdir(optionfilefiname))<0){ */
10306: /* perror("mkdir"); */
10307: /* } */
10308:
10309: /*-------- arguments in the command line --------*/
10310:
1.186 brouard 10311: /* Main Log file */
1.126 brouard 10312: strcat(filelog, optionfilefiname);
10313: strcat(filelog,".log"); /* */
10314: if((ficlog=fopen(filelog,"w"))==NULL) {
10315: printf("Problem with logfile %s\n",filelog);
10316: goto end;
10317: }
10318: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 10319: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 10320: fprintf(ficlog,"\nEnter the parameter file name: \n");
10321: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
10322: path=%s \n\
10323: optionfile=%s\n\
10324: optionfilext=%s\n\
1.156 brouard 10325: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 10326:
1.197 brouard 10327: syscompilerinfo(1);
1.167 brouard 10328:
1.126 brouard 10329: printf("Local time (at start):%s",strstart);
10330: fprintf(ficlog,"Local time (at start): %s",strstart);
10331: fflush(ficlog);
10332: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 10333: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 10334:
10335: /* */
10336: strcpy(fileres,"r");
10337: strcat(fileres, optionfilefiname);
1.201 brouard 10338: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 10339: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 10340: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 10341:
1.186 brouard 10342: /* Main ---------arguments file --------*/
1.126 brouard 10343:
10344: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 10345: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
10346: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 10347: fflush(ficlog);
1.149 brouard 10348: /* goto end; */
10349: exit(70);
1.126 brouard 10350: }
10351:
10352:
10353:
10354: strcpy(filereso,"o");
1.201 brouard 10355: strcat(filereso,fileresu);
1.126 brouard 10356: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
10357: printf("Problem with Output resultfile: %s\n", filereso);
10358: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
10359: fflush(ficlog);
10360: goto end;
10361: }
10362:
10363: /* Reads comments: lines beginning with '#' */
10364: numlinepar=0;
1.197 brouard 10365:
10366: /* First parameter line */
10367: while(fgets(line, MAXLINE, ficpar)) {
10368: /* If line starts with a # it is a comment */
10369: if (line[0] == '#') {
10370: numlinepar++;
10371: fputs(line,stdout);
10372: fputs(line,ficparo);
10373: fputs(line,ficlog);
10374: continue;
10375: }else
10376: break;
10377: }
10378: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
10379: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
10380: if (num_filled != 5) {
10381: printf("Should be 5 parameters\n");
10382: }
1.126 brouard 10383: numlinepar++;
1.197 brouard 10384: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
10385: }
10386: /* Second parameter line */
10387: while(fgets(line, MAXLINE, ficpar)) {
10388: /* If line starts with a # it is a comment */
10389: if (line[0] == '#') {
10390: numlinepar++;
10391: fputs(line,stdout);
10392: fputs(line,ficparo);
10393: fputs(line,ficlog);
10394: continue;
10395: }else
10396: break;
10397: }
1.223 brouard 10398: 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", \
10399: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
10400: if (num_filled != 11) {
10401: 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 10402: printf("but line=%s\n",line);
1.197 brouard 10403: }
1.223 brouard 10404: 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 10405: }
1.203 brouard 10406: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 10407: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 10408: /* Third parameter line */
10409: while(fgets(line, MAXLINE, ficpar)) {
10410: /* If line starts with a # it is a comment */
10411: if (line[0] == '#') {
10412: numlinepar++;
10413: fputs(line,stdout);
10414: fputs(line,ficparo);
10415: fputs(line,ficlog);
10416: continue;
10417: }else
10418: break;
10419: }
1.201 brouard 10420: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.263 brouard 10421: if (num_filled == 0){
10422: printf("ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10423: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' WITHOUT space:'%s'\n",num_filled, line);
10424: model[0]='\0';
10425: goto end;
10426: } else if (num_filled != 1){
1.197 brouard 10427: printf("ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10428: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age.' %s\n",num_filled, line);
10429: model[0]='\0';
10430: goto end;
10431: }
10432: else{
10433: if (model[0]=='+'){
10434: for(i=1; i<=strlen(model);i++)
10435: modeltemp[i-1]=model[i];
1.201 brouard 10436: strcpy(model,modeltemp);
1.197 brouard 10437: }
10438: }
1.199 brouard 10439: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 10440: printf("model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 10441: }
10442: /* 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); */
10443: /* numlinepar=numlinepar+3; /\* In general *\/ */
10444: /* 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 10445: 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);
10446: 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 10447: fflush(ficlog);
1.190 brouard 10448: /* if(model[0]=='#'|| model[0]== '\0'){ */
10449: if(model[0]=='#'){
1.187 brouard 10450: printf("Error in 'model' line: model should start with 'model=1+age+' and end with '.' \n \
10451: 'model=1+age+.' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age.' or \n \
10452: 'model=1+age+V1+V2.' or 'model=1+age+V1+V2+V1*V2.' etc. \n"); \
10453: if(mle != -1){
10454: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter file.\n");
10455: exit(1);
10456: }
10457: }
1.126 brouard 10458: while((c=getc(ficpar))=='#' && c!= EOF){
10459: ungetc(c,ficpar);
10460: fgets(line, MAXLINE, ficpar);
10461: numlinepar++;
1.195 brouard 10462: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
10463: z[0]=line[1];
10464: }
10465: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 10466: fputs(line, stdout);
10467: //puts(line);
1.126 brouard 10468: fputs(line,ficparo);
10469: fputs(line,ficlog);
10470: }
10471: ungetc(c,ficpar);
10472:
10473:
1.145 brouard 10474: covar=matrix(0,NCOVMAX,1,n); /**< used in readdata */
1.225 brouard 10475: coqvar=matrix(1,nqv,1,n); /**< Fixed quantitative covariate */
1.233 brouard 10476: cotvar=ma3x(1,maxwav,1,ntv+nqtv,1,n); /**< Time varying covariate (dummy and quantitative)*/
1.225 brouard 10477: cotqvar=ma3x(1,maxwav,1,nqtv,1,n); /**< Time varying quantitative covariate */
1.136 brouard 10478: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
10479: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
10480: v1+v2*age+v2*v3 makes cptcovn = 3
10481: */
10482: if (strlen(model)>1)
1.187 brouard 10483: 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 10484: else
1.187 brouard 10485: ncovmodel=2; /* Constant and age */
1.133 brouard 10486: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
10487: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 10488: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
10489: 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);
10490: 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);
10491: fflush(stdout);
10492: fclose (ficlog);
10493: goto end;
10494: }
1.126 brouard 10495: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10496: delti=delti3[1][1];
10497: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
10498: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 10499: /* We could also provide initial parameters values giving by simple logistic regression
10500: * only one way, that is without matrix product. We will have nlstate maximizations */
10501: /* for(i=1;i<nlstate;i++){ */
10502: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
10503: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
10504: /* } */
1.126 brouard 10505: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 10506: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
10507: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10508: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
10509: fclose (ficparo);
10510: fclose (ficlog);
10511: goto end;
10512: exit(0);
1.220 brouard 10513: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 10514: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 10515: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
10516: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 10517: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
10518: matcov=matrix(1,npar,1,npar);
1.203 brouard 10519: hess=matrix(1,npar,1,npar);
1.220 brouard 10520: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 10521: /* Read guessed parameters */
1.126 brouard 10522: /* Reads comments: lines beginning with '#' */
10523: while((c=getc(ficpar))=='#' && c!= EOF){
10524: ungetc(c,ficpar);
10525: fgets(line, MAXLINE, ficpar);
10526: numlinepar++;
1.141 brouard 10527: fputs(line,stdout);
1.126 brouard 10528: fputs(line,ficparo);
10529: fputs(line,ficlog);
10530: }
10531: ungetc(c,ficpar);
10532:
10533: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 10534: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 10535: for(i=1; i <=nlstate; i++){
1.234 brouard 10536: j=0;
1.126 brouard 10537: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 10538: if(jj==i) continue;
10539: j++;
10540: fscanf(ficpar,"%1d%1d",&i1,&j1);
10541: if ((i1 != i) || (j1 != jj)){
10542: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 10543: It might be a problem of design; if ncovcol and the model are correct\n \
10544: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 10545: exit(1);
10546: }
10547: fprintf(ficparo,"%1d%1d",i1,j1);
10548: if(mle==1)
10549: printf("%1d%1d",i,jj);
10550: fprintf(ficlog,"%1d%1d",i,jj);
10551: for(k=1; k<=ncovmodel;k++){
10552: fscanf(ficpar," %lf",¶m[i][j][k]);
10553: if(mle==1){
10554: printf(" %lf",param[i][j][k]);
10555: fprintf(ficlog," %lf",param[i][j][k]);
10556: }
10557: else
10558: fprintf(ficlog," %lf",param[i][j][k]);
10559: fprintf(ficparo," %lf",param[i][j][k]);
10560: }
10561: fscanf(ficpar,"\n");
10562: numlinepar++;
10563: if(mle==1)
10564: printf("\n");
10565: fprintf(ficlog,"\n");
10566: fprintf(ficparo,"\n");
1.126 brouard 10567: }
10568: }
10569: fflush(ficlog);
1.234 brouard 10570:
1.251 brouard 10571: /* Reads parameters values */
1.126 brouard 10572: p=param[1][1];
1.251 brouard 10573: pstart=paramstart[1][1];
1.126 brouard 10574:
10575: /* Reads comments: lines beginning with '#' */
10576: while((c=getc(ficpar))=='#' && c!= EOF){
10577: ungetc(c,ficpar);
10578: fgets(line, MAXLINE, ficpar);
10579: numlinepar++;
1.141 brouard 10580: fputs(line,stdout);
1.126 brouard 10581: fputs(line,ficparo);
10582: fputs(line,ficlog);
10583: }
10584: ungetc(c,ficpar);
10585:
10586: for(i=1; i <=nlstate; i++){
10587: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 10588: fscanf(ficpar,"%1d%1d",&i1,&j1);
10589: if ( (i1-i) * (j1-j) != 0){
10590: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
10591: exit(1);
10592: }
10593: printf("%1d%1d",i,j);
10594: fprintf(ficparo,"%1d%1d",i1,j1);
10595: fprintf(ficlog,"%1d%1d",i1,j1);
10596: for(k=1; k<=ncovmodel;k++){
10597: fscanf(ficpar,"%le",&delti3[i][j][k]);
10598: printf(" %le",delti3[i][j][k]);
10599: fprintf(ficparo," %le",delti3[i][j][k]);
10600: fprintf(ficlog," %le",delti3[i][j][k]);
10601: }
10602: fscanf(ficpar,"\n");
10603: numlinepar++;
10604: printf("\n");
10605: fprintf(ficparo,"\n");
10606: fprintf(ficlog,"\n");
1.126 brouard 10607: }
10608: }
10609: fflush(ficlog);
1.234 brouard 10610:
1.145 brouard 10611: /* Reads covariance matrix */
1.126 brouard 10612: delti=delti3[1][1];
1.220 brouard 10613:
10614:
1.126 brouard 10615: /* 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 10616:
1.126 brouard 10617: /* Reads comments: lines beginning with '#' */
10618: while((c=getc(ficpar))=='#' && c!= EOF){
10619: ungetc(c,ficpar);
10620: fgets(line, MAXLINE, ficpar);
10621: numlinepar++;
1.141 brouard 10622: fputs(line,stdout);
1.126 brouard 10623: fputs(line,ficparo);
10624: fputs(line,ficlog);
10625: }
10626: ungetc(c,ficpar);
1.220 brouard 10627:
1.126 brouard 10628: matcov=matrix(1,npar,1,npar);
1.203 brouard 10629: hess=matrix(1,npar,1,npar);
1.131 brouard 10630: for(i=1; i <=npar; i++)
10631: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 10632:
1.194 brouard 10633: /* Scans npar lines */
1.126 brouard 10634: for(i=1; i <=npar; i++){
1.226 brouard 10635: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 10636: if(count != 3){
1.226 brouard 10637: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10638: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10639: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10640: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 10641: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
10642: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 10643: exit(1);
1.220 brouard 10644: }else{
1.226 brouard 10645: if(mle==1)
10646: printf("%1d%1d%d",i1,j1,jk);
10647: }
10648: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
10649: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 10650: for(j=1; j <=i; j++){
1.226 brouard 10651: fscanf(ficpar," %le",&matcov[i][j]);
10652: if(mle==1){
10653: printf(" %.5le",matcov[i][j]);
10654: }
10655: fprintf(ficlog," %.5le",matcov[i][j]);
10656: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 10657: }
10658: fscanf(ficpar,"\n");
10659: numlinepar++;
10660: if(mle==1)
1.220 brouard 10661: printf("\n");
1.126 brouard 10662: fprintf(ficlog,"\n");
10663: fprintf(ficparo,"\n");
10664: }
1.194 brouard 10665: /* End of read covariance matrix npar lines */
1.126 brouard 10666: for(i=1; i <=npar; i++)
10667: for(j=i+1;j<=npar;j++)
1.226 brouard 10668: matcov[i][j]=matcov[j][i];
1.126 brouard 10669:
10670: if(mle==1)
10671: printf("\n");
10672: fprintf(ficlog,"\n");
10673:
10674: fflush(ficlog);
10675:
10676: /*-------- Rewriting parameter file ----------*/
10677: strcpy(rfileres,"r"); /* "Rparameterfile */
10678: strcat(rfileres,optionfilefiname); /* Parameter file first name*/
10679: strcat(rfileres,"."); /* */
10680: strcat(rfileres,optionfilext); /* Other files have txt extension */
10681: if((ficres =fopen(rfileres,"w"))==NULL) {
1.201 brouard 10682: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
10683: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
1.126 brouard 10684: }
10685: fprintf(ficres,"#%s\n",version);
10686: } /* End of mle != -3 */
1.218 brouard 10687:
1.186 brouard 10688: /* Main data
10689: */
1.126 brouard 10690: n= lastobs;
10691: num=lvector(1,n);
10692: moisnais=vector(1,n);
10693: annais=vector(1,n);
10694: moisdc=vector(1,n);
10695: andc=vector(1,n);
1.220 brouard 10696: weight=vector(1,n);
1.126 brouard 10697: agedc=vector(1,n);
10698: cod=ivector(1,n);
1.220 brouard 10699: for(i=1;i<=n;i++){
1.234 brouard 10700: num[i]=0;
10701: moisnais[i]=0;
10702: annais[i]=0;
10703: moisdc[i]=0;
10704: andc[i]=0;
10705: agedc[i]=0;
10706: cod[i]=0;
10707: weight[i]=1.0; /* Equal weights, 1 by default */
10708: }
1.126 brouard 10709: mint=matrix(1,maxwav,1,n);
10710: anint=matrix(1,maxwav,1,n);
1.131 brouard 10711: s=imatrix(1,maxwav+1,1,n); /* s[i][j] health state for wave i and individual j */
1.126 brouard 10712: tab=ivector(1,NCOVMAX);
1.144 brouard 10713: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 10714: 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 10715:
1.136 brouard 10716: /* Reads data from file datafile */
10717: if (readdata(datafile, firstobs, lastobs, &imx)==1)
10718: goto end;
10719:
10720: /* Calculation of the number of parameters from char model */
1.234 brouard 10721: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 10722: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
10723: k=3 V4 Tvar[k=3]= 4 (from V4)
10724: k=2 V1 Tvar[k=2]= 1 (from V1)
10725: k=1 Tvar[1]=2 (from V2)
1.234 brouard 10726: */
10727:
10728: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
10729: TvarsDind=ivector(1,NCOVMAX); /* */
10730: TvarsD=ivector(1,NCOVMAX); /* */
10731: TvarsQind=ivector(1,NCOVMAX); /* */
10732: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 10733: TvarF=ivector(1,NCOVMAX); /* */
10734: TvarFind=ivector(1,NCOVMAX); /* */
10735: TvarV=ivector(1,NCOVMAX); /* */
10736: TvarVind=ivector(1,NCOVMAX); /* */
10737: TvarA=ivector(1,NCOVMAX); /* */
10738: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 10739: TvarFD=ivector(1,NCOVMAX); /* */
10740: TvarFDind=ivector(1,NCOVMAX); /* */
10741: TvarFQ=ivector(1,NCOVMAX); /* */
10742: TvarFQind=ivector(1,NCOVMAX); /* */
10743: TvarVD=ivector(1,NCOVMAX); /* */
10744: TvarVDind=ivector(1,NCOVMAX); /* */
10745: TvarVQ=ivector(1,NCOVMAX); /* */
10746: TvarVQind=ivector(1,NCOVMAX); /* */
10747:
1.230 brouard 10748: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 10749: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 10750: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
10751: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
10752: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 10753: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
10754: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
10755: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
10756: */
10757: /* For model-covariate k tells which data-covariate to use but
10758: because this model-covariate is a construction we invent a new column
10759: ncovcol + k1
10760: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
10761: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 10762: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
10763: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 10764: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
10765: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 10766: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 10767: */
1.145 brouard 10768: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
10769: 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 10770: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
10771: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 10772: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 10773: 4 covariates (3 plus signs)
10774: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
10775: */
1.230 brouard 10776: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 10777: * individual dummy, fixed or varying:
10778: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
10779: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 10780: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
10781: * V1 df, V2 qf, V3 & V4 dv, V5 qv
10782: * Tmodelind[1]@9={9,0,3,2,}*/
10783: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
10784: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 10785: * individual quantitative, fixed or varying:
10786: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
10787: * 3, 1, 0, 0, 0, 0, 0, 0},
10788: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 10789: /* Main decodemodel */
10790:
1.187 brouard 10791:
1.223 brouard 10792: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 10793: goto end;
10794:
1.137 brouard 10795: if((double)(lastobs-imx)/(double)imx > 1.10){
10796: nbwarn++;
10797: 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);
10798: 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);
10799: }
1.136 brouard 10800: /* if(mle==1){*/
1.137 brouard 10801: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
10802: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 10803: }
10804:
10805: /*-calculation of age at interview from date of interview and age at death -*/
10806: agev=matrix(1,maxwav,1,imx);
10807:
10808: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
10809: goto end;
10810:
1.126 brouard 10811:
1.136 brouard 10812: agegomp=(int)agemin;
10813: free_vector(moisnais,1,n);
10814: free_vector(annais,1,n);
1.126 brouard 10815: /* free_matrix(mint,1,maxwav,1,n);
10816: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 10817: /* free_vector(moisdc,1,n); */
10818: /* free_vector(andc,1,n); */
1.145 brouard 10819: /* */
10820:
1.126 brouard 10821: wav=ivector(1,imx);
1.214 brouard 10822: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
10823: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
10824: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
10825: 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.*/
10826: bh=imatrix(1,lastpass-firstpass+2,1,imx);
10827: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 10828:
10829: /* Concatenates waves */
1.214 brouard 10830: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
10831: Death is a valid wave (if date is known).
10832: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
10833: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
10834: and mw[mi+1][i]. dh depends on stepm.
10835: */
10836:
1.126 brouard 10837: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 10838: /* Concatenates waves */
1.145 brouard 10839:
1.215 brouard 10840: free_vector(moisdc,1,n);
10841: free_vector(andc,1,n);
10842:
1.126 brouard 10843: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
10844: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
10845: ncodemax[1]=1;
1.145 brouard 10846: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 10847: cptcoveff=0;
1.220 brouard 10848: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
10849: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 10850: }
10851:
10852: ncovcombmax=pow(2,cptcoveff);
10853: invalidvarcomb=ivector(1, ncovcombmax);
10854: for(i=1;i<ncovcombmax;i++)
10855: invalidvarcomb[i]=0;
10856:
1.211 brouard 10857: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 10858: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 10859: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 10860:
1.200 brouard 10861: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 10862: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 10863: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 10864: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
10865: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
10866: * (currently 0 or 1) in the data.
10867: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
10868: * corresponding modality (h,j).
10869: */
10870:
1.145 brouard 10871: h=0;
10872: /*if (cptcovn > 0) */
1.126 brouard 10873: m=pow(2,cptcoveff);
10874:
1.144 brouard 10875: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 10876: * For k=4 covariates, h goes from 1 to m=2**k
10877: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
10878: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 10879: * h\k 1 2 3 4
1.143 brouard 10880: *______________________________
10881: * 1 i=1 1 i=1 1 i=1 1 i=1 1
10882: * 2 2 1 1 1
10883: * 3 i=2 1 2 1 1
10884: * 4 2 2 1 1
10885: * 5 i=3 1 i=2 1 2 1
10886: * 6 2 1 2 1
10887: * 7 i=4 1 2 2 1
10888: * 8 2 2 2 1
1.197 brouard 10889: * 9 i=5 1 i=3 1 i=2 1 2
10890: * 10 2 1 1 2
10891: * 11 i=6 1 2 1 2
10892: * 12 2 2 1 2
10893: * 13 i=7 1 i=4 1 2 2
10894: * 14 2 1 2 2
10895: * 15 i=8 1 2 2 2
10896: * 16 2 2 2 2
1.143 brouard 10897: */
1.212 brouard 10898: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 10899: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
10900: * and the value of each covariate?
10901: * V1=1, V2=1, V3=2, V4=1 ?
10902: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
10903: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
10904: * In order to get the real value in the data, we use nbcode
10905: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
10906: * We are keeping this crazy system in order to be able (in the future?)
10907: * to have more than 2 values (0 or 1) for a covariate.
10908: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
10909: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
10910: * bbbbbbbb
10911: * 76543210
10912: * h-1 00000101 (6-1=5)
1.219 brouard 10913: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 10914: * &
10915: * 1 00000001 (1)
1.219 brouard 10916: * 00000000 = 1 & ((h-1) >> (k-1))
10917: * +1= 00000001 =1
1.211 brouard 10918: *
10919: * h=14, k=3 => h'=h-1=13, k'=k-1=2
10920: * h' 1101 =2^3+2^2+0x2^1+2^0
10921: * >>k' 11
10922: * & 00000001
10923: * = 00000001
10924: * +1 = 00000010=2 = codtabm(14,3)
10925: * Reverse h=6 and m=16?
10926: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
10927: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
10928: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
10929: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
10930: * V3=decodtabm(14,3,2**4)=2
10931: * h'=13 1101 =2^3+2^2+0x2^1+2^0
10932: *(h-1) >> (j-1) 0011 =13 >> 2
10933: * &1 000000001
10934: * = 000000001
10935: * +1= 000000010 =2
10936: * 2211
10937: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
10938: * V3=2
1.220 brouard 10939: * codtabm and decodtabm are identical
1.211 brouard 10940: */
10941:
1.145 brouard 10942:
10943: free_ivector(Ndum,-1,NCOVMAX);
10944:
10945:
1.126 brouard 10946:
1.186 brouard 10947: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 10948: strcpy(optionfilegnuplot,optionfilefiname);
10949: if(mle==-3)
1.201 brouard 10950: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 10951: strcat(optionfilegnuplot,".gp");
10952:
10953: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
10954: printf("Problem with file %s",optionfilegnuplot);
10955: }
10956: else{
1.204 brouard 10957: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 10958: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 10959: //fprintf(ficgp,"set missing 'NaNq'\n");
10960: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 10961: }
10962: /* fclose(ficgp);*/
1.186 brouard 10963:
10964:
10965: /* Initialisation of --------- index.htm --------*/
1.126 brouard 10966:
10967: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
10968: if(mle==-3)
1.201 brouard 10969: strcat(optionfilehtm,"-MORT_");
1.126 brouard 10970: strcat(optionfilehtm,".htm");
10971: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 10972: printf("Problem with %s \n",optionfilehtm);
10973: exit(0);
1.126 brouard 10974: }
10975:
10976: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
10977: strcat(optionfilehtmcov,"-cov.htm");
10978: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
10979: printf("Problem with %s \n",optionfilehtmcov), exit(0);
10980: }
10981: else{
10982: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
10983: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10984: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 10985: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
10986: }
10987:
1.213 brouard 10988: 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 10989: <hr size=\"2\" color=\"#EC5E5E\"> \n\
10990: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 10991: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 10992: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 10993: \n\
10994: <hr size=\"2\" color=\"#EC5E5E\">\
10995: <ul><li><h4>Parameter files</h4>\n\
10996: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
10997: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
10998: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
10999: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11000: - Date and time at start: %s</ul>\n",\
11001: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11002: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11003: fileres,fileres,\
11004: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11005: fflush(fichtm);
11006:
11007: strcpy(pathr,path);
11008: strcat(pathr,optionfilefiname);
1.184 brouard 11009: #ifdef WIN32
11010: _chdir(optionfilefiname); /* Move to directory named optionfile */
11011: #else
1.126 brouard 11012: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11013: #endif
11014:
1.126 brouard 11015:
1.220 brouard 11016: /* Calculates basic frequencies. Computes observed prevalence at single age
11017: and for any valid combination of covariates
1.126 brouard 11018: and prints on file fileres'p'. */
1.251 brouard 11019: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11020: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11021:
11022: fprintf(fichtm,"\n");
11023: fprintf(fichtm,"<br>Total number of observations=%d <br>\n\
11024: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11025: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
11026: imx,agemin,agemax,jmin,jmax,jmean);
11027: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.220 brouard 11028: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11029: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11030: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11031: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11032:
1.126 brouard 11033: /* For Powell, parameters are in a vector p[] starting at p[1]
11034: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11035: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11036:
11037: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11038: /* For mortality only */
1.126 brouard 11039: if (mle==-3){
1.136 brouard 11040: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11041: for(i=1;i<=NDIM;i++)
11042: for(j=1;j<=NDIM;j++)
11043: ximort[i][j]=0.;
1.186 brouard 11044: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.126 brouard 11045: cens=ivector(1,n);
11046: ageexmed=vector(1,n);
11047: agecens=vector(1,n);
11048: dcwave=ivector(1,n);
1.223 brouard 11049:
1.126 brouard 11050: for (i=1; i<=imx; i++){
11051: dcwave[i]=-1;
11052: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11053: if (s[m][i]>nlstate) {
11054: dcwave[i]=m;
11055: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11056: break;
11057: }
1.126 brouard 11058: }
1.226 brouard 11059:
1.126 brouard 11060: for (i=1; i<=imx; i++) {
11061: if (wav[i]>0){
1.226 brouard 11062: ageexmed[i]=agev[mw[1][i]][i];
11063: j=wav[i];
11064: agecens[i]=1.;
11065:
11066: if (ageexmed[i]> 1 && wav[i] > 0){
11067: agecens[i]=agev[mw[j][i]][i];
11068: cens[i]= 1;
11069: }else if (ageexmed[i]< 1)
11070: cens[i]= -1;
11071: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
11072: cens[i]=0 ;
1.126 brouard 11073: }
11074: else cens[i]=-1;
11075: }
11076:
11077: for (i=1;i<=NDIM;i++) {
11078: for (j=1;j<=NDIM;j++)
1.226 brouard 11079: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 11080: }
11081:
1.145 brouard 11082: /*p[1]=0.0268; p[NDIM]=0.083;*/
1.126 brouard 11083: /*printf("%lf %lf", p[1], p[2]);*/
11084:
11085:
1.136 brouard 11086: #ifdef GSL
11087: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 11088: #else
1.126 brouard 11089: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 11090: #endif
1.201 brouard 11091: strcpy(filerespow,"POW-MORT_");
11092: strcat(filerespow,fileresu);
1.126 brouard 11093: if((ficrespow=fopen(filerespow,"w"))==NULL) {
11094: printf("Problem with resultfile: %s\n", filerespow);
11095: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
11096: }
1.136 brouard 11097: #ifdef GSL
11098: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 11099: #else
1.126 brouard 11100: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 11101: #endif
1.126 brouard 11102: /* for (i=1;i<=nlstate;i++)
11103: for(j=1;j<=nlstate+ndeath;j++)
11104: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
11105: */
11106: fprintf(ficrespow,"\n");
1.136 brouard 11107: #ifdef GSL
11108: /* gsl starts here */
11109: T = gsl_multimin_fminimizer_nmsimplex;
11110: gsl_multimin_fminimizer *sfm = NULL;
11111: gsl_vector *ss, *x;
11112: gsl_multimin_function minex_func;
11113:
11114: /* Initial vertex size vector */
11115: ss = gsl_vector_alloc (NDIM);
11116:
11117: if (ss == NULL){
11118: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
11119: }
11120: /* Set all step sizes to 1 */
11121: gsl_vector_set_all (ss, 0.001);
11122:
11123: /* Starting point */
1.126 brouard 11124:
1.136 brouard 11125: x = gsl_vector_alloc (NDIM);
11126:
11127: if (x == NULL){
11128: gsl_vector_free(ss);
11129: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
11130: }
11131:
11132: /* Initialize method and iterate */
11133: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 11134: /* gsl_vector_set(x, 0, 0.0268); */
11135: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 11136: gsl_vector_set(x, 0, p[1]);
11137: gsl_vector_set(x, 1, p[2]);
11138:
11139: minex_func.f = &gompertz_f;
11140: minex_func.n = NDIM;
11141: minex_func.params = (void *)&p; /* ??? */
11142:
11143: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
11144: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
11145:
11146: printf("Iterations beginning .....\n\n");
11147: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
11148:
11149: iteri=0;
11150: while (rval == GSL_CONTINUE){
11151: iteri++;
11152: status = gsl_multimin_fminimizer_iterate(sfm);
11153:
11154: if (status) printf("error: %s\n", gsl_strerror (status));
11155: fflush(0);
11156:
11157: if (status)
11158: break;
11159:
11160: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
11161: ssval = gsl_multimin_fminimizer_size (sfm);
11162:
11163: if (rval == GSL_SUCCESS)
11164: printf ("converged to a local maximum at\n");
11165:
11166: printf("%5d ", iteri);
11167: for (it = 0; it < NDIM; it++){
11168: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
11169: }
11170: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
11171: }
11172:
11173: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
11174:
11175: gsl_vector_free(x); /* initial values */
11176: gsl_vector_free(ss); /* inital step size */
11177: for (it=0; it<NDIM; it++){
11178: p[it+1]=gsl_vector_get(sfm->x,it);
11179: fprintf(ficrespow," %.12lf", p[it]);
11180: }
11181: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
11182: #endif
11183: #ifdef POWELL
11184: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
11185: #endif
1.126 brouard 11186: fclose(ficrespow);
11187:
1.203 brouard 11188: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 11189:
11190: for(i=1; i <=NDIM; i++)
11191: for(j=i+1;j<=NDIM;j++)
1.220 brouard 11192: matcov[i][j]=matcov[j][i];
1.126 brouard 11193:
11194: printf("\nCovariance matrix\n ");
1.203 brouard 11195: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 11196: for(i=1; i <=NDIM; i++) {
11197: for(j=1;j<=NDIM;j++){
1.220 brouard 11198: printf("%f ",matcov[i][j]);
11199: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 11200: }
1.203 brouard 11201: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 11202: }
11203:
11204: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 11205: for (i=1;i<=NDIM;i++) {
1.126 brouard 11206: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 11207: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
11208: }
1.126 brouard 11209: lsurv=vector(1,AGESUP);
11210: lpop=vector(1,AGESUP);
11211: tpop=vector(1,AGESUP);
11212: lsurv[agegomp]=100000;
11213:
11214: for (k=agegomp;k<=AGESUP;k++) {
11215: agemortsup=k;
11216: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
11217: }
11218:
11219: for (k=agegomp;k<agemortsup;k++)
11220: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
11221:
11222: for (k=agegomp;k<agemortsup;k++){
11223: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
11224: sumlpop=sumlpop+lpop[k];
11225: }
11226:
11227: tpop[agegomp]=sumlpop;
11228: for (k=agegomp;k<(agemortsup-3);k++){
11229: /* tpop[k+1]=2;*/
11230: tpop[k+1]=tpop[k]-lpop[k];
11231: }
11232:
11233:
11234: printf("\nAge lx qx dx Lx Tx e(x)\n");
11235: for (k=agegomp;k<(agemortsup-2);k++)
11236: 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]);
11237:
11238:
11239: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 11240: ageminpar=50;
11241: agemaxpar=100;
1.194 brouard 11242: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
11243: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11244: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11245: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
11246: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
11247: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11248: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11249: }else{
11250: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
11251: 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 11252: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 11253: }
1.201 brouard 11254: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 11255: stepm, weightopt,\
11256: model,imx,p,matcov,agemortsup);
11257:
11258: free_vector(lsurv,1,AGESUP);
11259: free_vector(lpop,1,AGESUP);
11260: free_vector(tpop,1,AGESUP);
1.220 brouard 11261: free_matrix(ximort,1,NDIM,1,NDIM);
1.136 brouard 11262: free_ivector(cens,1,n);
11263: free_vector(agecens,1,n);
11264: free_ivector(dcwave,1,n);
1.220 brouard 11265: #ifdef GSL
1.136 brouard 11266: #endif
1.186 brouard 11267: } /* Endof if mle==-3 mortality only */
1.205 brouard 11268: /* Standard */
11269: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
11270: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11271: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 11272: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 11273: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11274: for (k=1; k<=npar;k++)
11275: printf(" %d %8.5f",k,p[k]);
11276: printf("\n");
1.205 brouard 11277: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
11278: /* mlikeli uses func not funcone */
1.247 brouard 11279: /* for(i=1;i<nlstate;i++){ */
11280: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11281: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11282: /* } */
1.205 brouard 11283: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
11284: }
11285: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
11286: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
11287: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
11288: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11289: }
11290: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 11291: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
11292: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
11293: for (k=1; k<=npar;k++)
11294: printf(" %d %8.5f",k,p[k]);
11295: printf("\n");
11296:
11297: /*--------- results files --------------*/
1.224 brouard 11298: 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 11299:
11300:
11301: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11302: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11303: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
11304: for(i=1,jk=1; i <=nlstate; i++){
11305: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 11306: if (k != i) {
11307: printf("%d%d ",i,k);
11308: fprintf(ficlog,"%d%d ",i,k);
11309: fprintf(ficres,"%1d%1d ",i,k);
11310: for(j=1; j <=ncovmodel; j++){
11311: printf("%12.7f ",p[jk]);
11312: fprintf(ficlog,"%12.7f ",p[jk]);
11313: fprintf(ficres,"%12.7f ",p[jk]);
11314: jk++;
11315: }
11316: printf("\n");
11317: fprintf(ficlog,"\n");
11318: fprintf(ficres,"\n");
11319: }
1.126 brouard 11320: }
11321: }
1.203 brouard 11322: if(mle != 0){
11323: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 11324: ftolhess=ftol; /* Usually correct */
1.203 brouard 11325: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
11326: 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");
11327: 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");
11328: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 11329: for(k=1; k <=(nlstate+ndeath); k++){
11330: if (k != i) {
11331: printf("%d%d ",i,k);
11332: fprintf(ficlog,"%d%d ",i,k);
11333: for(j=1; j <=ncovmodel; j++){
11334: 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]));
11335: 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]));
11336: jk++;
11337: }
11338: printf("\n");
11339: fprintf(ficlog,"\n");
11340: }
11341: }
1.193 brouard 11342: }
1.203 brouard 11343: } /* end of hesscov and Wald tests */
1.225 brouard 11344:
1.203 brouard 11345: /* */
1.126 brouard 11346: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
11347: printf("# Scales (for hessian or gradient estimation)\n");
11348: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
11349: for(i=1,jk=1; i <=nlstate; i++){
11350: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 11351: if (j!=i) {
11352: fprintf(ficres,"%1d%1d",i,j);
11353: printf("%1d%1d",i,j);
11354: fprintf(ficlog,"%1d%1d",i,j);
11355: for(k=1; k<=ncovmodel;k++){
11356: printf(" %.5e",delti[jk]);
11357: fprintf(ficlog," %.5e",delti[jk]);
11358: fprintf(ficres," %.5e",delti[jk]);
11359: jk++;
11360: }
11361: printf("\n");
11362: fprintf(ficlog,"\n");
11363: fprintf(ficres,"\n");
11364: }
1.126 brouard 11365: }
11366: }
11367:
11368: 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 11369: if(mle >= 1) /* To big for the screen */
1.126 brouard 11370: 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");
11371: 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");
11372: /* # 121 Var(a12)\n\ */
11373: /* # 122 Cov(b12,a12) Var(b12)\n\ */
11374: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
11375: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
11376: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
11377: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
11378: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
11379: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
11380:
11381:
11382: /* Just to have a covariance matrix which will be more understandable
11383: even is we still don't want to manage dictionary of variables
11384: */
11385: for(itimes=1;itimes<=2;itimes++){
11386: jj=0;
11387: for(i=1; i <=nlstate; i++){
1.225 brouard 11388: for(j=1; j <=nlstate+ndeath; j++){
11389: if(j==i) continue;
11390: for(k=1; k<=ncovmodel;k++){
11391: jj++;
11392: ca[0]= k+'a'-1;ca[1]='\0';
11393: if(itimes==1){
11394: if(mle>=1)
11395: printf("#%1d%1d%d",i,j,k);
11396: fprintf(ficlog,"#%1d%1d%d",i,j,k);
11397: fprintf(ficres,"#%1d%1d%d",i,j,k);
11398: }else{
11399: if(mle>=1)
11400: printf("%1d%1d%d",i,j,k);
11401: fprintf(ficlog,"%1d%1d%d",i,j,k);
11402: fprintf(ficres,"%1d%1d%d",i,j,k);
11403: }
11404: ll=0;
11405: for(li=1;li <=nlstate; li++){
11406: for(lj=1;lj <=nlstate+ndeath; lj++){
11407: if(lj==li) continue;
11408: for(lk=1;lk<=ncovmodel;lk++){
11409: ll++;
11410: if(ll<=jj){
11411: cb[0]= lk +'a'-1;cb[1]='\0';
11412: if(ll<jj){
11413: if(itimes==1){
11414: if(mle>=1)
11415: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11416: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11417: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
11418: }else{
11419: if(mle>=1)
11420: printf(" %.5e",matcov[jj][ll]);
11421: fprintf(ficlog," %.5e",matcov[jj][ll]);
11422: fprintf(ficres," %.5e",matcov[jj][ll]);
11423: }
11424: }else{
11425: if(itimes==1){
11426: if(mle>=1)
11427: printf(" Var(%s%1d%1d)",ca,i,j);
11428: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
11429: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
11430: }else{
11431: if(mle>=1)
11432: printf(" %.7e",matcov[jj][ll]);
11433: fprintf(ficlog," %.7e",matcov[jj][ll]);
11434: fprintf(ficres," %.7e",matcov[jj][ll]);
11435: }
11436: }
11437: }
11438: } /* end lk */
11439: } /* end lj */
11440: } /* end li */
11441: if(mle>=1)
11442: printf("\n");
11443: fprintf(ficlog,"\n");
11444: fprintf(ficres,"\n");
11445: numlinepar++;
11446: } /* end k*/
11447: } /*end j */
1.126 brouard 11448: } /* end i */
11449: } /* end itimes */
11450:
11451: fflush(ficlog);
11452: fflush(ficres);
1.225 brouard 11453: while(fgets(line, MAXLINE, ficpar)) {
11454: /* If line starts with a # it is a comment */
11455: if (line[0] == '#') {
11456: numlinepar++;
11457: fputs(line,stdout);
11458: fputs(line,ficparo);
11459: fputs(line,ficlog);
11460: continue;
11461: }else
11462: break;
11463: }
11464:
1.209 brouard 11465: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
11466: /* ungetc(c,ficpar); */
11467: /* fgets(line, MAXLINE, ficpar); */
11468: /* fputs(line,stdout); */
11469: /* fputs(line,ficparo); */
11470: /* } */
11471: /* ungetc(c,ficpar); */
1.126 brouard 11472:
11473: estepm=0;
1.209 brouard 11474: 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 11475:
11476: if (num_filled != 6) {
11477: 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);
11478: 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);
11479: goto end;
11480: }
11481: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
11482: }
11483: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
11484: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
11485:
1.209 brouard 11486: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 11487: if (estepm==0 || estepm < stepm) estepm=stepm;
11488: if (fage <= 2) {
11489: bage = ageminpar;
11490: fage = agemaxpar;
11491: }
11492:
11493: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 11494: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
11495: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 11496:
1.186 brouard 11497: /* Other stuffs, more or less useful */
1.254 brouard 11498: while(fgets(line, MAXLINE, ficpar)) {
11499: /* If line starts with a # it is a comment */
11500: if (line[0] == '#') {
11501: numlinepar++;
11502: fputs(line,stdout);
11503: fputs(line,ficparo);
11504: fputs(line,ficlog);
11505: continue;
11506: }else
11507: break;
11508: }
11509:
11510: if((num_filled=sscanf(line,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav)) !=EOF){
11511:
11512: if (num_filled != 7) {
11513: printf("Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11514: fprintf(ficlog,"Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11515: goto end;
11516: }
11517: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
11518: 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);
11519: 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);
11520: fprintf(ficlog,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
1.126 brouard 11521: }
1.254 brouard 11522:
11523: while(fgets(line, MAXLINE, ficpar)) {
11524: /* If line starts with a # it is a comment */
11525: if (line[0] == '#') {
11526: numlinepar++;
11527: fputs(line,stdout);
11528: fputs(line,ficparo);
11529: fputs(line,ficlog);
11530: continue;
11531: }else
11532: break;
1.126 brouard 11533: }
11534:
11535:
11536: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
11537: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
11538:
1.254 brouard 11539: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
11540: if (num_filled != 1) {
11541: printf("Error: Not 1 (data)parameters in line but %d, for example:pop_based=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11542: fprintf(ficlog,"Error: Not 1 (data)parameters in line but %d, for example: pop_based=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11543: goto end;
11544: }
11545: printf("pop_based=%d\n",popbased);
11546: fprintf(ficlog,"pop_based=%d\n",popbased);
11547: fprintf(ficparo,"pop_based=%d\n",popbased);
11548: fprintf(ficres,"pop_based=%d\n",popbased);
11549: }
11550:
1.258 brouard 11551: /* Results */
11552: nresult=0;
11553: do{
11554: if(!fgets(line, MAXLINE, ficpar)){
11555: endishere=1;
11556: parameterline=14;
11557: }else if (line[0] == '#') {
11558: /* If line starts with a # it is a comment */
1.254 brouard 11559: numlinepar++;
11560: fputs(line,stdout);
11561: fputs(line,ficparo);
11562: fputs(line,ficlog);
11563: continue;
1.258 brouard 11564: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
11565: parameterline=11;
11566: else if(sscanf(line,"backcast=%[^\n]\n",modeltemp))
11567: parameterline=12;
11568: else if(sscanf(line,"result:%[^\n]\n",modeltemp))
11569: parameterline=13;
11570: else{
11571: parameterline=14;
1.254 brouard 11572: }
1.258 brouard 11573: switch (parameterline){
11574: case 11:
11575: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF){
11576: if (num_filled != 8) {
11577: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11578: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11579: goto end;
11580: }
11581: 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);
11582: 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);
11583: 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);
11584: 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);
11585: /* day and month of proj2 are not used but only year anproj2.*/
11586: }
1.254 brouard 11587: break;
1.258 brouard 11588: case 12:
11589: /*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);*/
11590: if((num_filled=sscanf(line,"backcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&backcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF){
11591: if (num_filled != 8) {
1.262 brouard 11592: printf("Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
11593: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:backcast=1 starting-back-date=1/1/1990 final-back-date=1/1/1970 mobil_average=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
1.258 brouard 11594: goto end;
11595: }
11596: printf("backcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",backcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
11597: 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);
11598: 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);
11599: 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);
11600: /* day and month of proj2 are not used but only year anproj2.*/
11601: }
1.230 brouard 11602: break;
1.258 brouard 11603: case 13:
11604: if((num_filled=sscanf(line,"result:%[^\n]\n",resultline)) !=EOF){
11605: if (num_filled == 0){
11606: resultline[0]='\0';
11607: printf("Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11608: fprintf(ficlog,"Warning %d: no result line! It should be at minimum 'result: V2=0 V1=1 or result:.\n%s\n", num_filled, line);
11609: break;
11610: } else if (num_filled != 1){
11611: printf("ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11612: fprintf(ficlog,"ERROR %d: result line! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",num_filled, line);
11613: }
11614: nresult++; /* Sum of resultlines */
11615: printf("Result %d: result=%s\n",nresult, resultline);
11616: if(nresult > MAXRESULTLINES){
11617: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11618: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\n",MAXRESULTLINES,nresult);
11619: goto end;
11620: }
11621: decoderesult(resultline, nresult); /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
11622: fprintf(ficparo,"result: %s\n",resultline);
11623: fprintf(ficres,"result: %s\n",resultline);
11624: fprintf(ficlog,"result: %s\n",resultline);
1.230 brouard 11625: break;
1.258 brouard 11626: case 14:
1.259 brouard 11627: if(ncovmodel >2 && nresult==0 ){
11628: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 11629: goto end;
11630: }
1.259 brouard 11631: break;
1.258 brouard 11632: default:
11633: nresult=1;
11634: decoderesult(".",nresult ); /* No covariate */
11635: }
11636: } /* End switch parameterline */
11637: }while(endishere==0); /* End do */
1.126 brouard 11638:
1.230 brouard 11639: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 11640: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 11641:
11642: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 11643: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 11644: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11645: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11646: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 11647: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 11648: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
11649: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 11650: }else{
1.266 brouard 11651: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)ageminpar);
1.220 brouard 11652: }
11653: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.258 brouard 11654: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,backcast, estepm, \
1.225 brouard 11655: jprev1,mprev1,anprev1,dateprev1,jprev2,mprev2,anprev2,dateprev2);
1.220 brouard 11656:
1.225 brouard 11657: /*------------ free_vector -------------*/
11658: /* chdir(path); */
1.220 brouard 11659:
1.215 brouard 11660: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
11661: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
11662: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
11663: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.126 brouard 11664: free_lvector(num,1,n);
11665: free_vector(agedc,1,n);
11666: /*free_matrix(covar,0,NCOVMAX,1,n);*/
11667: /*free_matrix(covar,1,NCOVMAX,1,n);*/
11668: fclose(ficparo);
11669: fclose(ficres);
1.220 brouard 11670:
11671:
1.186 brouard 11672: /* Other results (useful)*/
1.220 brouard 11673:
11674:
1.126 brouard 11675: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 11676: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
11677: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 11678: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 11679: fclose(ficrespl);
11680:
11681: /*------------- h Pij x at various ages ------------*/
1.180 brouard 11682: /*#include "hpijx.h"*/
11683: hPijx(p, bage, fage);
1.145 brouard 11684: fclose(ficrespij);
1.227 brouard 11685:
1.220 brouard 11686: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 11687: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 11688: k=1;
1.126 brouard 11689: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 11690:
1.219 brouard 11691: /* Prevalence for each covariates in probs[age][status][cov] */
1.218 brouard 11692: probs= ma3x(1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.126 brouard 11693: for(i=1;i<=AGESUP;i++)
1.219 brouard 11694: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 11695: for(k=1;k<=ncovcombmax;k++)
11696: probs[i][j][k]=0.;
1.219 brouard 11697: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
11698: if (mobilav!=0 ||mobilavproj !=0 ) {
11699: mobaverages= ma3x(1, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.227 brouard 11700: for(i=1;i<=AGESUP;i++)
11701: for(j=1;j<=nlstate;j++)
11702: for(k=1;k<=ncovcombmax;k++)
11703: mobaverages[i][j][k]=0.;
1.219 brouard 11704: mobaverage=mobaverages;
11705: if (mobilav!=0) {
1.235 brouard 11706: printf("Movingaveraging observed prevalence\n");
1.258 brouard 11707: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 11708: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
11709: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
11710: printf(" Error in movingaverage mobilav=%d\n",mobilav);
11711: }
1.219 brouard 11712: }
1.266 brouard 11713: /* else if(mobilavproj==-1){ /\* Forcing raw observed prevalences *\/ */
11714: /* for(i=1;i<=AGESUP;i++) */
11715: /* for(j=1;j<=nlstate;j++) */
11716: /* for(k=1;k<=ncovcombmax;k++) */
11717: /* mobaverages[i][j][k]=probs[i][j][k]; */
11718: /* /\* /\\* Prevalence for each covariates in probs[age][status][cov] *\\/ *\/ */
11719: /* /\* prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); *\/ */
11720: /* } */
1.219 brouard 11721: else if (mobilavproj !=0) {
1.235 brouard 11722: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 11723: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 11724: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
11725: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
11726: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
11727: }
1.219 brouard 11728: }
11729: }/* end if moving average */
1.227 brouard 11730:
1.126 brouard 11731: /*---------- Forecasting ------------------*/
11732: /*if((stepm == 1) && (strcmp(model,".")==0)){*/
11733: if(prevfcast==1){
11734: /* if(stepm ==1){*/
1.225 brouard 11735: prevforecast(fileresu, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);
1.126 brouard 11736: }
1.217 brouard 11737: if(backcast==1){
1.219 brouard 11738: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11739: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11740: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
11741:
11742: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
11743:
11744: bprlim=matrix(1,nlstate,1,nlstate);
11745: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
11746: fclose(ficresplb);
11747:
1.222 brouard 11748: hBijx(p, bage, fage, mobaverage);
11749: fclose(ficrespijb);
1.219 brouard 11750: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
11751:
1.267 ! brouard 11752: prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, mobilavproj,
! 11753: bage, fage, firstpass, lastpass, anback2, p, cptcoveff);
1.219 brouard 11754: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11755: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11756: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
11757: }
1.217 brouard 11758:
1.186 brouard 11759:
11760: /* ------ Other prevalence ratios------------ */
1.126 brouard 11761:
1.215 brouard 11762: free_ivector(wav,1,imx);
11763: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
11764: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
11765: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 11766:
11767:
1.127 brouard 11768: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 11769:
1.201 brouard 11770: strcpy(filerese,"E_");
11771: strcat(filerese,fileresu);
1.126 brouard 11772: if((ficreseij=fopen(filerese,"w"))==NULL) {
11773: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11774: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
11775: }
1.208 brouard 11776: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
11777: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 11778:
11779: pstamp(ficreseij);
1.219 brouard 11780:
1.235 brouard 11781: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11782: if (cptcovn < 1){i1=1;}
11783:
11784: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11785: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11786: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11787: continue;
1.219 brouard 11788: fprintf(ficreseij,"\n#****** ");
1.235 brouard 11789: printf("\n#****** ");
1.225 brouard 11790: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11791: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11792: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11793: }
11794: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11795: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11796: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 11797: }
11798: fprintf(ficreseij,"******\n");
1.235 brouard 11799: printf("******\n");
1.219 brouard 11800:
11801: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11802: oldm=oldms;savm=savms;
1.235 brouard 11803: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 11804:
1.219 brouard 11805: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 11806: }
11807: fclose(ficreseij);
1.208 brouard 11808: printf("done evsij\n");fflush(stdout);
11809: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.218 brouard 11810:
1.227 brouard 11811: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 11812:
11813:
1.201 brouard 11814: strcpy(filerest,"T_");
11815: strcat(filerest,fileresu);
1.127 brouard 11816: if((ficrest=fopen(filerest,"w"))==NULL) {
11817: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
11818: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
11819: }
1.208 brouard 11820: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
11821: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.218 brouard 11822:
1.126 brouard 11823:
1.201 brouard 11824: strcpy(fileresstde,"STDE_");
11825: strcat(fileresstde,fileresu);
1.126 brouard 11826: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 11827: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
11828: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 11829: }
1.227 brouard 11830: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
11831: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 11832:
1.201 brouard 11833: strcpy(filerescve,"CVE_");
11834: strcat(filerescve,fileresu);
1.126 brouard 11835: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 11836: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
11837: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 11838: }
1.227 brouard 11839: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
11840: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 11841:
1.201 brouard 11842: strcpy(fileresv,"V_");
11843: strcat(fileresv,fileresu);
1.126 brouard 11844: if((ficresvij=fopen(fileresv,"w"))==NULL) {
11845: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
11846: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
11847: }
1.227 brouard 11848: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
11849: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 11850:
1.145 brouard 11851: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11852: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
11853:
1.235 brouard 11854: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
11855: if (cptcovn < 1){i1=1;}
11856:
11857: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11858: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11859: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11860: continue;
1.242 brouard 11861: printf("\n#****** Result for:");
11862: fprintf(ficrest,"\n#****** Result for:");
11863: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 11864: for(j=1;j<=cptcoveff;j++){
11865: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11866: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11867: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11868: }
1.235 brouard 11869: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11870: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11871: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11872: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11873: }
1.208 brouard 11874: fprintf(ficrest,"******\n");
1.227 brouard 11875: fprintf(ficlog,"******\n");
11876: printf("******\n");
1.208 brouard 11877:
11878: fprintf(ficresstdeij,"\n#****** ");
11879: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 11880: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 11881: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11882: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 11883: }
1.235 brouard 11884: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11885: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11886: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11887: }
1.208 brouard 11888: fprintf(ficresstdeij,"******\n");
11889: fprintf(ficrescveij,"******\n");
11890:
11891: fprintf(ficresvij,"\n#****** ");
1.238 brouard 11892: /* pstamp(ficresvij); */
1.225 brouard 11893: for(j=1;j<=cptcoveff;j++)
1.227 brouard 11894: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11895: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11896: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11897: }
1.208 brouard 11898: fprintf(ficresvij,"******\n");
11899:
11900: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11901: oldm=oldms;savm=savms;
1.235 brouard 11902: printf(" cvevsij ");
11903: fprintf(ficlog, " cvevsij ");
11904: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 11905: printf(" end cvevsij \n ");
11906: fprintf(ficlog, " end cvevsij \n ");
11907:
11908: /*
11909: */
11910: /* goto endfree; */
11911:
11912: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
11913: pstamp(ficrest);
11914:
11915:
11916: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 11917: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
11918: cptcod= 0; /* To be deleted */
11919: printf("varevsij vpopbased=%d \n",vpopbased);
11920: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 11921: 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 11922: 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 ");
11923: if(vpopbased==1)
11924: 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);
11925: else
11926: fprintf(ficrest,"the age specific period (stable) prevalences in each health state \n");
11927: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
11928: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
11929: fprintf(ficrest,"\n");
11930: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
11931: epj=vector(1,nlstate+1);
11932: printf("Computing age specific period (stable) prevalences in each health state \n");
11933: fprintf(ficlog,"Computing age specific period (stable) prevalences in each health state \n");
11934: for(age=bage; age <=fage ;age++){
1.235 brouard 11935: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 11936: if (vpopbased==1) {
11937: if(mobilav ==0){
11938: for(i=1; i<=nlstate;i++)
11939: prlim[i][i]=probs[(int)age][i][k];
11940: }else{ /* mobilav */
11941: for(i=1; i<=nlstate;i++)
11942: prlim[i][i]=mobaverage[(int)age][i][k];
11943: }
11944: }
1.219 brouard 11945:
1.227 brouard 11946: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
11947: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
11948: /* printf(" age %4.0f ",age); */
11949: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
11950: for(i=1, epj[j]=0.;i <=nlstate;i++) {
11951: epj[j] += prlim[i][i]*eij[i][j][(int)age];
11952: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
11953: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
11954: }
11955: epj[nlstate+1] +=epj[j];
11956: }
11957: /* printf(" age %4.0f \n",age); */
1.219 brouard 11958:
1.227 brouard 11959: for(i=1, vepp=0.;i <=nlstate;i++)
11960: for(j=1;j <=nlstate;j++)
11961: vepp += vareij[i][j][(int)age];
11962: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
11963: for(j=1;j <=nlstate;j++){
11964: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
11965: }
11966: fprintf(ficrest,"\n");
11967: }
1.208 brouard 11968: } /* End vpopbased */
11969: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11970: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
11971: free_vector(epj,1,nlstate+1);
1.235 brouard 11972: printf("done selection\n");fflush(stdout);
11973: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 11974:
1.145 brouard 11975: /*}*/
1.235 brouard 11976: } /* End k selection */
1.227 brouard 11977:
11978: printf("done State-specific expectancies\n");fflush(stdout);
11979: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
11980:
1.126 brouard 11981: /*------- Variance of period (stable) prevalence------*/
1.227 brouard 11982:
1.201 brouard 11983: strcpy(fileresvpl,"VPL_");
11984: strcat(fileresvpl,fileresu);
1.126 brouard 11985: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
11986: printf("Problem with variance of period (stable) prevalence resultfile: %s\n", fileresvpl);
11987: exit(0);
11988: }
1.208 brouard 11989: printf("Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
11990: fprintf(ficlog, "Computing Variance-covariance of period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.227 brouard 11991:
1.145 brouard 11992: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
11993: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
1.227 brouard 11994:
1.235 brouard 11995: i1=pow(2,cptcoveff);
11996: if (cptcovn < 1){i1=1;}
11997:
11998: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11999: for(k=1; k<=i1;k++){
1.253 brouard 12000: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12001: continue;
1.227 brouard 12002: fprintf(ficresvpl,"\n#****** ");
12003: printf("\n#****** ");
12004: fprintf(ficlog,"\n#****** ");
12005: for(j=1;j<=cptcoveff;j++) {
12006: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12007: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12008: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12009: }
1.235 brouard 12010: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12011: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12012: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12013: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12014: }
1.227 brouard 12015: fprintf(ficresvpl,"******\n");
12016: printf("******\n");
12017: fprintf(ficlog,"******\n");
12018:
12019: varpl=matrix(1,nlstate,(int) bage, (int) fage);
12020: oldm=oldms;savm=savms;
1.235 brouard 12021: varprevlim(fileres, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, strstart, nres);
1.227 brouard 12022: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
1.145 brouard 12023: /*}*/
1.126 brouard 12024: }
1.227 brouard 12025:
1.126 brouard 12026: fclose(ficresvpl);
1.208 brouard 12027: printf("done variance-covariance of period prevalence\n");fflush(stdout);
12028: fprintf(ficlog,"done variance-covariance of period prevalence\n");fflush(ficlog);
1.227 brouard 12029:
12030: free_vector(weight,1,n);
12031: free_imatrix(Tvard,1,NCOVMAX,1,2);
12032: free_imatrix(s,1,maxwav+1,1,n);
12033: free_matrix(anint,1,maxwav,1,n);
12034: free_matrix(mint,1,maxwav,1,n);
12035: free_ivector(cod,1,n);
12036: free_ivector(tab,1,NCOVMAX);
12037: fclose(ficresstdeij);
12038: fclose(ficrescveij);
12039: fclose(ficresvij);
12040: fclose(ficrest);
12041: fclose(ficpar);
12042:
12043:
1.126 brouard 12044: /*---------- End : free ----------------*/
1.219 brouard 12045: if (mobilav!=0 ||mobilavproj !=0)
12046: 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 12047: free_ma3x(probs,1,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 12048: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
12049: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 12050: } /* mle==-3 arrives here for freeing */
1.227 brouard 12051: /* endfree:*/
12052: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
12053: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
12054: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
12055: free_ma3x(cotqvar,1,maxwav,1,nqtv,1,n);
1.233 brouard 12056: free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,1,n);
1.227 brouard 12057: free_matrix(coqvar,1,maxwav,1,n);
12058: free_matrix(covar,0,NCOVMAX,1,n);
12059: free_matrix(matcov,1,npar,1,npar);
12060: free_matrix(hess,1,npar,1,npar);
12061: /*free_vector(delti,1,npar);*/
12062: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12063: free_matrix(agev,1,maxwav,1,imx);
12064: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12065:
12066: free_ivector(ncodemax,1,NCOVMAX);
12067: free_ivector(ncodemaxwundef,1,NCOVMAX);
12068: free_ivector(Dummy,-1,NCOVMAX);
12069: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 12070: free_ivector(DummyV,1,NCOVMAX);
12071: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 12072: free_ivector(Typevar,-1,NCOVMAX);
12073: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 12074: free_ivector(TvarsQ,1,NCOVMAX);
12075: free_ivector(TvarsQind,1,NCOVMAX);
12076: free_ivector(TvarsD,1,NCOVMAX);
12077: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 12078: free_ivector(TvarFD,1,NCOVMAX);
12079: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 12080: free_ivector(TvarF,1,NCOVMAX);
12081: free_ivector(TvarFind,1,NCOVMAX);
12082: free_ivector(TvarV,1,NCOVMAX);
12083: free_ivector(TvarVind,1,NCOVMAX);
12084: free_ivector(TvarA,1,NCOVMAX);
12085: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 12086: free_ivector(TvarFQ,1,NCOVMAX);
12087: free_ivector(TvarFQind,1,NCOVMAX);
12088: free_ivector(TvarVD,1,NCOVMAX);
12089: free_ivector(TvarVDind,1,NCOVMAX);
12090: free_ivector(TvarVQ,1,NCOVMAX);
12091: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 12092: free_ivector(Tvarsel,1,NCOVMAX);
12093: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 12094: free_ivector(Tposprod,1,NCOVMAX);
12095: free_ivector(Tprod,1,NCOVMAX);
12096: free_ivector(Tvaraff,1,NCOVMAX);
12097: free_ivector(invalidvarcomb,1,ncovcombmax);
12098: free_ivector(Tage,1,NCOVMAX);
12099: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 12100: free_ivector(TmodelInvind,1,NCOVMAX);
12101: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 12102:
12103: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
12104: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 12105: fflush(fichtm);
12106: fflush(ficgp);
12107:
1.227 brouard 12108:
1.126 brouard 12109: if((nberr >0) || (nbwarn>0)){
1.216 brouard 12110: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
12111: 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 12112: }else{
12113: printf("End of Imach\n");
12114: fprintf(ficlog,"End of Imach\n");
12115: }
12116: printf("See log file on %s\n",filelog);
12117: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 12118: /*(void) gettimeofday(&end_time,&tzp);*/
12119: rend_time = time(NULL);
12120: end_time = *localtime(&rend_time);
12121: /* tml = *localtime(&end_time.tm_sec); */
12122: strcpy(strtend,asctime(&end_time));
1.126 brouard 12123: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
12124: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 12125: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 12126:
1.157 brouard 12127: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
12128: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
12129: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 12130: /* printf("Total time was %d uSec.\n", total_usecs);*/
12131: /* if(fileappend(fichtm,optionfilehtm)){ */
12132: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12133: fclose(fichtm);
12134: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
12135: fclose(fichtmcov);
12136: fclose(ficgp);
12137: fclose(ficlog);
12138: /*------ End -----------*/
1.227 brouard 12139:
12140:
12141: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 12142: #ifdef WIN32
1.227 brouard 12143: if (_chdir(pathcd) != 0)
12144: printf("Can't move to directory %s!\n",path);
12145: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 12146: #else
1.227 brouard 12147: if(chdir(pathcd) != 0)
12148: printf("Can't move to directory %s!\n", path);
12149: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 12150: #endif
1.126 brouard 12151: printf("Current directory %s!\n",pathcd);
12152: /*strcat(plotcmd,CHARSEPARATOR);*/
12153: sprintf(plotcmd,"gnuplot");
1.157 brouard 12154: #ifdef _WIN32
1.126 brouard 12155: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
12156: #endif
12157: if(!stat(plotcmd,&info)){
1.158 brouard 12158: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12159: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 12160: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 12161: }else
12162: strcpy(pplotcmd,plotcmd);
1.157 brouard 12163: #ifdef __unix
1.126 brouard 12164: strcpy(plotcmd,GNUPLOTPROGRAM);
12165: if(!stat(plotcmd,&info)){
1.158 brouard 12166: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 12167: }else
12168: strcpy(pplotcmd,plotcmd);
12169: #endif
12170: }else
12171: strcpy(pplotcmd,plotcmd);
12172:
12173: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 12174: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.227 brouard 12175:
1.126 brouard 12176: if((outcmd=system(plotcmd)) != 0){
1.158 brouard 12177: printf("gnuplot command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 12178: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 12179: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.150 brouard 12180: if((outcmd=system(plotcmd)) != 0)
1.153 brouard 12181: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.126 brouard 12182: }
1.158 brouard 12183: printf(" Successful, please wait...");
1.126 brouard 12184: while (z[0] != 'q') {
12185: /* chdir(path); */
1.154 brouard 12186: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 12187: scanf("%s",z);
12188: /* if (z[0] == 'c') system("./imach"); */
12189: if (z[0] == 'e') {
1.158 brouard 12190: #ifdef __APPLE__
1.152 brouard 12191: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 12192: #elif __linux
12193: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 12194: #else
1.152 brouard 12195: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 12196: #endif
12197: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
12198: system(pplotcmd);
1.126 brouard 12199: }
12200: else if (z[0] == 'g') system(plotcmd);
12201: else if (z[0] == 'q') exit(0);
12202: }
1.227 brouard 12203: end:
1.126 brouard 12204: while (z[0] != 'q') {
1.195 brouard 12205: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 12206: scanf("%s",z);
12207: }
12208: }
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